Add UPDATING entries and bump version.

Approved by: so
Fix multiple security issues in OpenSSL.
2026-10-30 12:25:10 +01:00 · 2025-09-30 08:36:37 -07:00 · 2025-09-30 08:31:54 -07:00 · 2025-09-29 21:58:01 -07:00 · 2025-09-16 09:28:35 -07:00 · 2025-09-13 17:21:16 -07:00
568 changed files with 34085 additions and 6736 deletions
@@ -4,7 +4,7 @@ compute_engine_instance:
  # gcloud compute images list --project freebsd-org-cloud-dev --no-standard-images
  platform: freebsd
  image_project: freebsd-org-cloud-dev
-  image: freebsd-13-2-release-amd64
+  image: freebsd-13-3-release-amd64
  cpu: 8
  memory: 8G
  disk: 40
@@ -51,6 +51,26 @@
 #   xargs -n1 | sort | uniq -d;
 # done

+# 20250310: caroot bundle updated
+OLD_FILES+=usr/share/certs/trusted/Entrust_Root_Certification_Authority_-_G4.pem
+OLD_FILES+=usr/share/certs/trusted/SecureSign_RootCA11.pem
+OLD_FILES+=usr/share/certs/trusted/Security_Communication_RootCA3.pem
+OLD_FILES+=usr/share/certs/trusted/SwissSign_Silver_CA_-_G2.pem
+OLD_FILES+=usr/share/certs/untrusted/AddTrust_External_Root.pem
+OLD_FILES+=usr/share/certs/untrusted/AddTrust_Low-Value_Services_Root.pem
+OLD_FILES+=usr/share/certs/untrusted/Staat_der_Nederlanden_Root_CA_-_G2.pem
+OLD_FILES+=usr/share/certs/untrusted/Cybertrust_Global_Root.pem
+OLD_FILES+=usr/share/certs/untrusted/DST_Root_CA_X3.pem
+OLD_FILES+=usr/share/certs/untrusted/GlobalSign_Root_CA_-_R2.pem
+OLD_FILES+=usr/share/certs/untrusted/QuoVadis_Root_CA.pem
+OLD_FILES+=usr/share/certs/untrusted/Sonera_Class_2_Root_CA.pem
+OLD_FILES+=usr/share/certs/untrusted/GeoTrust_Global_CA.pem
+OLD_FILES+=usr/share/certs/untrusted/Staat_der_Nederlanden_EV_Root_CA.pem
+OLD_FILES+=usr/share/certs/untrusted/E-Tugra_Certification_Authority.pem
+OLD_FILES+=usr/share/certs/untrusted/Hongkong_Post_Root_CA_1.pem
+OLD_FILES+=usr/share/certs/untrusted/Security_Communication_Root_CA.pem
+OLD_FILES+=usr/share/certs/untrusted/Trustis_FPS_Root_CA.pem
+
 # 20240914 libpcap upgrade from 1.10.4 to 1.10.5
 OLD_FILES+=contrib/libpcap/pcap-haiku.cpp
 OLD_FILES+=contrib/libpcap/pcap-rpcap-int.h
@@ -12,6 +12,85 @@ Items affecting the ports and packages system can be found in
 /usr/ports/UPDATING.  Please read that file before updating system packages
 and/or ports.

+20250930:
+	14.2-RELEASE-p7	EN-25:18.freebsd-update
+			SA-25:08.openssl
+
+	Fix freebsd-update(8) installing libraries in incorrect order. [EN-25:18.freebsd-update]
+
+	Fix multiple vulnerabilities in OpenSSL. [SA-25:08.openssl]
+
+20250916:
+	14.2-RELEASE-p6	EN-25:15.arm64
+
+	Fix arm64 syscall(2) allows unprivileged user to panic kernel. [EN-25:15.arm64]
+
+20250808:
+	14.2-RELEASE-p5	EN-25:14.route
+			SA-25:07.libarchive
+
+	Fix route(8) monitor buffering too much when redirected to a
+	file. [EN-25:14.route]
+
+	Update libarchive to 3.8.1 to fix integer overflow leading to double
+        free. [SA-25:07.libarchive]
+
+20250702:
+        14.2-RELEASE-p4	EN-25:09.libc
+			EN-25:10.zfs
+			EN-25:11.ena
+			SA-25:06.xz
+
+	Fix dynamically-loaded C++ libraries crashing at exit [EN-25:09.libc]
+
+	Fix corruption in ZFS replication streams from encrypted datasets [EN-25:10.zfs]
+
+	Fix ena resets and kernel panic on Nitro v4 or newer instances [EN-25:11.ena]
+
+	Fix use-after-free in multi-threaded xz decoder [SA-25:06.xz]
+
+20250410:
+	14.2-RELEASE-p3 EN-25:04.tzdata
+			EN-25:05.expat
+			EN-25:06.daemon
+			EN-25:07.openssl
+			EN-25:08.caroot
+
+	Timezone database information update [EN-25:04.tzdata]
+
+	Update expat to 2.7.1 [EN-25:05.expat]
+
+	daemon(8) missing signals [EN-25:06.daemon]
+
+	Update OpenSSL to 3.0.16 [EN-25:07.openssl]
+
+	Root certificate bundle update [EN-25:08.caroot]
+
+20250221:
+	14.2-RELEASE-p2 SA-25:05.openssh
+
+	Multiple vulnerabilities in OpenSSH [SA-25:05.openssh]
+
+20250129:
+	14.2-RELEASE-p1 SA-25:02.fs
+			SA-25:03.etcupdate
+			SA-25:04.ktrace
+			EN-25:02.audit
+			EN-25:03.tzdata
+
+	Buffer overflow in some filesystems via NFS [SA-25:02.fs]
+
+	Unprivileged access to system files [SA-25:03.etcupdate]
+
+	Uninitialized kernel memory disclosure via ktrace(2) [SA-25:04.ktrace]
+
+	System call auditing disabled by DTrace [EN-25:02.audit]
+
+	Timezone database information update [EN-25:03.tzdata]
+
+20241203:
+	14.2-RELEASE.
+
 20240419:
 	Clang, llvm, lld, lldb, compiler-rt, libc++, libunwind and openmp have
 	been upgraded to 18.1.6. It is important that you run `make delete-old`
@@ -1,5 +1,5 @@
 Copyright (c) 1998-2000 Thai Open Source Software Center Ltd and Clark Cooper
-Copyright (c) 2001-2022 Expat maintainers
+Copyright (c) 2001-2025 Expat maintainers

 Permission is hereby granted, free of charge, to any person obtaining
 a copy of this software and associated documentation files (the
@@ -1,6 +1,286 @@
-NOTE: We are looking for help with a few things:
-      https://github.com/libexpat/libexpat/labels/help%20wanted
-      If you can help, please get in touch.  Thanks!
+                           __  __            _
+                        ___\ \/ /_ __   __ _| |_
+                       / _ \\  /| '_ \ / _` | __|
+                      |  __//  \| |_) | (_| | |_
+                       \___/_/\_\ .__/ \__,_|\__|
+                                |_| XML parser
+
+!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
+!! <blink>Expat is UNDERSTAFFED and WITHOUT FUNDING.</blink>                 !!
+!!                 ~~~~~~~~~~~~                                              !!
+!! The following topics need *additional skilled C developers* to progress   !!
+!! in a timely manner or at all (loosely ordered by descending priority):    !!
+!!                                                                           !!
+!! - teaming up on researching and fixing future security reports and        !!
+!!   ClusterFuzz findings with few-days-max response times in communication  !!
+!!   in order to (1) have a sound fix ready before the end of a 90 days      !!
+!!   grace period and (2) in a sustainable manner,                           !!
+!! - helping CPython Expat bindings with supporting Expat's billion laughs   !!
+!!   attack protection API (https://github.com/python/cpython/issues/90949): !!
+!!   - XML_SetBillionLaughsAttackProtectionActivationThreshold               !!
+!!   - XML_SetBillionLaughsAttackProtectionMaximumAmplification              !!
+!! - helping Perl's XML::Parser Expat bindings with supporting Expat's       !!
+!!   security API (https://github.com/cpan-authors/XML-Parser/issues/102):   !!
+!!   - XML_SetBillionLaughsAttackProtectionActivationThreshold               !!
+!!   - XML_SetBillionLaughsAttackProtectionMaximumAmplification              !!
+!!   - XML_SetReparseDeferralEnabled                                         !!
+!! - implementing and auto-testing XML 1.0r5 support                         !!
+!!   (needs discussion before pull requests),                                !!
+!! - smart ideas on fixing the Autotools CMake files generation issue        !!
+!!   without breaking CI (needs discussion before pull requests),            !!
+!! - pushing migration from `int` to `size_t` further                        !!
+!!   including edge-cases test coverage (needs discussion before anything).  !!
+!!                                                                           !!
+!! For details, please reach out via e-mail to [email protected] so we   !!
+!! can schedule a voice call on the topic, in English or German.             !!
+!!                                                                           !!
+!! THANK YOU!                        Sebastian Pipping -- Berlin, 2024-03-09 !!
+!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
+
+Release 2.7.1 Thu March 27 2025
+        Bug fixes:
+       #980 #989  Restore event pointer behavior from Expat 2.6.4
+                    (that the fix to CVE-2024-8176 changed in 2.7.0);
+                    affected API functions are:
+                    - XML_GetCurrentByteCount
+                    - XML_GetCurrentByteIndex
+                    - XML_GetCurrentColumnNumber
+                    - XML_GetCurrentLineNumber
+                    - XML_GetInputContext
+
+        Other changes:
+       #976 #977  Autotools: Integrate files "fuzz/xml_lpm_fuzzer.{cpp,proto}"
+                    with Automake that were missing from 2.7.0 release tarballs
+       #983 #984  Fix printf format specifiers for 32bit Emscripten
+            #992  docs: Promote OpenSSF Best Practices self-certification
+            #978  tests/benchmark: Resolve mistaken double close
+            #986  Address compiler warnings
+       #990 #993  Version info bumped from 11:1:10 (libexpat*.so.1.10.1)
+                    to 11:2:10 (libexpat*.so.1.10.2); see https://verbump.de/
+                    for what these numbers do
+
+        Infrastructure:
+            #982  CI: Start running Perl XML::Parser integration tests
+            #987  CI: Enforce Clang Static Analyzer clean code
+            #991  CI: Re-enable warning clang-analyzer-valist.Uninitialized
+                    for clang-tidy
+            #981  CI: Cover compilation with musl
+       #983 #984  CI: Cover compilation with 32bit Emscripten
+       #976 #977  CI: Protect against fuzzer files missing from future
+                    release archives
+
+        Special thanks to:
+            Berkay Eren Ürün
+            Matthew Fernandez
+                 and
+            Perl XML::Parser
+
+Release 2.7.0 Thu March 13 2025
+        Security fixes:
+       #893 #973  CVE-2024-8176 -- Fix crash from chaining a large number
+                    of entities caused by stack overflow by resolving use of
+                    recursion, for all three uses of entities:
+                    - general entities in character data ("<e>&g1;</e>")
+                    - general entities in attribute values ("<e k1='&g1;'/>")
+                    - parameter entities ("%p1;")
+                    Known impact is (reliable and easy) denial of service:
+                    CVSS:3.1/AV:N/AC:L/PR:N/UI:N/S:U/C:N/I:N/A:H/E:H/RL:O/RC:C
+                    (Base Score: 7.5, Temporal Score: 7.2)
+                    Please note that a layer of compression around XML can
+                    significantly reduce the minimum attack payload size.
+
+        Other changes:
+       #935 #937  Autotools: Make generated CMake files look for
+                    libexpat.@[email protected] on macOS
+            #925  Autotools: Sync CMake templates with CMake 3.29
+  #945 #962 #966  CMake: Drop support for CMake <3.13
+            #942  CMake: Small fuzzing related improvements
+            #921  docs: Add missing documentation of error code
+                    XML_ERROR_NOT_STARTED that was introduced with 2.6.4
+            #941  docs: Document need for C++11 compiler for use from C++
+            #959  tests/benchmark: Fix a (harmless) TOCTTOU
+            #944  Windows: Fix installer target location of file xmlwf.xml
+                    for CMake
+            #953  Windows: Address warning -Wunknown-warning-option
+                    about -Wno-pedantic-ms-format from LLVM MinGW
+            #971  Address Cppcheck warnings
+       #969 #970  Mass-migrate links from http:// to https://
+    #947 #958 ..
+       #974 #975  Document changes since the previous release
+       #974 #975  Version info bumped from 11:0:10 (libexpat*.so.1.10.0)
+                    to 11:1:10 (libexpat*.so.1.10.1); see https://verbump.de/
+                    for what these numbers do
+
+        Infrastructure:
+            #926  tests: Increase robustness
+    #927 #932 ..
+       #930 #933  tests: Increase test coverage
+    #617 #950 ..
+    #951 #952 ..
+    #954 #955 ..  Fuzzing: Add new fuzzer "xml_lpm_fuzzer" based on
+            #961    Google's libprotobuf-mutator ("LPM")
+            #957  Fuzzing|CI: Start producing fuzzing code coverage reports
+            #936  CI: Pass -q -q for LCOV >=2.1 in coverage.sh
+            #942  CI: Small fuzzing related improvements
+    #139 #203 ..
+       #791 #946  CI: Make GitHub Actions build using MSVC on Windows and
+                      produce 32bit and 64bit Windows binaries
+            #956  CI: Get off of about-to-be-removed Ubuntu 20.04
+       #960 #964  CI: Start uploading to Coverity Scan for static analysis
+            #972  CI: Stop loading DTD from the internet to address flaky CI
+            #971  CI: Adapt to breaking changes in Cppcheck
+
+        Special thanks to:
+            Alexander Gieringer
+            Berkay Eren Ürün
+            Hanno Böck
+            Jann Horn
+            Mark Brand
+            Sebastian Andrzej Siewior
+            Snild Dolkow
+            Thomas Pröll
+            Tomas Korbar
+            valord577
+                 and
+            Google Project Zero
+            Linutronix
+            Red Hat
+            Siemens
+
+Release 2.6.4 Wed November 6 2024
+        Security fixes:
+            #915  CVE-2024-50602 -- Fix crash within function XML_ResumeParser
+                    from a NULL pointer dereference by disallowing function
+                    XML_StopParser to (stop or) suspend an unstarted parser.
+                    A new error code XML_ERROR_NOT_STARTED was introduced to
+                    properly communicate this situation.  // CWE-476 CWE-754
+
+        Other changes:
+            #903  CMake: Add alias target "expat::expat"
+            #905  docs: Document use via CMake >=3.18 with FetchContent
+                    and SOURCE_SUBDIR and its consequences
+            #902  tests: Reduce use of global parser instance
+            #904  tests: Resolve duplicate handler
+       #317 #918  tests: Improve tests on doctype closing (ex CVE-2019-15903)
+            #914  Fix signedness of format strings
+            #915  For use from C++, expat.h started requiring C++11 due to
+                    use of C99 features
+       #919 #920  Version info bumped from 10:3:9 (libexpat*.so.1.9.3)
+                    to 11:0:10 (libexpat*.so.1.10.0); see https://verbump.de/
+                    for what these numbers do
+
+        Infrastructure:
+            #907  CI: Upgrade Clang from 18 to 19
+            #913  CI: Drop macos-12 and add macos-15
+            #910  CI: Adapt to breaking changes in GitHub Actions
+            #898  Add missing entries to .gitignore
+
+        Special thanks to:
+            Hanno Böck
+            José Eduardo Gutiérrez Conejo
+            José Ricardo Cardona Quesada
+
+Release 2.6.3 Wed September 4 2024
+        Security fixes:
+       #887 #890  CVE-2024-45490 -- Calling function XML_ParseBuffer with
+                    len < 0 without noticing and then calling XML_GetBuffer
+                    will have XML_ParseBuffer fail to recognize the problem
+                    and XML_GetBuffer corrupt memory.
+                    With the fix, XML_ParseBuffer now complains with error
+                    XML_ERROR_INVALID_ARGUMENT just like sibling XML_Parse
+                    has been doing since Expat 2.2.1, and now documented.
+                    Impact is denial of service to potentially artitrary code
+                    execution.
+       #888 #891  CVE-2024-45491 -- Internal function dtdCopy can have an
+                    integer overflow for nDefaultAtts on 32-bit platforms
+                    (where UINT_MAX equals SIZE_MAX).
+                    Impact is denial of service to potentially artitrary code
+                    execution.
+       #889 #892  CVE-2024-45492 -- Internal function nextScaffoldPart can
+                    have an integer overflow for m_groupSize on 32-bit
+                    platforms (where UINT_MAX equals SIZE_MAX).
+                    Impact is denial of service to potentially artitrary code
+                    execution.
+
+        Other changes:
+       #851 #879  Autotools: Sync CMake templates with CMake 3.28
+            #853  Autotools: Always provide path to find(1) for portability
+            #861  Autotools: Ensure that the m4 directory always exists.
+            #870  Autotools: Simplify handling of SIZEOF_VOID_P
+            #869  Autotools: Support non-GNU sed
+            #856  Autotools|CMake: Fix main() to main(void)
+            #865  Autotools|CMake: Fix compile tests for HAVE_SYSCALL_GETRANDOM
+            #863  Autotools|CMake: Stop requiring dos2unix
+       #854 #855  CMake: Fix check for symbols size_t and off_t
+            #864  docs|tests: Convert README to Markdown and update
+            #741  Windows: Drop support for Visual Studio <=15.0/2017
+            #886  Drop needless XML_DTD guards around is_param access
+            #885  Fix typo in a code comment
+       #894 #896  Version info bumped from 10:2:9 (libexpat*.so.1.9.2)
+                    to 10:3:9 (libexpat*.so.1.9.3); see https://verbump.de/
+                    for what these numbers do
+
+        Infrastructure:
+            #880  Readme: Promote the call for help
+            #868  CI: Fix various issues
+            #849  CI: Allow triggering GitHub Actions workflows manually
+    #851 #872 ..
+       #873 #879  CI: Adapt to breaking changes in GitHub Actions
+
+        Special thanks to:
+            Alexander Bluhm
+            Berkay Eren Ürün
+            Dag-Erling Smørgrav
+            Ferenc Géczi
+            TaiYou
+
+Release 2.6.2 Wed March 13 2024
+        Security fixes:
+       #839 #842  CVE-2024-28757 -- Prevent billion laughs attacks with
+                    isolated use of external parsers.  Please see the commit
+                    message of commit 1d50b80cf31de87750103656f6eb693746854aa8
+                    for details.
+
+        Bug fixes:
+       #839 #841  Reject direct parameter entity recursion
+                    and avoid the related undefined behavior
+
+        Other changes:
+            #847  Autotools: Fix build for DOCBOOK_TO_MAN containing spaces
+            #837  Add missing #821 and #824 to 2.6.1 change log
+       #838 #843  Version info bumped from 10:1:9 (libexpat*.so.1.9.1)
+                    to 10:2:9 (libexpat*.so.1.9.2); see https://verbump.de/
+                    for what these numbers do
+
+        Special thanks to:
+            Philippe Antoine
+            Tomas Korbar
+                 and
+            Clang UndefinedBehaviorSanitizer
+            OSS-Fuzz / ClusterFuzz
+
+Release 2.6.1 Thu February 29 2024
+        Bug fixes:
+            #817  Make tests independent of CPU speed, and thus more robust
+       #828 #836  Expose billion laughs API with XML_DTD defined and
+                    XML_GE undefined, regression from 2.6.0
+
+        Other changes:
+            #829  Hide test-only code behind new internal macro
+            #833  Autotools: Reject expat_config.h.in defining SIZEOF_VOID_P
+       #821 #824  Autotools: Fix "make clean" for case:
+                    ./configure --without-docbook && make clean all
+            #819  Address compiler warnings
+       #832 #834  Version info bumped from 10:0:9 (libexpat*.so.1.9.0)
+                    to 10:1:9 (libexpat*.so.1.9.1); see https://verbump.de/
+                    for what these numbers do
+
+        Infrastructure:
+            #818  CI: Adapt to breaking changes in clang-format
+
+        Special thanks to:
+            David Hall
+            Snild Dolkow

 Release 2.6.0 Tue February 6 2024
        Security fixes:
@@ -1,4 +1,3 @@
-# $FreeBSD$
 *.MPW
 *.cmake
 *.def
@@ -6,10 +6,12 @@
 #                      \___/_/\_\ .__/ \__,_|\__|
 #                               |_| XML parser
 #
-# Copyright (c) 2017-2023 Sebastian Pipping <[email protected]>
+# Copyright (c) 2017-2025 Sebastian Pipping <[email protected]>
 # Copyright (c) 2018      KangLin <[email protected]>
 # Copyright (c) 2022      Johnny Jazeix <[email protected]>
 # Copyright (c) 2023      Sony Corporation / Snild Dolkow <[email protected]>
+# Copyright (c) 2024      Alexander Bluhm <[email protected]>
+# Copyright (c) 2024      Dag-Erling Smørgrav <[email protected]>
 # Licensed under the MIT license:
 #
 # Permission is  hereby granted,  free of charge,  to any  person obtaining
@@ -94,6 +96,8 @@ EXTRA_DIST = \
    conftools/expat.m4 \
    conftools/get-version.sh \
    \
+    fuzz/xml_lpm_fuzzer.cpp \
+    fuzz/xml_lpm_fuzzer.proto \
    fuzz/xml_parsebuffer_fuzzer.c \
    fuzz/xml_parse_fuzzer.c \
    \
@@ -114,10 +118,10 @@ buildlib:
 	@echo 'ERROR: is no longer supported.  INSTEAD please:' >&2
 	@echo 'ERROR:' >&2
 	@echo 'ERROR:  * Mass-patch Makefile.am, e.g.' >&2
-	@echo 'ERROR:    # find -name Makefile.am -exec sed \' >&2
+	@echo 'ERROR:    # find . -name Makefile.am -exec sed \' >&2
 	@echo 'ERROR:          -e "s,libexpat\.la,libexpatw.la," \' >&2
 	@echo 'ERROR:          -e "s,libexpat_la,libexpatw_la," \' >&2
-	@echo 'ERROR:          -i {} +' >&2
+	@echo 'ERROR:          -i.bak {} +' >&2
 	@echo 'ERROR:' >&2
 	@echo 'ERROR:  * Run automake to re-generate Makefile.in files' >&2
 	@echo 'ERROR:' >&2
@@ -22,10 +22,12 @@
 #                      \___/_/\_\ .__/ \__,_|\__|
 #                               |_| XML parser
 #
-# Copyright (c) 2017-2023 Sebastian Pipping <[email protected]>
+# Copyright (c) 2017-2025 Sebastian Pipping <[email protected]>
 # Copyright (c) 2018      KangLin <[email protected]>
 # Copyright (c) 2022      Johnny Jazeix <[email protected]>
 # Copyright (c) 2023      Sony Corporation / Snild Dolkow <[email protected]>
+# Copyright (c) 2024      Alexander Bluhm <[email protected]>
+# Copyright (c) 2024      Dag-Erling Smørgrav <[email protected]>
 # Licensed under the MIT license:
 #
 # Permission is  hereby granted,  free of charge,  to any  person obtaining
@@ -384,6 +386,7 @@ RANLIB = @RANLIB@
 SED = @SED@
 SET_MAKE = @SET_MAKE@
 SHELL = @SHELL@
+SIZEOF_VOID_P = @SIZEOF_VOID_P@
 SO_MAJOR = @SO_MAJOR@
 SO_MINOR = @SO_MINOR@
 SO_PATCH = @SO_PATCH@
@@ -397,7 +400,6 @@ ac_ct_AR = @ac_ct_AR@
 ac_ct_CC = @ac_ct_CC@
 ac_ct_CXX = @ac_ct_CXX@
 ac_ct_DUMPBIN = @ac_ct_DUMPBIN@
-ac_cv_sizeof_void_p = @ac_cv_sizeof_void_p@
 am__include = @am__include@
 am__leading_dot = @am__leading_dot@
 am__quote = @am__quote@
@@ -492,6 +494,8 @@ EXTRA_DIST = \
    conftools/expat.m4 \
    conftools/get-version.sh \
    \
+    fuzz/xml_lpm_fuzzer.cpp \
+    fuzz/xml_lpm_fuzzer.proto \
    fuzz/xml_parsebuffer_fuzzer.c \
    fuzz/xml_parse_fuzzer.c \
    \
@@ -1080,10 +1084,10 @@ buildlib:
 	@echo 'ERROR: is no longer supported.  INSTEAD please:' >&2
 	@echo 'ERROR:' >&2
 	@echo 'ERROR:  * Mass-patch Makefile.am, e.g.' >&2
-	@echo 'ERROR:    # find -name Makefile.am -exec sed \' >&2
+	@echo 'ERROR:    # find . -name Makefile.am -exec sed \' >&2
 	@echo 'ERROR:          -e "s,libexpat\.la,libexpatw.la," \' >&2
 	@echo 'ERROR:          -e "s,libexpat_la,libexpatw_la," \' >&2
-	@echo 'ERROR:          -i {} +' >&2
+	@echo 'ERROR:          -i.bak {} +' >&2
 	@echo 'ERROR:' >&2
 	@echo 'ERROR:  * Run automake to re-generate Makefile.in files' >&2
 	@echo 'ERROR:' >&2
@@ -3,9 +3,16 @@
 [![Packaging status](https://repology.org/badge/tiny-repos/expat.svg)](https://repology.org/metapackage/expat/versions)
 [![Downloads SourceForge](https://img.shields.io/sourceforge/dt/expat?label=Downloads%20SourceForge)](https://sourceforge.net/projects/expat/files/)
 [![Downloads GitHub](https://img.shields.io/github/downloads/libexpat/libexpat/total?label=Downloads%20GitHub)](https://github.com/libexpat/libexpat/releases)
+[![OpenSSF Best Practices](https://www.bestpractices.dev/projects/10205/badge)](https://www.bestpractices.dev/projects/10205)
+
+> [!CAUTION]
+>
+> Expat is **understaffed** and without funding.
+> There is a [call for help with details](https://github.com/libexpat/libexpat/blob/master/expat/Changes)
+> at the top of the `Changes` file.


-# Expat, Release 2.6.0
+# Expat, Release 2.7.1

 This is Expat, a C99 library for parsing
 [XML 1.0 Fourth Edition](https://www.w3.org/TR/2006/REC-xml-20060816/), started by
@@ -16,11 +23,11 @@ are called when the parser discovers the associated structures in the
 document being parsed.  A start tag is an example of the kind of
 structures for which you may register handlers.

-Expat supports the following compilers:
+Expat supports the following C99 compilers:

- GNU GCC >=4.5
+- GNU GCC >=4.5 (for use from C) or GNU GCC >=4.8.1 (for use from C++)
 - LLVM Clang >=3.5
- Microsoft Visual Studio >=15.0/2017 (rolling `${today} minus 5 years`)
+- Microsoft Visual Studio >=16.0/2019 (rolling `${today} minus 5 years`)

 Windows users can use the
 [`expat-win32bin-*.*.*.{exe,zip}` download](https://github.com/libexpat/libexpat/releases),
@@ -37,16 +44,16 @@ This license is the same as the MIT/X Consortium license.

 ## Using libexpat in your CMake-Based Project

-There are two ways of using libexpat with CMake:
+There are three documented ways of using libexpat with CMake:

-### a) Module Mode
+### a) `find_package` with Module Mode

 This approach leverages CMake's own [module `FindEXPAT`](https://cmake.org/cmake/help/latest/module/FindEXPAT.html).

 Notice the *uppercase* `EXPAT` in the following example:

 ```cmake
-cmake_minimum_required(VERSION 3.0)  # or 3.10, see below
+cmake_minimum_required(VERSION 3.10)

 project(hello VERSION 1.0.0)

@@ -56,15 +63,10 @@ add_executable(hello
    hello.c
 )

-# a) for CMake >=3.10 (see CMake's FindEXPAT docs)
 target_link_libraries(hello PUBLIC EXPAT::EXPAT)
-
-# b) for CMake >=3.0
-target_include_directories(hello PRIVATE ${EXPAT_INCLUDE_DIRS})
-target_link_libraries(hello PUBLIC ${EXPAT_LIBRARIES})
 ```

-### b) Config Mode
+### b) `find_package` with Config Mode

 This approach requires files from…

@@ -79,7 +81,7 @@ or
 Notice the *lowercase* `expat` in the following example:

 ```cmake
-cmake_minimum_required(VERSION 3.0)
+cmake_minimum_required(VERSION 3.10)

 project(hello VERSION 1.0.0)

@@ -92,6 +94,45 @@ add_executable(hello
 target_link_libraries(hello PUBLIC expat::expat)
 ```

+### c) The `FetchContent` module
+
+This approach — as demonstrated below — requires CMake >=3.18 for both the
+[`FetchContent` module](https://cmake.org/cmake/help/latest/module/FetchContent.html)
+and its support for the `SOURCE_SUBDIR` option to be available.
+
+Please note that:
+- Use of the `FetchContent` module with *non-release* SHA1s or `master`
+  of libexpat is neither advised nor considered officially supported.
+- Pinning to a specific commit is great for robust CI.
+- Pinning to a specific commit needs updating every time there is a new
+  release of libexpat — either manually or through automation —,
+  to not miss out on libexpat security updates.
+
+For an example that pulls in libexpat via Git:
+
+```cmake
+cmake_minimum_required(VERSION 3.18)
+
+include(FetchContent)
+
+project(hello VERSION 1.0.0)
+
+FetchContent_Declare(
+    expat
+    GIT_REPOSITORY https://github.com/libexpat/libexpat/
+    GIT_TAG        000000000_GIT_COMMIT_SHA1_HERE_000000000  # i.e. Git tag R_0_Y_Z
+    SOURCE_SUBDIR  expat/
+)
+
+FetchContent_MakeAvailable(expat)
+
+add_executable(hello
+    hello.c
+)
+
+target_link_libraries(hello PUBLIC expat)
+```
+

 ## Building from a Git Clone

@@ -158,10 +199,10 @@ support this mode of compilation (yet):

 1. Mass-patch `Makefile.am` files to use `libexpatw.la` for a library name:
   <br/>
-   `find -name Makefile.am -exec sed
+   `find . -name Makefile.am -exec sed
       -e 's,libexpat\.la,libexpatw.la,'
       -e 's,libexpat_la,libexpatw_la,'
-       -i {} +`
+       -i.bak {} +`

 1. Run `automake` to re-write `Makefile.in` files:<br/>
   `automake`
@@ -250,7 +291,7 @@ EXPAT_ENABLE_INSTALL:BOOL=ON
 // Use /MT flag (static CRT) when compiling in MSVC
 EXPAT_MSVC_STATIC_CRT:BOOL=OFF

-// Build fuzzers via ossfuzz for the expat library
+// Build fuzzers via OSS-Fuzz for the expat library
 EXPAT_OSSFUZZ_BUILD:BOOL=OFF

 // Build a shared expat library
@@ -8,6 +8,7 @@
 #
 # Copyright (c) 2017-2022 Sebastian Pipping <[email protected]>
 # Copyright (c) 2018      Marco Maggi <[email protected]>
+# Copyright (c) 2024      Dag-Erling Smørgrav <[email protected]>
 # Licensed under the MIT license:
 #
 # Permission is  hereby granted,  free of charge,  to any  person obtaining
@@ -31,25 +32,4 @@

 set -e

-# File expat_config.h.in (as generated by autoheader by autoreconf) contains
-# macro SIZEOF_VOID_P which is (1) not really needed by Expat as of today and
-# (2) a problem to "multilib" systems with one shared installed
-# /usr/include/expat_config.h for two Expats with different "void *" sizes
-# installed in e.g. /usr/lib32 and /usr/lib64.  Hence we patch macro
-# SIZEOF_VOID_P out of template expat_config.h.in so that configure will
-# not put SIZEOF_VOID_P in the eventual expat_config.h.
-patch_expat_config_h_in() {
-    local filename="$1"
-    local sizeof_void_p_line_number="$(grep -F -n SIZEOF_VOID_P "${filename}" | awk -F: '{print $1}')"
-    [[ ${sizeof_void_p_line_number} =~ ^[0-9]+$ ]]  # cheap assert
-    local first_line_to_delete=$(( sizeof_void_p_line_number - 1 ))
-    local last_line_to_delete=$(( sizeof_void_p_line_number + 1 ))
-    # Note: Avoiding "sed -i" only for macOS portability.
-    local tempfile="$(mktemp)"
-    sed "${first_line_to_delete},${last_line_to_delete}d" "${filename}" > "${tempfile}"
-    mv "${tempfile}" "${filename}"
-}
-
-autoreconf --warnings=all --install --verbose "$@"
-
-patch_expat_config_h_in expat_config.h.in
+exec autoreconf --warnings=all --install --verbose "$@"
@@ -11,7 +11,7 @@ dnl   Copyright (c) 2000      Clark Cooper <[email protected]>
 dnl   Copyright (c) 2000-2005 Fred L. Drake, Jr. <[email protected]>
 dnl   Copyright (c) 2001-2003 Greg Stein <[email protected]>
 dnl   Copyright (c) 2006-2012 Karl Waclawek <[email protected]>
-dnl   Copyright (c) 2016-2024 Sebastian Pipping <[email protected]>
+dnl   Copyright (c) 2016-2025 Sebastian Pipping <[email protected]>
 dnl   Copyright (c) 2017      S. P. Zeidler <[email protected]>
 dnl   Copyright (c) 2017      Stephen Groat <[email protected]>
 dnl   Copyright (c) 2017-2020 Joe Orton <[email protected]>
@@ -22,6 +22,8 @@ dnl   Copyright (c) 2018      KangLin <[email protected]>
 dnl   Copyright (c) 2019      Mohammed Khajapasha <[email protected]>
 dnl   Copyright (c) 2019      Kishore Kunche <[email protected]>
 dnl   Copyright (c) 2020      Jeffrey Walton <[email protected]>
+dnl   Copyright (c) 2024      Ferenc Géczi <[email protected]>
+dnl   Copyright (c) 2024      Dag-Erling Smørgrav <[email protected]>
 dnl   Licensed under the MIT license:
 dnl
 dnl   Permission is  hereby granted,  free of charge,  to any  person obtaining
@@ -82,9 +84,9 @@ dnl
 dnl If the API changes incompatibly set LIBAGE back to 0
 dnl

-LIBCURRENT=10  # sync
-LIBREVISION=0  # with
-LIBAGE=9       # CMakeLists.txt!
+LIBCURRENT=11  # sync
+LIBREVISION=2  # with
+LIBAGE=10      # CMakeLists.txt!

 AC_CONFIG_HEADERS([expat_config.h])
 AH_TOP([#ifndef EXPAT_CONFIG_H
@@ -160,7 +162,6 @@ AC_C_BIGENDIAN([AC_DEFINE([WORDS_BIGENDIAN], 1)
 AC_DEFINE_UNQUOTED([BYTEORDER], $BYTEORDER, [1234 = LILENDIAN, 4321 = BIGENDIAN])

 AC_C_CONST
-AC_TYPE_SIZE_T

 AC_ARG_WITH([xmlwf],
  [AS_HELP_STRING([--without-xmlwf], [do not build xmlwf])],
@@ -215,7 +216,7 @@ AC_LINK_IFELSE([AC_LANG_SOURCE([
    #else
    # include <stdlib.h>  /* for arc4random_buf on BSD */
    #endif
-    int main() {
+    int main(void) {
      char dummy[[123]];  // double brackets for m4
      arc4random_buf(dummy, 0U);
      return 0;
@@ -232,7 +233,7 @@ AC_LINK_IFELSE([AC_LANG_SOURCE([
       #else
       # include <stdlib.h>
       #endif
-       int main() {
+       int main(void) {
          arc4random();
          return 0;
       }
@@ -254,7 +255,7 @@ AS_IF([test "x$with_getrandom" != xno],
   AC_LINK_IFELSE([AC_LANG_SOURCE([
       #include <stdlib.h>  /* for NULL */
       #include <sys/random.h>
-       int main() {
+       int main(void) {
         return getrandom(NULL, 0U, 0U);
       }
     ])],
@@ -275,10 +276,11 @@ AS_HELP_STRING([--without-sys-getrandom],
 AS_IF([test "x$with_sys_getrandom" != xno],
  [AC_MSG_CHECKING([for syscall SYS_getrandom (Linux 3.17+)])
   AC_LINK_IFELSE([AC_LANG_SOURCE([
+       #define _GNU_SOURCE
       #include <stdlib.h>  /* for NULL */
       #include <unistd.h>  /* for syscall */
       #include <sys/syscall.h>  /* for SYS_getrandom */
-       int main() {
+       int main(void) {
         syscall(SYS_getrandom, NULL, 0, 0);
         return 0;
     }
@@ -357,11 +359,22 @@ AS_IF([test "x${DOCBOOK_TO_MAN}" != x -a "x$with_docbook" != xno],
  page for xmlwf.])])])

 dnl This will make sure that a release tarball shipping a pre-rendered xmlwf man page will
-dnl get it installed, independent of whether some flavor of docbook2man is available.
+dnl get it installed, when no working flavor of docbook2man is available (or wanted).
 dnl This relies on file xmlwf.1 being at least as recent as its source file xmlwf.xml.
 AS_IF([test -f "${srcdir}"/doc/xmlwf.1],
-  [AM_CONDITIONAL(WITH_DOCBOOK, [true])],
-  [AM_CONDITIONAL(WITH_DOCBOOK, [test "x${DOCBOOK_TO_MAN}" != x])])
+  [AM_CONDITIONAL(WITH_MANPAGE, [true])
+   AS_IF([test "x$with_docbook" = xno -o "x${DOCBOOK_TO_MAN}" = x],
+     [AM_CONDITIONAL(WITH_PREBUILT_MANPAGE, [true])
+      AM_CONDITIONAL(WITH_DISTRIBUTABLE_MANPAGE, [false])],
+     [AM_CONDITIONAL(WITH_PREBUILT_MANPAGE, [false])
+      AM_CONDITIONAL(WITH_DISTRIBUTABLE_MANPAGE, [true])])
+   ],
+  [AS_IF([test "x$with_docbook" != xno -a "x${DOCBOOK_TO_MAN}" != x],
+     [AM_CONDITIONAL(WITH_MANPAGE, [true])
+      AM_CONDITIONAL(WITH_DISTRIBUTABLE_MANPAGE, [true])],
+     [AM_CONDITIONAL(WITH_MANPAGE, [false])
+      AM_CONDITIONAL(WITH_DISTRIBUTABLE_MANPAGE, [false])])
+   AM_CONDITIONAL(WITH_PREBUILT_MANPAGE, [false])])

 dnl Configure CMake file templates
 dnl NOTE: The *_TRUE variables read here are Automake conditionals
@@ -392,7 +405,6 @@ LIBDIR_BASENAME="$(basename "${libdir}")"
 SO_MAJOR="$(expr "${LIBCURRENT}" - "${LIBAGE}")"
 SO_MINOR="${LIBAGE}"
 SO_PATCH="${LIBREVISION}"
-AC_CHECK_SIZEOF([void *])  # sets ac_cv_sizeof_void_p
 AC_SUBST([EXPAT_ATTR_INFO])
 AC_SUBST([EXPAT_DTD])
 AC_SUBST([EXPAT_LARGE_SIZE])
@@ -405,8 +417,13 @@ AC_SUBST([LIBDIR_BASENAME])
 AC_SUBST([SO_MAJOR])
 AC_SUBST([SO_MINOR])
 AC_SUBST([SO_PATCH])
-AC_SUBST([ac_cv_sizeof_void_p])

+dnl The canonical way of doing this is AC_CHECK_SIZEOF(void *), but
+dnl that adds SIZEOF_VOID_P to expat_config.h.in, making it difficult
+dnl to have 32-bit and 64-bit versions of libexpat installed on the
+dnl same system with a single, shared copy of the header.
+AC_COMPUTE_INT(SIZEOF_VOID_P, [sizeof(void *)])
+AC_SUBST([SIZEOF_VOID_P])

 dnl write the Automake flags we set
 AC_SUBST([AM_CPPFLAGS])
@@ -6,9 +6,10 @@
 #                      \___/_/\_\ .__/ \__,_|\__|
 #                               |_| XML parser
 #
-# Copyright (c) 2017-2022 Sebastian Pipping <[email protected]>
+# Copyright (c) 2017-2024 Sebastian Pipping <[email protected]>
 # Copyright (c) 2017      Stephen Groat <[email protected]>
 # Copyright (c) 2017      Joe Orton <[email protected]>
+# Copyright (c) 2024      Tomas Korbar <[email protected]>
 # Licensed under the MIT license:
 #
 # Permission is  hereby granted,  free of charge,  to any  person obtaining
@@ -32,26 +33,24 @@

 .PHONY: dist-hook  # not inside conditional to avoid automake warning

-if WITH_DOCBOOK
+if WITH_MANPAGE
 dist_man_MANS = xmlwf.1

 xmlwf.1: xmlwf.xml
 	-rm -f $@
-	$(DOCBOOK_TO_MAN) $<
+	test "x$(DOCBOOK_TO_MAN)" != x && $(DOCBOOK_TO_MAN) $<
 	test -f $@ || mv XMLWF.1 $@
-else
+endif
+
+if !WITH_DISTRIBUTABLE_MANPAGE
 dist-hook:
 	@echo 'ERROR: Configure with --with-docbook for "make dist".' 1>&2
 	@false
 endif

-# https://www.gnu.org/software/automake/manual/automake.html#What-Gets-Cleaned
-.PHONY: clean-local
-clean-local: clean-local-check
-
-.PHONY: clean-local-check
-clean-local-check:
-	$(RM) xmlwf.1
+if !WITH_PREBUILT_MANPAGE
+CLEANFILES = xmlwf.1
+endif

 EXTRA_DIST = \
    ok.min.css \
@@ -22,9 +22,10 @@
 #                      \___/_/\_\ .__/ \__,_|\__|
 #                               |_| XML parser
 #
-# Copyright (c) 2017-2022 Sebastian Pipping <[email protected]>
+# Copyright (c) 2017-2024 Sebastian Pipping <[email protected]>
 # Copyright (c) 2017      Stephen Groat <[email protected]>
 # Copyright (c) 2017      Joe Orton <[email protected]>
+# Copyright (c) 2024      Tomas Korbar <[email protected]>
 # Licensed under the MIT license:
 #
 # Permission is  hereby granted,  free of charge,  to any  person obtaining
@@ -285,6 +286,7 @@ RANLIB = @RANLIB@
 SED = @SED@
 SET_MAKE = @SET_MAKE@
 SHELL = @SHELL@
+SIZEOF_VOID_P = @SIZEOF_VOID_P@
 SO_MAJOR = @SO_MAJOR@
 SO_MINOR = @SO_MINOR@
 SO_PATCH = @SO_PATCH@
@@ -298,7 +300,6 @@ ac_ct_AR = @ac_ct_AR@
 ac_ct_CC = @ac_ct_CC@
 ac_ct_CXX = @ac_ct_CXX@
 ac_ct_DUMPBIN = @ac_ct_DUMPBIN@
-ac_cv_sizeof_void_p = @ac_cv_sizeof_void_p@
 am__include = @am__include@
 am__leading_dot = @am__leading_dot@
 am__quote = @am__quote@
@@ -345,7 +346,8 @@ target_alias = @target_alias@
 top_build_prefix = @top_build_prefix@
 top_builddir = @top_builddir@
 top_srcdir = @top_srcdir@
-@WITH_DOCBOOK_TRUE@dist_man_MANS = xmlwf.1
+@WITH_MANPAGE_TRUE@dist_man_MANS = xmlwf.1
+@WITH_PREBUILT_MANPAGE_FALSE@CLEANFILES = xmlwf.1
 EXTRA_DIST = \
    ok.min.css \
    reference.html \
@@ -439,7 +441,7 @@ ctags CTAGS:

 cscope cscopelist:

-@WITH_DOCBOOK_TRUE@dist-hook:
+@WITH_DISTRIBUTABLE_MANPAGE_TRUE@dist-hook:
 distdir: $(BUILT_SOURCES)
 	$(MAKE) $(AM_MAKEFLAGS) distdir-am

@@ -505,6 +507,7 @@ install-strip:
 mostlyclean-generic:

 clean-generic:
+	-test -z "$(CLEANFILES)" || rm -f $(CLEANFILES)

 distclean-generic:
 	-test -z "$(CONFIG_CLEAN_FILES)" || rm -f $(CONFIG_CLEAN_FILES)
@@ -515,7 +518,7 @@ maintainer-clean-generic:
 	@echo "it deletes files that may require special tools to rebuild."
 clean: clean-am

-clean-am: clean-generic clean-libtool clean-local mostlyclean-am
+clean-am: clean-generic clean-libtool mostlyclean-am

 distclean: distclean-am
 	-rm -f Makefile
@@ -584,38 +587,31 @@ uninstall-man: uninstall-man1
 .MAKE: install-am install-strip

 .PHONY: all all-am check check-am clean clean-generic clean-libtool \
-	clean-local cscopelist-am ctags-am dist-hook distclean \
-	distclean-generic distclean-libtool distdir dvi dvi-am html \
-	html-am info info-am install install-am install-data \
-	install-data-am install-dvi install-dvi-am install-exec \
-	install-exec-am install-html install-html-am install-info \
-	install-info-am install-man install-man1 install-pdf \
-	install-pdf-am install-ps install-ps-am install-strip \
-	installcheck installcheck-am installdirs maintainer-clean \
-	maintainer-clean-generic mostlyclean mostlyclean-generic \
-	mostlyclean-libtool pdf pdf-am ps ps-am tags-am uninstall \
-	uninstall-am uninstall-man uninstall-man1
+	cscopelist-am ctags-am dist-hook distclean distclean-generic \
+	distclean-libtool distdir dvi dvi-am html html-am info info-am \
+	install install-am install-data install-data-am install-dvi \
+	install-dvi-am install-exec install-exec-am install-html \
+	install-html-am install-info install-info-am install-man \
+	install-man1 install-pdf install-pdf-am install-ps \
+	install-ps-am install-strip installcheck installcheck-am \
+	installdirs maintainer-clean maintainer-clean-generic \
+	mostlyclean mostlyclean-generic mostlyclean-libtool pdf pdf-am \
+	ps ps-am tags-am uninstall uninstall-am uninstall-man \
+	uninstall-man1

 .PRECIOUS: Makefile


 .PHONY: dist-hook  # not inside conditional to avoid automake warning

-@WITH_DOCBOOK[email protected]: xmlwf.xml
-@WITH_DOCBOOK_TRUE@	-rm -f $@
-@WITH_DOCBOOK_TRUE@	$(DOCBOOK_TO_MAN) $<
-@WITH_DOCBOOK_TRUE@	test -f $@ || mv XMLWF.1 $@
-@WITH_DOCBOOK_FALSE@dist-hook:
-@WITH_DOCBOOK_FALSE@	@echo 'ERROR: Configure with --with-docbook for "make dist".' 1>&2
-@WITH_DOCBOOK_FALSE@	@false
+@WITH_MANPAGE[email protected]: xmlwf.xml
+@WITH_MANPAGE_TRUE@	-rm -f $@
+@WITH_MANPAGE_TRUE@	test "x$(DOCBOOK_TO_MAN)" != x && $(DOCBOOK_TO_MAN) $<
+@WITH_MANPAGE_TRUE@	test -f $@ || mv XMLWF.1 $@

-# https://www.gnu.org/software/automake/manual/automake.html#What-Gets-Cleaned
-.PHONY: clean-local
-clean-local: clean-local-check
-
-.PHONY: clean-local-check
-clean-local-check:
-	$(RM) xmlwf.1
+@WITH_DISTRIBUTABLE_MANPAGE_FALSE@dist-hook:
+@WITH_DISTRIBUTABLE_MANPAGE_FALSE@	@echo 'ERROR: Configure with --with-docbook for "make dist".' 1>&2
+@WITH_DISTRIBUTABLE_MANPAGE_FALSE@	@false

 # Tell versions [3.59,3.63) of GNU make to not export all variables.
 # Otherwise a system limit (for SysV at least) may be exceeded.
@@ -14,7 +14,7 @@
   Copyright (c) 2000      Clark Cooper <[email protected]>
   Copyright (c) 2000-2004 Fred L. Drake, Jr. <[email protected]>
   Copyright (c) 2002-2012 Karl Waclawek <[email protected]>
-   Copyright (c) 2017-2024 Sebastian Pipping <[email protected]>
+   Copyright (c) 2017-2025 Sebastian Pipping <[email protected]>
   Copyright (c) 2017      Jakub Wilk <[email protected]>
   Copyright (c) 2021      Tomas Korbar <[email protected]>
   Copyright (c) 2021      Nicolas Cavallari <nicolas.cavallari@green-communications.fr>
@@ -52,7 +52,7 @@
  <div>
    <h1>
      The Expat XML Parser
-      <small>Release 2.6.0</small>
+      <small>Release 2.7.1</small>
    </h1>
  </div>
 <div class="content">
@@ -319,7 +319,7 @@ directions in the next section. Otherwise if you have Microsoft's
 Developer Studio installed,
 you can use CMake to generate a <code>.sln</code> file, e.g.
 <code>
-cmake -G"Visual Studio 15 2017" -DCMAKE_BUILD_TYPE=RelWithDebInfo .
+cmake -G"Visual Studio 16 2019" -DCMAKE_BUILD_TYPE=RelWithDebInfo .
 </code>, and build Expat using <code>msbuild /m expat.sln</code> after.</p>

 <p>Alternatively, you may download the Win32 binary package that
@@ -356,10 +356,7 @@ library and header would get installed in
 <h3>Configuring Expat Using the Pre-Processor</h3>

 <p>Expat's feature set can be configured using a small number of
-pre-processor definitions.  The definition of this symbols does not
-affect the set of entry points for Expat, only the behavior of the API
-and the definition of character types in the case of
-<code>XML_UNICODE_WCHAR_T</code>.  The symbols are:</p>
+pre-processor definitions.  The symbols are:</p>

 <dl class="cpp-symbols">
 <dt><a name="XML_GE">XML_GE</a></dt>
@@ -1138,7 +1135,9 @@ containing part (or perhaps all) of the document. The number of bytes of s
 that are part of the document is indicated by <code>len</code>. This means
 that <code>s</code> doesn't have to be null-terminated. It also means that
 if <code>len</code> is larger than the number of bytes in the block of
-memory that <code>s</code> points at, then a memory fault is likely. The
+memory that <code>s</code> points at, then a memory fault is likely.
+Negative values for <code>len</code> are rejected since Expat 2.2.1.
+The
 <code>isFinal</code> parameter informs the parser that this is the last
 piece of the document. Frequently, the last piece is empty (i.e.
 <code>len</code> is zero.)
@@ -1186,11 +1185,17 @@ XML_ParseBuffer(XML_Parser p,
                int isFinal);
 </pre>
 <div class="fcndef">
+<p>
 This is just like <code><a href= "#XML_Parse" >XML_Parse</a></code>,
 except in this case Expat provides the buffer.  By obtaining the
 buffer from Expat with the <code><a href= "#XML_GetBuffer"
 >XML_GetBuffer</a></code> function, the application can avoid double
 copying of the input.
+</p>
+
+<p>
+Negative values for <code>len</code> are rejected since Expat 2.6.3.
+</p>
 </div>

 <h4 id="XML_GetBuffer">XML_GetBuffer</h4>
@@ -1262,6 +1267,11 @@ call-backs, except when parsing an external parameter entity and
 <code>XML_STATUS_ERROR</code> otherwise.  The possible error codes
 are:</p>
 <dl>
+  <dt><code>XML_ERROR_NOT_STARTED</code></dt>
+  <dd>
+    when stopping or suspending a parser before it has started,
+    added in Expat 2.6.4.
+  </dd>
  <dt><code>XML_ERROR_SUSPENDED</code></dt>
  <dd>when suspending an already suspended parser.</dd>
  <dt><code>XML_ERROR_FINISHED</code></dt>
@@ -5,7 +5,7 @@
 \\$2 \(la\\$1\(ra\\$3
 ..
 .if \n(.g .mso www.tmac
-.TH XMLWF 1 "February 6, 2024" "" ""
+.TH XMLWF 1 "March 27, 2025" "" ""
 .SH NAME
 xmlwf \- Determines if an XML document is well-formed
 .SH SYNOPSIS
@@ -9,7 +9,7 @@
   Copyright (c) 2001      Scott Bronson <[email protected]>
   Copyright (c) 2002-2003 Fred L. Drake, Jr. <[email protected]>
   Copyright (c) 2009      Karl Waclawek <[email protected]>
-   Copyright (c) 2016-2024 Sebastian Pipping <[email protected]>
+   Copyright (c) 2016-2025 Sebastian Pipping <[email protected]>
   Copyright (c) 2016      Ardo van Rangelrooij <[email protected]>
   Copyright (c) 2017      Rhodri James <[email protected]>
   Copyright (c) 2020      Joe Orton <[email protected]>
@@ -21,7 +21,7 @@
          "http://www.oasis-open.org/docbook/xml/4.2/docbookx.dtd" [
  <!ENTITY dhfirstname "<firstname>Scott</firstname>">
  <!ENTITY dhsurname   "<surname>Bronson</surname>">
-  <!ENTITY dhdate      "<date>February 6, 2024</date>">
+  <!ENTITY dhdate      "<date>March 27, 2025</date>">
  <!-- Please adjust this^^ date whenever cutting a new release. -->
  <!ENTITY dhsection   "<manvolnum>1</manvolnum>">
  <!ENTITY dhemail     "<email>[email protected]</email>">
@@ -313,6 +313,7 @@ RANLIB = @RANLIB@
 SED = @SED@
 SET_MAKE = @SET_MAKE@
 SHELL = @SHELL@
+SIZEOF_VOID_P = @SIZEOF_VOID_P@
 SO_MAJOR = @SO_MAJOR@
 SO_MINOR = @SO_MINOR@
 SO_PATCH = @SO_PATCH@
@@ -326,7 +327,6 @@ ac_ct_AR = @ac_ct_AR@
 ac_ct_CC = @ac_ct_CC@
 ac_ct_CXX = @ac_ct_CXX@
 ac_ct_DUMPBIN = @ac_ct_DUMPBIN@
-ac_cv_sizeof_void_p = @ac_cv_sizeof_void_p@
 am__include = @am__include@
 am__leading_dot = @am__leading_dot@
 am__quote = @am__quote@
@@ -15,6 +15,7 @@
   Copyright (c) 2016-2024 Sebastian Pipping <sebastian@pipping.org>
   Copyright (c) 2017      Rhodri James <rhodri@wildebeest.org.uk>
   Copyright (c) 2019      Zhongyuan Zhou <zhouzhongyuan@huawei.com>
+   Copyright (c) 2024      Hanno Böck <hanno@gentoo.org>
   Licensed under the MIT license:

   Permission is  hereby granted,  free of charge,  to any  person obtaining
@@ -127,15 +128,15 @@ dumpContentModelElement(const XML_Content *model, unsigned level,
  }

  // Node
-  printf("[%u] type=%s(%d), quant=%s(%d)", (unsigned)(model - root),
-         contentTypeName(model->type), model->type,
-         contentQuantName(model->quant), model->quant);
+  printf("[%u] type=%s(%u), quant=%s(%u)", (unsigned)(model - root),
+         contentTypeName(model->type), (unsigned int)model->type,
+         contentQuantName(model->quant), (unsigned int)model->quant);
  if (model->name) {
    printf(", name=\"%" XML_FMT_STR "\"", model->name);
  } else {
    printf(", name=NULL");
  }
-  printf(", numchildren=%d", model->numchildren);
+  printf(", numchildren=%u", model->numchildren);
  printf("\n");
 }

@@ -139,7 +139,4 @@
 /* Define to `long int' if <sys/types.h> does not define. */
 #undef off_t

-/* Define to `unsigned int' if <sys/types.h> does not define. */
-#undef size_t
-
 #endif // ndef EXPAT_CONFIG_H
@@ -7,6 +7,7 @@
 #                               |_| XML parser
 #
 # Copyright (c) 2019-2022 Sebastian Pipping <[email protected]>
+# Copyright (c) 2024      Dag-Erling Smørgrav <[email protected]>
 # Licensed under the MIT license:
 #
 # Permission is  hereby granted,  free of charge,  to any  person obtaining
@@ -32,10 +33,10 @@ set -e

 filename="${1:-tests/xmltest.log}"

-dos2unix "${filename}"
-
-tempfile="$(mktemp)"
-sed \
+sed -i.bak \
+        -e '# convert DOS line endings to Unix without resorting to dos2unix' \
+        -e $'s/\r//' \
+        \
        -e 's/^wine: Call .* msvcrt\.dll\._wperror, aborting$/ibm49i02.dtd: No such file or directory/' \
        \
        -e '/^wine: /d' \
@@ -46,5 +47,4 @@ sed \
        -e '/^wine client error:/d' \
        -e '/^In ibm\/invalid\/P49\/: Unhandled exception: unimplemented .\+/d' \
        \
-        "${filename}" > "${tempfile}"
-mv "${tempfile}" "${filename}"
+        "${filename}"
@@ -0,0 +1,464 @@
+/*
+                            __  __            _
+                         ___\ \/ /_ __   __ _| |_
+                        / _ \\  /| '_ \ / _` | __|
+                       |  __//  \| |_) | (_| | |_
+                        \___/_/\_\ .__/ \__,_|\__|
+                                 |_| XML parser
+
+   Copyright (c) 2022 Mark Brand <markbrand@google.com>
+   Copyright (c) 2025 Sebastian Pipping <sebastian@pipping.org>
+   Licensed under the MIT license:
+
+   Permission is  hereby granted,  free of charge,  to any  person obtaining
+   a  copy  of  this  software   and  associated  documentation  files  (the
+   "Software"),  to  deal in  the  Software  without restriction,  including
+   without  limitation the  rights  to use,  copy,  modify, merge,  publish,
+   distribute, sublicense, and/or sell copies of the Software, and to permit
+   persons  to whom  the Software  is  furnished to  do so,  subject to  the
+   following conditions:
+
+   The above copyright  notice and this permission notice  shall be included
+   in all copies or substantial portions of the Software.
+
+   THE  SOFTWARE  IS  PROVIDED  "AS  IS",  WITHOUT  WARRANTY  OF  ANY  KIND,
+   EXPRESS  OR IMPLIED,  INCLUDING  BUT  NOT LIMITED  TO  THE WARRANTIES  OF
+   MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN
+   NO EVENT SHALL THE AUTHORS OR  COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM,
+   DAMAGES OR  OTHER LIABILITY, WHETHER  IN AN  ACTION OF CONTRACT,  TORT OR
+   OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE
+   USE OR OTHER DEALINGS IN THE SOFTWARE.
+*/
+
+#if defined(NDEBUG)
+#  undef NDEBUG // because checks below rely on assert(...)
+#endif
+
+#include <assert.h>
+#include <stdint.h>
+#include <vector>
+
+#include "expat.h"
+#include "xml_lpm_fuzzer.pb.h"
+#include "src/libfuzzer/libfuzzer_macro.h"
+
+static const char *g_encoding = nullptr;
+static const char *g_external_entity = nullptr;
+static size_t g_external_entity_size = 0;
+
+void
+SetEncoding(const xml_lpm_fuzzer::Encoding &e) {
+  switch (e) {
+  case xml_lpm_fuzzer::Encoding::UTF8:
+    g_encoding = "UTF-8";
+    break;
+
+  case xml_lpm_fuzzer::Encoding::UTF16:
+    g_encoding = "UTF-16";
+    break;
+
+  case xml_lpm_fuzzer::Encoding::ISO88591:
+    g_encoding = "ISO-8859-1";
+    break;
+
+  case xml_lpm_fuzzer::Encoding::ASCII:
+    g_encoding = "US-ASCII";
+    break;
+
+  case xml_lpm_fuzzer::Encoding::NONE:
+    g_encoding = NULL;
+    break;
+
+  default:
+    g_encoding = "UNKNOWN";
+    break;
+  }
+}
+
+static int g_allocation_count = 0;
+static std::vector<int> g_fail_allocations = {};
+
+void *
+MallocHook(size_t size) {
+  g_allocation_count += 1;
+  for (auto index : g_fail_allocations) {
+    if (index == g_allocation_count) {
+      return NULL;
+    }
+  }
+  return malloc(size);
+}
+
+void *
+ReallocHook(void *ptr, size_t size) {
+  g_allocation_count += 1;
+  for (auto index : g_fail_allocations) {
+    if (index == g_allocation_count) {
+      return NULL;
+    }
+  }
+  return realloc(ptr, size);
+}
+
+void
+FreeHook(void *ptr) {
+  free(ptr);
+}
+
+XML_Memory_Handling_Suite memory_handling_suite
+    = {MallocHook, ReallocHook, FreeHook};
+
+void InitializeParser(XML_Parser parser);
+
+// We want a parse function that supports resumption, so that we can cover the
+// suspend/resume code.
+enum XML_Status
+Parse(XML_Parser parser, const char *input, int input_len, int is_final) {
+  enum XML_Status status = XML_Parse(parser, input, input_len, is_final);
+  while (status == XML_STATUS_SUSPENDED) {
+    status = XML_ResumeParser(parser);
+  }
+  return status;
+}
+
+// When the fuzzer is compiled with instrumentation such as ASan, then the
+// accesses in TouchString will fault if they access invalid memory (ie. detect
+// either a use-after-free or buffer-overflow). By calling TouchString in each
+// of the callbacks, we can check that the arguments meet the API specifications
+// in terms of length/null-termination. no_optimize is used to ensure that the
+// compiler has to emit actual memory reads, instead of removing them.
+static volatile size_t no_optimize = 0;
+static void
+TouchString(const XML_Char *ptr, int len = -1) {
+  if (! ptr) {
+    return;
+  }
+
+  if (len == -1) {
+    for (XML_Char value = *ptr++; value; value = *ptr++) {
+      no_optimize += value;
+    }
+  } else {
+    for (int i = 0; i < len; ++i) {
+      no_optimize += ptr[i];
+    }
+  }
+}
+
+static void
+TouchNodeAndRecurse(XML_Content *content) {
+  switch (content->type) {
+  case XML_CTYPE_EMPTY:
+  case XML_CTYPE_ANY:
+    assert(content->quant == XML_CQUANT_NONE);
+    assert(content->name == NULL);
+    assert(content->numchildren == 0);
+    assert(content->children == NULL);
+    break;
+
+  case XML_CTYPE_MIXED:
+    assert(content->quant == XML_CQUANT_NONE
+           || content->quant == XML_CQUANT_REP);
+    assert(content->name == NULL);
+    for (unsigned int i = 0; i < content->numchildren; ++i) {
+      assert(content->children[i].type == XML_CTYPE_NAME);
+      assert(content->children[i].quant == XML_CQUANT_NONE);
+      assert(content->children[i].numchildren == 0);
+      assert(content->children[i].children == NULL);
+      TouchString(content->children[i].name);
+    }
+    break;
+
+  case XML_CTYPE_NAME:
+    assert((content->quant == XML_CQUANT_NONE)
+           || (content->quant == XML_CQUANT_OPT)
+           || (content->quant == XML_CQUANT_REP)
+           || (content->quant == XML_CQUANT_PLUS));
+    assert(content->numchildren == 0);
+    assert(content->children == NULL);
+    TouchString(content->name);
+    break;
+
+  case XML_CTYPE_CHOICE:
+  case XML_CTYPE_SEQ:
+    assert((content->quant == XML_CQUANT_NONE)
+           || (content->quant == XML_CQUANT_OPT)
+           || (content->quant == XML_CQUANT_REP)
+           || (content->quant == XML_CQUANT_PLUS));
+    assert(content->name == NULL);
+    for (unsigned int i = 0; i < content->numchildren; ++i) {
+      TouchNodeAndRecurse(&content->children[i]);
+    }
+    break;
+
+  default:
+    assert(false);
+  }
+}
+
+static void XMLCALL
+ElementDeclHandler(void *userData, const XML_Char *name, XML_Content *model) {
+  TouchString(name);
+  TouchNodeAndRecurse(model);
+  XML_FreeContentModel((XML_Parser)userData, model);
+}
+
+static void XMLCALL
+AttlistDeclHandler(void *userData, const XML_Char *elname,
+                   const XML_Char *attname, const XML_Char *atttype,
+                   const XML_Char *dflt, int isrequired) {
+  (void)userData;
+  TouchString(elname);
+  TouchString(attname);
+  TouchString(atttype);
+  TouchString(dflt);
+  (void)isrequired;
+}
+
+static void XMLCALL
+XmlDeclHandler(void *userData, const XML_Char *version,
+               const XML_Char *encoding, int standalone) {
+  (void)userData;
+  TouchString(version);
+  TouchString(encoding);
+  (void)standalone;
+}
+
+static void XMLCALL
+StartElementHandler(void *userData, const XML_Char *name,
+                    const XML_Char **atts) {
+  (void)userData;
+  TouchString(name);
+  for (size_t i = 0; atts[i] != NULL; ++i) {
+    TouchString(atts[i]);
+  }
+}
+
+static void XMLCALL
+EndElementHandler(void *userData, const XML_Char *name) {
+  (void)userData;
+  TouchString(name);
+}
+
+static void XMLCALL
+CharacterDataHandler(void *userData, const XML_Char *s, int len) {
+  (void)userData;
+  TouchString(s, len);
+}
+
+static void XMLCALL
+ProcessingInstructionHandler(void *userData, const XML_Char *target,
+                             const XML_Char *data) {
+  (void)userData;
+  TouchString(target);
+  TouchString(data);
+}
+
+static void XMLCALL
+CommentHandler(void *userData, const XML_Char *data) {
+  TouchString(data);
+  // Use the comment handler to trigger parser suspend, so that we can get
+  // coverage of that code.
+  XML_StopParser((XML_Parser)userData, XML_TRUE);
+}
+
+static void XMLCALL
+StartCdataSectionHandler(void *userData) {
+  (void)userData;
+}
+
+static void XMLCALL
+EndCdataSectionHandler(void *userData) {
+  (void)userData;
+}
+
+static void XMLCALL
+DefaultHandler(void *userData, const XML_Char *s, int len) {
+  (void)userData;
+  TouchString(s, len);
+}
+
+static void XMLCALL
+StartDoctypeDeclHandler(void *userData, const XML_Char *doctypeName,
+                        const XML_Char *sysid, const XML_Char *pubid,
+                        int has_internal_subset) {
+  (void)userData;
+  TouchString(doctypeName);
+  TouchString(sysid);
+  TouchString(pubid);
+  (void)has_internal_subset;
+}
+
+static void XMLCALL
+EndDoctypeDeclHandler(void *userData) {
+  (void)userData;
+}
+
+static void XMLCALL
+EntityDeclHandler(void *userData, const XML_Char *entityName,
+                  int is_parameter_entity, const XML_Char *value,
+                  int value_length, const XML_Char *base,
+                  const XML_Char *systemId, const XML_Char *publicId,
+                  const XML_Char *notationName) {
+  (void)userData;
+  TouchString(entityName);
+  (void)is_parameter_entity;
+  TouchString(value, value_length);
+  TouchString(base);
+  TouchString(systemId);
+  TouchString(publicId);
+  TouchString(notationName);
+}
+
+static void XMLCALL
+NotationDeclHandler(void *userData, const XML_Char *notationName,
+                    const XML_Char *base, const XML_Char *systemId,
+                    const XML_Char *publicId) {
+  (void)userData;
+  TouchString(notationName);
+  TouchString(base);
+  TouchString(systemId);
+  TouchString(publicId);
+}
+
+static void XMLCALL
+StartNamespaceDeclHandler(void *userData, const XML_Char *prefix,
+                          const XML_Char *uri) {
+  (void)userData;
+  TouchString(prefix);
+  TouchString(uri);
+}
+
+static void XMLCALL
+EndNamespaceDeclHandler(void *userData, const XML_Char *prefix) {
+  (void)userData;
+  TouchString(prefix);
+}
+
+static int XMLCALL
+NotStandaloneHandler(void *userData) {
+  (void)userData;
+  return XML_STATUS_OK;
+}
+
+static int XMLCALL
+ExternalEntityRefHandler(XML_Parser parser, const XML_Char *context,
+                         const XML_Char *base, const XML_Char *systemId,
+                         const XML_Char *publicId) {
+  int rc = XML_STATUS_ERROR;
+  TouchString(context);
+  TouchString(base);
+  TouchString(systemId);
+  TouchString(publicId);
+
+  if (g_external_entity) {
+    XML_Parser ext_parser
+        = XML_ExternalEntityParserCreate(parser, context, g_encoding);
+    rc = Parse(ext_parser, g_external_entity, g_external_entity_size, 1);
+    XML_ParserFree(ext_parser);
+  }
+
+  return rc;
+}
+
+static void XMLCALL
+SkippedEntityHandler(void *userData, const XML_Char *entityName,
+                     int is_parameter_entity) {
+  (void)userData;
+  TouchString(entityName);
+  (void)is_parameter_entity;
+}
+
+static int XMLCALL
+UnknownEncodingHandler(void *encodingHandlerData, const XML_Char *name,
+                       XML_Encoding *info) {
+  (void)encodingHandlerData;
+  TouchString(name);
+  (void)info;
+  return XML_STATUS_ERROR;
+}
+
+void
+InitializeParser(XML_Parser parser) {
+  XML_SetUserData(parser, (void *)parser);
+  XML_SetHashSalt(parser, 0x41414141);
+  XML_SetParamEntityParsing(parser, XML_PARAM_ENTITY_PARSING_ALWAYS);
+
+  XML_SetElementDeclHandler(parser, ElementDeclHandler);
+  XML_SetAttlistDeclHandler(parser, AttlistDeclHandler);
+  XML_SetXmlDeclHandler(parser, XmlDeclHandler);
+  XML_SetElementHandler(parser, StartElementHandler, EndElementHandler);
+  XML_SetCharacterDataHandler(parser, CharacterDataHandler);
+  XML_SetProcessingInstructionHandler(parser, ProcessingInstructionHandler);
+  XML_SetCommentHandler(parser, CommentHandler);
+  XML_SetCdataSectionHandler(parser, StartCdataSectionHandler,
+                             EndCdataSectionHandler);
+  // XML_SetDefaultHandler disables entity expansion
+  XML_SetDefaultHandlerExpand(parser, DefaultHandler);
+  XML_SetDoctypeDeclHandler(parser, StartDoctypeDeclHandler,
+                            EndDoctypeDeclHandler);
+  // Note: This is mutually exclusive with XML_SetUnparsedEntityDeclHandler,
+  //       and there isn't any significant code change between the two.
+  XML_SetEntityDeclHandler(parser, EntityDeclHandler);
+  XML_SetNotationDeclHandler(parser, NotationDeclHandler);
+  XML_SetNamespaceDeclHandler(parser, StartNamespaceDeclHandler,
+                              EndNamespaceDeclHandler);
+  XML_SetNotStandaloneHandler(parser, NotStandaloneHandler);
+  XML_SetExternalEntityRefHandler(parser, ExternalEntityRefHandler);
+  XML_SetSkippedEntityHandler(parser, SkippedEntityHandler);
+  XML_SetUnknownEncodingHandler(parser, UnknownEncodingHandler, (void *)parser);
+}
+
+DEFINE_TEXT_PROTO_FUZZER(const xml_lpm_fuzzer::Testcase &testcase) {
+  g_external_entity = nullptr;
+
+  if (! testcase.actions_size()) {
+    return;
+  }
+
+  g_allocation_count = 0;
+  g_fail_allocations.clear();
+  for (int i = 0; i < testcase.fail_allocations_size(); ++i) {
+    g_fail_allocations.push_back(testcase.fail_allocations(i));
+  }
+
+  SetEncoding(testcase.encoding());
+  XML_Parser parser
+      = XML_ParserCreate_MM(g_encoding, &memory_handling_suite, "|");
+  InitializeParser(parser);
+
+  for (int i = 0; i < testcase.actions_size(); ++i) {
+    const auto &action = testcase.actions(i);
+    switch (action.action_case()) {
+    case xml_lpm_fuzzer::Action::kChunk:
+      if (XML_STATUS_ERROR
+          == Parse(parser, action.chunk().data(), action.chunk().size(), 0)) {
+        // Force a reset after parse error.
+        XML_ParserReset(parser, g_encoding);
+        InitializeParser(parser);
+      }
+      break;
+
+    case xml_lpm_fuzzer::Action::kLastChunk:
+      Parse(parser, action.last_chunk().data(), action.last_chunk().size(), 1);
+      XML_ParserReset(parser, g_encoding);
+      InitializeParser(parser);
+      break;
+
+    case xml_lpm_fuzzer::Action::kReset:
+      XML_ParserReset(parser, g_encoding);
+      InitializeParser(parser);
+      break;
+
+    case xml_lpm_fuzzer::Action::kExternalEntity:
+      g_external_entity = action.external_entity().data();
+      g_external_entity_size = action.external_entity().size();
+      break;
+
+    default:
+      break;
+    }
+  }
+
+  XML_ParserFree(parser);
+}
@@ -0,0 +1,58 @@
+/*
+                            __  __            _
+                         ___\ \/ /_ __   __ _| |_
+                        / _ \\  /| '_ \ / _` | __|
+                       |  __//  \| |_) | (_| | |_
+                        \___/_/\_\ .__/ \__,_|\__|
+                                 |_| XML parser
+
+   Copyright (c) 2022 Mark Brand <[email protected]>
+   Copyright (c) 2025 Sebastian Pipping <[email protected]>
+   Licensed under the MIT license:
+
+   Permission is  hereby granted,  free of charge,  to any  person obtaining
+   a  copy  of  this  software   and  associated  documentation  files  (the
+   "Software"),  to  deal in  the  Software  without restriction,  including
+   without  limitation the  rights  to use,  copy,  modify, merge,  publish,
+   distribute, sublicense, and/or sell copies of the Software, and to permit
+   persons  to whom  the Software  is  furnished to  do so,  subject to  the
+   following conditions:
+
+   The above copyright  notice and this permission notice  shall be included
+   in all copies or substantial portions of the Software.
+
+   THE  SOFTWARE  IS  PROVIDED  "AS  IS",  WITHOUT  WARRANTY  OF  ANY  KIND,
+   EXPRESS  OR IMPLIED,  INCLUDING  BUT  NOT LIMITED  TO  THE WARRANTIES  OF
+   MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN
+   NO EVENT SHALL THE AUTHORS OR  COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM,
+   DAMAGES OR  OTHER LIABILITY, WHETHER  IN AN  ACTION OF CONTRACT,  TORT OR
+   OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE
+   USE OR OTHER DEALINGS IN THE SOFTWARE.
+*/
+
+syntax = "proto2";
+package xml_lpm_fuzzer;
+
+enum Encoding {
+  UTF8 = 0;
+  UTF16 = 1;
+  ISO88591 = 2;
+  ASCII = 3;
+  UNKNOWN = 4;
+  NONE = 5;
+}
+
+message Action {
+  oneof action {
+    string chunk = 1;
+    string last_chunk = 2;
+    bool reset = 3;
+    string external_entity = 4;
+  }
+}
+
+message Testcase {
+  required Encoding encoding = 1;
+  repeated Action actions = 2;
+  repeated int32 fail_allocations = 3;
+}
@@ -5,7 +5,7 @@
 * you may not use this file except in compliance with the License.
 * You may obtain a copy of the License at
 *
- *      http://www.apache.org/licenses/LICENSE-2.0
+ *      https://www.apache.org/licenses/LICENSE-2.0
 *
 * Unless required by applicable law or agreed to in writing, software
 * distributed under the License is distributed on an "AS IS" BASIS,
@@ -5,7 +5,7 @@
 * you may not use this file except in compliance with the License.
 * You may obtain a copy of the License at
 *
- *      http://www.apache.org/licenses/LICENSE-2.0
+ *      https://www.apache.org/licenses/LICENSE-2.0
 *
 * Unless required by applicable law or agreed to in writing, software
 * distributed under the License is distributed on an "AS IS" BASIS,
@@ -6,7 +6,7 @@
 #                      \___/_/\_\ .__/ \__,_|\__|
 #                               |_| XML parser
 #
-# Copyright (c) 2017-2022 Sebastian Pipping <[email protected]>
+# Copyright (c) 2017-2024 Sebastian Pipping <[email protected]>
 # Copyright (c) 2017      Tomasz Kłoczko <[email protected]>
 # Copyright (c) 2019      David Loffredo <[email protected]>
 # Licensed under the MIT license:
@@ -36,7 +36,9 @@ include_HEADERS = \
    expat_external.h

 lib_LTLIBRARIES = libexpat.la
-noinst_LTLIBRARIES = libexpatinternal.la
+if WITH_TESTS
+noinst_LTLIBRARIES = libtestpat.la
+endif

 libexpat_la_LDFLAGS = \
    @AM_LDFLAGS@ \
@@ -44,17 +46,16 @@ libexpat_la_LDFLAGS = \
    -no-undefined \
    -version-info @LIBCURRENT@:@LIBREVISION@:@LIBAGE@

-libexpat_la_SOURCES =
-
-# This layer of indirection allows
-# the test suite to access internal symbols
-# despite compiling with -fvisibility=hidden
-libexpatinternal_la_SOURCES = \
+libexpat_la_SOURCES = \
    xmlparse.c \
    xmltok.c \
    xmlrole.c

-libexpat_la_LIBADD = libexpatinternal.la
+if WITH_TESTS
+libtestpat_la_CPPFLAGS = -DXML_TESTING
+
+libtestpat_la_SOURCES = $(libexpat_la_SOURCES)
+endif

 doc_DATA = \
    ../AUTHORS \
@@ -22,7 +22,7 @@
 #                      \___/_/\_\ .__/ \__,_|\__|
 #                               |_| XML parser
 #
-# Copyright (c) 2017-2022 Sebastian Pipping <[email protected]>
+# Copyright (c) 2017-2024 Sebastian Pipping <[email protected]>
 # Copyright (c) 2017      Tomasz Kłoczko <[email protected]>
 # Copyright (c) 2019      David Loffredo <[email protected]>
 # Licensed under the MIT license:
@@ -176,8 +176,8 @@ am__uninstall_files_from_dir = { \
 am__installdirs = "$(DESTDIR)$(libdir)" "$(DESTDIR)$(docdir)" \
 	"$(DESTDIR)$(includedir)"
 LTLIBRARIES = $(lib_LTLIBRARIES) $(noinst_LTLIBRARIES)
-libexpat_la_DEPENDENCIES = libexpatinternal.la
-am_libexpat_la_OBJECTS =
+libexpat_la_LIBADD =
+am_libexpat_la_OBJECTS = xmlparse.lo xmltok.lo xmlrole.lo
 libexpat_la_OBJECTS = $(am_libexpat_la_OBJECTS)
 AM_V_lt = $(am__v_lt_@AM_V@)
 am__v_lt_ = $(am__v_lt_@AM_DEFAULT_V@)
@@ -186,9 +186,13 @@ am__v_lt_1 =
 libexpat_la_LINK = $(LIBTOOL) $(AM_V_lt) --tag=CC $(AM_LIBTOOLFLAGS) \
 	$(LIBTOOLFLAGS) --mode=link $(CCLD) $(AM_CFLAGS) $(CFLAGS) \
 	$(libexpat_la_LDFLAGS) $(LDFLAGS) -o $@
-libexpatinternal_la_LIBADD =
-am_libexpatinternal_la_OBJECTS = xmlparse.lo xmltok.lo xmlrole.lo
-libexpatinternal_la_OBJECTS = $(am_libexpatinternal_la_OBJECTS)
+libtestpat_la_LIBADD =
+am__libtestpat_la_SOURCES_DIST = xmlparse.c xmltok.c xmlrole.c
+am__objects_1 = libtestpat_la-xmlparse.lo libtestpat_la-xmltok.lo \
+	libtestpat_la-xmlrole.lo
+@WITH_TESTS_TRUE@am_libtestpat_la_OBJECTS = $(am__objects_1)
+libtestpat_la_OBJECTS = $(am_libtestpat_la_OBJECTS)
+@WITH_TESTS_TRUE@am_libtestpat_la_rpath =
 AM_V_P = $(am__v_P_@AM_V@)
 am__v_P_ = $(am__v_P_@AM_DEFAULT_V@)
 am__v_P_0 = false
@@ -204,8 +208,10 @@ am__v_at_1 =
 DEFAULT_INCLUDES = -I.@am__isrc@ -I$(top_builddir)
 depcomp = $(SHELL) $(top_srcdir)/conftools/depcomp
 am__maybe_remake_depfiles = depfiles
-am__depfiles_remade = ./$(DEPDIR)/xmlparse.Plo ./$(DEPDIR)/xmlrole.Plo \
-	./$(DEPDIR)/xmltok.Plo
+am__depfiles_remade = ./$(DEPDIR)/libtestpat_la-xmlparse.Plo \
+	./$(DEPDIR)/libtestpat_la-xmlrole.Plo \
+	./$(DEPDIR)/libtestpat_la-xmltok.Plo ./$(DEPDIR)/xmlparse.Plo \
+	./$(DEPDIR)/xmlrole.Plo ./$(DEPDIR)/xmltok.Plo
 am__mv = mv -f
 COMPILE = $(CC) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) \
 	$(CPPFLAGS) $(AM_CFLAGS) $(CFLAGS)
@@ -225,8 +231,9 @@ AM_V_CCLD = $(am__v_CCLD_@AM_V@)
 am__v_CCLD_ = $(am__v_CCLD_@AM_DEFAULT_V@)
 am__v_CCLD_0 = @echo "  CCLD    " $@;
 am__v_CCLD_1 = 
-SOURCES = $(libexpat_la_SOURCES) $(libexpatinternal_la_SOURCES)
-DIST_SOURCES = $(libexpat_la_SOURCES) $(libexpatinternal_la_SOURCES)
+SOURCES = $(libexpat_la_SOURCES) $(libtestpat_la_SOURCES)
+DIST_SOURCES = $(libexpat_la_SOURCES) \
+	$(am__libtestpat_la_SOURCES_DIST)
 am__can_run_installinfo = \
  case $$AM_UPDATE_INFO_DIR in \
    n|no|NO) false;; \
@@ -344,6 +351,7 @@ RANLIB = @RANLIB@
 SED = @SED@
 SET_MAKE = @SET_MAKE@
 SHELL = @SHELL@
+SIZEOF_VOID_P = @SIZEOF_VOID_P@
 SO_MAJOR = @SO_MAJOR@
 SO_MINOR = @SO_MINOR@
 SO_PATCH = @SO_PATCH@
@@ -357,7 +365,6 @@ ac_ct_AR = @ac_ct_AR@
 ac_ct_CC = @ac_ct_CC@
 ac_ct_CXX = @ac_ct_CXX@
 ac_ct_DUMPBIN = @ac_ct_DUMPBIN@
-ac_cv_sizeof_void_p = @ac_cv_sizeof_void_p@
 am__include = @am__include@
 am__leading_dot = @am__leading_dot@
 am__quote = @am__quote@
@@ -410,24 +417,20 @@ include_HEADERS = \
    expat_external.h

 lib_LTLIBRARIES = libexpat.la
-noinst_LTLIBRARIES = libexpatinternal.la
+@WITH_TESTS_TRUE@noinst_LTLIBRARIES = libtestpat.la
 libexpat_la_LDFLAGS = \
    @AM_LDFLAGS@ \
    @LIBM@ \
    -no-undefined \
    -version-info @LIBCURRENT@:@LIBREVISION@:@LIBAGE@

-libexpat_la_SOURCES = 
-
-# This layer of indirection allows
-# the test suite to access internal symbols
-# despite compiling with -fvisibility=hidden
-libexpatinternal_la_SOURCES = \
+libexpat_la_SOURCES = \
    xmlparse.c \
    xmltok.c \
    xmlrole.c

-libexpat_la_LIBADD = libexpatinternal.la
+@WITH_TESTS_TRUE@libtestpat_la_CPPFLAGS = -DXML_TESTING
+@WITH_TESTS_TRUE@libtestpat_la_SOURCES = $(libexpat_la_SOURCES)
 doc_DATA = \
    ../AUTHORS \
    ../Changes
@@ -534,8 +537,8 @@ clean-noinstLTLIBRARIES:
 libexpat.la: $(libexpat_la_OBJECTS) $(libexpat_la_DEPENDENCIES) $(EXTRA_libexpat_la_DEPENDENCIES) 
 	$(AM_V_CCLD)$(libexpat_la_LINK) -rpath $(libdir) $(libexpat_la_OBJECTS) $(libexpat_la_LIBADD) $(LIBS)

-libexpatinternal.la: $(libexpatinternal_la_OBJECTS) $(libexpatinternal_la_DEPENDENCIES) $(EXTRA_libexpatinternal_la_DEPENDENCIES) 
-	$(AM_V_CCLD)$(LINK)  $(libexpatinternal_la_OBJECTS) $(libexpatinternal_la_LIBADD) $(LIBS)
+libtestpat.la: $(libtestpat_la_OBJECTS) $(libtestpat_la_DEPENDENCIES) $(EXTRA_libtestpat_la_DEPENDENCIES) 
+	$(AM_V_CCLD)$(LINK) $(am_libtestpat_la_rpath) $(libtestpat_la_OBJECTS) $(libtestpat_la_LIBADD) $(LIBS)

 mostlyclean-compile:
 	-rm -f *.$(OBJEXT)
@@ -543,6 +546,9 @@ mostlyclean-compile:
 distclean-compile:
 	-rm -f *.tab.c

+@AMDEP_TRUE@@am__include@ @am__quote@./$(DEPDIR)/libtestpat_la-xmlparse.Plo@am__quote@ # am--include-marker
+@AMDEP_TRUE@@am__include@ @am__quote@./$(DEPDIR)/libtestpat_la-xmlrole.Plo@am__quote@ # am--include-marker
+@AMDEP_TRUE@@am__include@ @am__quote@./$(DEPDIR)/libtestpat_la-xmltok.Plo@am__quote@ # am--include-marker
@AMDEP_TRUE@@am__include@ @am__quote@./$(DEPDIR)/xmlparse.Plo@am__quote@ # am--include-marker
@AMDEP_TRUE@@am__include@ @am__quote@./$(DEPDIR)/xmlrole.Plo@am__quote@ # am--include-marker
@AMDEP_TRUE@@am__include@ @am__quote@./$(DEPDIR)/xmltok.Plo@am__quote@ # am--include-marker
@@ -574,6 +580,27 @@ am--depfiles: $(am__depfiles_remade)
@AMDEP_TRUE@@am__fastdepCC_FALSE@	DEPDIR=$(DEPDIR) $(CCDEPMODE) $(depcomp) @AMDEPBACKSLASH@
@am__fastdepCC_FALSE@	$(AM_V_CC@am__nodep@)$(LTCOMPILE) -c -o $@ $<

+libtestpat_la-xmlparse.lo: xmlparse.c
+@am__fastdepCC_TRUE@	$(AM_V_CC)$(LIBTOOL) $(AM_V_lt) --tag=CC $(AM_LIBTOOLFLAGS) $(LIBTOOLFLAGS) --mode=compile $(CC) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(libtestpat_la_CPPFLAGS) $(CPPFLAGS) $(AM_CFLAGS) $(CFLAGS) -MT libtestpat_la-xmlparse.lo -MD -MP -MF $(DEPDIR)/libtestpat_la-xmlparse.Tpo -c -o libtestpat_la-xmlparse.lo `test -f 'xmlparse.c' || echo '$(srcdir)/'`xmlparse.c
+@am__fastdepCC_TRUE@	$(AM_V_at)$(am__mv) $(DEPDIR)/libtestpat_la-xmlparse.Tpo $(DEPDIR)/libtestpat_la-xmlparse.Plo
+@AMDEP_TRUE@@am__fastdepCC_FALSE@	$(AM_V_CC)source='xmlparse.c' object='libtestpat_la-xmlparse.lo' libtool=yes @AMDEPBACKSLASH@
+@AMDEP_TRUE@@am__fastdepCC_FALSE@	DEPDIR=$(DEPDIR) $(CCDEPMODE) $(depcomp) @AMDEPBACKSLASH@
+@am__fastdepCC_FALSE@	$(AM_V_CC@am__nodep@)$(LIBTOOL) $(AM_V_lt) --tag=CC $(AM_LIBTOOLFLAGS) $(LIBTOOLFLAGS) --mode=compile $(CC) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(libtestpat_la_CPPFLAGS) $(CPPFLAGS) $(AM_CFLAGS) $(CFLAGS) -c -o libtestpat_la-xmlparse.lo `test -f 'xmlparse.c' || echo '$(srcdir)/'`xmlparse.c
+
+libtestpat_la-xmltok.lo: xmltok.c
+@am__fastdepCC_TRUE@	$(AM_V_CC)$(LIBTOOL) $(AM_V_lt) --tag=CC $(AM_LIBTOOLFLAGS) $(LIBTOOLFLAGS) --mode=compile $(CC) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(libtestpat_la_CPPFLAGS) $(CPPFLAGS) $(AM_CFLAGS) $(CFLAGS) -MT libtestpat_la-xmltok.lo -MD -MP -MF $(DEPDIR)/libtestpat_la-xmltok.Tpo -c -o libtestpat_la-xmltok.lo `test -f 'xmltok.c' || echo '$(srcdir)/'`xmltok.c
+@am__fastdepCC_TRUE@	$(AM_V_at)$(am__mv) $(DEPDIR)/libtestpat_la-xmltok.Tpo $(DEPDIR)/libtestpat_la-xmltok.Plo
+@AMDEP_TRUE@@am__fastdepCC_FALSE@	$(AM_V_CC)source='xmltok.c' object='libtestpat_la-xmltok.lo' libtool=yes @AMDEPBACKSLASH@
+@AMDEP_TRUE@@am__fastdepCC_FALSE@	DEPDIR=$(DEPDIR) $(CCDEPMODE) $(depcomp) @AMDEPBACKSLASH@
+@am__fastdepCC_FALSE@	$(AM_V_CC@am__nodep@)$(LIBTOOL) $(AM_V_lt) --tag=CC $(AM_LIBTOOLFLAGS) $(LIBTOOLFLAGS) --mode=compile $(CC) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(libtestpat_la_CPPFLAGS) $(CPPFLAGS) $(AM_CFLAGS) $(CFLAGS) -c -o libtestpat_la-xmltok.lo `test -f 'xmltok.c' || echo '$(srcdir)/'`xmltok.c
+
+libtestpat_la-xmlrole.lo: xmlrole.c
+@am__fastdepCC_TRUE@	$(AM_V_CC)$(LIBTOOL) $(AM_V_lt) --tag=CC $(AM_LIBTOOLFLAGS) $(LIBTOOLFLAGS) --mode=compile $(CC) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(libtestpat_la_CPPFLAGS) $(CPPFLAGS) $(AM_CFLAGS) $(CFLAGS) -MT libtestpat_la-xmlrole.lo -MD -MP -MF $(DEPDIR)/libtestpat_la-xmlrole.Tpo -c -o libtestpat_la-xmlrole.lo `test -f 'xmlrole.c' || echo '$(srcdir)/'`xmlrole.c
+@am__fastdepCC_TRUE@	$(AM_V_at)$(am__mv) $(DEPDIR)/libtestpat_la-xmlrole.Tpo $(DEPDIR)/libtestpat_la-xmlrole.Plo
+@AMDEP_TRUE@@am__fastdepCC_FALSE@	$(AM_V_CC)source='xmlrole.c' object='libtestpat_la-xmlrole.lo' libtool=yes @AMDEPBACKSLASH@
+@AMDEP_TRUE@@am__fastdepCC_FALSE@	DEPDIR=$(DEPDIR) $(CCDEPMODE) $(depcomp) @AMDEPBACKSLASH@
+@am__fastdepCC_FALSE@	$(AM_V_CC@am__nodep@)$(LIBTOOL) $(AM_V_lt) --tag=CC $(AM_LIBTOOLFLAGS) $(LIBTOOLFLAGS) --mode=compile $(CC) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(libtestpat_la_CPPFLAGS) $(CPPFLAGS) $(AM_CFLAGS) $(CFLAGS) -c -o libtestpat_la-xmlrole.lo `test -f 'xmlrole.c' || echo '$(srcdir)/'`xmlrole.c
+
 mostlyclean-libtool:
 	-rm -f *.lo

@@ -749,7 +776,10 @@ clean-am: clean-generic clean-libLTLIBRARIES clean-libtool \
 	clean-noinstLTLIBRARIES mostlyclean-am

 distclean: distclean-am
-		-rm -f ./$(DEPDIR)/xmlparse.Plo
+		-rm -f ./$(DEPDIR)/libtestpat_la-xmlparse.Plo
+	-rm -f ./$(DEPDIR)/libtestpat_la-xmlrole.Plo
+	-rm -f ./$(DEPDIR)/libtestpat_la-xmltok.Plo
+	-rm -f ./$(DEPDIR)/xmlparse.Plo
 	-rm -f ./$(DEPDIR)/xmlrole.Plo
 	-rm -f ./$(DEPDIR)/xmltok.Plo
 	-rm -f Makefile
@@ -798,7 +828,10 @@ install-ps-am:
 installcheck-am:

 maintainer-clean: maintainer-clean-am
-		-rm -f ./$(DEPDIR)/xmlparse.Plo
+		-rm -f ./$(DEPDIR)/libtestpat_la-xmlparse.Plo
+	-rm -f ./$(DEPDIR)/libtestpat_la-xmlrole.Plo
+	-rm -f ./$(DEPDIR)/libtestpat_la-xmltok.Plo
+	-rm -f ./$(DEPDIR)/xmlparse.Plo
 	-rm -f ./$(DEPDIR)/xmlrole.Plo
 	-rm -f ./$(DEPDIR)/xmltok.Plo
 	-rm -f Makefile
@@ -11,13 +11,14 @@
   Copyright (c) 2000-2005 Fred L. Drake, Jr. <fdrake@users.sourceforge.net>
   Copyright (c) 2001-2002 Greg Stein <gstein@users.sourceforge.net>
   Copyright (c) 2002-2016 Karl Waclawek <karl@waclawek.net>
-   Copyright (c) 2016-2024 Sebastian Pipping <sebastian@pipping.org>
+   Copyright (c) 2016-2025 Sebastian Pipping <sebastian@pipping.org>
   Copyright (c) 2016      Cristian Rodríguez <crrodriguez@opensuse.org>
   Copyright (c) 2016      Thomas Beutlich <tc@tbeu.de>
   Copyright (c) 2017      Rhodri James <rhodri@wildebeest.org.uk>
   Copyright (c) 2022      Thijs Schreijer <thijs@thijsschreijer.nl>
   Copyright (c) 2023      Hanno Böck <hanno@gentoo.org>
   Copyright (c) 2023      Sony Corporation / Snild Dolkow <snild@sony.com>
+   Copyright (c) 2024      Taichi Haradaguchi <20001722@ymail.ne.jp>
   Licensed under the MIT license:

   Permission is  hereby granted,  free of charge,  to any  person obtaining
@@ -129,7 +130,9 @@ enum XML_Error {
  /* Added in 2.3.0. */
  XML_ERROR_NO_BUFFER,
  /* Added in 2.4.0. */
-  XML_ERROR_AMPLIFICATION_LIMIT_BREACH
+  XML_ERROR_AMPLIFICATION_LIMIT_BREACH,
+  /* Added in 2.6.4. */
+  XML_ERROR_NOT_STARTED,
 };

 enum XML_Content_Type {
@@ -1042,7 +1045,7 @@ typedef struct {
 XMLPARSEAPI(const XML_Feature *)
 XML_GetFeatureList(void);

-#if XML_GE == 1
+#if defined(XML_DTD) || (defined(XML_GE) && XML_GE == 1)
 /* Added in Expat 2.4.0 for XML_DTD defined and
 * added in Expat 2.6.0 for XML_GE == 1. */
 XMLPARSEAPI(XML_Bool)
@@ -1064,8 +1067,8 @@ XML_SetReparseDeferralEnabled(XML_Parser parser, XML_Bool enabled);
   See https://semver.org
 */
 #define XML_MAJOR_VERSION 2
-#define XML_MINOR_VERSION 6
-#define XML_MICRO_VERSION 0
+#define XML_MINOR_VERSION 7
+#define XML_MICRO_VERSION 1

 #ifdef __cplusplus
 }
@@ -28,10 +28,11 @@
   Copyright (c) 2002-2003 Fred L. Drake, Jr. <fdrake@users.sourceforge.net>
   Copyright (c) 2002-2006 Karl Waclawek <karl@waclawek.net>
   Copyright (c) 2003      Greg Stein <gstein@users.sourceforge.net>
-   Copyright (c) 2016-2023 Sebastian Pipping <sebastian@pipping.org>
+   Copyright (c) 2016-2025 Sebastian Pipping <sebastian@pipping.org>
   Copyright (c) 2018      Yury Gribov <tetra2005@gmail.com>
   Copyright (c) 2019      David Loffredo <loffredo@steptools.com>
-   Copyright (c) 2023      Sony Corporation / Snild Dolkow <snild@sony.com>
+   Copyright (c) 2023-2024 Sony Corporation / Snild Dolkow <snild@sony.com>
+   Copyright (c) 2024      Taichi Haradaguchi <20001722@ymail.ne.jp>
   Licensed under the MIT license:

   Permission is  hereby granted,  free of charge,  to any  person obtaining
@@ -126,6 +127,9 @@
 #  elif ULONG_MAX == 18446744073709551615u // 2^64-1
 #    define EXPAT_FMT_PTRDIFF_T(midpart) "%" midpart "ld"
 #    define EXPAT_FMT_SIZE_T(midpart) "%" midpart "lu"
+#  elif defined(EMSCRIPTEN) // 32bit mode Emscripten
+#    define EXPAT_FMT_PTRDIFF_T(midpart) "%" midpart "ld"
+#    define EXPAT_FMT_SIZE_T(midpart) "%" midpart "zu"
 #  else
 #    define EXPAT_FMT_PTRDIFF_T(midpart) "%" midpart "d"
 #    define EXPAT_FMT_SIZE_T(midpart) "%" midpart "u"
@@ -155,14 +159,20 @@ extern "C" {
 void _INTERNAL_trim_to_complete_utf8_characters(const char *from,
                                                const char **fromLimRef);

-#if XML_GE == 1
+#if defined(XML_GE) && XML_GE == 1
 unsigned long long testingAccountingGetCountBytesDirect(XML_Parser parser);
 unsigned long long testingAccountingGetCountBytesIndirect(XML_Parser parser);
 const char *unsignedCharToPrintable(unsigned char c);
 #endif

-extern XML_Bool g_reparseDeferralEnabledDefault; // written ONLY in runtests.c
-extern unsigned int g_parseAttempts;             // used for testing only
+extern
+#if ! defined(XML_TESTING)
+    const
+#endif
+    XML_Bool g_reparseDeferralEnabledDefault; // written ONLY in runtests.c
+#if defined(XML_TESTING)
+extern unsigned int g_bytesScanned; // used for testing only
+#endif

 #ifdef __cplusplus
 }
@@ -126,8 +126,7 @@
   | ((uint64_t)((p)[4]) << 32) | ((uint64_t)((p)[5]) << 40)                   \
   | ((uint64_t)((p)[6]) << 48) | ((uint64_t)((p)[7]) << 56))

-#define SIPHASH_INITIALIZER                                                    \
-  { 0, 0, 0, 0, {0}, 0, 0 }
+#define SIPHASH_INITIALIZER {0, 0, 0, 0, {0}, 0, 0}

 struct siphash {
  uint64_t v0, v1, v2, v3;
@@ -6,9 +6,10 @@
 #                      \___/_/\_\ .__/ \__,_|\__|
 #                               |_| XML parser
 #
-# Copyright (c) 2017-2023 Sebastian Pipping <[email protected]>
+# Copyright (c) 2017-2024 Sebastian Pipping <[email protected]>
 # Copyright (c) 2017-2022 Rhodri James <[email protected]>
 # Copyright (c) 2020      Jeffrey Walton <[email protected]>
+# Copyright (c) 2024      Dag-Erling Smørgrav <[email protected]>
 # Licensed under the MIT license:
 #
 # Permission is  hereby granted,  free of charge,  to any  person obtaining
@@ -32,7 +33,7 @@

 SUBDIRS = . benchmark

-AM_CPPFLAGS = @AM_CPPFLAGS@ -I$(srcdir)/../lib
+AM_CPPFLAGS = @AM_CPPFLAGS@ -I$(srcdir)/../lib -DXML_TESTING

 check_PROGRAMS = runtests runtests_cxx
 TESTS = runtests runtests_cxx
@@ -72,8 +73,8 @@ runtests_cxx_SOURCES = \
    runtests_cxx.cpp \
    structdata_cxx.cpp

-runtests_LDADD = ../lib/libexpatinternal.la
-runtests_cxx_LDADD = ../lib/libexpatinternal.la
+runtests_LDADD = ../lib/libtestpat.la
+runtests_cxx_LDADD = ../lib/libtestpat.la

 runtests_LDFLAGS = @AM_LDFLAGS@ @LIBM@
 runtests_cxx_LDFLAGS = @AM_LDFLAGS@ @LIBM@
@@ -92,7 +93,7 @@ EXTRA_DIST = \
    structdata.h \
    minicheck.h \
    memcheck.h \
-    README.txt \
+    README.md \
    udiffer.py \
    xmltest.log.expected \
    xmltest.sh
@@ -22,9 +22,10 @@
 #                      \___/_/\_\ .__/ \__,_|\__|
 #                               |_| XML parser
 #
-# Copyright (c) 2017-2023 Sebastian Pipping <[email protected]>
+# Copyright (c) 2017-2024 Sebastian Pipping <[email protected]>
 # Copyright (c) 2017-2022 Rhodri James <[email protected]>
 # Copyright (c) 2020      Jeffrey Walton <[email protected]>
+# Copyright (c) 2024      Dag-Erling Smørgrav <[email protected]>
 # Licensed under the MIT license:
 #
 # Permission is  hereby granted,  free of charge,  to any  person obtaining
@@ -151,7 +152,7 @@ am_runtests_OBJECTS = acc_tests.$(OBJEXT) alloc_tests.$(OBJEXT) \
 	nsalloc_tests.$(OBJEXT) runtests.$(OBJEXT) \
 	structdata.$(OBJEXT)
 runtests_OBJECTS = $(am_runtests_OBJECTS)
-runtests_DEPENDENCIES = ../lib/libexpatinternal.la
+runtests_DEPENDENCIES = ../lib/libtestpat.la
 AM_V_lt = $(am__v_lt_@AM_V@)
 am__v_lt_ = $(am__v_lt_@AM_DEFAULT_V@)
 am__v_lt_0 = --silent
@@ -168,7 +169,7 @@ am_runtests_cxx_OBJECTS = acc_tests_cxx.$(OBJEXT) \
 	ns_tests_cxx.$(OBJEXT) runtests_cxx.$(OBJEXT) \
 	structdata_cxx.$(OBJEXT)
 runtests_cxx_OBJECTS = $(am_runtests_cxx_OBJECTS)
-runtests_cxx_DEPENDENCIES = ../lib/libexpatinternal.la
+runtests_cxx_DEPENDENCIES = ../lib/libtestpat.la
 runtests_cxx_LINK = $(LIBTOOL) $(AM_V_lt) --tag=CXX $(AM_LIBTOOLFLAGS) \
 	$(LIBTOOLFLAGS) --mode=link $(CXXLD) $(AM_CXXFLAGS) \
 	$(CXXFLAGS) $(runtests_cxx_LDFLAGS) $(LDFLAGS) -o $@
@@ -485,7 +486,7 @@ TEST_LOG_COMPILE = $(TEST_LOG_COMPILER) $(AM_TEST_LOG_FLAGS) \
 DIST_SUBDIRS = $(SUBDIRS)
 am__DIST_COMMON = $(srcdir)/Makefile.in \
 	$(top_srcdir)/conftools/depcomp \
-	$(top_srcdir)/conftools/test-driver
+	$(top_srcdir)/conftools/test-driver README.md
 DISTFILES = $(DIST_COMMON) $(DIST_SOURCES) $(TEXINFOS) $(EXTRA_DIST)
 am__relativize = \
  dir0=`pwd`; \
@@ -515,7 +516,7 @@ am__relativize = \
 ACLOCAL = @ACLOCAL@
 AMTAR = @AMTAR@
 AM_CFLAGS = @AM_CFLAGS@
-AM_CPPFLAGS = @AM_CPPFLAGS@ -I$(srcdir)/../lib
+AM_CPPFLAGS = @AM_CPPFLAGS@ -I$(srcdir)/../lib -DXML_TESTING
 AM_CXXFLAGS = @AM_CXXFLAGS@
 AM_DEFAULT_VERBOSITY = @AM_DEFAULT_VERBOSITY@
 AM_LDFLAGS = @AM_LDFLAGS@
@@ -602,6 +603,7 @@ RANLIB = @RANLIB@
 SED = @SED@
 SET_MAKE = @SET_MAKE@
 SHELL = @SHELL@
+SIZEOF_VOID_P = @SIZEOF_VOID_P@
 SO_MAJOR = @SO_MAJOR@
 SO_MINOR = @SO_MINOR@
 SO_PATCH = @SO_PATCH@
@@ -615,7 +617,6 @@ ac_ct_AR = @ac_ct_AR@
 ac_ct_CC = @ac_ct_CC@
 ac_ct_CXX = @ac_ct_CXX@
 ac_ct_DUMPBIN = @ac_ct_DUMPBIN@
-ac_cv_sizeof_void_p = @ac_cv_sizeof_void_p@
 am__include = @am__include@
 am__leading_dot = @am__leading_dot@
 am__quote = @am__quote@
@@ -698,8 +699,8 @@ runtests_cxx_SOURCES = \
    runtests_cxx.cpp \
    structdata_cxx.cpp

-runtests_LDADD = ../lib/libexpatinternal.la
-runtests_cxx_LDADD = ../lib/libexpatinternal.la
+runtests_LDADD = ../lib/libtestpat.la
+runtests_cxx_LDADD = ../lib/libtestpat.la
 runtests_LDFLAGS = @AM_LDFLAGS@ @LIBM@
 runtests_cxx_LDFLAGS = @AM_LDFLAGS@ @LIBM@
 EXTRA_DIST = \
@@ -716,7 +717,7 @@ EXTRA_DIST = \
    structdata.h \
    minicheck.h \
    memcheck.h \
-    README.txt \
+    README.md \
    udiffer.py \
    xmltest.log.expected \
    xmltest.sh
@@ -0,0 +1,11 @@
+This directory contains the test suite for Expat.  The tests provide
+general unit testing and regression coverage.  The tests are not
+expected to be useful examples of Expat usage; see the
+[examples](../examples) directory for that.
+
+The Expat tests use a partial internal implementation of the
+[Check](https://libcheck.github.io/check/) unit testing framework for
+C.
+
+Expat must be built and, on some platforms, installed, before the
+tests can be run.
@@ -1,13 +0,0 @@
-This directory contains the (fledgling) test suite for Expat.  The
-tests provide general unit testing and regression coverage.  The tests
-are not expected to be useful examples of Expat usage; see the
-examples/ directory for that.
-
-The Expat tests use a partial internal implementation of the "Check"
-unit testing framework for C. More information on Check can be found at:
-
-        http://check.sourceforge.net/
-
-Expat must be built and, depending on platform, must be installed, before "make check" can be executed.
-
-This test suite can all change in a later version.
@@ -360,13 +360,16 @@ END_TEST
 START_TEST(test_helper_unsigned_char_to_printable) {
  // Smoke test
  unsigned char uc = 0;
-  for (; uc < (unsigned char)-1; uc++) {
+  for (;; uc++) {
    set_subtest("char %u", (unsigned)uc);
    const char *const printable = unsignedCharToPrintable(uc);
    if (printable == NULL)
      fail("unsignedCharToPrintable returned NULL");
    else if (strlen(printable) < (size_t)1)
      fail("unsignedCharToPrintable returned empty string");
+    if (uc == (unsigned char)-1) {
+      break;
+    }
  }

  // Two concrete samples
@@ -378,6 +381,63 @@ START_TEST(test_helper_unsigned_char_to_printable) {
    fail("unsignedCharToPrintable result mistaken");
 }
 END_TEST
+
+START_TEST(test_amplification_isolated_external_parser) {
+  // NOTE: Length 44 is precisely twice the length of "<!ENTITY a SYSTEM 'b'>"
+  // (22) that is used in function accountingGetCurrentAmplification in
+  // xmlparse.c.
+  //                  1.........1.........1.........1.........1..4 => 44
+  const char doc[] = "<!ENTITY % p1 '123456789_123456789_1234567'>";
+  const int docLen = (int)sizeof(doc) - 1;
+  const float maximumToleratedAmplification = 2.0f;
+
+  struct TestCase {
+    int offsetOfThreshold;
+    enum XML_Status expectedStatus;
+  };
+
+  struct TestCase cases[] = {
+      {-2, XML_STATUS_ERROR}, {-1, XML_STATUS_ERROR}, {0, XML_STATUS_ERROR},
+      {+1, XML_STATUS_OK},    {+2, XML_STATUS_OK},
+  };
+
+  for (size_t i = 0; i < sizeof(cases) / sizeof(cases[0]); i++) {
+    const int offsetOfThreshold = cases[i].offsetOfThreshold;
+    const enum XML_Status expectedStatus = cases[i].expectedStatus;
+    const unsigned long long activationThresholdBytes
+        = docLen + offsetOfThreshold;
+
+    set_subtest("offsetOfThreshold=%d, expectedStatus=%d", offsetOfThreshold,
+                expectedStatus);
+
+    XML_Parser parser = XML_ParserCreate(NULL);
+    assert_true(parser != NULL);
+
+    assert_true(XML_SetBillionLaughsAttackProtectionMaximumAmplification(
+                    parser, maximumToleratedAmplification)
+                == XML_TRUE);
+    assert_true(XML_SetBillionLaughsAttackProtectionActivationThreshold(
+                    parser, activationThresholdBytes)
+                == XML_TRUE);
+
+    XML_Parser ext_parser = XML_ExternalEntityParserCreate(parser, NULL, NULL);
+    assert_true(ext_parser != NULL);
+
+    const enum XML_Status actualStatus
+        = _XML_Parse_SINGLE_BYTES(ext_parser, doc, docLen, XML_TRUE);
+
+    assert_true(actualStatus == expectedStatus);
+    if (actualStatus != XML_STATUS_OK) {
+      assert_true(XML_GetErrorCode(ext_parser)
+                  == XML_ERROR_AMPLIFICATION_LIMIT_BREACH);
+    }
+
+    XML_ParserFree(ext_parser);
+    XML_ParserFree(parser);
+  }
+}
+END_TEST
+
 #endif // XML_GE == 1

 void
@@ -390,6 +450,8 @@ make_accounting_test_case(Suite *s) {
  tcase_add_test(tc_accounting, test_accounting_precision);
  tcase_add_test(tc_accounting, test_billion_laughs_attack_protection_api);
  tcase_add_test(tc_accounting, test_helper_unsigned_char_to_printable);
+  tcase_add_test__ifdef_xml_dtd(tc_accounting,
+                                test_amplification_isolated_external_parser);
 #else
  UNUSED_P(s);
 #endif /* XML_GE == 1 */
@@ -19,6 +19,7 @@
   Copyright (c) 2020      Tim Gates <tim.gates@iress.com>
   Copyright (c) 2021      Donghee Na <donghee.na@python.org>
   Copyright (c) 2023      Sony Corporation / Snild Dolkow <snild@sony.com>
+   Copyright (c) 2025      Berkay Eren Ürün <berkay.ueruen@siemens.com>
   Licensed under the MIT license:

   Permission is  hereby granted,  free of charge,  to any  person obtaining
@@ -450,6 +451,31 @@ START_TEST(test_alloc_internal_entity) {
 }
 END_TEST

+START_TEST(test_alloc_parameter_entity) {
+  const char *text = "<!DOCTYPE foo ["
+                     "<!ENTITY % param1 \"<!ENTITY internal 'some_text'>\">"
+                     "%param1;"
+                     "]> <foo>&internal;content</foo>";
+  int i;
+  const int alloc_test_max_repeats = 30;
+
+  for (i = 0; i < alloc_test_max_repeats; i++) {
+    g_allocation_count = i;
+    XML_SetParamEntityParsing(g_parser, XML_PARAM_ENTITY_PARSING_ALWAYS);
+    if (_XML_Parse_SINGLE_BYTES(g_parser, text, (int)strlen(text), XML_TRUE)
+        != XML_STATUS_ERROR)
+      break;
+    alloc_teardown();
+    alloc_setup();
+  }
+  g_allocation_count = -1;
+  if (i == 0)
+    fail("Parameter entity processed despite duff allocator");
+  if (i == alloc_test_max_repeats)
+    fail("Parameter entity not processed at max allocation count");
+}
+END_TEST
+
 /* Test the robustness against allocation failure of element handling
 * Based on test_dtd_default_handling().
 */
@@ -2079,6 +2105,7 @@ make_alloc_test_case(Suite *s) {
  tcase_add_test__ifdef_xml_dtd(tc_alloc, test_alloc_external_entity);
  tcase_add_test__ifdef_xml_dtd(tc_alloc, test_alloc_ext_entity_set_encoding);
  tcase_add_test__ifdef_xml_dtd(tc_alloc, test_alloc_internal_entity);
+  tcase_add_test__ifdef_xml_dtd(tc_alloc, test_alloc_parameter_entity);
  tcase_add_test__ifdef_xml_dtd(tc_alloc, test_alloc_dtd_default_handling);
  tcase_add_test(tc_alloc, test_alloc_explicit_encoding);
  tcase_add_test(tc_alloc, test_alloc_set_base);
@@ -10,7 +10,7 @@
   Copyright (c) 2003      Greg Stein <gstein@users.sourceforge.net>
   Copyright (c) 2005-2007 Steven Solie <steven@solie.ca>
   Copyright (c) 2005-2012 Karl Waclawek <karl@waclawek.net>
-   Copyright (c) 2016-2024 Sebastian Pipping <sebastian@pipping.org>
+   Copyright (c) 2016-2025 Sebastian Pipping <sebastian@pipping.org>
   Copyright (c) 2017-2022 Rhodri James <rhodri@wildebeest.org.uk>
   Copyright (c) 2017      Joe Orton <jorton@redhat.com>
   Copyright (c) 2017      José Gutiérrez de la Concha <jose@zeroc.com>
@@ -19,6 +19,7 @@
   Copyright (c) 2020      Tim Gates <tim.gates@iress.com>
   Copyright (c) 2021      Donghee Na <donghee.na@python.org>
   Copyright (c) 2023-2024 Sony Corporation / Snild Dolkow <snild@sony.com>
+   Copyright (c) 2024-2025 Berkay Eren Ürün <berkay.ueruen@siemens.com>
   Licensed under the MIT license:

   Permission is  hereby granted,  free of charge,  to any  person obtaining
@@ -1191,6 +1192,22 @@ START_TEST(test_not_standalone_handler_accept) {
 }
 END_TEST

+START_TEST(test_entity_start_tag_level_greater_than_one) {
+  const char *const text = "<!DOCTYPE t1 [\n"
+                           "  <!ENTITY e1 'hello'>\n"
+                           "]>\n"
+                           "<t1>\n"
+                           "  <t2>&e1;</t2>\n"
+                           "</t1>\n";
+
+  XML_Parser parser = XML_ParserCreate(NULL);
+  assert_true(_XML_Parse_SINGLE_BYTES(parser, text, (int)strlen(text),
+                                      /*isFinal*/ XML_TRUE)
+              == XML_STATUS_OK);
+  XML_ParserFree(parser);
+}
+END_TEST
+
 START_TEST(test_wfc_no_recursive_entity_refs) {
  const char *text = "<!DOCTYPE doc [\n"
                     "  <!ENTITY entity '&#38;entity;'>\n"
@@ -1202,6 +1219,136 @@ START_TEST(test_wfc_no_recursive_entity_refs) {
 }
 END_TEST

+START_TEST(test_no_indirectly_recursive_entity_refs) {
+  struct TestCase {
+    const char *doc;
+    bool usesParameterEntities;
+  };
+
+  const struct TestCase cases[] = {
+      // general entity + character data
+      {"<!DOCTYPE a [\n"
+       "  <!ENTITY e1 '&e2;'>\n"
+       "  <!ENTITY e2 '&e1;'>\n"
+       "]><a>&e2;</a>\n",
+       false},
+
+      // general entity + attribute value
+      {"<!DOCTYPE a [\n"
+       "  <!ENTITY e1 '&e2;'>\n"
+       "  <!ENTITY e2 '&e1;'>\n"
+       "]><a k1='&e2;' />\n",
+       false},
+
+      // parameter entity
+      {"<!DOCTYPE doc [\n"
+       "  <!ENTITY % p1 '&#37;p2;'>\n"
+       "  <!ENTITY % p2 '&#37;p1;'>\n"
+       "  <!ENTITY % define_g \"<!ENTITY g '&#37;p2;'>\">\n"
+       "  %define_g;\n"
+       "]>\n"
+       "<doc/>\n",
+       true},
+  };
+  const XML_Bool reset_or_not[] = {XML_TRUE, XML_FALSE};
+
+  for (size_t i = 0; i < sizeof(cases) / sizeof(cases[0]); i++) {
+    for (size_t j = 0; j < sizeof(reset_or_not) / sizeof(reset_or_not[0]);
+         j++) {
+      const XML_Bool reset_wanted = reset_or_not[j];
+      const char *const doc = cases[i].doc;
+      const bool usesParameterEntities = cases[i].usesParameterEntities;
+
+      set_subtest("[%i,reset=%i] %s", (int)i, (int)j, doc);
+
+#ifdef XML_DTD // both GE and DTD
+      const bool rejection_expected = true;
+#elif XML_GE == 1 // GE but not DTD
+      const bool rejection_expected = ! usesParameterEntities;
+#else             // neither DTD nor GE
+      const bool rejection_expected = false;
+#endif
+
+      XML_Parser parser = XML_ParserCreate(NULL);
+
+#ifdef XML_DTD
+      if (usesParameterEntities) {
+        assert_true(
+            XML_SetParamEntityParsing(parser, XML_PARAM_ENTITY_PARSING_ALWAYS)
+            == 1);
+      }
+#else
+      UNUSED_P(usesParameterEntities);
+#endif // XML_DTD
+
+      const enum XML_Status status
+          = _XML_Parse_SINGLE_BYTES(parser, doc, (int)strlen(doc),
+                                    /*isFinal*/ XML_TRUE);
+
+      if (rejection_expected) {
+        assert_true(status == XML_STATUS_ERROR);
+        assert_true(XML_GetErrorCode(parser) == XML_ERROR_RECURSIVE_ENTITY_REF);
+      } else {
+        assert_true(status == XML_STATUS_OK);
+      }
+
+      if (reset_wanted) {
+        // This covers free'ing of (eventually) all three open entity lists by
+        // XML_ParserReset.
+        XML_ParserReset(parser, NULL);
+      }
+
+      // This covers free'ing of (eventually) all three open entity lists by
+      // XML_ParserFree (unless XML_ParserReset has already done that above).
+      XML_ParserFree(parser);
+    }
+  }
+}
+END_TEST
+
+START_TEST(test_recursive_external_parameter_entity_2) {
+  struct TestCase {
+    const char *doc;
+    enum XML_Status expectedStatus;
+  };
+
+  struct TestCase cases[] = {
+      {"<!ENTITY % p1 '%p1;'>", XML_STATUS_ERROR},
+      {"<!ENTITY % p1 '%p1;'>"
+       "<!ENTITY % p1 'first declaration wins'>",
+       XML_STATUS_ERROR},
+      {"<!ENTITY % p1 'first declaration wins'>"
+       "<!ENTITY % p1 '%p1;'>",
+       XML_STATUS_OK},
+      {"<!ENTITY % p1 '&#37;p1;'>", XML_STATUS_OK},
+  };
+
+  for (size_t i = 0; i < sizeof(cases) / sizeof(cases[0]); i++) {
+    const char *const doc = cases[i].doc;
+    const enum XML_Status expectedStatus = cases[i].expectedStatus;
+    set_subtest("%s", doc);
+
+    XML_Parser parser = XML_ParserCreate(NULL);
+    assert_true(parser != NULL);
+
+    XML_Parser ext_parser = XML_ExternalEntityParserCreate(parser, NULL, NULL);
+    assert_true(ext_parser != NULL);
+
+    const enum XML_Status actualStatus
+        = _XML_Parse_SINGLE_BYTES(ext_parser, doc, (int)strlen(doc), XML_TRUE);
+
+    assert_true(actualStatus == expectedStatus);
+    if (actualStatus != XML_STATUS_OK) {
+      assert_true(XML_GetErrorCode(ext_parser)
+                  == XML_ERROR_RECURSIVE_ENTITY_REF);
+    }
+
+    XML_ParserFree(ext_parser);
+    XML_ParserFree(parser);
+  }
+}
+END_TEST
+
 /* Test incomplete external entities are faulted */
 START_TEST(test_ext_entity_invalid_parse) {
  const char *text = "<!DOCTYPE doc [\n"
@@ -1374,7 +1521,9 @@ START_TEST(test_suspend_parser_between_char_data_calls) {

  XML_SetCharacterDataHandler(g_parser, clearing_aborting_character_handler);
  g_resumable = XML_TRUE;
-  if (_XML_Parse_SINGLE_BYTES(g_parser, text, (int)strlen(text), XML_TRUE)
+  // can't use SINGLE_BYTES here, because it'll return early on suspension, and
+  // we won't know exactly how much input we actually managed to give Expat.
+  if (XML_Parse(g_parser, text, (int)strlen(text), XML_TRUE)
      != XML_STATUS_SUSPENDED)
    xml_failure(g_parser);
  if (XML_GetErrorCode(g_parser) != XML_ERROR_NONE)
@@ -1403,7 +1552,9 @@ START_TEST(test_repeated_stop_parser_between_char_data_calls) {
  XML_SetCharacterDataHandler(g_parser, parser_stop_character_handler);
  g_resumable = XML_TRUE;
  g_abortable = XML_FALSE;
-  if (_XML_Parse_SINGLE_BYTES(g_parser, text, (int)strlen(text), XML_TRUE)
+  // can't use SINGLE_BYTES here, because it'll return early on suspension, and
+  // we won't know exactly how much input we actually managed to give Expat.
+  if (XML_Parse(g_parser, text, (int)strlen(text), XML_TRUE)
      != XML_STATUS_SUSPENDED)
    fail("Failed to double-suspend parser");

@@ -1787,12 +1938,19 @@ END_TEST

 /* Test suspending the parser in cdata handler */
 START_TEST(test_suspend_parser_between_cdata_calls) {
+  if (g_chunkSize != 0) {
+    // this test does not use SINGLE_BYTES, because of suspension
+    return;
+  }
+
  const char *text = long_cdata_text;
  enum XML_Status result;

  XML_SetCharacterDataHandler(g_parser, clearing_aborting_character_handler);
  g_resumable = XML_TRUE;
-  result = _XML_Parse_SINGLE_BYTES(g_parser, text, (int)strlen(text), XML_TRUE);
+  // can't use SINGLE_BYTES here, because it'll return early on suspension, and
+  // we won't know exactly how much input we actually managed to give Expat.
+  result = XML_Parse(g_parser, text, (int)strlen(text), XML_TRUE);
  if (result != XML_STATUS_SUSPENDED) {
    if (result == XML_STATUS_ERROR)
      xml_failure(g_parser);
@@ -2314,11 +2472,20 @@ START_TEST(test_attributes) {
  info[0].attributes = doc_info;
  info[1].attributes = tag_info;

-  XML_SetStartElementHandler(g_parser, counting_start_element_handler);
-  XML_SetUserData(g_parser, info);
-  if (_XML_Parse_SINGLE_BYTES(g_parser, text, (int)strlen(text), XML_TRUE)
+  XML_Parser parser = XML_ParserCreate(NULL);
+  assert_true(parser != NULL);
+  ParserAndElementInfo parserAndElementInfos = {
+      parser,
+      info,
+  };
+
+  XML_SetStartElementHandler(parser, counting_start_element_handler);
+  XML_SetUserData(parser, &parserAndElementInfos);
+  if (_XML_Parse_SINGLE_BYTES(parser, text, (int)strlen(text), XML_TRUE)
      == XML_STATUS_ERROR)
-    xml_failure(g_parser);
+    xml_failure(parser);
+
+  XML_ParserFree(parser);
 }
 END_TEST

@@ -2326,6 +2493,11 @@ END_TEST
 * entity.  Exercises some obscure code in XML_ParserReset().
 */
 START_TEST(test_reset_in_entity) {
+  if (g_chunkSize != 0) {
+    // this test does not use SINGLE_BYTES, because of suspension
+    return;
+  }
+
  const char *text = "<!DOCTYPE doc [\n"
                     "<!ENTITY wombat 'wom'>\n"
                     "<!ENTITY entity 'hi &wom; there'>\n"
@@ -2335,7 +2507,9 @@ START_TEST(test_reset_in_entity) {

  g_resumable = XML_TRUE;
  XML_SetCharacterDataHandler(g_parser, clearing_aborting_character_handler);
-  if (_XML_Parse_SINGLE_BYTES(g_parser, text, (int)strlen(text), XML_TRUE)
+  // can't use SINGLE_BYTES here, because it'll return early on suspension, and
+  // we won't know exactly how much input we actually managed to give Expat.
+  if (XML_Parse(g_parser, text, (int)strlen(text), XML_TRUE)
      == XML_STATUS_ERROR)
    xml_failure(g_parser);
  XML_GetParsingStatus(g_parser, &status);
@@ -2761,6 +2935,61 @@ START_TEST(test_empty_parse) {
 }
 END_TEST

+/* Test XML_Parse for len < 0 */
+START_TEST(test_negative_len_parse) {
+  const char *const doc = "<root/>";
+  for (int isFinal = 0; isFinal < 2; isFinal++) {
+    set_subtest("isFinal=%d", isFinal);
+
+    XML_Parser parser = XML_ParserCreate(NULL);
+
+    if (XML_GetErrorCode(parser) != XML_ERROR_NONE)
+      fail("There was not supposed to be any initial parse error.");
+
+    const enum XML_Status status = XML_Parse(parser, doc, -1, isFinal);
+
+    if (status != XML_STATUS_ERROR)
+      fail("Negative len was expected to fail the parse but did not.");
+
+    if (XML_GetErrorCode(parser) != XML_ERROR_INVALID_ARGUMENT)
+      fail("Parse error does not match XML_ERROR_INVALID_ARGUMENT.");
+
+    XML_ParserFree(parser);
+  }
+}
+END_TEST
+
+/* Test XML_ParseBuffer for len < 0 */
+START_TEST(test_negative_len_parse_buffer) {
+  const char *const doc = "<root/>";
+  for (int isFinal = 0; isFinal < 2; isFinal++) {
+    set_subtest("isFinal=%d", isFinal);
+
+    XML_Parser parser = XML_ParserCreate(NULL);
+
+    if (XML_GetErrorCode(parser) != XML_ERROR_NONE)
+      fail("There was not supposed to be any initial parse error.");
+
+    void *const buffer = XML_GetBuffer(parser, (int)strlen(doc));
+
+    if (buffer == NULL)
+      fail("XML_GetBuffer failed.");
+
+    memcpy(buffer, doc, strlen(doc));
+
+    const enum XML_Status status = XML_ParseBuffer(parser, -1, isFinal);
+
+    if (status != XML_STATUS_ERROR)
+      fail("Negative len was expected to fail the parse but did not.");
+
+    if (XML_GetErrorCode(parser) != XML_ERROR_INVALID_ARGUMENT)
+      fail("Parse error does not match XML_ERROR_INVALID_ARGUMENT.");
+
+    XML_ParserFree(parser);
+  }
+}
+END_TEST
+
 /* Test odd corners of the XML_GetBuffer interface */
 static enum XML_Status
 get_feature(enum XML_FeatureEnum feature_id, long *presult) {
@@ -3527,7 +3756,9 @@ START_TEST(test_suspend_xdecl) {
  XML_SetXmlDeclHandler(g_parser, entity_suspending_xdecl_handler);
  XML_SetUserData(g_parser, g_parser);
  g_resumable = XML_TRUE;
-  if (_XML_Parse_SINGLE_BYTES(g_parser, text, (int)strlen(text), XML_TRUE)
+  // can't use SINGLE_BYTES here, because it'll return early on suspension, and
+  // we won't know exactly how much input we actually managed to give Expat.
+  if (XML_Parse(g_parser, text, (int)strlen(text), XML_TRUE)
      != XML_STATUS_SUSPENDED)
    xml_failure(g_parser);
  if (XML_GetErrorCode(g_parser) != XML_ERROR_NONE)
@@ -3723,13 +3954,20 @@ END_TEST

 /* Test syntax error is caught at parse resumption */
 START_TEST(test_resume_entity_with_syntax_error) {
+  if (g_chunkSize != 0) {
+    // this test does not use SINGLE_BYTES, because of suspension
+    return;
+  }
+
  const char *text = "<!DOCTYPE doc [\n"
                     "<!ENTITY foo '<suspend>Hi</wombat>'>\n"
                     "]>\n"
                     "<doc>&foo;</doc>\n";

  XML_SetStartElementHandler(g_parser, start_element_suspender);
-  if (_XML_Parse_SINGLE_BYTES(g_parser, text, (int)strlen(text), XML_TRUE)
+  // can't use SINGLE_BYTES here, because it'll return early on suspension, and
+  // we won't know exactly how much input we actually managed to give Expat.
+  if (XML_Parse(g_parser, text, (int)strlen(text), XML_TRUE)
      != XML_STATUS_SUSPENDED)
    xml_failure(g_parser);
  if (XML_ResumeParser(g_parser) != XML_STATUS_ERROR)
@@ -3853,7 +4091,7 @@ START_TEST(test_skipped_null_loaded_ext_entity) {
      = {"<!ENTITY % pe1 SYSTEM 'http://example.org/two.ent'>\n"
         "<!ENTITY % pe2 '%pe1;'>\n"
         "%pe2;\n",
-         external_entity_null_loader};
+         external_entity_null_loader, NULL};

  XML_SetUserData(g_parser, &test_data);
  XML_SetParamEntityParsing(g_parser, XML_PARAM_ENTITY_PARSING_ALWAYS);
@@ -3871,7 +4109,7 @@ START_TEST(test_skipped_unloaded_ext_entity) {
      = {"<!ENTITY % pe1 SYSTEM 'http://example.org/two.ent'>\n"
         "<!ENTITY % pe2 '%pe1;'>\n"
         "%pe2;\n",
-         NULL};
+         NULL, NULL};

  XML_SetUserData(g_parser, &test_data);
  XML_SetParamEntityParsing(g_parser, XML_PARAM_ENTITY_PARSING_ALWAYS);
@@ -5171,6 +5409,151 @@ START_TEST(test_pool_integrity_with_unfinished_attr) {
 }
 END_TEST

+/* Test a possible early return location in internalEntityProcessor */
+START_TEST(test_entity_ref_no_elements) {
+  const char *const text = "<!DOCTYPE foo [\n"
+                           "<!ENTITY e1 \"test\">\n"
+                           "]> <foo>&e1;"; // intentionally missing newline
+
+  XML_Parser parser = XML_ParserCreate(NULL);
+  assert_true(_XML_Parse_SINGLE_BYTES(parser, text, (int)strlen(text), XML_TRUE)
+              == XML_STATUS_ERROR);
+  assert_true(XML_GetErrorCode(parser) == XML_ERROR_NO_ELEMENTS);
+  XML_ParserFree(parser);
+}
+END_TEST
+
+/* Tests if chained entity references lead to unbounded recursion */
+START_TEST(test_deep_nested_entity) {
+  const size_t N_LINES = 60000;
+  const size_t SIZE_PER_LINE = 50;
+
+  char *const text = (char *)malloc((N_LINES + 4) * SIZE_PER_LINE);
+  if (text == NULL) {
+    fail("malloc failed");
+  }
+
+  char *textPtr = text;
+
+  // Create the XML
+  textPtr += snprintf(textPtr, SIZE_PER_LINE,
+                      "<!DOCTYPE foo [\n"
+                      "	<!ENTITY s0 'deepText'>\n");
+
+  for (size_t i = 1; i < N_LINES; ++i) {
+    textPtr += snprintf(textPtr, SIZE_PER_LINE, "  <!ENTITY s%lu '&s%lu;'>\n",
+                        (long unsigned)i, (long unsigned)(i - 1));
+  }
+
+  snprintf(textPtr, SIZE_PER_LINE, "]> <foo>&s%lu;</foo>\n",
+           (long unsigned)(N_LINES - 1));
+
+  const XML_Char *const expected = XCS("deepText");
+
+  CharData storage;
+  CharData_Init(&storage);
+
+  XML_Parser parser = XML_ParserCreate(NULL);
+
+  XML_SetCharacterDataHandler(parser, accumulate_characters);
+  XML_SetUserData(parser, &storage);
+
+  if (_XML_Parse_SINGLE_BYTES(parser, text, (int)strlen(text), XML_TRUE)
+      == XML_STATUS_ERROR)
+    xml_failure(parser);
+
+  CharData_CheckXMLChars(&storage, expected);
+  XML_ParserFree(parser);
+  free(text);
+}
+END_TEST
+
+/* Tests if chained entity references in attributes
+lead to unbounded recursion */
+START_TEST(test_deep_nested_attribute_entity) {
+  const size_t N_LINES = 60000;
+  const size_t SIZE_PER_LINE = 100;
+
+  char *const text = (char *)malloc((N_LINES + 4) * SIZE_PER_LINE);
+  if (text == NULL) {
+    fail("malloc failed");
+  }
+
+  char *textPtr = text;
+
+  // Create the XML
+  textPtr += snprintf(textPtr, SIZE_PER_LINE,
+                      "<!DOCTYPE foo [\n"
+                      "	<!ENTITY s0 'deepText'>\n");
+
+  for (size_t i = 1; i < N_LINES; ++i) {
+    textPtr += snprintf(textPtr, SIZE_PER_LINE, "  <!ENTITY s%lu '&s%lu;'>\n",
+                        (long unsigned)i, (long unsigned)(i - 1));
+  }
+
+  snprintf(textPtr, SIZE_PER_LINE, "]> <foo name='&s%lu;'>mainText</foo>\n",
+           (long unsigned)(N_LINES - 1));
+
+  AttrInfo doc_info[] = {{XCS("name"), XCS("deepText")}, {NULL, NULL}};
+  ElementInfo info[] = {{XCS("foo"), 1, NULL, NULL}, {NULL, 0, NULL, NULL}};
+  info[0].attributes = doc_info;
+
+  XML_Parser parser = XML_ParserCreate(NULL);
+  ParserAndElementInfo parserPlusElemenInfo = {parser, info};
+
+  XML_SetStartElementHandler(parser, counting_start_element_handler);
+  XML_SetUserData(parser, &parserPlusElemenInfo);
+
+  if (_XML_Parse_SINGLE_BYTES(parser, text, (int)strlen(text), XML_TRUE)
+      == XML_STATUS_ERROR)
+    xml_failure(parser);
+
+  XML_ParserFree(parser);
+  free(text);
+}
+END_TEST
+
+START_TEST(test_deep_nested_entity_delayed_interpretation) {
+  const size_t N_LINES = 70000;
+  const size_t SIZE_PER_LINE = 100;
+
+  char *const text = (char *)malloc((N_LINES + 4) * SIZE_PER_LINE);
+  if (text == NULL) {
+    fail("malloc failed");
+  }
+
+  char *textPtr = text;
+
+  // Create the XML
+  textPtr += snprintf(textPtr, SIZE_PER_LINE,
+                      "<!DOCTYPE foo [\n"
+                      "	<!ENTITY %% s0 'deepText'>\n");
+
+  for (size_t i = 1; i < N_LINES; ++i) {
+    textPtr += snprintf(textPtr, SIZE_PER_LINE,
+                        "  <!ENTITY %% s%lu '&#37;s%lu;'>\n", (long unsigned)i,
+                        (long unsigned)(i - 1));
+  }
+
+  snprintf(textPtr, SIZE_PER_LINE,
+           "  <!ENTITY %% define_g \"<!ENTITY g '&#37;s%lu;'>\">\n"
+           "  %%define_g;\n"
+           "]>\n"
+           "<foo/>\n",
+           (long unsigned)(N_LINES - 1));
+
+  XML_Parser parser = XML_ParserCreate(NULL);
+
+  XML_SetParamEntityParsing(parser, XML_PARAM_ENTITY_PARSING_ALWAYS);
+  if (_XML_Parse_SINGLE_BYTES(parser, text, (int)strlen(text), XML_TRUE)
+      == XML_STATUS_ERROR)
+    xml_failure(parser);
+
+  XML_ParserFree(parser);
+  free(text);
+}
+END_TEST
+
 START_TEST(test_nested_entity_suspend) {
  const char *const text = "<!DOCTYPE a [\n"
                           "  <!ENTITY e1 '<!--e1-->'>\n"
@@ -5201,14 +5584,37 @@ START_TEST(test_nested_entity_suspend) {
 }
 END_TEST

+START_TEST(test_nested_entity_suspend_2) {
+  const char *const text = "<!DOCTYPE doc [\n"
+                           "  <!ENTITY ge1 'head1Ztail1'>\n"
+                           "  <!ENTITY ge2 'head2&ge1;tail2'>\n"
+                           "  <!ENTITY ge3 'head3&ge2;tail3'>\n"
+                           "]>\n"
+                           "<doc>&ge3;</doc>";
+  const XML_Char *const expected = XCS("head3") XCS("head2") XCS("head1")
+      XCS("Z") XCS("tail1") XCS("tail2") XCS("tail3");
+  CharData storage;
+  CharData_Init(&storage);
+  XML_Parser parser = XML_ParserCreate(NULL);
+  ParserPlusStorage parserPlusStorage = {parser, &storage};
+
+  XML_SetCharacterDataHandler(parser, accumulate_char_data_and_suspend);
+  XML_SetUserData(parser, &parserPlusStorage);
+
+  enum XML_Status status = XML_Parse(parser, text, (int)strlen(text), XML_TRUE);
+  while (status == XML_STATUS_SUSPENDED) {
+    status = XML_ResumeParser(parser);
+  }
+  if (status != XML_STATUS_OK)
+    xml_failure(parser);
+
+  CharData_CheckXMLChars(&storage, expected);
+  XML_ParserFree(parser);
+}
+END_TEST
+
 /* Regression test for quadratic parsing on large tokens */
-START_TEST(test_big_tokens_take_linear_time) {
-  const char *const too_slow_failure_message
-      = "Compared to the baseline runtime of the first test, this test has a "
-        "slowdown of more than <max_slowdown>. "
-        "Please keep increasing the value by 1 until it reliably passes the "
-        "test on your hardware and open a bug sharing that number with us. "
-        "Thanks in advance!";
+START_TEST(test_big_tokens_scale_linearly) {
  const struct {
    const char *pre;
    const char *post;
@@ -5220,65 +5626,57 @@ START_TEST(test_big_tokens_take_linear_time) {
      {"<e><", "/></e>"},                   // big elem name, used to be O(N²)
  };
  const int num_cases = sizeof(text) / sizeof(text[0]);
-  // For the test we need a <max_slowdown> value that is:
-  // (1) big enough that the test passes reliably (avoiding flaky tests), and
-  // (2) small enough that the test actually catches regressions.
-  const int max_slowdown = 15;
  char aaaaaa[4096];
  const int fillsize = (int)sizeof(aaaaaa);
  const int fillcount = 100;
+  const unsigned approx_bytes = fillsize * fillcount; // ignore pre/post.
+  const unsigned max_factor = 4;
+  const unsigned max_scanned = max_factor * approx_bytes;

  memset(aaaaaa, 'a', fillsize);

  if (! g_reparseDeferralEnabledDefault) {
    return; // heuristic is disabled; we would get O(n^2) and fail.
  }
-#if ! defined(__linux__)
-  if (CLOCKS_PER_SEC < 100000) {
-    // Skip this test if clock() doesn't have reasonably good resolution.
-    // This workaround is primarily targeting Windows and FreeBSD, since
-    // XSI requires the value to be 1.000.000 (10x the condition here), and
-    // we want to be very sure that at least one platform in CI can catch
-    // regressions (through a failing test).
-    return;
-  }
-#endif

-  clock_t baseline = 0;
  for (int i = 0; i < num_cases; ++i) {
    XML_Parser parser = XML_ParserCreate(NULL);
    assert_true(parser != NULL);
    enum XML_Status status;
-    set_subtest("max_slowdown=%d text=\"%saaaaaa%s\"", max_slowdown,
-                text[i].pre, text[i].post);
-    const clock_t start = clock();
+    set_subtest("text=\"%saaaaaa%s\"", text[i].pre, text[i].post);

    // parse the start text
+    g_bytesScanned = 0;
    status = _XML_Parse_SINGLE_BYTES(parser, text[i].pre,
                                     (int)strlen(text[i].pre), XML_FALSE);
    if (status != XML_STATUS_OK) {
      xml_failure(parser);
    }
+
    // parse lots of 'a', failing the test early if it takes too long
+    unsigned past_max_count = 0;
    for (int f = 0; f < fillcount; ++f) {
      status = _XML_Parse_SINGLE_BYTES(parser, aaaaaa, fillsize, XML_FALSE);
      if (status != XML_STATUS_OK) {
        xml_failure(parser);
      }
-      // i == 0 means we're still calculating the baseline value
-      if (i > 0) {
-        const clock_t now = clock();
-        const clock_t clocks_so_far = now - start;
-        const int slowdown = clocks_so_far / baseline;
-        if (slowdown >= max_slowdown) {
-          fprintf(
-              stderr,
-              "fill#%d: clocks_so_far=%d baseline=%d slowdown=%d max_slowdown=%d\n",
-              f, (int)clocks_so_far, (int)baseline, slowdown, max_slowdown);
-          fail(too_slow_failure_message);
-        }
+      if (g_bytesScanned > max_scanned) {
+        // We're not done, and have already passed the limit -- the test will
+        // definitely fail. This block allows us to save time by failing early.
+        const unsigned pushed
+            = (unsigned)strlen(text[i].pre) + (f + 1) * fillsize;
+        fprintf(
+            stderr,
+            "after %d/%d loops: pushed=%u scanned=%u (factor ~%.2f) max_scanned: %u (factor ~%u)\n",
+            f + 1, fillcount, pushed, g_bytesScanned,
+            g_bytesScanned / (double)pushed, max_scanned, max_factor);
+        past_max_count++;
+        // We are failing, but allow a few log prints first. If we don't reach
+        // a count of five, the test will fail after the loop instead.
+        assert_true(past_max_count < 5);
      }
    }
+
    // parse the end text
    status = _XML_Parse_SINGLE_BYTES(parser, text[i].post,
                                     (int)strlen(text[i].post), XML_TRUE);
@@ -5286,18 +5684,14 @@ START_TEST(test_big_tokens_take_linear_time) {
      xml_failure(parser);
    }

-    // how long did it take in total?
-    const clock_t end = clock();
-    const clock_t taken = end - start;
-    if (i == 0) {
-      assert_true(taken > 0); // just to make sure we don't div-by-0 later
-      baseline = taken;
-    }
-    const int slowdown = taken / baseline;
-    if (slowdown >= max_slowdown) {
-      fprintf(stderr, "taken=%d baseline=%d slowdown=%d max_slowdown=%d\n",
-              (int)taken, (int)baseline, slowdown, max_slowdown);
-      fail(too_slow_failure_message);
+    assert_true(g_bytesScanned > approx_bytes); // or the counter isn't working
+    if (g_bytesScanned > max_scanned) {
+      fprintf(
+          stderr,
+          "after all input: scanned=%u (factor ~%.2f) max_scanned: %u (factor ~%u)\n",
+          g_bytesScanned, g_bytesScanned / (double)approx_bytes, max_scanned,
+          max_factor);
+      fail("scanned too many bytes");
    }

    XML_ParserFree(parser);
@@ -5774,19 +6168,17 @@ START_TEST(test_varying_buffer_fills) {
                fillsize[2], fillsize[3]);
    XML_Parser parser = XML_ParserCreate(NULL);
    assert_true(parser != NULL);
-    g_parseAttempts = 0;

    CharData storage;
    CharData_Init(&storage);
    XML_SetUserData(parser, &storage);
    XML_SetStartElementHandler(parser, start_element_event_handler);

+    g_bytesScanned = 0;
    int worstcase_bytes = 0; // sum of (buffered bytes at each XML_Parse call)
-    int scanned_bytes = 0;   // sum of (buffered bytes at each actual parse)
    int offset = 0;
    while (*fillsize >= 0) {
      assert_true(offset + *fillsize <= document_length); // or test is invalid
-      const unsigned attempts_before = g_parseAttempts;
      const enum XML_Status status
          = XML_Parse(parser, &document[offset], *fillsize, XML_FALSE);
      if (status != XML_STATUS_OK) {
@@ -5796,28 +6188,20 @@ START_TEST(test_varying_buffer_fills) {
      fillsize++;
      assert_true(offset <= INT_MAX - worstcase_bytes); // avoid overflow
      worstcase_bytes += offset; // we might've tried to parse all pending bytes
-      if (g_parseAttempts != attempts_before) {
-        assert_true(g_parseAttempts == attempts_before + 1); // max 1/XML_Parse
-        assert_true(offset <= INT_MAX - scanned_bytes);      // avoid overflow
-        scanned_bytes += offset; // we *did* try to parse all pending bytes
-      }
    }
    assert_true(storage.count == 1); // the big token should've been parsed
-    assert_true(scanned_bytes > 0);  // test-the-test: does our counter work?
+    assert_true(g_bytesScanned > 0); // test-the-test: does our counter work?
    if (g_reparseDeferralEnabledDefault) {
      // heuristic is enabled; some XML_Parse calls may have deferred reparsing
-      const int max_bytes_scanned = -*fillsize;
-      if (scanned_bytes > max_bytes_scanned) {
+      const unsigned max_bytes_scanned = -*fillsize;
+      if (g_bytesScanned > max_bytes_scanned) {
        fprintf(stderr,
-                "bytes scanned in parse attempts: actual=%d limit=%d \n",
-                scanned_bytes, max_bytes_scanned);
+                "bytes scanned in parse attempts: actual=%u limit=%u \n",
+                g_bytesScanned, max_bytes_scanned);
        fail("too many bytes scanned in parse attempts");
      }
-      assert_true(scanned_bytes <= worstcase_bytes);
-    } else {
-      // heuristic is disabled; every XML_Parse() will have reparsed
-      assert_true(scanned_bytes == worstcase_bytes);
    }
+    assert_true(g_bytesScanned <= (unsigned)worstcase_bytes);

    XML_ParserFree(parser);
  }
@@ -5889,7 +6273,9 @@ make_basic_test_case(Suite *s) {
  tcase_add_test(tc_basic, test_wfc_undeclared_entity_with_external_subset);
  tcase_add_test(tc_basic, test_not_standalone_handler_reject);
  tcase_add_test(tc_basic, test_not_standalone_handler_accept);
+  tcase_add_test(tc_basic, test_entity_start_tag_level_greater_than_one);
  tcase_add_test__if_xml_ge(tc_basic, test_wfc_no_recursive_entity_refs);
+  tcase_add_test(tc_basic, test_no_indirectly_recursive_entity_refs);
  tcase_add_test__ifdef_xml_dtd(tc_basic, test_ext_entity_invalid_parse);
  tcase_add_test__if_xml_ge(tc_basic, test_dtd_default_handling);
  tcase_add_test(tc_basic, test_dtd_attr_handling);
@@ -5940,6 +6326,8 @@ make_basic_test_case(Suite *s) {
  tcase_add_test__ifdef_xml_dtd(tc_basic, test_user_parameters);
  tcase_add_test__ifdef_xml_dtd(tc_basic, test_ext_entity_ref_parameter);
  tcase_add_test(tc_basic, test_empty_parse);
+  tcase_add_test(tc_basic, test_negative_len_parse);
+  tcase_add_test(tc_basic, test_negative_len_parse_buffer);
  tcase_add_test(tc_basic, test_get_buffer_1);
  tcase_add_test(tc_basic, test_get_buffer_2);
 #if XML_CONTEXT_BYTES > 0
@@ -5972,6 +6360,8 @@ make_basic_test_case(Suite *s) {
  tcase_add_test__ifdef_xml_dtd(tc_basic, test_skipped_parameter_entity);
  tcase_add_test__ifdef_xml_dtd(tc_basic,
                                test_recursive_external_parameter_entity);
+  tcase_add_test__ifdef_xml_dtd(tc_basic,
+                                test_recursive_external_parameter_entity_2);
  tcase_add_test(tc_basic, test_undefined_ext_entity_in_external_dtd);
  tcase_add_test(tc_basic, test_suspend_xdecl);
  tcase_add_test(tc_basic, test_abort_epilog);
@@ -6064,8 +6454,14 @@ make_basic_test_case(Suite *s) {
  tcase_add_test(tc_basic, test_empty_element_abort);
  tcase_add_test__ifdef_xml_dtd(tc_basic,
                                test_pool_integrity_with_unfinished_attr);
+  tcase_add_test__if_xml_ge(tc_basic, test_entity_ref_no_elements);
+  tcase_add_test__if_xml_ge(tc_basic, test_deep_nested_entity);
+  tcase_add_test__if_xml_ge(tc_basic, test_deep_nested_attribute_entity);
+  tcase_add_test__if_xml_ge(tc_basic,
+                            test_deep_nested_entity_delayed_interpretation);
  tcase_add_test__if_xml_ge(tc_basic, test_nested_entity_suspend);
-  tcase_add_test(tc_basic, test_big_tokens_take_linear_time);
+  tcase_add_test__if_xml_ge(tc_basic, test_nested_entity_suspend_2);
+  tcase_add_test(tc_basic, test_big_tokens_scale_linearly);
  tcase_add_test(tc_basic, test_set_reparse_deferral);
  tcase_add_test(tc_basic, test_reparse_deferral_is_inherited);
  tcase_add_test(tc_basic, test_set_reparse_deferral_on_null_parser);
@@ -303,6 +303,7 @@ RANLIB = @RANLIB@
 SED = @SED@
 SET_MAKE = @SET_MAKE@
 SHELL = @SHELL@
+SIZEOF_VOID_P = @SIZEOF_VOID_P@
 SO_MAJOR = @SO_MAJOR@
 SO_MINOR = @SO_MINOR@
 SO_PATCH = @SO_PATCH@
@@ -316,7 +317,6 @@ ac_ct_AR = @ac_ct_AR@
 ac_ct_CC = @ac_ct_CC@
 ac_ct_CXX = @ac_ct_CXX@
 ac_ct_DUMPBIN = @ac_ct_DUMPBIN@
-ac_cv_sizeof_void_p = @ac_cv_sizeof_void_p@
 am__include = @am__include@
 am__leading_dot = @am__leading_dot@
 am__quote = @am__quote@
@@ -8,7 +8,7 @@

   Copyright (c) 2003-2006 Karl Waclawek <karl@waclawek.net>
   Copyright (c) 2005-2007 Steven Solie <steven@solie.ca>
-   Copyright (c) 2017-2023 Sebastian Pipping <sebastian@pipping.org>
+   Copyright (c) 2017-2025 Sebastian Pipping <sebastian@pipping.org>
   Copyright (c) 2017      Rhodri James <rhodri@wildebeest.org.uk>
   Licensed under the MIT license:

@@ -32,10 +32,18 @@
   USE OR OTHER DEALINGS IN THE SOFTWARE.
 */

+#define _POSIX_C_SOURCE 1 // fdopen
+
+#if defined(_MSC_VER)
+#  include <io.h> // _open, _close
+#else
+#  include <unistd.h> // close
+#endif
+
+#include <fcntl.h> // open
 #include <sys/stat.h>
 #include <assert.h>
 #include <stddef.h> // ptrdiff_t
-#include <stdlib.h>
 #include <stdio.h>
 #include <time.h>
 #include "expat.h"
@@ -52,17 +60,18 @@
 #  define XML_FMT_STR "s"
 #endif

-static void
+static int
 usage(const char *prog, int rc) {
  fprintf(stderr, "usage: %s [-n] filename bufferSize nr_of_loops\n", prog);
-  exit(rc);
+  return rc;
 }

 int
 main(int argc, char *argv[]) {
  XML_Parser parser;
  char *XMLBuf, *XMLBufEnd, *XMLBufPtr;
-  FILE *fd;
+  int fd;
+  FILE *file;
  struct stat fileAttr;
  int nrOfLoops, bufferSize, i, isFinal;
  size_t fileSize;
@@ -76,34 +85,48 @@ main(int argc, char *argv[]) {
        ns = 1;
        j = 1;
      } else
-        usage(argv[0], 1);
+        return usage(argv[0], 1);
    }
  }

  if (argc != j + 4)
-    usage(argv[0], 1);
+    return usage(argv[0], 1);

-  if (stat(argv[j + 1], &fileAttr) != 0) {
-    fprintf(stderr, "could not access file '%s'\n", argv[j + 1]);
+  fd = open(argv[j + 1], O_RDONLY);
+  if (fd == -1) {
+    fprintf(stderr, "could not open file '%s'\n", argv[j + 1]);
    return 2;
  }

-  fd = fopen(argv[j + 1], "r");
-  if (! fd) {
-    fprintf(stderr, "could not open file '%s'\n", argv[j + 1]);
-    exit(2);
+  if (fstat(fd, &fileAttr) != 0) {
+    close(fd);
+    fprintf(stderr, "could not fstat file '%s'\n", argv[j + 1]);
+    return 2;
+  }
+
+  file = fdopen(fd, "r");
+  if (! file) {
+    close(fd);
+    fprintf(stderr, "could not fdopen file '%s'\n", argv[j + 1]);
+    return 2;
  }

  bufferSize = atoi(argv[j + 2]);
  nrOfLoops = atoi(argv[j + 3]);
  if (bufferSize <= 0 || nrOfLoops <= 0) {
+    fclose(file); // NOTE: this closes fd as well
    fprintf(stderr, "buffer size and nr of loops must be greater than zero.\n");
-    exit(3);
+    return 3;
  }

  XMLBuf = malloc(fileAttr.st_size);
-  fileSize = fread(XMLBuf, sizeof(char), fileAttr.st_size, fd);
-  fclose(fd);
+  if (XMLBuf == NULL) {
+    fclose(file); // NOTE: this closes fd as well
+    fprintf(stderr, "ouf of memory.\n");
+    return 5;
+  }
+  fileSize = fread(XMLBuf, sizeof(char), fileAttr.st_size, file);
+  fclose(file); // NOTE: this closes fd as well

  if (ns)
    parser = XML_ParserCreateNS(NULL, '!');
@@ -132,7 +155,7 @@ main(int argc, char *argv[]) {
                XML_GetCurrentColumnNumber(parser));
        free(XMLBuf);
        XML_ParserFree(parser);
-        exit(4);
+        return 4;
      }
      XMLBufPtr += bufferSize;
    } while (! isFinal);
@@ -10,7 +10,7 @@
   Copyright (c) 2003      Greg Stein <gstein@users.sourceforge.net>
   Copyright (c) 2005-2007 Steven Solie <steven@solie.ca>
   Copyright (c) 2005-2012 Karl Waclawek <karl@waclawek.net>
-   Copyright (c) 2016-2023 Sebastian Pipping <sebastian@pipping.org>
+   Copyright (c) 2016-2025 Sebastian Pipping <sebastian@pipping.org>
   Copyright (c) 2017-2022 Rhodri James <rhodri@wildebeest.org.uk>
   Copyright (c) 2017      Joe Orton <jorton@redhat.com>
   Copyright (c) 2017      José Gutiérrez de la Concha <jose@zeroc.com>
@@ -42,6 +42,8 @@
 */

 #include <assert.h>
+#include <errno.h>
+#include <stdint.h> // for SIZE_MAX
 #include <stdio.h>
 #include <string.h>

@@ -51,6 +53,7 @@
 #include "chardata.h"
 #include "minicheck.h"
 #include "common.h"
+#include "handlers.h"

 /* Common test data */

@@ -201,6 +204,12 @@ _XML_Parse_SINGLE_BYTES(XML_Parser parser, const char *s, int len,
    for (; len > chunksize; len -= chunksize, s += chunksize) {
      enum XML_Status res = XML_Parse(parser, s, chunksize, XML_FALSE);
      if (res != XML_STATUS_OK) {
+        if ((res == XML_STATUS_SUSPENDED) && (len > chunksize)) {
+          fail("Use of function _XML_Parse_SINGLE_BYTES with a chunk size "
+               "greater than 0 (from g_chunkSize) does not work well with "
+               "suspension. Please consider use of plain XML_Parse at this "
+               "place in your test, instead.");
+        }
        return res;
      }
    }
@@ -221,30 +230,6 @@ _expect_failure(const char *text, enum XML_Error errorCode,
    _xml_failure(g_parser, file, lineno);
 }

-/* Character data support for handlers, built on top of the code in
- * chardata.c
- */
-void XMLCALL
-accumulate_characters(void *userData, const XML_Char *s, int len) {
-  CharData_AppendXMLChars((CharData *)userData, s, len);
-}
-
-void XMLCALL
-accumulate_attribute(void *userData, const XML_Char *name,
-                     const XML_Char **atts) {
-  CharData *storage = (CharData *)userData;
-  UNUSED_P(name);
-  /* Check there are attributes to deal with */
-  if (atts == NULL)
-    return;
-
-  while (storage->count < 0 && atts[0] != NULL) {
-    /* "accumulate" the value of the first attribute we see */
-    CharData_AppendXMLChars(storage, atts[1], -1);
-    atts += 2;
-  }
-}
-
 void
 _run_character_check(const char *text, const XML_Char *expected,
                     const char *file, int line) {
@@ -273,12 +258,6 @@ _run_attribute_check(const char *text, const XML_Char *expected,
  CharData_CheckXMLChars(&storage, expected);
 }

-void XMLCALL
-ext_accumulate_characters(void *userData, const XML_Char *s, int len) {
-  ExtTest *test_data = (ExtTest *)userData;
-  accumulate_characters(test_data->storage, s, len);
-}
-
 void
 _run_ext_character_check(const char *text, ExtTest *test_data,
                         const XML_Char *expected, const char *file, int line) {
@@ -323,3 +302,26 @@ duff_reallocator(void *ptr, size_t size) {
    g_reallocation_count--;
  return realloc(ptr, size);
 }
+
+// Portable remake of strndup(3) for C99; does not care about space efficiency
+char *
+portable_strndup(const char *s, size_t n) {
+  if ((s == NULL) || (n == SIZE_MAX)) {
+    errno = EINVAL;
+    return NULL;
+  }
+
+  char *const buffer = (char *)malloc(n + 1);
+  if (buffer == NULL) {
+    errno = ENOMEM;
+    return NULL;
+  }
+
+  errno = 0;
+
+  memcpy(buffer, s, n);
+
+  buffer[n] = '\0';
+
+  return buffer;
+}
@@ -10,7 +10,7 @@
   Copyright (c) 2003      Greg Stein <gstein@users.sourceforge.net>
   Copyright (c) 2005-2007 Steven Solie <steven@solie.ca>
   Copyright (c) 2005-2012 Karl Waclawek <karl@waclawek.net>
-   Copyright (c) 2016-2023 Sebastian Pipping <sebastian@pipping.org>
+   Copyright (c) 2016-2025 Sebastian Pipping <sebastian@pipping.org>
   Copyright (c) 2017-2022 Rhodri James <rhodri@wildebeest.org.uk>
   Copyright (c) 2017      Joe Orton <jorton@redhat.com>
   Copyright (c) 2017      José Gutiérrez de la Concha <jose@zeroc.com>
@@ -111,12 +111,6 @@ extern void _expect_failure(const char *text, enum XML_Error errorCode,
 /* Support functions for handlers to collect up character and attribute data.
 */

-extern void XMLCALL accumulate_characters(void *userData, const XML_Char *s,
-                                          int len);
-
-extern void XMLCALL accumulate_attribute(void *userData, const XML_Char *name,
-                                         const XML_Char **atts);
-
 extern void _run_character_check(const char *text, const XML_Char *expected,
                                 const char *file, int line);

@@ -135,9 +129,6 @@ typedef struct ExtTest {
  CharData *storage;
 } ExtTest;

-extern void XMLCALL ext_accumulate_characters(void *userData, const XML_Char *s,
-                                              int len);
-
 extern void _run_ext_character_check(const char *text, ExtTest *test_data,
                                     const XML_Char *expected, const char *file,
                                     int line);
@@ -155,6 +146,8 @@ extern void *duff_allocator(size_t size);

 extern void *duff_reallocator(void *ptr, size_t size);

+extern char *portable_strndup(const char *s, size_t n);
+
 #endif /* XML_COMMON_H */

 #ifdef __cplusplus
@@ -103,7 +103,9 @@ end_element_event_handler2(void *userData, const XML_Char *name) {
 void XMLCALL
 counting_start_element_handler(void *userData, const XML_Char *name,
                               const XML_Char **atts) {
-  ElementInfo *info = (ElementInfo *)userData;
+  ParserAndElementInfo *const parserAndElementInfos
+      = (ParserAndElementInfo *)userData;
+  ElementInfo *info = parserAndElementInfos->info;
  AttrInfo *attr;
  int count, id, i;

@@ -120,12 +122,12 @@ counting_start_element_handler(void *userData, const XML_Char *name,
   * is possibly a little unexpected, but it is what the
   * documentation in expat.h tells us to expect.
   */
-  count = XML_GetSpecifiedAttributeCount(g_parser);
+  count = XML_GetSpecifiedAttributeCount(parserAndElementInfos->parser);
  if (info->attr_count * 2 != count) {
    fail("Not got expected attribute count");
    return;
  }
-  id = XML_GetIdAttributeIndex(g_parser);
+  id = XML_GetIdAttributeIndex(parserAndElementInfos->parser);
  if (id == -1 && info->id_name != NULL) {
    fail("ID not present");
    return;
@@ -1840,6 +1842,15 @@ element_decl_suspender(void *userData, const XML_Char *name,
  XML_FreeContentModel(g_parser, model);
 }

+void XMLCALL
+suspend_after_element_declaration(void *userData, const XML_Char *name,
+                                  XML_Content *model) {
+  UNUSED_P(name);
+  XML_Parser parser = (XML_Parser)userData;
+  assert_true(XML_StopParser(parser, /*resumable*/ XML_TRUE) == XML_STATUS_OK);
+  XML_FreeContentModel(parser, model);
+}
+
 void XMLCALL
 accumulate_pi_characters(void *userData, const XML_Char *target,
                         const XML_Char *data) {
@@ -1881,9 +1892,17 @@ accumulate_entity_decl(void *userData, const XML_Char *entityName,
 }

 void XMLCALL
-accumulate_char_data(void *userData, const XML_Char *s, int len) {
-  CharData *const storage = (CharData *)userData;
-  CharData_AppendXMLChars(storage, s, len);
+accumulate_char_data_and_suspend(void *userData, const XML_Char *s, int len) {
+  ParserPlusStorage *const parserPlusStorage = (ParserPlusStorage *)userData;
+
+  CharData_AppendXMLChars(parserPlusStorage->storage, s, len);
+
+  for (int i = 0; i < len; i++) {
+    if (s[i] == 'Z') {
+      XML_StopParser(parserPlusStorage->parser, /*resumable=*/XML_TRUE);
+      break;
+    }
+  }
 }

 void XMLCALL
@@ -1910,6 +1929,34 @@ accumulate_start_element(void *userData, const XML_Char *name,
  CharData_AppendXMLChars(storage, XCS(")\n"), 2);
 }

+void XMLCALL
+accumulate_characters(void *userData, const XML_Char *s, int len) {
+  CharData *const storage = (CharData *)userData;
+  CharData_AppendXMLChars(storage, s, len);
+}
+
+void XMLCALL
+accumulate_attribute(void *userData, const XML_Char *name,
+                     const XML_Char **atts) {
+  CharData *const storage = (CharData *)userData;
+  UNUSED_P(name);
+  /* Check there are attributes to deal with */
+  if (atts == NULL)
+    return;
+
+  while (storage->count < 0 && atts[0] != NULL) {
+    /* "accumulate" the value of the first attribute we see */
+    CharData_AppendXMLChars(storage, atts[1], -1);
+    atts += 2;
+  }
+}
+
+void XMLCALL
+ext_accumulate_characters(void *userData, const XML_Char *s, int len) {
+  ExtTest *const test_data = (ExtTest *)userData;
+  accumulate_characters(test_data->storage, s, len);
+}
+
 void XMLCALL
 checking_default_handler(void *userData, const XML_Char *s, int len) {
  DefaultCheck *data = (DefaultCheck *)userData;
@@ -92,6 +92,11 @@ typedef struct elementInfo {
  AttrInfo *attributes;
 } ElementInfo;

+typedef struct StructParserAndElementInfo {
+  XML_Parser parser;
+  ElementInfo *info;
+} ParserAndElementInfo;
+
 extern void XMLCALL counting_start_element_handler(void *userData,
                                                   const XML_Char *name,
                                                   const XML_Char **atts);
@@ -320,6 +325,7 @@ extern int XMLCALL external_entity_devaluer(XML_Parser parser,
 typedef struct ext_hdlr_data {
  const char *parse_text;
  XML_ExternalEntityRefHandler handler;
+  CharData *storage;
 } ExtHdlrData;

 extern int XMLCALL external_entity_oneshot_loader(XML_Parser parser,
@@ -552,6 +558,10 @@ extern void XMLCALL suspending_comment_handler(void *userData,
 extern void XMLCALL element_decl_suspender(void *userData, const XML_Char *name,
                                           XML_Content *model);

+extern void XMLCALL suspend_after_element_declaration(void *userData,
+                                                      const XML_Char *name,
+                                                      XML_Content *model);
+
 extern void XMLCALL accumulate_pi_characters(void *userData,
                                             const XML_Char *target,
                                             const XML_Char *data);
@@ -564,13 +574,23 @@ extern void XMLCALL accumulate_entity_decl(
    const XML_Char *systemId, const XML_Char *publicId,
    const XML_Char *notationName);

-extern void XMLCALL accumulate_char_data(void *userData, const XML_Char *s,
-                                         int len);
+extern void XMLCALL accumulate_char_data_and_suspend(void *userData,
+                                                     const XML_Char *s,
+                                                     int len);

 extern void XMLCALL accumulate_start_element(void *userData,
                                             const XML_Char *name,
                                             const XML_Char **atts);

+extern void XMLCALL accumulate_characters(void *userData, const XML_Char *s,
+                                          int len);
+
+extern void XMLCALL accumulate_attribute(void *userData, const XML_Char *name,
+                                         const XML_Char **atts);
+
+extern void XMLCALL ext_accumulate_characters(void *userData, const XML_Char *s,
+                                              int len);
+
 typedef struct default_check {
  const XML_Char *expected;
  const int expectedLen;
@@ -14,7 +14,7 @@

   Copyright (c) 2004-2006 Fred L. Drake, Jr. <fdrake@users.sourceforge.net>
   Copyright (c) 2006-2012 Karl Waclawek <karl@waclawek.net>
-   Copyright (c) 2016-2024 Sebastian Pipping <sebastian@pipping.org>
+   Copyright (c) 2016-2025 Sebastian Pipping <sebastian@pipping.org>
   Copyright (c) 2022      Rhodri James <rhodri@wildebeest.org.uk>
   Copyright (c) 2023-2024 Sony Corporation / Snild Dolkow <snild@sony.com>
   Licensed under the MIT license:
@@ -129,8 +129,10 @@ void _check_set_test_info(char const *function, char const *filename,
 * Prototypes for the actual implementation.
 */

-#  if defined(__GNUC__)
+#  if defined(__has_attribute)
+#    if __has_attribute(noreturn)
 __attribute__((noreturn))
+#    endif
 #  endif
 void
 _fail(const char *file, int line, const char *msg);
@@ -10,7 +10,7 @@
   Copyright (c) 2003      Greg Stein <gstein@users.sourceforge.net>
   Copyright (c) 2005-2007 Steven Solie <steven@solie.ca>
   Copyright (c) 2005-2012 Karl Waclawek <karl@waclawek.net>
-   Copyright (c) 2016-2024 Sebastian Pipping <sebastian@pipping.org>
+   Copyright (c) 2016-2025 Sebastian Pipping <sebastian@pipping.org>
   Copyright (c) 2017-2022 Rhodri James <rhodri@wildebeest.org.uk>
   Copyright (c) 2017      Joe Orton <jorton@redhat.com>
   Copyright (c) 2017      José Gutiérrez de la Concha <jose@zeroc.com>
@@ -59,6 +59,9 @@
 #include "handlers.h"
 #include "misc_tests.h"

+void XMLCALL accumulate_characters_ext_handler(void *userData,
+                                               const XML_Char *s, int len);
+
 /* Test that a failure to allocate the parser structure fails gracefully */
 START_TEST(test_misc_alloc_create_parser) {
  XML_Memory_Handling_Suite memsuite = {duff_allocator, realloc, free};
@@ -208,7 +211,7 @@ START_TEST(test_misc_version) {
  if (! versions_equal(&read_version, &parsed_version))
    fail("Version mismatch");

-  if (xcstrcmp(version_text, XCS("expat_2.6.0"))) /* needs bump on releases */
+  if (xcstrcmp(version_text, XCS("expat_2.7.1"))) /* needs bump on releases */
    fail("XML_*_VERSION in expat.h out of sync?\n");
 }
 END_TEST
@@ -294,6 +297,7 @@ START_TEST(test_misc_stop_during_end_handler_issue_240_1) {
  parser = XML_ParserCreate(NULL);
  XML_SetElementHandler(parser, start_element_issue_240, end_element_issue_240);
  mydata = (DataIssue240 *)malloc(sizeof(DataIssue240));
+  assert_true(mydata != NULL);
  mydata->parser = parser;
  mydata->deep = 0;
  XML_SetUserData(parser, mydata);
@@ -315,6 +319,7 @@ START_TEST(test_misc_stop_during_end_handler_issue_240_2) {
  parser = XML_ParserCreate(NULL);
  XML_SetElementHandler(parser, start_element_issue_240, end_element_issue_240);
  mydata = (DataIssue240 *)malloc(sizeof(DataIssue240));
+  assert_true(mydata != NULL);
  mydata->parser = parser;
  mydata->deep = 0;
  XML_SetUserData(parser, mydata);
@@ -328,63 +333,119 @@ START_TEST(test_misc_stop_during_end_handler_issue_240_2) {
 END_TEST

 START_TEST(test_misc_deny_internal_entity_closing_doctype_issue_317) {
-  const char *const inputOne = "<!DOCTYPE d [\n"
-                               "<!ENTITY % e ']><d/>'>\n"
-                               "\n"
-                               "%e;";
-  const char *const inputTwo = "<!DOCTYPE d [\n"
-                               "<!ENTITY % e1 ']><d/>'><!ENTITY % e2 '&e1;'>\n"
-                               "\n"
-                               "%e2;";
-  const char *const inputThree = "<!DOCTYPE d [\n"
-                                 "<!ENTITY % e ']><d'>\n"
-                                 "\n"
-                                 "%e;";
-  const char *const inputIssue317 = "<!DOCTYPE doc [\n"
-                                    "<!ENTITY % foo ']>\n"
-                                    "<doc>Hell<oc (#PCDATA)*>'>\n"
-                                    "%foo;\n"
-                                    "]>\n"
-                                    "<doc>Hello, world</dVc>";
+  const char *const inputOne
+      = "<!DOCTYPE d [\n"
+        "<!ENTITY % element_d '<!ELEMENT d (#PCDATA)*>'>\n"
+        "%element_d;\n"
+        "<!ENTITY % e ']><d/>'>\n"
+        "\n"
+        "%e;";
+  const char *const inputTwo
+      = "<!DOCTYPE d [\n"
+        "<!ENTITY % element_d '<!ELEMENT d (#PCDATA)*>'>\n"
+        "%element_d;\n"
+        "<!ENTITY % e1 ']><d/>'><!ENTITY % e2 '&#37;e1;'>\n"
+        "\n"
+        "%e2;";
+  const char *const inputThree
+      = "<!DOCTYPE d [\n"
+        "<!ENTITY % element_d '<!ELEMENT d (#PCDATA)*>'>\n"
+        "%element_d;\n"
+        "<!ENTITY % e ']><d'>\n"
+        "\n"
+        "%e;/>";
+  const char *const inputIssue317
+      = "<!DOCTYPE doc [\n"
+        "<!ENTITY % element_doc '<!ELEMENT doc (#PCDATA)*>'>\n"
+        "%element_doc;\n"
+        "<!ENTITY % foo ']>\n"
+        "<doc>Hell<oc (#PCDATA)*>'>\n"
+        "%foo;\n"
+        "]>\n"
+        "<doc>Hello, world</dVc>";

  const char *const inputs[] = {inputOne, inputTwo, inputThree, inputIssue317};
+  const XML_Bool suspendOrNot[] = {XML_FALSE, XML_TRUE};
  size_t inputIndex = 0;

  for (; inputIndex < sizeof(inputs) / sizeof(inputs[0]); inputIndex++) {
-    set_subtest("%s", inputs[inputIndex]);
-    XML_Parser parser;
-    enum XML_Status parseResult;
-    int setParamEntityResult;
-    XML_Size lineNumber;
-    XML_Size columnNumber;
-    const char *const input = inputs[inputIndex];
+    for (size_t suspendOrNotIndex = 0;
+         suspendOrNotIndex < sizeof(suspendOrNot) / sizeof(suspendOrNot[0]);
+         suspendOrNotIndex++) {
+      const char *const input = inputs[inputIndex];
+      const XML_Bool suspend = suspendOrNot[suspendOrNotIndex];
+      if (suspend && (g_chunkSize > 0)) {
+        // We cannot use _XML_Parse_SINGLE_BYTES below due to suspension, and
+        // so chunk sizes >0 would only repeat the very same test
+        // due to use of plain XML_Parse; we are saving upon that runtime:
+        return;
+      }

-    parser = XML_ParserCreate(NULL);
-    setParamEntityResult
-        = XML_SetParamEntityParsing(parser, XML_PARAM_ENTITY_PARSING_ALWAYS);
-    if (setParamEntityResult != 1)
-      fail("Failed to set XML_PARAM_ENTITY_PARSING_ALWAYS.");
+      set_subtest("[input=%d suspend=%s] %s", (int)inputIndex,
+                  suspend ? "true" : "false", input);
+      XML_Parser parser;
+      enum XML_Status parseResult;
+      int setParamEntityResult;
+      XML_Size lineNumber;
+      XML_Size columnNumber;
+
+      parser = XML_ParserCreate(NULL);
+      setParamEntityResult
+          = XML_SetParamEntityParsing(parser, XML_PARAM_ENTITY_PARSING_ALWAYS);
+      if (setParamEntityResult != 1)
+        fail("Failed to set XML_PARAM_ENTITY_PARSING_ALWAYS.");
+
+      if (suspend) {
+        XML_SetUserData(parser, parser);
+        XML_SetElementDeclHandler(parser, suspend_after_element_declaration);
+      }
+
+      if (suspend) {
+        // can't use SINGLE_BYTES here, because it'll return early on
+        // suspension, and we won't know exactly how much input we actually
+        // managed to give Expat.
+        parseResult = XML_Parse(parser, input, (int)strlen(input), 0);
+
+        while (parseResult == XML_STATUS_SUSPENDED) {
+          parseResult = XML_ResumeParser(parser);
+        }
+
+        if (parseResult != XML_STATUS_ERROR) {
+          // can't use SINGLE_BYTES here, because it'll return early on
+          // suspension, and we won't know exactly how much input we actually
+          // managed to give Expat.
+          parseResult = XML_Parse(parser, "", 0, 1);
+        }
+
+        while (parseResult == XML_STATUS_SUSPENDED) {
+          parseResult = XML_ResumeParser(parser);
+        }
+      } else {
+        parseResult
+            = _XML_Parse_SINGLE_BYTES(parser, input, (int)strlen(input), 0);
+
+        if (parseResult != XML_STATUS_ERROR) {
+          parseResult = _XML_Parse_SINGLE_BYTES(parser, "", 0, 1);
+        }
+      }

-    parseResult = _XML_Parse_SINGLE_BYTES(parser, input, (int)strlen(input), 0);
-    if (parseResult != XML_STATUS_ERROR) {
-      parseResult = _XML_Parse_SINGLE_BYTES(parser, "", 0, 1);
      if (parseResult != XML_STATUS_ERROR) {
        fail("Parsing was expected to fail but succeeded.");
      }
+
+      if (XML_GetErrorCode(parser) != XML_ERROR_INVALID_TOKEN)
+        fail("Error code does not match XML_ERROR_INVALID_TOKEN");
+
+      lineNumber = XML_GetCurrentLineNumber(parser);
+      if (lineNumber != 6)
+        fail("XML_GetCurrentLineNumber does not work as expected.");
+
+      columnNumber = XML_GetCurrentColumnNumber(parser);
+      if (columnNumber != 0)
+        fail("XML_GetCurrentColumnNumber does not work as expected.");
+
+      XML_ParserFree(parser);
    }
-
-    if (XML_GetErrorCode(parser) != XML_ERROR_INVALID_TOKEN)
-      fail("Error code does not match XML_ERROR_INVALID_TOKEN");
-
-    lineNumber = XML_GetCurrentLineNumber(parser);
-    if (lineNumber != 4)
-      fail("XML_GetCurrentLineNumber does not work as expected.");
-
-    columnNumber = XML_GetCurrentColumnNumber(parser);
-    if (columnNumber != 0)
-      fail("XML_GetCurrentColumnNumber does not work as expected.");
-
-    XML_ParserFree(parser);
  }
 }
 END_TEST
@@ -447,7 +508,7 @@ START_TEST(test_misc_general_entities_support) {
  XML_SetExternalEntityRefHandler(parser,
                                  external_entity_failer__if_not_xml_ge);
  XML_SetEntityDeclHandler(parser, accumulate_entity_decl);
-  XML_SetCharacterDataHandler(parser, accumulate_char_data);
+  XML_SetCharacterDataHandler(parser, accumulate_characters);

  if (_XML_Parse_SINGLE_BYTES(parser, doc, (int)strlen(doc), XML_TRUE)
      != XML_STATUS_OK) {
@@ -496,6 +557,127 @@ START_TEST(test_misc_char_handler_stop_without_leak) {
 }
 END_TEST

+START_TEST(test_misc_resumeparser_not_crashing) {
+  XML_Parser parser = XML_ParserCreate(NULL);
+  XML_GetBuffer(parser, 1);
+  XML_StopParser(parser, /*resumable=*/XML_TRUE);
+  XML_ResumeParser(parser); // could crash here, previously
+  XML_ParserFree(parser);
+}
+END_TEST
+
+START_TEST(test_misc_stopparser_rejects_unstarted_parser) {
+  const XML_Bool cases[] = {XML_TRUE, XML_FALSE};
+  for (size_t i = 0; i < sizeof(cases) / sizeof(cases[0]); i++) {
+    const XML_Bool resumable = cases[i];
+    XML_Parser parser = XML_ParserCreate(NULL);
+    assert_true(XML_GetErrorCode(parser) == XML_ERROR_NONE);
+    assert_true(XML_StopParser(parser, resumable) == XML_STATUS_ERROR);
+    assert_true(XML_GetErrorCode(parser) == XML_ERROR_NOT_STARTED);
+    XML_ParserFree(parser);
+  }
+}
+END_TEST
+
+/* Adaptation of accumulate_characters that takes ExtHdlrData input to work with
+ * test_renter_loop_finite_content below */
+void XMLCALL
+accumulate_characters_ext_handler(void *userData, const XML_Char *s, int len) {
+  ExtHdlrData *const test_data = (ExtHdlrData *)userData;
+  CharData_AppendXMLChars(test_data->storage, s, len);
+}
+
+/* Test that internalEntityProcessor does not re-enter forever;
+ * based on files tests/xmlconf/xmltest/valid/ext-sa/012.{xml,ent} */
+START_TEST(test_renter_loop_finite_content) {
+  CharData storage;
+  CharData_Init(&storage);
+  const char *const text = "<!DOCTYPE doc [\n"
+                           "<!ENTITY e1 '&e2;'>\n"
+                           "<!ENTITY e2 '&e3;'>\n"
+                           "<!ENTITY e3 SYSTEM '012.ent'>\n"
+                           "<!ENTITY e4 '&e5;'>\n"
+                           "<!ENTITY e5 '(e5)'>\n"
+                           "<!ELEMENT doc (#PCDATA)>\n"
+                           "]>\n"
+                           "<doc>&e1;</doc>\n";
+  ExtHdlrData test_data = {"&e4;\n", external_entity_null_loader, &storage};
+  const XML_Char *const expected = XCS("(e5)\n");
+
+  XML_Parser parser = XML_ParserCreate(NULL);
+  assert_true(parser != NULL);
+  XML_SetUserData(parser, &test_data);
+  XML_SetExternalEntityRefHandler(parser, external_entity_oneshot_loader);
+  XML_SetCharacterDataHandler(parser, accumulate_characters_ext_handler);
+  if (_XML_Parse_SINGLE_BYTES(parser, text, (int)strlen(text), XML_TRUE)
+      == XML_STATUS_ERROR)
+    xml_failure(parser);
+
+  CharData_CheckXMLChars(&storage, expected);
+  XML_ParserFree(parser);
+}
+END_TEST
+
+// Inspired by function XML_OriginalString of Perl's XML::Parser
+static char *
+dup_original_string(XML_Parser parser) {
+  const int byte_count = XML_GetCurrentByteCount(parser);
+
+  assert_true(byte_count >= 0);
+
+  int offset = -1;
+  int size = -1;
+
+  const char *const context = XML_GetInputContext(parser, &offset, &size);
+
+#if XML_CONTEXT_BYTES > 0
+  assert_true(context != NULL);
+  assert_true(offset >= 0);
+  assert_true(size >= 0);
+  return portable_strndup(context + offset, byte_count);
+#else
+  assert_true(context == NULL);
+  return NULL;
+#endif
+}
+
+static void
+on_characters_issue_980(void *userData, const XML_Char *s, int len) {
+  (void)s;
+  (void)len;
+  XML_Parser parser = (XML_Parser)userData;
+
+  char *const original_string = dup_original_string(parser);
+
+#if XML_CONTEXT_BYTES > 0
+  assert_true(original_string != NULL);
+  assert_true(strcmp(original_string, "&draft.day;") == 0);
+  free(original_string);
+#else
+  assert_true(original_string == NULL);
+#endif
+}
+
+START_TEST(test_misc_expected_event_ptr_issue_980) {
+  // NOTE: This is a tiny subset of sample "REC-xml-19980210.xml"
+  //       from Perl's XML::Parser
+  const char *const doc = "<!DOCTYPE day [\n"
+                          "  <!ENTITY draft.day '10'>\n"
+                          "]>\n"
+                          "<day>&draft.day;</day>\n";
+
+  XML_Parser parser = XML_ParserCreate(NULL);
+  XML_SetUserData(parser, parser);
+  XML_SetCharacterDataHandler(parser, on_characters_issue_980);
+
+  assert_true(_XML_Parse_SINGLE_BYTES(parser, doc, (int)strlen(doc),
+                                      /*isFinal=*/XML_TRUE)
+              == XML_STATUS_OK);
+
+  XML_ParserFree(parser);
+}
+END_TEST
+
 void
 make_miscellaneous_test_case(Suite *s) {
  TCase *tc_misc = tcase_create("miscellaneous tests");
@@ -520,4 +702,8 @@ make_miscellaneous_test_case(Suite *s) {
                 test_misc_create_external_entity_parser_with_null_context);
  tcase_add_test(tc_misc, test_misc_general_entities_support);
  tcase_add_test(tc_misc, test_misc_char_handler_stop_without_leak);
+  tcase_add_test(tc_misc, test_misc_resumeparser_not_crashing);
+  tcase_add_test(tc_misc, test_misc_stopparser_rejects_unstarted_parser);
+  tcase_add_test__if_xml_ge(tc_misc, test_renter_loop_finite_content);
+  tcase_add_test(tc_misc, test_misc_expected_event_ptr_issue_980);
 }
@@ -2,8 +2,8 @@
 # EXPAT TEST SCRIPT FOR W3C XML TEST SUITE
 #
 # This script can be used to exercise Expat against the
-# w3c.org xml test suite, available from
-# http://www.w3.org/XML/Test/xmlts20020606.zip.
+# w3c.org xml test suite, available from:
+# https://www.w3.org/XML/Test/xmlts20020606.zip
 #
 # To run this script, first set XMLWF below so that xmlwf can be
 # found, then set the output directory with OUTPUT.
@@ -30,6 +30,7 @@
 # Copyright (c) 2002      Karl Waclawek <[email protected]>
 # Copyright (c) 2008-2019 Sebastian Pipping <[email protected]>
 # Copyright (c) 2017      Rhodri James <[email protected]>
+# Copyright (c) 2025      Hanno Böck <[email protected]>
 # Licensed under the MIT license:
 #
 # Permission is  hereby granted,  free of charge,  to any  person obtaining
@@ -311,6 +311,7 @@ RANLIB = @RANLIB@
 SED = @SED@
 SET_MAKE = @SET_MAKE@
 SHELL = @SHELL@
+SIZEOF_VOID_P = @SIZEOF_VOID_P@
 SO_MAJOR = @SO_MAJOR@
 SO_MINOR = @SO_MINOR@
 SO_PATCH = @SO_PATCH@
@@ -324,7 +325,6 @@ ac_ct_AR = @ac_ct_AR@
 ac_ct_CC = @ac_ct_CC@
 ac_ct_CXX = @ac_ct_CXX@
 ac_ct_DUMPBIN = @ac_ct_DUMPBIN@
-ac_cv_sizeof_void_p = @ac_cv_sizeof_void_p@
 am__include = @am__include@
 am__leading_dot = @am__leading_dot@
 am__quote = @am__quote@
@@ -14,6 +14,7 @@
   Copyright (c) 2017      Rhodri James <rhodri@wildebeest.org.uk>
   Copyright (c) 2017      Franek Korta <fkorta@gmail.com>
   Copyright (c) 2022      Sean McBride <sean@rogue-research.com>
+   Copyright (c) 2025      Hanno Böck <hanno@gentoo.org>
   Licensed under the MIT license:

   Permission is  hereby granted,  free of charge,  to any  person obtaining
@@ -55,7 +56,7 @@
 #  define EXPAT_read_count_t int
 #  define EXPAT_read_req_t unsigned int
 #else /* POSIX */
-/* http://pubs.opengroup.org/onlinepubs/009695399/functions/read.html */
+/* https://pubs.opengroup.org/onlinepubs/009695399/functions/read.html */
 #  define EXPAT_read read
 #  define EXPAT_read_count_t ssize_t
 #  define EXPAT_read_req_t size_t
@@ -15,6 +15,7 @@
   Copyright (c) 2017      Rhodri James <rhodri@wildebeest.org.uk>
   Copyright (c) 2019      David Loffredo <loffredo@steptools.com>
   Copyright (c) 2021      Donghee Na <donghee.na@python.org>
+   Copyright (c) 2024      Hanno Böck <hanno@gentoo.org>
   Licensed under the MIT license:

   Permission is  hereby granted,  free of charge,  to any  person obtaining
@@ -91,7 +92,8 @@ reportError(XML_Parser parser, const XML_Char *filename) {
             filename, XML_GetErrorLineNumber(parser),
             XML_GetErrorColumnNumber(parser), message);
  else
-    ftprintf(stderr, T("%s: (unknown message %d)\n"), filename, code);
+    ftprintf(stderr, T("%s: (unknown message %u)\n"), filename,
+             (unsigned int)code);
 }

 /* This implementation will give problems on files larger than INT_MAX. */
@@ -20,7 +20,7 @@ the actual statements in the files are controlling.
   libarchive/mtree.5

 * The following source files are in the public domain:
-   libarchive/archive_getdate.c
+   libarchive/archive_parse_date.c

 * The following source files are triple-licensed with the ability to choose
  from CC0 1.0 Universal, OpenSSL or Apache 2.0 licenses:
@@ -1,3 +1,11 @@
+Jun 01, 2026: libarchive 3.8.1 released
+
+May 20, 2025: libarchive 3.8.0 released
+
+Mar 30, 2025: libarchive 3.7.9 released
+
+Mar 20, 2025: libarchive 3.7.8 released
+
 Oct 13, 2024: libarchive 3.7.7 released

 Sep 23, 2024: libarchive 3.7.6 released
@@ -85,13 +85,14 @@ Currently, the library automatically detects and reads the following formats:
  * PWB binary cpio
  * ISO9660 CD-ROM images (with optional Rockridge or Joliet extensions)
  * ZIP archives (with uncompressed or "deflate" compressed entries, including support for encrypted Zip archives)
-  * ZIPX archives (with support for bzip2, ppmd8, lzma and xz compressed entries)
+  * ZIPX archives (with support for bzip2, zstd, ppmd8, lzma and xz compressed entries)
  * GNU and BSD 'ar' archives
  * 'mtree' format
  * 7-Zip archives (including archives that use zstandard compression)
  * Microsoft CAB format
  * LHA and LZH archives
  * RAR and RAR 5.0 archives (with some limitations due to RAR's proprietary status)
+  * WARC archives
  * XAR archives

 The library also detects and handles any of the following before evaluating the archive:
@@ -120,15 +121,18 @@ The library can create archives in any of the following formats:
  * PWB binary cpio
  * shar archives
  * ZIP archives (with uncompressed or "deflate" compressed entries)
+  * ZIPX archives (with bzip2, zstd, lzma or xz compressed entries)
  * GNU and BSD 'ar' archives
  * 'mtree' format
  * ISO9660 format
-  * 7-Zip archives
+  * 7-Zip archives (including archives that use zstandard compression)
+  * WARC archives
  * XAR archives

 When creating archives, the result can be filtered with any of the following:

  * uuencode
+  * base64
  * gzip compression
  * bzip2 compression
  * compress/LZW compression
@@ -241,4 +245,3 @@ questions we are asked about libarchive:
  appropriate.  It has many advantages over other tar formats
  (including the legacy GNU tar format) and is widely supported by
  current tar implementations.
-
@@ -0,0 +1,19 @@
+#!/bin/sh
+set -eux
+
+# Uncommenting these adds a full minute to the CI time
+#brew update > /dev/null
+#brew upgrade > /dev/null
+
+# This does an upgrade if the package is already installed
+brew install \
+	autoconf \
+	automake \
+	libtool \
+	pkg-config \
+	cmake \
+	xz \
+	lz4 \
+	zstd \
+	libxml2 \
+	openssl
@@ -1206,7 +1206,7 @@ list_item_verbose(struct cpio *cpio, struct archive_entry *entry)
 	else
 		strcpy(date, "invalid mtime");

-	fprintf(out, "%s%3d %-8s %-8s %8s %12s %s",
+	fprintf(out, "%s%3u %-8s %-8s %8s %12s %s",
 	    archive_entry_strmode(entry),
 	    archive_entry_nlink(entry),
 	    uname, gname, size, date,
@@ -189,10 +189,10 @@ DEFINE_TEST(test_format_newc)
 	gid = from_hex(e + 30, 8); /* gid */
 	assertEqualMem(e + 38, "00000003", 8); /* nlink */
 	t = from_hex(e + 46, 8); /* mtime */
-	failure("t=%#08jx now=%#08jx=%jd", (intmax_t)t, (intmax_t)now,
+	failure("t=%#08jx now=%#08jx=%jd", (uintmax_t)t, (uintmax_t)now,
 	    (intmax_t)now);
 	assert(t <= now); /* File wasn't created in future. */
-	failure("t=%#08jx now - 2=%#08jx=%jd", (intmax_t)t, (intmax_t)now - 2,
+	failure("t=%#08jx now - 2=%#08jx=%jd", (uintmax_t)t, (uintmax_t)now - 2,
 	    (intmax_t)now - 2);
 	assert(t >= now - 2); /* File was created w/in last 2 secs. */
 	failure("newc format stores body only with last appearance of a link\n"
@@ -219,7 +219,7 @@ DEFINE_TEST(test_format_newc)
 		assert(is_hex(e, 110));
 		assertEqualMem(e + 0, "070701", 6); /* Magic */
 		assert(is_hex(e + 6, 8)); /* ino */
-#if defined(_WIN32) && !defined(CYGWIN)
+#if defined(_WIN32) && !defined(__CYGWIN__)
 		/* Mode: Group members bits and others bits do not work. */
 		assertEqualInt(0xa180, from_hex(e + 14, 8) & 0xffc0);
 #else
@@ -230,7 +230,7 @@ DEFINE_TEST(test_format_newc)
 		assertEqualMem(e + 38, "00000001", 8); /* nlink */
 		t2 = from_hex(e + 46, 8); /* mtime */
 		failure("First entry created at t=%#08jx this entry created"
-		    " at t2=%#08jx", (intmax_t)t, (intmax_t)t2);
+		    " at t2=%#08jx", (uintmax_t)t, (uintmax_t)t2);
 		assert(t2 == t || t2 == t + 1); /* Almost same as first entry. */
 		assertEqualMem(e + 54, "00000005", 8); /* File size */
 		fs = (uint64_t)from_hex(e + 54, 8);
@@ -266,7 +266,7 @@ DEFINE_TEST(test_format_newc)
 #endif
 	t2 = from_hex(e + 46, 8); /* mtime */
 	failure("First entry created at t=%#08jx this entry created at"
-	    "t2=%#08jx", (intmax_t)t, (intmax_t)t2);
+	    "t2=%#08jx", (uintmax_t)t, (uintmax_t)t2);
 	assert(t2 == t || t2 == t + 1); /* Almost same as first entry. */
 	assertEqualMem(e + 54, "00000000", 8); /* File size */
 	fs = (uint64_t)from_hex(e + 54, 8);
@@ -300,7 +300,7 @@ DEFINE_TEST(test_format_newc)
 	assertEqualMem(e + 38, "00000003", 8); /* nlink */
 	t2 = from_hex(e + 46, 8); /* mtime */
 	failure("First entry created at t=%#08jx this entry created at"
-	    "t2=%#08jx", (intmax_t)t, (intmax_t)t2);
+	    "t2=%#08jx", (uintmax_t)t, (uintmax_t)t2);
 	assert(t2 == t || t2 == t + 1); /* Almost same as first entry. */
 	assertEqualInt(10, from_hex(e + 54, 8)); /* File size */
 	fs = (uint64_t)from_hex(e + 54, 8);
@@ -52,7 +52,7 @@ test_create(void)
 		 * #ifdef this section out.  Most of the test below is
 		 * still valid. */
 		memset(&times, 0, sizeof(times));
-#if defined(_WIN32) && !defined(CYGWIN)
+#if defined(_WIN32) && !defined(__CYGWIN__)
 		times.actime = 86400;
 		times.modtime = 86400;
 #else
@@ -18,7 +18,7 @@ is_octal(const char *p, size_t l)
 	return (1);
 }

-static long long int
+static unsigned long long int
 from_octal(const char *p, size_t l)
 {
 	long long int r = 0;
@@ -117,7 +117,7 @@ DEFINE_TEST(test_option_c)
 	gid = (int)from_octal(e + 30, 6);
 	assertEqualMem(e + 36, "000001", 6); /* nlink */
 	failure("file entries should not have rdev set (dev field was 0%o)",
-	    dev);
+	    (unsigned int)dev);
 	assertEqualMem(e + 42, "000000", 6); /* rdev */
 	t = from_octal(e + 48, 11); /* mtime */
 	assert(t <= now); /* File wasn't created in future. */
@@ -133,7 +133,7 @@ DEFINE_TEST(test_option_c)
 		assert(is_octal(e, 76)); /* Entire header is octal digits. */
 		assertEqualMem(e + 0, "070707", 6); /* Magic */
 		assertEqualInt(dev, from_octal(e + 6, 6)); /* dev */
-		assert(ino != from_octal(e + 12, 6)); /* ino */
+		assert(ino != (int)from_octal(e + 12, 6)); /* ino */
 #if !defined(_WIN32) || defined(__CYGWIN__)
 		/* On Windows, symbolic link and group members bits and
 		 * others bits do not work. */
@@ -163,7 +163,7 @@ DEFINE_TEST(test_option_c)
 	assertEqualInt(dev, from_octal(e + 6, 6));
 	/* Ino must be different from first entry. */
 	assert(is_octal(e + 12, 6)); /* ino */
-	assert(ino != from_octal(e + 12, 6));
+	assert(ino != (int)from_octal(e + 12, 6));
 #if defined(_WIN32) && !defined(__CYGWIN__)
 	/* Group members bits and others bits do not work. */
 	assertEqualMem(e + 18, "040777", 6); /* Mode */
@@ -34,12 +34,15 @@
 * assert that ARCHIVE_VERSION_NUMBER >= 2012108.
 */
 /* Note: Compiler will complain if this does not match archive_entry.h! */
-#define	ARCHIVE_VERSION_NUMBER 3007007
+#define	ARCHIVE_VERSION_NUMBER 3008001

 #include <sys/stat.h>
 #include <stddef.h>  /* for wchar_t */
 #include <stdio.h> /* For FILE * */
+#if ARCHIVE_VERSION_NUMBER < 4000000
+/* time_t is slated to be removed from public includes in 4.0 */
 #include <time.h> /* For time_t */
+#endif

 /*
 * Note: archive.h is for use outside of libarchive; the configuration
@@ -63,12 +66,15 @@
 #define __LA_INT64_T_DEFINED
 # if defined(_WIN32) && !defined(__CYGWIN__) && !defined(__WATCOMC__)
 typedef __int64 la_int64_t;
+typedef unsigned __int64 la_uint64_t;
 # else
 # include <unistd.h>  /* ssize_t */
 #  if defined(_SCO_DS) || defined(__osf__)
 typedef long long la_int64_t;
+typedef unsigned long long la_uint64_t;
 #  else
 typedef int64_t la_int64_t;
+typedef uint64_t la_uint64_t;
 #  endif
 # endif
 #endif
@@ -94,6 +100,22 @@ typedef ssize_t la_ssize_t;
 # endif
 #endif

+#if ARCHIVE_VERSION_NUMBER < 4000000
+/* Use the platform types for time_t */
+#define __LA_TIME_T time_t
+#else
+/* Use 64-bits integer types for time_t */
+#define __LA_TIME_T la_int64_t
+#endif
+
+#if ARCHIVE_VERSION_NUMBER < 4000000
+/* Use the platform types for dev_t */
+#define __LA_DEV_T dev_t
+#else
+/* Use 64-bits integer types for dev_t */
+#define __LA_DEV_T la_int64_t
+#endif
+
 /* Large file support for Android */
 #if defined(__LIBARCHIVE_BUILD) && defined(__ANDROID__)
 #include "android_lf.h"
@@ -132,7 +154,7 @@ typedef ssize_t la_ssize_t;
 #define	__LA_PRINTF(fmtarg, firstvararg)	/* nothing */
 #endif

-#if defined(__GNUC__) && __GNUC__ >= 3 && __GNUC_MINOR__ >= 1
+#if defined(__GNUC__) && (__GNUC__ > 3 || (__GNUC__ == 3 && __GNUC_MINOR__ >= 1))
 # define __LA_DEPRECATED __attribute__((deprecated))
 #else
 # define __LA_DEPRECATED
@@ -155,7 +177,7 @@ __LA_DECL int		archive_version_number(void);
 /*
 * Textual name/version of the library, useful for version displays.
 */
-#define	ARCHIVE_VERSION_ONLY_STRING "3.7.7"
+#define	ARCHIVE_VERSION_ONLY_STRING "3.8.1"
 #define	ARCHIVE_VERSION_STRING "libarchive " ARCHIVE_VERSION_ONLY_STRING
 __LA_DECL const char *	archive_version_string(void);

@@ -178,6 +200,23 @@ __LA_DECL const char *  archive_liblzma_version(void);
 __LA_DECL const char *  archive_bzlib_version(void);
 __LA_DECL const char *  archive_liblz4_version(void);
 __LA_DECL const char *  archive_libzstd_version(void);
+__LA_DECL const char *  archive_liblzo2_version(void);
+__LA_DECL const char *  archive_libexpat_version(void);
+__LA_DECL const char *  archive_libbsdxml_version(void);
+__LA_DECL const char *  archive_libxml2_version(void);
+__LA_DECL const char *  archive_mbedtls_version(void);
+__LA_DECL const char *  archive_nettle_version(void);
+__LA_DECL const char *  archive_openssl_version(void);
+__LA_DECL const char *  archive_libmd_version(void);
+__LA_DECL const char *  archive_commoncrypto_version(void);
+__LA_DECL const char *  archive_cng_version(void);
+__LA_DECL const char *  archive_wincrypt_version(void);
+__LA_DECL const char *  archive_librichacl_version(void);
+__LA_DECL const char *  archive_libacl_version(void);
+__LA_DECL const char *  archive_libattr_version(void);
+__LA_DECL const char *  archive_libiconv_version(void);
+__LA_DECL const char *  archive_libpcre_version(void);
+__LA_DECL const char *  archive_libpcre2_version(void);

 /* Declare our basic types. */
 struct archive;
@@ -449,6 +488,8 @@ __LA_DECL int archive_read_support_format_by_code(struct archive *, int);
 __LA_DECL int archive_read_support_format_cab(struct archive *);
 __LA_DECL int archive_read_support_format_cpio(struct archive *);
 __LA_DECL int archive_read_support_format_empty(struct archive *);
+/* archive_read_support_format_gnutar() is an alias for historical reasons
+ * of archive_read_support_format_tar(). */
 __LA_DECL int archive_read_support_format_gnutar(struct archive *);
 __LA_DECL int archive_read_support_format_iso9660(struct archive *);
 __LA_DECL int archive_read_support_format_lha(struct archive *);
@@ -828,6 +869,10 @@ __LA_DECL int archive_write_set_format_filter_by_ext(struct archive *a, const ch
 __LA_DECL int archive_write_set_format_filter_by_ext_def(struct archive *a, const char *filename, const char * def_ext);
 __LA_DECL int archive_write_zip_set_compression_deflate(struct archive *);
 __LA_DECL int archive_write_zip_set_compression_store(struct archive *);
+__LA_DECL int archive_write_zip_set_compression_lzma(struct archive *);
+__LA_DECL int archive_write_zip_set_compression_xz(struct archive *);
+__LA_DECL int archive_write_zip_set_compression_bzip2(struct archive *);
+__LA_DECL int archive_write_zip_set_compression_zstd(struct archive *);
 /* Deprecated; use archive_write_open2 instead */
 __LA_DECL int archive_write_open(struct archive *, void *,
 		     archive_open_callback *, archive_write_callback *,
@@ -1083,6 +1128,10 @@ __LA_DECL int		 archive_compression(struct archive *)
 				__LA_DEPRECATED;
 #endif

+/* Parses a date string relative to the current time.
+ * NOTE: This is not intended for general date parsing, and the resulting timestamp should only be used for libarchive. */
+__LA_DECL time_t	archive_parse_date(time_t now, const char *datestr);
+
 __LA_DECL int		 archive_errno(struct archive *);
 __LA_DECL const char	*archive_error_string(struct archive *);
 __LA_DECL const char	*archive_format_name(struct archive *);
@@ -1201,8 +1250,10 @@ __LA_DECL int	archive_match_include_gname_w(struct archive *,
 		    const wchar_t *);

 /* Utility functions */
+#if ARCHIVE_VERSION_NUMBER < 4000000
 /* Convenience function to sort a NULL terminated list of strings */
 __LA_DECL int archive_utility_string_sort(char **);
+#endif

 #ifdef __cplusplus
 }
@@ -1185,8 +1185,13 @@ archive_acl_from_text_w(struct archive_acl *acl, const wchar_t *text,
 		/* Set remaining fields to blank. */
 		for (n = fields; n < numfields; ++n)
 			field[n].start = field[n].end = NULL;
+		
+		if (field[0].start == NULL || field[0].end == NULL) {
+			/* This should never happen */
+			return (ARCHIVE_FATAL);
+		}

-		if (field[0].start != NULL && *(field[0].start) == L'#') {
+		if (*(field[0].start) == L'#') {
 			/* Comment, skip entry */
 			continue;
 		}
@@ -1676,7 +1681,12 @@ archive_acl_from_text_nl(struct archive_acl *acl, const char *text,
 		for (n = fields; n < numfields; ++n)
 			field[n].start = field[n].end = NULL;

-		if (field[0].start != NULL && *(field[0].start) == '#') {
+		if (field[0].start == NULL || field[0].end == NULL) {
+			/* This should never happen */
+			return (ARCHIVE_FATAL);
+		}
+
+		if (*(field[0].start) == '#') {
 			/* Comment, skip entry */
 			continue;
 		}
@@ -80,4 +80,4 @@ int archive_acl_from_text_l(struct archive_acl *, const char * /* text */,
 int archive_acl_from_text_nl(struct archive_acl *, const char * /* text */,
    size_t /* size of text */, int /* type */, struct archive_string_conv *);

-#endif /* ARCHIVE_ENTRY_PRIVATE_H_INCLUDED */
+#endif /* !ARCHIVE_ACL_PRIVATE_H_INCLUDED */
@@ -71,7 +71,7 @@ get_argument(struct archive_string *as, const char *p)
 	archive_string_empty(as);

 	/* Skip beginning space characters. */
-	while (*s != '\0' && *s == ' ')
+	while (*s == ' ')
 		s++;
 	/* Copy non-space characters. */
 	while (*s != '\0' && *s != ' ') {
@@ -64,6 +64,7 @@ typedef struct {

 #elif defined(_WIN32) && !defined(__CYGWIN__) && defined(HAVE_BCRYPT_H) && _WIN32_WINNT >= _WIN32_WINNT_VISTA
 #include <bcrypt.h>
+#define	ARCHIVE_CRYPTOR_USE_CNG 1

 /* Common in other bcrypt implementations, but missing from VS2008. */
 #ifndef BCRYPT_SUCCESS
@@ -86,6 +87,7 @@ typedef struct {
 #include <mbedtls/aes.h>
 #include <mbedtls/md.h>
 #include <mbedtls/pkcs5.h>
+#define	ARCHIVE_CRYPTOR_USE_MBED 1

 #define AES_MAX_KEY_SIZE 32
 #define AES_BLOCK_SIZE 16
@@ -105,6 +107,7 @@ typedef struct {
 #endif
 #include <nettle/aes.h>
 #include <nettle/version.h>
+#define	ARCHIVE_CRYPTOR_USE_NETTLE 1

 typedef struct {
 #if NETTLE_VERSION_MAJOR < 3
@@ -125,6 +128,7 @@ typedef struct {

 #elif defined(HAVE_LIBCRYPTO)
 #include "archive_openssl_evp_private.h"
+#define	ARCHIVE_CRYPTOR_USE_OPENSSL 1
 #define AES_BLOCK_SIZE	16
 #define AES_MAX_KEY_SIZE 32

@@ -140,6 +144,10 @@ typedef struct {

 #else

+#if defined(_WIN32) && !defined(__CYGWIN__) && !(defined(HAVE_BCRYPT_H) && _WIN32_WINNT >= _WIN32_WINNT_VISTA)
+#define ARCHIVE_CRYPTOR_USE_WINCRYPT 1
+#endif
+
 #define AES_BLOCK_SIZE	16
 #define AES_MAX_KEY_SIZE 32
 typedef int archive_crypto_ctx;
@@ -196,6 +196,13 @@ __archive_md5final(archive_md5_ctx *ctx, void *md)

 #elif defined(ARCHIVE_CRYPTO_MD5_LIBSYSTEM)

+// These functions are available in macOS 10.4 and later, but deprecated from 10.15 onwards.
+// We need to continue supporting this feature regardless, so suppress the warnings.
+#if defined(__clang__)
+#pragma clang diagnostic push
+#pragma clang diagnostic ignored "-Wdeprecated-declarations"
+#endif
+
 static int
 __archive_md5init(archive_md5_ctx *ctx)
 {
@@ -218,13 +225,46 @@ __archive_md5final(archive_md5_ctx *ctx, void *md)
  return (ARCHIVE_OK);
 }

+#if defined(__clang__)
+#pragma clang diagnostic pop
+#endif
+
+#elif defined(ARCHIVE_CRYPTO_MD5_WIN)
+
+static int
+__archive_md5init(archive_md5_ctx *ctx)
+{
+#if defined(HAVE_BCRYPT_H) && _WIN32_WINNT >= _WIN32_WINNT_VISTA
+  return (win_crypto_init(ctx, BCRYPT_MD5_ALGORITHM));
+#else
+  return (win_crypto_init(ctx, PROV_RSA_FULL, CALG_MD5));
+#endif
+}
+
+static int
+__archive_md5update(archive_md5_ctx *ctx, const void *indata,
+    size_t insize)
+{
+  return (win_crypto_Update(ctx, indata, insize));
+}
+
+static int
+__archive_md5final(archive_md5_ctx *ctx, void *md)
+{
+  return (win_crypto_Final(md, 16, ctx));
+}
+
 #elif defined(ARCHIVE_CRYPTO_MD5_MBEDTLS)

 static int
 __archive_md5init(archive_md5_ctx *ctx)
 {
  mbedtls_md5_init(ctx);
+#if MBEDTLS_VERSION_NUMBER > 0x03000000
+  if (mbedtls_md5_starts(ctx) == 0)
+#else
  if (mbedtls_md5_starts_ret(ctx) == 0)
+#endif
    return (ARCHIVE_OK);
  else
    return (ARCHIVE_FATAL);
@@ -234,7 +274,11 @@ static int
 __archive_md5update(archive_md5_ctx *ctx, const void *indata,
    size_t insize)
 {
+#if MBEDTLS_VERSION_NUMBER > 0x03000000
+  if (mbedtls_md5_update(ctx, indata, insize) == 0)
+#else
  if (mbedtls_md5_update_ret(ctx, indata, insize) == 0)
+#endif
    return (ARCHIVE_OK);
  else
    return (ARCHIVE_FATAL);
@@ -243,7 +287,11 @@ __archive_md5update(archive_md5_ctx *ctx, const void *indata,
 static int
 __archive_md5final(archive_md5_ctx *ctx, void *md)
 {
+#if MBEDTLS_VERSION_NUMBER > 0x03000000
+  if (mbedtls_md5_finish(ctx, md) == 0) {
+#else
  if (mbedtls_md5_finish_ret(ctx, md) == 0) {
+#endif
    mbedtls_md5_free(ctx);
    return (ARCHIVE_OK);
  } else {
@@ -311,31 +359,6 @@ __archive_md5final(archive_md5_ctx *ctx, void *md)
  return (ARCHIVE_OK);
 }

-#elif defined(ARCHIVE_CRYPTO_MD5_WIN)
-
-static int
-__archive_md5init(archive_md5_ctx *ctx)
-{
-#if defined(HAVE_BCRYPT_H) && _WIN32_WINNT >= _WIN32_WINNT_VISTA
-  return (win_crypto_init(ctx, BCRYPT_MD5_ALGORITHM));
-#else
-  return (win_crypto_init(ctx, PROV_RSA_FULL, CALG_MD5));
-#endif
-}
-
-static int
-__archive_md5update(archive_md5_ctx *ctx, const void *indata,
-    size_t insize)
-{
-  return (win_crypto_Update(ctx, indata, insize));
-}
-
-static int
-__archive_md5final(archive_md5_ctx *ctx, void *md)
-{
-  return (win_crypto_Final(md, 16, ctx));
-}
-
 #else

 static int
@@ -420,7 +443,11 @@ static int
 __archive_ripemd160init(archive_rmd160_ctx *ctx)
 {
  mbedtls_ripemd160_init(ctx);
+#if MBEDTLS_VERSION_NUMBER > 0x03000000
+  if (mbedtls_ripemd160_starts(ctx) == 0)
+#else
  if (mbedtls_ripemd160_starts_ret(ctx) == 0)
+#endif
    return (ARCHIVE_OK);
  else
    return (ARCHIVE_FATAL);
@@ -430,7 +457,11 @@ static int
 __archive_ripemd160update(archive_rmd160_ctx *ctx, const void *indata,
    size_t insize)
 {
+#if MBEDTLS_VERSION_NUMBER > 0x03000000
+  if (mbedtls_ripemd160_update(ctx, indata, insize) == 0)
+#else
  if (mbedtls_ripemd160_update_ret(ctx, indata, insize) == 0)
+#endif
    return (ARCHIVE_OK);
  else
    return (ARCHIVE_FATAL);
@@ -439,7 +470,11 @@ __archive_ripemd160update(archive_rmd160_ctx *ctx, const void *indata,
 static int
 __archive_ripemd160final(archive_rmd160_ctx *ctx, void *md)
 {
+#if MBEDTLS_VERSION_NUMBER > 0x03000000
+  if (mbedtls_ripemd160_finish(ctx, md) == 0) {
+#else
  if (mbedtls_ripemd160_finish_ret(ctx, md) == 0) {
+#endif
    mbedtls_ripemd160_free(ctx);
    return (ARCHIVE_OK);
  } else {
@@ -605,13 +640,42 @@ __archive_sha1final(archive_sha1_ctx *ctx, void *md)
  return (ARCHIVE_OK);
 }

+#elif defined(ARCHIVE_CRYPTO_SHA1_WIN)
+
+static int
+__archive_sha1init(archive_sha1_ctx *ctx)
+{
+#if defined(HAVE_BCRYPT_H) && _WIN32_WINNT >= _WIN32_WINNT_VISTA
+  return (win_crypto_init(ctx, BCRYPT_SHA1_ALGORITHM));
+#else
+  return (win_crypto_init(ctx, PROV_RSA_FULL, CALG_SHA1));
+#endif
+}
+
+static int
+__archive_sha1update(archive_sha1_ctx *ctx, const void *indata,
+    size_t insize)
+{
+  return (win_crypto_Update(ctx, indata, insize));
+}
+
+static int
+__archive_sha1final(archive_sha1_ctx *ctx, void *md)
+{
+  return (win_crypto_Final(md, 20, ctx));
+}
+
 #elif defined(ARCHIVE_CRYPTO_SHA1_MBEDTLS)

 static int
 __archive_sha1init(archive_sha1_ctx *ctx)
 {
  mbedtls_sha1_init(ctx);
+#if MBEDTLS_VERSION_NUMBER > 0x03000000
+  if (mbedtls_sha1_starts(ctx) == 0)
+#else
  if (mbedtls_sha1_starts_ret(ctx) == 0)
+#endif
    return (ARCHIVE_OK);
  else
    return (ARCHIVE_FATAL);
@@ -621,7 +685,11 @@ static int
 __archive_sha1update(archive_sha1_ctx *ctx, const void *indata,
    size_t insize)
 {
+#if MBEDTLS_VERSION_NUMBER > 0x03000000
+  if (mbedtls_sha1_update(ctx, indata, insize) == 0)
+#else
  if (mbedtls_sha1_update_ret(ctx, indata, insize) == 0)
+#endif
    return (ARCHIVE_OK);
  else
    return (ARCHIVE_FATAL);
@@ -630,7 +698,11 @@ __archive_sha1update(archive_sha1_ctx *ctx, const void *indata,
 static int
 __archive_sha1final(archive_sha1_ctx *ctx, void *md)
 {
+#if MBEDTLS_VERSION_NUMBER > 0x03000000
+  if (mbedtls_sha1_finish(ctx, md) == 0) {
+#else
  if (mbedtls_sha1_finish_ret(ctx, md) == 0) {
+#endif
    mbedtls_sha1_free(ctx);
    return (ARCHIVE_OK);
  } else {
@@ -698,31 +770,6 @@ __archive_sha1final(archive_sha1_ctx *ctx, void *md)
  return (ARCHIVE_OK);
 }

-#elif defined(ARCHIVE_CRYPTO_SHA1_WIN)
-
-static int
-__archive_sha1init(archive_sha1_ctx *ctx)
-{
-#if defined(HAVE_BCRYPT_H) && _WIN32_WINNT >= _WIN32_WINNT_VISTA
-  return (win_crypto_init(ctx, BCRYPT_SHA1_ALGORITHM));
-#else
-  return (win_crypto_init(ctx, PROV_RSA_FULL, CALG_SHA1));
-#endif
-}
-
-static int
-__archive_sha1update(archive_sha1_ctx *ctx, const void *indata,
-    size_t insize)
-{
-  return (win_crypto_Update(ctx, indata, insize));
-}
-
-static int
-__archive_sha1final(archive_sha1_ctx *ctx, void *md)
-{
-  return (win_crypto_Final(md, 20, ctx));
-}
-
 #else

 static int
@@ -873,13 +920,42 @@ __archive_sha256final(archive_sha256_ctx *ctx, void *md)
  return (ARCHIVE_OK);
 }

+#elif defined(ARCHIVE_CRYPTO_SHA256_WIN)
+
+static int
+__archive_sha256init(archive_sha256_ctx *ctx)
+{
+#if defined(HAVE_BCRYPT_H) && _WIN32_WINNT >= _WIN32_WINNT_VISTA
+  return (win_crypto_init(ctx, BCRYPT_SHA256_ALGORITHM));
+#else
+  return (win_crypto_init(ctx, PROV_RSA_AES, CALG_SHA_256));
+#endif
+}
+
+static int
+__archive_sha256update(archive_sha256_ctx *ctx, const void *indata,
+    size_t insize)
+{
+  return (win_crypto_Update(ctx, indata, insize));
+}
+
+static int
+__archive_sha256final(archive_sha256_ctx *ctx, void *md)
+{
+  return (win_crypto_Final(md, 32, ctx));
+}
+
 #elif defined(ARCHIVE_CRYPTO_SHA256_MBEDTLS)

 static int
 __archive_sha256init(archive_sha256_ctx *ctx)
 {
  mbedtls_sha256_init(ctx);
+#if MBEDTLS_VERSION_NUMBER > 0x03000000
+  if (mbedtls_sha256_starts(ctx, 0) == 0)
+#else
  if (mbedtls_sha256_starts_ret(ctx, 0) == 0)
+#endif
    return (ARCHIVE_OK);
  else
    return (ARCHIVE_FATAL);
@@ -889,7 +965,11 @@ static int
 __archive_sha256update(archive_sha256_ctx *ctx, const void *indata,
    size_t insize)
 {
+#if MBEDTLS_VERSION_NUMBER > 0x03000000
+  if (mbedtls_sha256_update(ctx, indata, insize) == 0)
+#else
  if (mbedtls_sha256_update_ret(ctx, indata, insize) == 0)
+#endif
    return (ARCHIVE_OK);
  else
    return (ARCHIVE_FATAL);
@@ -898,7 +978,11 @@ __archive_sha256update(archive_sha256_ctx *ctx, const void *indata,
 static int
 __archive_sha256final(archive_sha256_ctx *ctx, void *md)
 {
+#if MBEDTLS_VERSION_NUMBER > 0x03000000
+  if (mbedtls_sha256_finish(ctx, md) == 0) {
+#else
  if (mbedtls_sha256_finish_ret(ctx, md) == 0) {
+#endif
    mbedtls_sha256_free(ctx);
    return (ARCHIVE_OK);
  } else {
@@ -962,31 +1046,6 @@ __archive_sha256final(archive_sha256_ctx *ctx, void *md)
  return (ARCHIVE_OK);
 }

-#elif defined(ARCHIVE_CRYPTO_SHA256_WIN)
-
-static int
-__archive_sha256init(archive_sha256_ctx *ctx)
-{
-#if defined(HAVE_BCRYPT_H) && _WIN32_WINNT >= _WIN32_WINNT_VISTA
-  return (win_crypto_init(ctx, BCRYPT_SHA256_ALGORITHM));
-#else
-  return (win_crypto_init(ctx, PROV_RSA_AES, CALG_SHA_256));
-#endif
-}
-
-static int
-__archive_sha256update(archive_sha256_ctx *ctx, const void *indata,
-    size_t insize)
-{
-  return (win_crypto_Update(ctx, indata, insize));
-}
-
-static int
-__archive_sha256final(archive_sha256_ctx *ctx, void *md)
-{
-  return (win_crypto_Final(md, 32, ctx));
-}
-
 #else

 static int
@@ -1113,13 +1172,42 @@ __archive_sha384final(archive_sha384_ctx *ctx, void *md)
  return (ARCHIVE_OK);
 }

+#elif defined(ARCHIVE_CRYPTO_SHA384_WIN)
+
+static int
+__archive_sha384init(archive_sha384_ctx *ctx)
+{
+#if defined(HAVE_BCRYPT_H) && _WIN32_WINNT >= _WIN32_WINNT_VISTA
+  return (win_crypto_init(ctx, BCRYPT_SHA384_ALGORITHM));
+#else
+  return (win_crypto_init(ctx, PROV_RSA_AES, CALG_SHA_384));
+#endif
+}
+
+static int
+__archive_sha384update(archive_sha384_ctx *ctx, const void *indata,
+    size_t insize)
+{
+  return (win_crypto_Update(ctx, indata, insize));
+}
+
+static int
+__archive_sha384final(archive_sha384_ctx *ctx, void *md)
+{
+  return (win_crypto_Final(md, 48, ctx));
+}
+
 #elif defined(ARCHIVE_CRYPTO_SHA384_MBEDTLS)

 static int
 __archive_sha384init(archive_sha384_ctx *ctx)
 {
  mbedtls_sha512_init(ctx);
+#if MBEDTLS_VERSION_NUMBER > 0x03000000
+  if (mbedtls_sha512_starts(ctx, 1) == 0)
+#else
  if (mbedtls_sha512_starts_ret(ctx, 1) == 0)
+#endif
    return (ARCHIVE_OK);
  else
    return (ARCHIVE_FATAL);
@@ -1129,7 +1217,11 @@ static int
 __archive_sha384update(archive_sha384_ctx *ctx, const void *indata,
    size_t insize)
 {
+#if MBEDTLS_VERSION_NUMBER > 0x03000000
+  if (mbedtls_sha512_update(ctx, indata, insize) == 0)
+#else
  if (mbedtls_sha512_update_ret(ctx, indata, insize) == 0)
+#endif
    return (ARCHIVE_OK);
  else
    return (ARCHIVE_FATAL);
@@ -1138,7 +1230,11 @@ __archive_sha384update(archive_sha384_ctx *ctx, const void *indata,
 static int
 __archive_sha384final(archive_sha384_ctx *ctx, void *md)
 {
+#if MBEDTLS_VERSION_NUMBER > 0x03000000
+  if (mbedtls_sha512_finish(ctx, md) == 0) {
+#else
  if (mbedtls_sha512_finish_ret(ctx, md) == 0) {
+#endif
    mbedtls_sha512_free(ctx);
    return (ARCHIVE_OK);
  } else {
@@ -1202,31 +1298,6 @@ __archive_sha384final(archive_sha384_ctx *ctx, void *md)
  return (ARCHIVE_OK);
 }

-#elif defined(ARCHIVE_CRYPTO_SHA384_WIN)
-
-static int
-__archive_sha384init(archive_sha384_ctx *ctx)
-{
-#if defined(HAVE_BCRYPT_H) && _WIN32_WINNT >= _WIN32_WINNT_VISTA
-  return (win_crypto_init(ctx, BCRYPT_SHA384_ALGORITHM));
-#else
-  return (win_crypto_init(ctx, PROV_RSA_AES, CALG_SHA_384));
-#endif
-}
-
-static int
-__archive_sha384update(archive_sha384_ctx *ctx, const void *indata,
-    size_t insize)
-{
-  return (win_crypto_Update(ctx, indata, insize));
-}
-
-static int
-__archive_sha384final(archive_sha384_ctx *ctx, void *md)
-{
-  return (win_crypto_Final(md, 48, ctx));
-}
-
 #else

 static int
@@ -1377,13 +1448,42 @@ __archive_sha512final(archive_sha512_ctx *ctx, void *md)
  return (ARCHIVE_OK);
 }

+#elif defined(ARCHIVE_CRYPTO_SHA512_WIN)
+
+static int
+__archive_sha512init(archive_sha512_ctx *ctx)
+{
+#if defined(HAVE_BCRYPT_H) && _WIN32_WINNT >= _WIN32_WINNT_VISTA
+  return (win_crypto_init(ctx, BCRYPT_SHA512_ALGORITHM));
+#else
+  return (win_crypto_init(ctx, PROV_RSA_AES, CALG_SHA_512));
+#endif
+}
+
+static int
+__archive_sha512update(archive_sha512_ctx *ctx, const void *indata,
+    size_t insize)
+{
+  return (win_crypto_Update(ctx, indata, insize));
+}
+
+static int
+__archive_sha512final(archive_sha512_ctx *ctx, void *md)
+{
+  return (win_crypto_Final(md, 64, ctx));
+}
+
 #elif defined(ARCHIVE_CRYPTO_SHA512_MBEDTLS)

 static int
 __archive_sha512init(archive_sha512_ctx *ctx)
 {
  mbedtls_sha512_init(ctx);
+#if MBEDTLS_VERSION_NUMBER > 0x03000000
+  if (mbedtls_sha512_starts(ctx, 0) == 0)
+#else
  if (mbedtls_sha512_starts_ret(ctx, 0) == 0)
+#endif
    return (ARCHIVE_OK);
  else
    return (ARCHIVE_FATAL);
@@ -1393,7 +1493,11 @@ static int
 __archive_sha512update(archive_sha512_ctx *ctx, const void *indata,
    size_t insize)
 {
+#if MBEDTLS_VERSION_NUMBER > 0x03000000
+  if (mbedtls_sha512_update(ctx, indata, insize) == 0)
+#else
  if (mbedtls_sha512_update_ret(ctx, indata, insize) == 0)
+#endif
    return (ARCHIVE_OK);
  else
    return (ARCHIVE_FATAL);
@@ -1402,7 +1506,11 @@ __archive_sha512update(archive_sha512_ctx *ctx, const void *indata,
 static int
 __archive_sha512final(archive_sha512_ctx *ctx, void *md)
 {
+#if MBEDTLS_VERSION_NUMBER > 0x03000000
+  if (mbedtls_sha512_finish(ctx, md) == 0) {
+#else
  if (mbedtls_sha512_finish_ret(ctx, md) == 0) {
+#endif
    mbedtls_sha512_free(ctx);
    return (ARCHIVE_OK);
  } else {
@@ -1466,31 +1574,6 @@ __archive_sha512final(archive_sha512_ctx *ctx, void *md)
  return (ARCHIVE_OK);
 }

-#elif defined(ARCHIVE_CRYPTO_SHA512_WIN)
-
-static int
-__archive_sha512init(archive_sha512_ctx *ctx)
-{
-#if defined(HAVE_BCRYPT_H) && _WIN32_WINNT >= _WIN32_WINNT_VISTA
-  return (win_crypto_init(ctx, BCRYPT_SHA512_ALGORITHM));
-#else
-  return (win_crypto_init(ctx, PROV_RSA_AES, CALG_SHA_512));
-#endif
-}
-
-static int
-__archive_sha512update(archive_sha512_ctx *ctx, const void *indata,
-    size_t insize)
-{
-  return (win_crypto_Update(ctx, indata, insize));
-}
-
-static int
-__archive_sha512final(archive_sha512_ctx *ctx, void *md)
-{
-  return (win_crypto_Final(md, 64, ctx));
-}
-
 #else

 static int
@@ -1525,11 +1608,12 @@ __archive_sha512final(archive_sha512_ctx *ctx, void *md)
 * 1. libc
 * 2. libc2
 * 3. libc3
- * 4. libSystem
- * 5. Nettle
- * 6. OpenSSL
- * 7. libmd
- * 8. Windows API
+ * 4. libmd
+ * 5. libSystem
+ * 6. Windows API
+ * 7. mbedTLS
+ * 8. Nettle
+ * 9. OpenSSL
 */
 const struct archive_digest __archive_digest =
 {
@@ -113,6 +113,7 @@
  defined(ARCHIVE_CRYPTO_SHA384_LIBSYSTEM) ||\
  defined(ARCHIVE_CRYPTO_SHA512_LIBSYSTEM)
 #include <CommonCrypto/CommonDigest.h>
+#define	ARCHIVE_CRYPTO_CommonCrypto 1
 #endif

 /* mbed TLS crypto headers */
@@ -167,6 +168,7 @@
 #if defined(HAVE_BCRYPT_H) && _WIN32_WINNT >= _WIN32_WINNT_VISTA
 /* don't use bcrypt when XP needs to be supported */
 #include <bcrypt.h>
+#define	ARCHIVE_CRYPTO_CNG 1
 typedef struct {
  int   valid;
  BCRYPT_ALG_HANDLE  hAlg;
@@ -175,6 +177,7 @@ typedef struct {
 #else
 #include <windows.h>
 #include <wincrypt.h>
+#define	ARCHIVE_CRYPTO_WINCRYPT 1
 typedef struct {
  int   valid;
  HCRYPTPROV  cryptProv;
@@ -190,14 +193,16 @@ typedef MD5_CTX archive_md5_ctx;
 typedef MD5_CTX archive_md5_ctx;
 #elif defined(ARCHIVE_CRYPTO_MD5_LIBSYSTEM)
 typedef CC_MD5_CTX archive_md5_ctx;
+#elif defined(ARCHIVE_CRYPTO_MD5_WIN)
+typedef Digest_CTX archive_md5_ctx;
 #elif defined(ARCHIVE_CRYPTO_MD5_MBEDTLS)
+#define	ARCHIVE_CRYPTO_MBED 1
 typedef mbedtls_md5_context archive_md5_ctx;
 #elif defined(ARCHIVE_CRYPTO_MD5_NETTLE)
+#define	ARCHIVE_CRYPTO_NETTLE 1
 typedef struct md5_ctx archive_md5_ctx;
 #elif defined(ARCHIVE_CRYPTO_MD5_OPENSSL)
 typedef EVP_MD_CTX *archive_md5_ctx;
-#elif defined(ARCHIVE_CRYPTO_MD5_WIN)
-typedef Digest_CTX archive_md5_ctx;
 #else
 typedef unsigned char archive_md5_ctx;
 #endif
@@ -207,8 +212,10 @@ typedef RMD160_CTX archive_rmd160_ctx;
 #elif defined(ARCHIVE_CRYPTO_RMD160_LIBMD)
 typedef RIPEMD160_CTX archive_rmd160_ctx;
 #elif defined(ARCHIVE_CRYPTO_RMD160_MBEDTLS)
+#define	ARCHIVE_CRYPTO_MBED 1
 typedef mbedtls_ripemd160_context archive_rmd160_ctx;
 #elif defined(ARCHIVE_CRYPTO_RMD160_NETTLE)
+#define	ARCHIVE_CRYPTO_NETTLE 1
 typedef struct ripemd160_ctx archive_rmd160_ctx;
 #elif defined(ARCHIVE_CRYPTO_RMD160_OPENSSL)
 typedef EVP_MD_CTX *archive_rmd160_ctx;
@@ -222,14 +229,16 @@ typedef SHA1_CTX archive_sha1_ctx;
 typedef SHA1_CTX archive_sha1_ctx;
 #elif defined(ARCHIVE_CRYPTO_SHA1_LIBSYSTEM)
 typedef CC_SHA1_CTX archive_sha1_ctx;
+#elif defined(ARCHIVE_CRYPTO_SHA1_WIN)
+typedef Digest_CTX archive_sha1_ctx;
 #elif defined(ARCHIVE_CRYPTO_SHA1_MBEDTLS)
+#define	ARCHIVE_CRYPTO_MBED 1
 typedef mbedtls_sha1_context archive_sha1_ctx;
 #elif defined(ARCHIVE_CRYPTO_SHA1_NETTLE)
+#define	ARCHIVE_CRYPTO_NETTLE 1
 typedef struct sha1_ctx archive_sha1_ctx;
 #elif defined(ARCHIVE_CRYPTO_SHA1_OPENSSL)
 typedef EVP_MD_CTX *archive_sha1_ctx;
-#elif defined(ARCHIVE_CRYPTO_SHA1_WIN)
-typedef Digest_CTX archive_sha1_ctx;
 #else
 typedef unsigned char archive_sha1_ctx;
 #endif
@@ -244,14 +253,16 @@ typedef SHA2_CTX archive_sha256_ctx;
 typedef SHA256_CTX archive_sha256_ctx;
 #elif defined(ARCHIVE_CRYPTO_SHA256_LIBSYSTEM)
 typedef CC_SHA256_CTX archive_sha256_ctx;
+#elif defined(ARCHIVE_CRYPTO_SHA256_WIN)
+typedef Digest_CTX archive_sha256_ctx;
 #elif defined(ARCHIVE_CRYPTO_SHA256_MBEDTLS)
+#define	ARCHIVE_CRYPTO_MBED 1
 typedef mbedtls_sha256_context archive_sha256_ctx;
 #elif defined(ARCHIVE_CRYPTO_SHA256_NETTLE)
+#define	ARCHIVE_CRYPTO_NETTLE 1
 typedef struct sha256_ctx archive_sha256_ctx;
 #elif defined(ARCHIVE_CRYPTO_SHA256_OPENSSL)
 typedef EVP_MD_CTX *archive_sha256_ctx;
-#elif defined(ARCHIVE_CRYPTO_SHA256_WIN)
-typedef Digest_CTX archive_sha256_ctx;
 #else
 typedef unsigned char archive_sha256_ctx;
 #endif
@@ -264,14 +275,16 @@ typedef SHA384_CTX archive_sha384_ctx;
 typedef SHA2_CTX archive_sha384_ctx;
 #elif defined(ARCHIVE_CRYPTO_SHA384_LIBSYSTEM)
 typedef CC_SHA512_CTX archive_sha384_ctx;
+#elif defined(ARCHIVE_CRYPTO_SHA384_WIN)
+typedef Digest_CTX archive_sha384_ctx;
 #elif defined(ARCHIVE_CRYPTO_SHA384_MBEDTLS)
+#define	ARCHIVE_CRYPTO_MBED 1
 typedef mbedtls_sha512_context archive_sha384_ctx;
 #elif defined(ARCHIVE_CRYPTO_SHA384_NETTLE)
+#define	ARCHIVE_CRYPTO_NETTLE 1
 typedef struct sha384_ctx archive_sha384_ctx;
 #elif defined(ARCHIVE_CRYPTO_SHA384_OPENSSL)
 typedef EVP_MD_CTX *archive_sha384_ctx;
-#elif defined(ARCHIVE_CRYPTO_SHA384_WIN)
-typedef Digest_CTX archive_sha384_ctx;
 #else
 typedef unsigned char archive_sha384_ctx;
 #endif
@@ -286,14 +299,16 @@ typedef SHA2_CTX archive_sha512_ctx;
 typedef SHA512_CTX archive_sha512_ctx;
 #elif defined(ARCHIVE_CRYPTO_SHA512_LIBSYSTEM)
 typedef CC_SHA512_CTX archive_sha512_ctx;
+#elif defined(ARCHIVE_CRYPTO_SHA512_WIN)
+typedef Digest_CTX archive_sha512_ctx;
 #elif defined(ARCHIVE_CRYPTO_SHA512_MBEDTLS)
+#define	ARCHIVE_CRYPTO_MBED 1
 typedef mbedtls_sha512_context archive_sha512_ctx;
 #elif defined(ARCHIVE_CRYPTO_SHA512_NETTLE)
+#define	ARCHIVE_CRYPTO_NETTLE 1
 typedef struct sha512_ctx archive_sha512_ctx;
 #elif defined(ARCHIVE_CRYPTO_SHA512_OPENSSL)
 typedef EVP_MD_CTX *archive_sha512_ctx;
-#elif defined(ARCHIVE_CRYPTO_SHA512_WIN)
-typedef Digest_CTX archive_sha512_ctx;
 #else
 typedef unsigned char archive_sha512_ctx;
 #endif
@@ -262,7 +262,7 @@ translate_acl(struct archive_read_disk *a,
 			}
 			for (i = 0; i < acl_nfs4_flag_map_size; ++i) {
 				r = acl_get_flag_np(acl_flagset,
-				    acl_nfs4_flag_map[i].p_perm);
+				    (acl_flag_t)acl_nfs4_flag_map[i].p_perm);
 				if (r == -1) {
 					archive_set_error(&a->archive, errno,
 					    "Failed to check flag in a NFSv4 "
@@ -517,7 +517,7 @@ set_acl(struct archive *a, int fd, const char *name,
 			for (i = 0; i < acl_nfs4_flag_map_size; ++i) {
 				if (ae_permset & acl_nfs4_flag_map[i].a_perm) {
 					if (acl_add_flag_np(acl_flagset,
-					    acl_nfs4_flag_map[i].p_perm) != 0) {
+					    (acl_flag_t)acl_nfs4_flag_map[i].p_perm) != 0) {
 						archive_set_error(a, errno,
 						    "Failed to add flag to "
 						    "NFSv4 ACL flagset");
@@ -275,7 +275,7 @@ archive_entry_new2(struct archive *a)
 * Functions for reading fields from an archive_entry.
 */

-time_t
+__LA_TIME_T
 archive_entry_atime(struct archive_entry *entry)
 {
 	return (entry->ae_stat.aest_atime);
@@ -293,7 +293,7 @@ archive_entry_atime_is_set(struct archive_entry *entry)
 	return (entry->ae_set & AE_SET_ATIME);
 }

-time_t
+__LA_TIME_T
 archive_entry_birthtime(struct archive_entry *entry)
 {
 	return (entry->ae_stat.aest_birthtime);
@@ -311,7 +311,7 @@ archive_entry_birthtime_is_set(struct archive_entry *entry)
 	return (entry->ae_set & AE_SET_BIRTHTIME);
 }

-time_t
+__LA_TIME_T
 archive_entry_ctime(struct archive_entry *entry)
 {
 	return (entry->ae_stat.aest_ctime);
@@ -329,7 +329,7 @@ archive_entry_ctime_nsec(struct archive_entry *entry)
 	return (entry->ae_stat.aest_ctime_nsec);
 }

-dev_t
+__LA_DEV_T
 archive_entry_dev(struct archive_entry *entry)
 {
 	if (entry->ae_stat.aest_dev_is_broken_down)
@@ -345,7 +345,7 @@ archive_entry_dev_is_set(struct archive_entry *entry)
 	return (entry->ae_set & AE_SET_DEV);
 }

-dev_t
+__LA_DEV_T
 archive_entry_devmajor(struct archive_entry *entry)
 {
 	if (entry->ae_stat.aest_dev_is_broken_down)
@@ -354,7 +354,7 @@ archive_entry_devmajor(struct archive_entry *entry)
 		return major(entry->ae_stat.aest_dev);
 }

-dev_t
+__LA_DEV_T
 archive_entry_devminor(struct archive_entry *entry)
 {
 	if (entry->ae_stat.aest_dev_is_broken_down)
@@ -568,7 +568,7 @@ archive_entry_mode(struct archive_entry *entry)
 	return (entry->acl.mode);
 }

-time_t
+__LA_TIME_T
 archive_entry_mtime(struct archive_entry *entry)
 {
 	return (entry->ae_stat.aest_mtime);
@@ -667,7 +667,7 @@ archive_entry_rdev_is_set(struct archive_entry *entry)
 	return (entry->ae_set & AE_SET_RDEV);
 }

-dev_t
+__LA_DEV_T
 archive_entry_rdev(struct archive_entry *entry)
 {
 	if (archive_entry_rdev_is_set(entry)) {
@@ -681,7 +681,7 @@ archive_entry_rdev(struct archive_entry *entry)
 	}
 }

-dev_t
+__LA_DEV_T
 archive_entry_rdevmajor(struct archive_entry *entry)
 {
 	if (archive_entry_rdev_is_set(entry)) {
@@ -694,7 +694,7 @@ archive_entry_rdevmajor(struct archive_entry *entry)
 	}
 }

-dev_t
+__LA_DEV_T
 archive_entry_rdevminor(struct archive_entry *entry)
 {
 	if (archive_entry_rdev_is_set(entry)) {
@@ -984,7 +984,9 @@ void
 archive_entry_set_ino(struct archive_entry *entry, la_int64_t ino)
 {
 	if (ino < 0) {
-		ino = 0;
+		entry->stat_valid = 0;
+		entry->ae_set &= ~AE_SET_INO;
+		return;
 	}
 	entry->stat_valid = 0;
 	entry->ae_set |= AE_SET_INO;
@@ -995,7 +997,9 @@ void
 archive_entry_set_ino64(struct archive_entry *entry, la_int64_t ino)
 {
 	if (ino < 0) {
-		ino = 0;
+		entry->stat_valid = 0;
+		entry->ae_set &= ~AE_SET_INO;
+		return;
 	}
 	entry->stat_valid = 0;
 	entry->ae_set |= AE_SET_INO;
@@ -1088,7 +1092,7 @@ _archive_entry_copy_hardlink_l(struct archive_entry *entry,
 }

 void
-archive_entry_set_atime(struct archive_entry *entry, time_t t, long ns)
+archive_entry_set_atime(struct archive_entry *entry, __LA_TIME_T t, long ns)
 {
 	FIX_NS(t, ns);
 	entry->stat_valid = 0;
@@ -1105,7 +1109,7 @@ archive_entry_unset_atime(struct archive_entry *entry)
 }

 void
-archive_entry_set_birthtime(struct archive_entry *entry, time_t t, long ns)
+archive_entry_set_birthtime(struct archive_entry *entry, __LA_TIME_T t, long ns)
 {
 	FIX_NS(t, ns);
 	entry->stat_valid = 0;
@@ -1122,7 +1126,7 @@ archive_entry_unset_birthtime(struct archive_entry *entry)
 }

 void
-archive_entry_set_ctime(struct archive_entry *entry, time_t t, long ns)
+archive_entry_set_ctime(struct archive_entry *entry, __LA_TIME_T t, long ns)
 {
 	FIX_NS(t, ns);
 	entry->stat_valid = 0;
@@ -1139,7 +1143,7 @@ archive_entry_unset_ctime(struct archive_entry *entry)
 }

 void
-archive_entry_set_dev(struct archive_entry *entry, dev_t d)
+archive_entry_set_dev(struct archive_entry *entry, __LA_DEV_T d)
 {
 	entry->stat_valid = 0;
 	entry->ae_set |= AE_SET_DEV;
@@ -1148,7 +1152,7 @@ archive_entry_set_dev(struct archive_entry *entry, dev_t d)
 }

 void
-archive_entry_set_devmajor(struct archive_entry *entry, dev_t m)
+archive_entry_set_devmajor(struct archive_entry *entry, __LA_DEV_T m)
 {
 	entry->stat_valid = 0;
 	entry->ae_set |= AE_SET_DEV;
@@ -1157,7 +1161,7 @@ archive_entry_set_devmajor(struct archive_entry *entry, dev_t m)
 }

 void
-archive_entry_set_devminor(struct archive_entry *entry, dev_t m)
+archive_entry_set_devminor(struct archive_entry *entry, __LA_DEV_T m)
 {
 	entry->stat_valid = 0;
 	entry->ae_set |= AE_SET_DEV;
@@ -1243,7 +1247,7 @@ archive_entry_set_mode(struct archive_entry *entry, mode_t m)
 }

 void
-archive_entry_set_mtime(struct archive_entry *entry, time_t t, long ns)
+archive_entry_set_mtime(struct archive_entry *entry, __LA_TIME_T t, long ns)
 {
 	FIX_NS(t, ns);
 	entry->stat_valid = 0;
@@ -1319,7 +1323,7 @@ archive_entry_set_perm(struct archive_entry *entry, mode_t p)
 }

 void
-archive_entry_set_rdev(struct archive_entry *entry, dev_t m)
+archive_entry_set_rdev(struct archive_entry *entry, __LA_DEV_T m)
 {
 	entry->stat_valid = 0;
 	entry->ae_stat.aest_rdev = m;
@@ -1330,7 +1334,7 @@ archive_entry_set_rdev(struct archive_entry *entry, dev_t m)
 }

 void
-archive_entry_set_rdevmajor(struct archive_entry *entry, dev_t m)
+archive_entry_set_rdevmajor(struct archive_entry *entry, __LA_DEV_T m)
 {
 	entry->stat_valid = 0;
 	entry->ae_stat.aest_rdev_is_broken_down = 1;
@@ -1340,7 +1344,7 @@ archive_entry_set_rdevmajor(struct archive_entry *entry, dev_t m)
 }

 void
-archive_entry_set_rdevminor(struct archive_entry *entry, dev_t m)
+archive_entry_set_rdevminor(struct archive_entry *entry, __LA_DEV_T m)
 {
 	entry->stat_valid = 0;
 	entry->ae_stat.aest_rdev_is_broken_down = 1;
@@ -1603,21 +1607,27 @@ archive_entry_set_digest(struct archive_entry *entry, int type,
 	switch (type) {
 	case ARCHIVE_ENTRY_DIGEST_MD5:
 		copy_digest(entry, md5, digest);
+		entry->mset_digest |= AE_MSET_DIGEST_MD5;
 		break;
 	case ARCHIVE_ENTRY_DIGEST_RMD160:
 		copy_digest(entry, rmd160, digest);
+		entry->mset_digest |= AE_MSET_DIGEST_RMD160;
 		break;
 	case ARCHIVE_ENTRY_DIGEST_SHA1:
 		copy_digest(entry, sha1, digest);
+		entry->mset_digest |= AE_MSET_DIGEST_SHA1;
 		break;
 	case ARCHIVE_ENTRY_DIGEST_SHA256:
 		copy_digest(entry, sha256, digest);
+		entry->mset_digest |= AE_MSET_DIGEST_SHA256;
 		break;
 	case ARCHIVE_ENTRY_DIGEST_SHA384:
 		copy_digest(entry, sha384, digest);
+		entry->mset_digest |= AE_MSET_DIGEST_SHA384;
 		break;
 	case ARCHIVE_ENTRY_DIGEST_SHA512:
 		copy_digest(entry, sha512, digest);
+		entry->mset_digest |= AE_MSET_DIGEST_SHA512;
 		break;
 	default:
 		return ARCHIVE_WARN;
@@ -28,7 +28,7 @@
 #define	ARCHIVE_ENTRY_H_INCLUDED

 /* Note: Compiler will complain if this does not match archive.h! */
-#define	ARCHIVE_VERSION_NUMBER 3007007
+#define	ARCHIVE_VERSION_NUMBER 3008001

 /*
 * Note: archive_entry.h is for use outside of libarchive; the
@@ -40,8 +40,11 @@

 #include <sys/types.h>
 #include <stddef.h>  /* for wchar_t */
-#include <stdint.h>
+#include <stdint.h>  /* for C99 int64_t, etc. */
+#if ARCHIVE_VERSION_NUMBER < 4000000
+/* time_t is slated to be removed from public includes in 4.0 */
 #include <time.h>
+#endif

 #if defined(_WIN32) && !defined(__CYGWIN__)
 #include <windows.h>
@@ -55,12 +58,15 @@
 #define __LA_INT64_T_DEFINED
 # if defined(_WIN32) && !defined(__CYGWIN__) && !defined(__WATCOMC__)
 typedef __int64 la_int64_t;
+typedef unsigned __int64 la_uint64_t;
 # else
 #include <unistd.h>
 #  if defined(_SCO_DS) || defined(__osf__)
 typedef long long la_int64_t;
+typedef unsigned long long la_uint64_t;
 #  else
 typedef int64_t la_int64_t;
+typedef uint64_t la_uint64_t;
 #  endif
 # endif
 #endif
@@ -96,6 +102,30 @@ typedef ssize_t la_ssize_t;
 # define	__LA_MODE_T	mode_t
 #endif

+#if ARCHIVE_VERSION_NUMBER < 4000000
+/* Use the platform types for time_t */
+#define __LA_TIME_T time_t
+#else
+/* Use 64-bits integer types for time_t */
+#define __LA_TIME_T la_int64_t
+#endif
+
+#if ARCHIVE_VERSION_NUMBER < 4000000
+/* Use the platform types for dev_t */
+#define __LA_DEV_T dev_t
+#else
+/* Use 64-bits integer types for dev_t */
+#define __LA_DEV_T la_int64_t
+#endif
+
+#if ARCHIVE_VERSION_NUMBER < 4000000
+/* Libarchive 3.x used signed int64 for inode numbers */
+#define __LA_INO_T la_int64_t
+#else
+/* Switch to unsigned for libarchive 4.0 */
+#define __LA_INO_T la_uint64_t
+#endif
+
 /* Large file support for Android */
 #if defined(__LIBARCHIVE_BUILD) && defined(__ANDROID__)
 #include "android_lf.h"
@@ -127,7 +157,8 @@ typedef ssize_t la_ssize_t;
 # define __LA_DECL
 #endif

-#if defined(__GNUC__) && __GNUC__ >= 3 && __GNUC_MINOR__ >= 1
+#if defined(__GNUC__) && (__GNUC__ > 3 || \
+    (__GNUC__ == 3 && __GNUC_MINOR__ >= 1))
 # define __LA_DEPRECATED __attribute__((deprecated))
 #else
 # define __LA_DEPRECATED
@@ -235,19 +266,19 @@ __LA_DECL struct archive_entry	*archive_entry_new2(struct archive *);
 * also return NULL when implicit character set conversions fail.
 * This is usually what you want.
 */
-__LA_DECL time_t	 archive_entry_atime(struct archive_entry *);
+__LA_DECL __LA_TIME_T	 archive_entry_atime(struct archive_entry *);
 __LA_DECL long		 archive_entry_atime_nsec(struct archive_entry *);
 __LA_DECL int		 archive_entry_atime_is_set(struct archive_entry *);
-__LA_DECL time_t	 archive_entry_birthtime(struct archive_entry *);
+__LA_DECL __LA_TIME_T	 archive_entry_birthtime(struct archive_entry *);
 __LA_DECL long		 archive_entry_birthtime_nsec(struct archive_entry *);
 __LA_DECL int		 archive_entry_birthtime_is_set(struct archive_entry *);
-__LA_DECL time_t	 archive_entry_ctime(struct archive_entry *);
+__LA_DECL __LA_TIME_T	 archive_entry_ctime(struct archive_entry *);
 __LA_DECL long		 archive_entry_ctime_nsec(struct archive_entry *);
 __LA_DECL int		 archive_entry_ctime_is_set(struct archive_entry *);
-__LA_DECL dev_t		 archive_entry_dev(struct archive_entry *);
+__LA_DECL __LA_DEV_T		 archive_entry_dev(struct archive_entry *);
 __LA_DECL int		 archive_entry_dev_is_set(struct archive_entry *);
-__LA_DECL dev_t		 archive_entry_devmajor(struct archive_entry *);
-__LA_DECL dev_t		 archive_entry_devminor(struct archive_entry *);
+__LA_DECL __LA_DEV_T		 archive_entry_devmajor(struct archive_entry *);
+__LA_DECL __LA_DEV_T		 archive_entry_devminor(struct archive_entry *);
 __LA_DECL __LA_MODE_T	 archive_entry_filetype(struct archive_entry *);
 __LA_DECL int		 archive_entry_filetype_is_set(struct archive_entry *);
 __LA_DECL void		 archive_entry_fflags(struct archive_entry *,
@@ -264,8 +295,8 @@ __LA_DECL const char	*archive_entry_hardlink(struct archive_entry *);
 __LA_DECL const char	*archive_entry_hardlink_utf8(struct archive_entry *);
 __LA_DECL const wchar_t	*archive_entry_hardlink_w(struct archive_entry *);
 __LA_DECL int		 archive_entry_hardlink_is_set(struct archive_entry *);
-__LA_DECL la_int64_t	 archive_entry_ino(struct archive_entry *);
-__LA_DECL la_int64_t	 archive_entry_ino64(struct archive_entry *);
+__LA_DECL __LA_INO_T	 archive_entry_ino(struct archive_entry *);
+__LA_DECL __LA_INO_T	 archive_entry_ino64(struct archive_entry *);
 __LA_DECL int		 archive_entry_ino_is_set(struct archive_entry *);
 __LA_DECL __LA_MODE_T	 archive_entry_mode(struct archive_entry *);
 __LA_DECL time_t	 archive_entry_mtime(struct archive_entry *);
@@ -278,9 +309,9 @@ __LA_DECL const wchar_t	*archive_entry_pathname_w(struct archive_entry *);
 __LA_DECL __LA_MODE_T	 archive_entry_perm(struct archive_entry *);
 __LA_DECL int		 archive_entry_perm_is_set(struct archive_entry *);
 __LA_DECL int		 archive_entry_rdev_is_set(struct archive_entry *);
-__LA_DECL dev_t		 archive_entry_rdev(struct archive_entry *);
-__LA_DECL dev_t		 archive_entry_rdevmajor(struct archive_entry *);
-__LA_DECL dev_t		 archive_entry_rdevminor(struct archive_entry *);
+__LA_DECL __LA_DEV_T		 archive_entry_rdev(struct archive_entry *);
+__LA_DECL __LA_DEV_T		 archive_entry_rdevmajor(struct archive_entry *);
+__LA_DECL __LA_DEV_T		 archive_entry_rdevminor(struct archive_entry *);
 __LA_DECL const char	*archive_entry_sourcepath(struct archive_entry *);
 __LA_DECL const wchar_t	*archive_entry_sourcepath_w(struct archive_entry *);
 __LA_DECL la_int64_t	 archive_entry_size(struct archive_entry *);
@@ -309,18 +340,18 @@ __LA_DECL int archive_entry_is_encrypted(struct archive_entry *);
 * always copied.
 */

-__LA_DECL void	archive_entry_set_atime(struct archive_entry *, time_t, long);
+__LA_DECL void	archive_entry_set_atime(struct archive_entry *, __LA_TIME_T, long);
 __LA_DECL void  archive_entry_unset_atime(struct archive_entry *);
 #if defined(_WIN32) && !defined(__CYGWIN__)
 __LA_DECL void archive_entry_copy_bhfi(struct archive_entry *, BY_HANDLE_FILE_INFORMATION *);
 #endif
-__LA_DECL void	archive_entry_set_birthtime(struct archive_entry *, time_t, long);
+__LA_DECL void	archive_entry_set_birthtime(struct archive_entry *, __LA_TIME_T, long);
 __LA_DECL void  archive_entry_unset_birthtime(struct archive_entry *);
-__LA_DECL void	archive_entry_set_ctime(struct archive_entry *, time_t, long);
+__LA_DECL void	archive_entry_set_ctime(struct archive_entry *, __LA_TIME_T, long);
 __LA_DECL void  archive_entry_unset_ctime(struct archive_entry *);
-__LA_DECL void	archive_entry_set_dev(struct archive_entry *, dev_t);
-__LA_DECL void	archive_entry_set_devmajor(struct archive_entry *, dev_t);
-__LA_DECL void	archive_entry_set_devminor(struct archive_entry *, dev_t);
+__LA_DECL void	archive_entry_set_dev(struct archive_entry *, __LA_DEV_T);
+__LA_DECL void	archive_entry_set_devmajor(struct archive_entry *, __LA_DEV_T);
+__LA_DECL void	archive_entry_set_devminor(struct archive_entry *, __LA_DEV_T);
 __LA_DECL void	archive_entry_set_filetype(struct archive_entry *, unsigned int);
 __LA_DECL void	archive_entry_set_fflags(struct archive_entry *,
 	    unsigned long /* set */, unsigned long /* clear */);
@@ -343,15 +374,15 @@ __LA_DECL void	archive_entry_set_hardlink_utf8(struct archive_entry *, const cha
 __LA_DECL void	archive_entry_copy_hardlink(struct archive_entry *, const char *);
 __LA_DECL void	archive_entry_copy_hardlink_w(struct archive_entry *, const wchar_t *);
 __LA_DECL int	archive_entry_update_hardlink_utf8(struct archive_entry *, const char *);
-__LA_DECL void	archive_entry_set_ino(struct archive_entry *, la_int64_t);
-__LA_DECL void	archive_entry_set_ino64(struct archive_entry *, la_int64_t);
+__LA_DECL void	archive_entry_set_ino(struct archive_entry *, __LA_INO_T);
+__LA_DECL void	archive_entry_set_ino64(struct archive_entry *, __LA_INO_T);
 __LA_DECL void	archive_entry_set_link(struct archive_entry *, const char *);
 __LA_DECL void	archive_entry_set_link_utf8(struct archive_entry *, const char *);
 __LA_DECL void	archive_entry_copy_link(struct archive_entry *, const char *);
 __LA_DECL void	archive_entry_copy_link_w(struct archive_entry *, const wchar_t *);
 __LA_DECL int	archive_entry_update_link_utf8(struct archive_entry *, const char *);
 __LA_DECL void	archive_entry_set_mode(struct archive_entry *, __LA_MODE_T);
-__LA_DECL void	archive_entry_set_mtime(struct archive_entry *, time_t, long);
+__LA_DECL void	archive_entry_set_mtime(struct archive_entry *, __LA_TIME_T, long);
 __LA_DECL void  archive_entry_unset_mtime(struct archive_entry *);
 __LA_DECL void	archive_entry_set_nlink(struct archive_entry *, unsigned int);
 __LA_DECL void	archive_entry_set_pathname(struct archive_entry *, const char *);
@@ -360,9 +391,9 @@ __LA_DECL void	archive_entry_copy_pathname(struct archive_entry *, const char *)
 __LA_DECL void	archive_entry_copy_pathname_w(struct archive_entry *, const wchar_t *);
 __LA_DECL int	archive_entry_update_pathname_utf8(struct archive_entry *, const char *);
 __LA_DECL void	archive_entry_set_perm(struct archive_entry *, __LA_MODE_T);
-__LA_DECL void	archive_entry_set_rdev(struct archive_entry *, dev_t);
-__LA_DECL void	archive_entry_set_rdevmajor(struct archive_entry *, dev_t);
-__LA_DECL void	archive_entry_set_rdevminor(struct archive_entry *, dev_t);
+__LA_DECL void	archive_entry_set_rdev(struct archive_entry *, __LA_DEV_T);
+__LA_DECL void	archive_entry_set_rdevmajor(struct archive_entry *, __LA_DEV_T);
+__LA_DECL void	archive_entry_set_rdevminor(struct archive_entry *, __LA_DEV_T);
 __LA_DECL void	archive_entry_set_size(struct archive_entry *, la_int64_t);
 __LA_DECL void	archive_entry_unset_size(struct archive_entry *);
 __LA_DECL void	archive_entry_copy_sourcepath(struct archive_entry *, const char *);
@@ -418,6 +449,7 @@ __LA_DECL void archive_entry_copy_mac_metadata(struct archive_entry *, const voi
 #define ARCHIVE_ENTRY_DIGEST_SHA512           0x00000006

 __LA_DECL const unsigned char * archive_entry_digest(struct archive_entry *, int /* type */);
+__LA_DECL int archive_entry_set_digest(struct archive_entry *, int, const unsigned char *);

 /*
 * ACL routines.  This used to simply store and return text-format ACL
@@ -280,6 +280,10 @@ find_entry(struct archive_entry_linkresolver *res,
 	dev_t			 dev;
 	int64_t			 ino;

+	if (!archive_entry_ino_is_set(entry) || !archive_entry_dev_is_set(entry)) {
+		return (NULL);
+	}
+
 	/* Free a held entry. */
 	if (res->spare != NULL) {
 		archive_entry_free(res->spare->canonical);
@@ -369,6 +373,10 @@ insert_entry(struct archive_entry_linkresolver *res,
 	struct links_entry *le;
 	size_t hash, bucket;

+	if (!archive_entry_ino_is_set(entry) || !archive_entry_dev_is_set(entry)) {
+		return (NULL);
+	}
+
 	/* Add this entry to the links cache. */
 	le = calloc(1, sizeof(struct links_entry));
 	if (le == NULL)
@@ -87,4 +87,4 @@ int _archive_entry_copy_symlink_l(struct archive_entry *,
 int _archive_entry_copy_uname_l(struct archive_entry *,
    const char *, size_t, struct archive_string_conv *);

-#endif /* ARCHIVE_ENTRY_LOCALE_H_INCLUDED */
+#endif /* !ARCHIVE_ENTRY_LOCALE_H_INCLUDED */
@@ -174,6 +174,13 @@ struct archive_entry {
 	size_t mac_metadata_size;

 	/* Digest support. */
+#define AE_MSET_DIGEST_MD5	 1
+#define AE_MSET_DIGEST_RMD160	 2
+#define AE_MSET_DIGEST_SHA1	 4
+#define AE_MSET_DIGEST_SHA256	 8
+#define AE_MSET_DIGEST_SHA384	16
+#define AE_MSET_DIGEST_SHA512	32
+	uint_least32_t mset_digest;
 	struct ae_digest digest;

 	/* ACL support. */
@@ -195,8 +202,4 @@ struct archive_entry {
 	int ae_symlink_type;
 };

-int
-archive_entry_set_digest(struct archive_entry *entry, int type,
-    const unsigned char *digest);
-
-#endif /* ARCHIVE_ENTRY_PRIVATE_H_INCLUDED */
+#endif /* !ARCHIVE_ENTRY_PRIVATE_H_INCLUDED */
@@ -116,4 +116,4 @@ struct archive_hmac {
 };

 extern const struct archive_hmac __archive_hmac;
-#endif /* ARCHIVE_HMAC_PRIVATE_H_INCLUDED */
+#endif /* !ARCHIVE_HMAC_PRIVATE_H_INCLUDED */
@@ -35,14 +35,17 @@
 #ifdef HAVE_STRING_H
 #include <string.h>
 #endif
+#ifdef HAVE_LIMITS_H
+#include <limits.h>
+#endif

 #include "archive.h"
 #include "archive_private.h"
 #include "archive_entry.h"
-#include "archive_getdate.h"
 #include "archive_pathmatch.h"
 #include "archive_rb.h"
 #include "archive_string.h"
+#include "archive_time_private.h"

 struct match {
 	struct match		*next;
@@ -53,8 +56,7 @@ struct match {
 struct match_list {
 	struct match		*first;
 	struct match		**last;
-	int			 count;
-	int			 unmatched_count;
+	size_t			 unmatched_count;
 	struct match		*unmatched_next;
 	int			 unmatched_eof;
 };
@@ -73,7 +75,6 @@ struct match_file {
 struct entry_list {
 	struct match_file	*first;
 	struct match_file	**last;
-	int			 count;
 };

 struct id_array {
@@ -144,12 +145,15 @@ static int	add_pattern_mbs(struct archive_match *, struct match_list *,
 		    const char *);
 static int	add_pattern_wcs(struct archive_match *, struct match_list *,
 		    const wchar_t *);
+#if !defined(_WIN32) || defined(__CYGWIN__)
 static int	cmp_key_mbs(const struct archive_rb_node *, const void *);
-static int	cmp_key_wcs(const struct archive_rb_node *, const void *);
 static int	cmp_node_mbs(const struct archive_rb_node *,
 		    const struct archive_rb_node *);
+#else
+static int	cmp_key_wcs(const struct archive_rb_node *, const void *);
 static int	cmp_node_wcs(const struct archive_rb_node *,
 		    const struct archive_rb_node *);
+#endif
 static void	entry_list_add(struct entry_list *, struct match_file *);
 static void	entry_list_free(struct entry_list *);
 static void	entry_list_init(struct entry_list *);
@@ -187,14 +191,14 @@ static int	time_excluded(struct archive_match *,
 		    struct archive_entry *);
 static int	validate_time_flag(struct archive *, int, const char *);

-#define get_date __archive_get_date
+#define get_date archive_parse_date

-static const struct archive_rb_tree_ops rb_ops_mbs = {
+static const struct archive_rb_tree_ops rb_ops = {
+#if !defined(_WIN32) || defined(__CYGWIN__)
 	cmp_node_mbs, cmp_key_mbs
-};
-
-static const struct archive_rb_tree_ops rb_ops_wcs = {
+#else
 	cmp_node_wcs, cmp_key_wcs
+#endif
 };

 /*
@@ -228,7 +232,7 @@ archive_match_new(void)
 	a->recursive_include = 1;
 	match_list_init(&(a->inclusions));
 	match_list_init(&(a->exclusions));
-	__archive_rb_tree_init(&(a->exclusion_tree), &rb_ops_mbs);
+	__archive_rb_tree_init(&(a->exclusion_tree), &rb_ops);
 	entry_list_init(&(a->exclusion_entry_list));
 	match_list_init(&(a->inclusion_unames));
 	match_list_init(&(a->inclusion_gnames));
@@ -507,7 +511,9 @@ archive_match_path_unmatched_inclusions(struct archive *_a)
 	    ARCHIVE_STATE_NEW, "archive_match_unmatched_inclusions");
 	a = (struct archive_match *)_a;

-	return (a->inclusions.unmatched_count);
+	if (a->inclusions.unmatched_count > (size_t)INT_MAX)
+		return INT_MAX;
+	return (int)(a->inclusions.unmatched_count);
 }

 int
@@ -650,7 +656,7 @@ add_pattern_from_file(struct archive_match *a, struct match_list *mlist,
 						break;
 					}
 				} else {
-			            	if (*b == 0x0d || *b == 0x0a) {
+					if (*b == 0x0d || *b == 0x0a) {
 						found_separator = 1;
 						break;
 					}
@@ -735,7 +741,7 @@ path_excluded(struct archive_match *a, int mbs, const void *pathname)
 		}
 	}

-	/* Exclusions take priority */
+	/* Exclusions take priority. */
 	for (match = a->exclusions.first; match != NULL;
 	    match = match->next){
 		r = match_path_exclusion(a, match, mbs, pathname);
@@ -834,7 +840,6 @@ match_list_init(struct match_list *list)
 {
 	list->first = NULL;
 	list->last = &(list->first);
-	list->count = 0;
 }

 static void
@@ -855,7 +860,6 @@ match_list_add(struct match_list *list, struct match *m)
 {
 	*list->last = m;
 	list->last = &(m->next);
-	list->count++;
 	list->unmatched_count++;
 }

@@ -1145,36 +1149,15 @@ set_timefilter_date_w(struct archive_match *a, int timetype,
 }

 #if defined(_WIN32) && !defined(__CYGWIN__)
-#define EPOC_TIME ARCHIVE_LITERAL_ULL(116444736000000000)
 static int
 set_timefilter_find_data(struct archive_match *a, int timetype,
-    DWORD ftLastWriteTime_dwHighDateTime, DWORD ftLastWriteTime_dwLowDateTime,
-    DWORD ftCreationTime_dwHighDateTime, DWORD ftCreationTime_dwLowDateTime)
+    const FILETIME* ftLastWriteTime, const FILETIME* ftCreationTime)
 {
-	ULARGE_INTEGER utc;
 	time_t ctime_sec, mtime_sec;
-	long ctime_ns, mtime_ns;
+	uint32_t ctime_ns, mtime_ns;

-	utc.HighPart = ftCreationTime_dwHighDateTime;
-	utc.LowPart = ftCreationTime_dwLowDateTime;
-	if (utc.QuadPart >= EPOC_TIME) {
-		utc.QuadPart -= EPOC_TIME;
-		ctime_sec = (time_t)(utc.QuadPart / 10000000);
-		ctime_ns = (long)(utc.QuadPart % 10000000) * 100;
-	} else {
-		ctime_sec = 0;
-		ctime_ns = 0;
-	}
-	utc.HighPart = ftLastWriteTime_dwHighDateTime;
-	utc.LowPart = ftLastWriteTime_dwLowDateTime;
-	if (utc.QuadPart >= EPOC_TIME) {
-		utc.QuadPart -= EPOC_TIME;
-		mtime_sec = (time_t)(utc.QuadPart / 10000000);
-		mtime_ns = (long)(utc.QuadPart % 10000000) * 100;
-	} else {
-		mtime_sec = 0;
-		mtime_ns = 0;
-	}
+	ntfs_to_unix(FILETIME_to_ntfs(ftLastWriteTime), &mtime_sec, &mtime_ns);
+	ntfs_to_unix(FILETIME_to_ntfs(ftCreationTime), &ctime_sec, &ctime_ns);
 	return set_timefilter(a, timetype,
 			mtime_sec, mtime_ns, ctime_sec, ctime_ns);
 }
@@ -1199,9 +1182,7 @@ set_timefilter_pathname_mbs(struct archive_match *a, int timetype,
 		return (ARCHIVE_FAILED);
 	}
 	FindClose(h);
-	return set_timefilter_find_data(a, timetype,
-	    d.ftLastWriteTime.dwHighDateTime, d.ftLastWriteTime.dwLowDateTime,
-	    d.ftCreationTime.dwHighDateTime, d.ftCreationTime.dwLowDateTime);
+	return set_timefilter_find_data(a, timetype, &d.ftLastWriteTime, &d.ftCreationTime);
 }

 static int
@@ -1223,9 +1204,7 @@ set_timefilter_pathname_wcs(struct archive_match *a, int timetype,
 		return (ARCHIVE_FAILED);
 	}
 	FindClose(h);
-	return set_timefilter_find_data(a, timetype,
-	    d.ftLastWriteTime.dwHighDateTime, d.ftLastWriteTime.dwLowDateTime,
-	    d.ftCreationTime.dwHighDateTime, d.ftCreationTime.dwLowDateTime);
+	return set_timefilter_find_data(a, timetype, &d.ftLastWriteTime, &d.ftCreationTime);
 }

 #else /* _WIN32 && !__CYGWIN__ */
@@ -1300,6 +1279,7 @@ set_timefilter_pathname_wcs(struct archive_match *a, int timetype,
 /*
 * Call back functions for archive_rb.
 */
+#if !defined(_WIN32) || defined(__CYGWIN__)
 static int
 cmp_node_mbs(const struct archive_rb_node *n1,
    const struct archive_rb_node *n2)
@@ -1316,7 +1296,7 @@ cmp_node_mbs(const struct archive_rb_node *n1,
 		return (-1);
 	return (strcmp(p1, p2));
 }
-        
+
 static int
 cmp_key_mbs(const struct archive_rb_node *n, const void *key)
 {
@@ -1328,7 +1308,7 @@ cmp_key_mbs(const struct archive_rb_node *n, const void *key)
 		return (-1);
 	return (strcmp(p, (const char *)key));
 }
-
+#else
 static int
 cmp_node_wcs(const struct archive_rb_node *n1,
    const struct archive_rb_node *n2)
@@ -1345,7 +1325,7 @@ cmp_node_wcs(const struct archive_rb_node *n1,
 		return (-1);
 	return (wcscmp(p1, p2));
 }
-        
+
 static int
 cmp_key_wcs(const struct archive_rb_node *n, const void *key)
 {
@@ -1357,13 +1337,13 @@ cmp_key_wcs(const struct archive_rb_node *n, const void *key)
 		return (-1);
 	return (wcscmp(p, (const wchar_t *)key));
 }
+#endif

 static void
 entry_list_init(struct entry_list *list)
 {
 	list->first = NULL;
 	list->last = &(list->first);
-	list->count = 0;
 }

 static void
@@ -1384,7 +1364,6 @@ entry_list_add(struct entry_list *list, struct match_file *file)
 {
 	*list->last = file;
 	list->last = &(file->next);
-	list->count++;
 }

 static int
@@ -1407,9 +1386,7 @@ add_entry(struct archive_match *a, int flag,
 		return (ARCHIVE_FAILED);
 	}
 	archive_mstring_copy_wcs(&(f->pathname), pathname);
-	a->exclusion_tree.rbt_ops = &rb_ops_wcs;
 #else
-	(void)rb_ops_wcs;
 	pathname = archive_entry_pathname(entry);
 	if (pathname == NULL) {
 		free(f);
@@ -1417,7 +1394,6 @@ add_entry(struct archive_match *a, int flag,
 		return (ARCHIVE_FAILED);
 	}
 	archive_mstring_copy_mbs(&(f->pathname), pathname);
-	a->exclusion_tree.rbt_ops = &rb_ops_mbs;
 #endif
 	f->flag = flag;
 	f->mtime_sec = archive_entry_mtime(entry);
@@ -1542,16 +1518,13 @@ time_excluded(struct archive_match *a, struct archive_entry *entry)
 	}

 	/* If there is no exclusion list, include the file. */
-	if (a->exclusion_entry_list.count == 0)
+	if (a->exclusion_entry_list.first == NULL)
 		return (0);

 #if defined(_WIN32) && !defined(__CYGWIN__)
 	pathname = archive_entry_pathname_w(entry);
-	a->exclusion_tree.rbt_ops = &rb_ops_wcs;
 #else
-	(void)rb_ops_wcs;
 	pathname = archive_entry_pathname(entry);
-	a->exclusion_tree.rbt_ops = &rb_ops_mbs;
 #endif
 	if (pathname == NULL)
 		return (0);
@@ -1707,7 +1680,7 @@ archive_match_owner_excluded(struct archive *_a,
 static int
 add_owner_id(struct archive_match *a, struct id_array *ids, int64_t id)
 {
-	unsigned i;
+	size_t i;

 	if (ids->count + 1 >= ids->size) {
 		void *p;
@@ -1744,10 +1717,10 @@ add_owner_id(struct archive_match *a, struct id_array *ids, int64_t id)
 static int
 match_owner_id(struct id_array *ids, int64_t id)
 {
-	unsigned b, m, t;
+	size_t b, m, t;

 	t = 0;
-	b = (unsigned)ids->count;
+	b = ids->count;
 	while (t < b) {
 		m = (t + b)>>1;
 		if (ids->ids[m] == id)
@@ -1842,7 +1815,7 @@ owner_excluded(struct archive_match *a, struct archive_entry *entry)
 			return (1);
 	}

-	if (a->inclusion_unames.count) {
+	if (a->inclusion_unames.first != NULL) {
 #if defined(_WIN32) && !defined(__CYGWIN__)
 		r = match_owner_name_wcs(a, &(a->inclusion_unames),
 			archive_entry_uname_w(entry));
@@ -1856,7 +1829,7 @@ owner_excluded(struct archive_match *a, struct archive_entry *entry)
 			return (r);
 	}

-	if (a->inclusion_gnames.count) {
+	if (a->inclusion_gnames.first != NULL) {
 #if defined(_WIN32) && !defined(__CYGWIN__)
 		r = match_owner_name_wcs(a, &(a->inclusion_gnames),
 			archive_entry_gname_w(entry));
@@ -1871,4 +1844,3 @@ owner_excluded(struct archive_match *a, struct archive_entry *entry)
 	}
 	return (0);
 }
-
@@ -38,7 +38,7 @@ parse_option(const char **str,
 int
 _archive_set_option(struct archive *a,
    const char *m, const char *o, const char *v,
-    int magic, const char *fn, option_handler use_option)
+    unsigned int magic, const char *fn, option_handler use_option)
 {
 	const char *mp, *op, *vp;
 	int r;
@@ -97,7 +97,7 @@ _archive_set_either_option(struct archive *a, const char *m, const char *o, cons

 int
 _archive_set_options(struct archive *a, const char *options,
-    int magic, const char *fn, option_handler use_option)
+    unsigned int magic, const char *fn, option_handler use_option)
 {
 	int allok = 1, anyok = 0, ignore_mod_err = 0, r;
 	char *data;
@@ -35,11 +35,11 @@ typedef int (*option_handler)(struct archive *a,
 int
 _archive_set_option(struct archive *a,
    const char *mod, const char *opt, const char *val,
-    int magic, const char *fn, option_handler use_option);
+    unsigned int magic, const char *fn, option_handler use_option);

 int
 _archive_set_options(struct archive *a, const char *options,
-    int magic, const char *fn, option_handler use_option);
+    unsigned int magic, const char *fn, option_handler use_option);

 int
 _archive_set_either_option(struct archive *a,
@@ -46,4 +46,4 @@ pack_t	 pack_native;
 					 (((y) << 12) & 0xfff00000) | \
 					 (((y) <<  0) & 0x000000ff)))

-#endif	/* ARCHIVE_PACK_DEV_H */
+#endif	/* !ARCHIVE_PACK_DEV_H */
@@ -35,8 +35,7 @@
 #include <string.h>
 #include <time.h>

-#define __LIBARCHIVE_BUILD 1
-#include "archive_getdate.h"
+#include "archive.h"

 /* Basic time units. */
 #define	EPOCH		1970
@@ -820,6 +819,23 @@ RelativeMonth(time_t Start, time_t Timezone, time_t RelMonth)
 		Timezone, DSTmaybe));
 }

+/*
+ * Parses and consumes an unsigned number.
+ * Returns 1 if any number is parsed. Otherwise, *value is unchanged.
+ */
+static char
+consume_unsigned_number(const char **in, time_t *value)
+{
+	char c;
+	if (isdigit((unsigned char)(c = **in))) {
+		for (*value = 0; isdigit((unsigned char)(c = *(*in)++)); )
+			*value = 10 * *value + c - '0';
+		(*in)--;
+		return 1;
+	}
+	return 0;
+}
+
 /*
 * Tokenizer.
 */
@@ -861,7 +877,8 @@ nexttoken(const char **in, time_t *value)
 			    && i < sizeof(buff)-1) {
 				if (*src != '.') {
 					if (isupper((unsigned char)*src))
-						buff[i++] = tolower((unsigned char)*src);
+						buff[i++] = (char)tolower(
+						    (unsigned char)*src);
 					else
 						buff[i++] = *src;
 				}
@@ -895,10 +912,7 @@ nexttoken(const char **in, time_t *value)
 		 * Because '-' and '+' have other special meanings, I
 		 * don't deal with signed numbers here.
 		 */
-		if (isdigit((unsigned char)(c = **in))) {
-			for (*value = 0; isdigit((unsigned char)(c = *(*in)++)); )
-				*value = 10 * *value + c - '0';
-			(*in)--;
+		if (consume_unsigned_number(in, value)) {
 			return (tUNUMBER);
 		}

@@ -914,7 +928,7 @@ difftm (struct tm *a, struct tm *b)
 {
 	int ay = a->tm_year + (TM_YEAR_ORIGIN - 1);
 	int by = b->tm_year + (TM_YEAR_ORIGIN - 1);
-	int days = (
+	long days = (
 		/* difference in day of year */
 		a->tm_yday - b->tm_yday
 		/* + intervening leap days */
@@ -929,6 +943,32 @@ difftm (struct tm *a, struct tm *b)
 	    + (a->tm_sec - b->tm_sec));
 }

+/*
+ * Parses a Unix epoch timestamp (seconds).
+ * This supports a subset of what GNU tar accepts from black box testing,
+ * but covers common use cases.
+ */
+static time_t
+parse_unix_epoch(const char *p)
+{
+	time_t epoch;
+
+	/* may begin with + */
+	if (*p == '+') {
+		p++;
+	}
+
+	/* followed by some number */
+	if (!consume_unsigned_number(&p, &epoch))
+		return (time_t)-1;
+
+	/* ...and nothing else */
+	if (*p != '\0')
+		return (time_t)-1;
+
+	return epoch;
+}
+
 /*
 *
 * The public function.
@@ -936,7 +976,7 @@ difftm (struct tm *a, struct tm *b)
 * TODO: tokens[] array should be dynamically sized.
 */
 time_t
-__archive_get_date(time_t now, const char *p)
+archive_parse_date(time_t now, const char *p)
 {
 	struct token	tokens[256];
 	struct gdstate	_gds;
@@ -948,6 +988,13 @@ __archive_get_date(time_t now, const char *p)
 	time_t		tod;
 	long		tzone;

+	/*
+	 * @-prefixed Unix epoch timestamps (seconds)
+	 * Skip the complex tokenizer - We do not want to accept strings like "@tenth"
+	 */
+	if (*p == '@')
+		return parse_unix_epoch(p + 1);
+
 	/* Clear out the parsed token array. */
 	memset(tokens, 0, sizeof(tokens));
 	/* Initialize the parser state. */
@@ -61,6 +61,11 @@
 # endif
 #endif

+/* For cygwin, to avoid missing LONG, ULONG, PUCHAR, ... definitions */
+#ifdef __CYGWIN__
+#include <windef.h>
+#endif
+
 /* It should be possible to get rid of this by extending the feature-test
 * macros to cover Windows API functions, probably along with non-trivial
 * refactoring of code to find structures that sit more cleanly on top of
@@ -50,4 +50,4 @@
 #define ARCHIVE_ACL_SUPPORT     1
 #endif

-#endif	/* ARCHIVE_PLATFORM_ACL_H_INCLUDED */
+#endif	/* !ARCHIVE_PLATFORM_ACL_H_INCLUDED */
@@ -42,4 +42,4 @@
 #define ARCHIVE_XATTR_SUPPORT     1
 #endif

-#endif	/* ARCHIVE_PLATFORM_XATTR_H_INCLUDED */
+#endif	/* !ARCHIVE_PLATFORM_XATTR_H_INCLUDED */
@@ -287,9 +287,14 @@ static void *AllocUnits(CPpmd7 *p, unsigned indx)
  return AllocUnitsRare(p, indx);
 }

-#define MyMem12Cpy(dest, src, num) \
-  { UInt32 *d = (UInt32 *)dest; const UInt32 *s = (const UInt32 *)src; UInt32 n = num; \
-    do { d[0] = s[0]; d[1] = s[1]; d[2] = s[2]; s += 3; d += 3; } while(--n); }
+#define MyMem12Cpy(dest, src, num) do {					\
+	UInt32 *d = (UInt32 *)dest;					\
+	const UInt32 *s = (const UInt32 *)src;				\
+	UInt32 n = num;							\
+	do {								\
+		d[0] = s[0]; d[1] = s[1]; d[2] = s[2]; s += 3; d += 3;	\
+	} while(--n);							\
+} while (0)

 static void *ShrinkUnits(CPpmd7 *p, void *oldPtr, unsigned oldNU, unsigned newNU)
 {
@@ -242,9 +242,14 @@ static void *AllocUnits(CPpmd8 *p, unsigned indx)
  return AllocUnitsRare(p, indx);
 }

-#define MyMem12Cpy(dest, src, num) \
-  { UInt32 *d = (UInt32 *)dest; const UInt32 *z = (const UInt32 *)src; UInt32 n = num; \
-    do { d[0] = z[0]; d[1] = z[1]; d[2] = z[2]; z += 3; d += 3; } while (--n); }
+#define MyMem12Cpy(dest, src, num) do {					\
+	UInt32 *d = (UInt32 *)dest;					\
+	const UInt32 *z = (const UInt32 *)src;				\
+	UInt32 n = num;							\
+	do {								\
+		d[0] = z[0]; d[1] = z[1]; d[2] = z[2]; z += 3; d += 3;	\
+	} while (--n);							\
+} while (0)

 static void *ShrinkUnits(CPpmd8 *p, void *oldPtr, unsigned oldNU, unsigned newNU)
 {
@@ -341,7 +346,9 @@ static void SetSuccessor(CPpmd_State *p, CPpmd_Void_Ref v)
  (p)->SuccessorHigh = (UInt16)(((UInt32)(v) >> 16) & 0xFFFF);
 }

-#define RESET_TEXT(offs) { p->Text = p->Base + p->AlignOffset + (offs); }
+#define RESET_TEXT(offs) do {						\
+	p->Text = p->Base + p->AlignOffset + (offs);			\
+} while (0)

 static void RestartModel(CPpmd8 *p)
 {
@@ -109,8 +109,12 @@ typedef struct
  Byte Count;  /* Count to next change of Shift */
 } CPpmd_See;

-#define Ppmd_See_Update(p)  if ((p)->Shift < PPMD_PERIOD_BITS && --(p)->Count == 0) \
-    { (p)->Summ <<= 1; (p)->Count = (Byte)(3 << (p)->Shift++); }
+#define Ppmd_See_Update(p) do {						\
+	if ((p)->Shift < PPMD_PERIOD_BITS && --(p)->Count == 0) {	\
+		(p)->Summ <<= 1;					\
+		(p)->Count = (Byte)(3 << (p)->Shift++);			\
+    	}								\
+} while (0)

 typedef struct
 {
@@ -144,8 +148,12 @@ typedef
  #endif
  CPpmd_Byte_Ref;

-#define PPMD_SetAllBitsIn256Bytes(p) \
-  { unsigned j; for (j = 0; j < 256 / sizeof(p[0]); j += 8) { \
-  p[j+7] = p[j+6] = p[j+5] = p[j+4] = p[j+3] = p[j+2] = p[j+1] = p[j+0] = ~(size_t)0; }}
+#define PPMD_SetAllBitsIn256Bytes(p) do {				\
+	unsigned j;							\
+	for (j = 0; j < 256 / sizeof(p[0]); j += 8) {			\
+		p[j+7] = p[j+6] = p[j+5] = p[j+4] =			\
+		    p[j+3] = p[j+2] = p[j+1] = p[j+0] = ~(size_t)0;	\
+	}								\
+} while (0)

 #endif
@@ -33,4 +33,4 @@
 /* Random number generator. */
 int archive_random(void *buf, size_t nbytes);

-#endif /* ARCHIVE_RANDOM_PRIVATE_H_INCLUDED */
+#endif /* !ARCHIVE_RANDOM_PRIVATE_H_INCLUDED */
@@ -110,4 +110,4 @@ struct archive_rb_node *
 	__archive_rb_tree_iterate(struct archive_rb_tree *,
 	struct archive_rb_node *, const unsigned int);

-#endif	/* ARCHIVE_RB_H_*/
+#endif	/* !ARCHIVE_RB_H_INCLUDED */
@@ -176,15 +176,9 @@ client_skip_proxy(struct archive_read_filter *self, int64_t request)
 		return 0;

 	if (self->archive->client.skipper != NULL) {
-		/* Seek requests over 1GiB are broken down into
-		 * multiple seeks.  This avoids overflows when the
-		 * requests get passed through 32-bit arguments. */
-		int64_t skip_limit = (int64_t)1 << 30;
 		int64_t total = 0;
 		for (;;) {
 			int64_t get, ask = request;
-			if (ask > skip_limit)
-				ask = skip_limit;
 			get = (self->archive->client.skipper)
 				(&self->archive->archive, self->data, ask);
 			total += get;
@@ -96,6 +96,10 @@ archive_read_append_filter(struct archive *_a, int code)
      strcpy(str, "lzip");
      r1 = archive_read_support_filter_lzip(_a);
      break;
+    case ARCHIVE_FILTER_LZOP:
+      strcpy(str, "lzop");
+      r1 = archive_read_support_filter_lzop(_a);
+      break;
    case ARCHIVE_FILTER_LRZIP:
      strcpy(str, "lrzip");
      r1 = archive_read_support_filter_lrzip(_a);
@@ -888,7 +888,7 @@ setup_sparse_fiemap(struct archive_read_disk *a,
 	count = (sizeof(buff) - sizeof(*fm))/sizeof(*fe);
 	fm = (struct fiemap *)buff;
 	fm->fm_start = 0;
-	fm->fm_length = ~0ULL;;
+	fm->fm_length = ~0ULL;
 	fm->fm_flags = FIEMAP_FLAG_SYNC;
 	fm->fm_extent_count = count;
 	do_fiemap = 1;
@@ -276,7 +276,7 @@ tree_dir_next_posix(struct tree *t);
 #endif

 /* Initiate/terminate a tree traversal. */
-static struct tree *tree_open(const char *, int, int);
+static struct tree *tree_open(const char *, char, int);
 static struct tree *tree_reopen(struct tree *, const char *, int);
 static void tree_close(struct tree *);
 static void tree_free(struct tree *);
@@ -514,7 +514,7 @@ _archive_read_close(struct archive *_a)

 static void
 setup_symlink_mode(struct archive_read_disk *a, char symlink_mode,
-    int follow_symlinks)
+    char follow_symlinks)
 {
 	a->symlink_mode = symlink_mode;
 	a->follow_symlinks = follow_symlinks;
@@ -778,7 +778,8 @@ _archive_read_data_block(struct archive *_a, const void **buff,
 	 */
 	if (t->current_sparse->offset > t->entry_total) {
 		if (lseek(t->entry_fd,
-		    (off_t)t->current_sparse->offset, SEEK_SET) < 0) {
+		    (off_t)t->current_sparse->offset, SEEK_SET) !=
+		    t->current_sparse->offset) {
 			archive_set_error(&a->archive, errno, "Seek error");
 			r = ARCHIVE_FATAL;
 			a->archive.state = ARCHIVE_STATE_FATAL;
@@ -2180,7 +2181,7 @@ tree_append(struct tree *t, const char *name, size_t name_length)
 * Open a directory tree for traversal.
 */
 static struct tree *
-tree_open(const char *path, int symlink_mode, int restore_time)
+tree_open(const char *path, char symlink_mode, int restore_time)
 {
 	struct tree *t;

@@ -52,10 +52,11 @@ struct archive_read_disk {

 	/*
 	 * Since symlink interaction changes, we need to track whether
-	 * we're following symlinks for the current item.  'L' mode above
-	 * sets this true, 'P' sets it false, 'H' changes it as we traverse.
+	 * we're following symlinks for the current item, governed by the above
+	 * symlink_mode.  'L' sets this true, 'P' sets it false, 'H' changes it
+	 * as we traverse.
 	 */
-	char	follow_symlinks;  /* Either 'L' or 'P'. */
+	char	follow_symlinks;  /* Either 0 or 1. */

 	/* Directory traversals. */
 	struct tree *tree;
@@ -37,8 +37,10 @@
 .Nm archive_read_support_format_lha ,
 .Nm archive_read_support_format_mtree ,
 .Nm archive_read_support_format_rar ,
+.Nm archive_read_support_format_rar5 ,
 .Nm archive_read_support_format_raw ,
 .Nm archive_read_support_format_tar ,
+.Nm archive_read_support_format_warc ,
 .Nm archive_read_support_format_xar ,
 .Nm archive_read_support_format_zip
 .Nd functions for reading streaming archives
@@ -70,10 +72,14 @@ Streaming Archive Library (libarchive, -larchive)
 .Ft int
 .Fn archive_read_support_format_rar "struct archive *"
 .Ft int
+.Fn archive_read_support_format_rar5 "struct archive *"
+.Ft int
 .Fn archive_read_support_format_raw "struct archive *"
 .Ft int
 .Fn archive_read_support_format_tar "struct archive *"
 .Ft int
+.Fn archive_read_support_format_warc "struct archive *"
+.Ft int
 .Fn archive_read_support_format_xar "struct archive *"
 .Ft int
 .Fn archive_read_support_format_zip "struct archive *"
@@ -89,8 +95,10 @@ Streaming Archive Library (libarchive, -larchive)
 .Fn archive_read_support_format_lha ,
 .Fn archive_read_support_format_mtree ,
 .Fn archive_read_support_format_rar ,
+.Fn archive_read_support_format_rar5 ,
 .Fn archive_read_support_format_raw ,
 .Fn archive_read_support_format_tar ,
+.Fn archive_read_support_format_warc ,
 .Fn archive_read_support_format_xar ,
 .Fn archive_read_support_format_zip
 .Xc
@@ -100,6 +108,13 @@ For example,
 .Fn archive_read_support_format_tar
 enables support for a variety of standard tar formats, old-style tar,
 ustar, pax interchange format, and many common variants.
+.Fn archive_read_support_format_zip
+enables support for both the streaming and the seeking zip readers,
+which can separately be enabled by respectively
+.Fn archive_read_support_format_zip_streamable
+and
+.Fn archive_read_support_format_zip_seekable
+.
 .It Fn archive_read_support_format_all
 Enables support for all available formats except the
 .Dq raw
@@ -52,6 +52,7 @@
 struct read_fd_data {
 	int	 fd;
 	size_t	 block_size;
+	int64_t	 size;
 	char	 use_lseek;
 	void	*buffer;
 };
@@ -95,6 +96,7 @@ archive_read_open_fd(struct archive *a, int fd, size_t block_size)
 	if (S_ISREG(st.st_mode)) {
 		archive_read_extract_set_skip_file(a, st.st_dev, st.st_ino);
 		mine->use_lseek = 1;
+		mine->size = st.st_size;
 	}
 #if defined(__CYGWIN__) || defined(_WIN32)
 	setmode(mine->fd, O_BINARY);
@@ -131,7 +133,7 @@ static int64_t
 file_skip(struct archive *a, void *client_data, int64_t request)
 {
 	struct read_fd_data *mine = (struct read_fd_data *)client_data;
-	int64_t skip = request;
+	off_t skip = (off_t)request;
 	int64_t old_offset, new_offset;
 	int skip_bits = sizeof(skip) * 8 - 1;  /* off_t is a signed type. */

@@ -140,20 +142,25 @@ file_skip(struct archive *a, void *client_data, int64_t request)

 	/* Reduce a request that would overflow the 'skip' variable. */
 	if (sizeof(request) > sizeof(skip)) {
-		int64_t max_skip =
+		const int64_t max_skip =
 		    (((int64_t)1 << (skip_bits - 1)) - 1) * 2 + 1;
 		if (request > max_skip)
-			skip = max_skip;
+			skip = (off_t)max_skip;
 	}

-	/* Reduce request to the next smallest multiple of block_size */
-	request = (request / mine->block_size) * mine->block_size;
-	if (request == 0)
+	/* Reduce 'skip' to the next smallest multiple of block_size */
+	skip = (off_t)(((int64_t)skip / mine->block_size) * mine->block_size);
+	if (skip == 0)
 		return (0);

-	if (((old_offset = lseek(mine->fd, 0, SEEK_CUR)) >= 0) &&
-	    ((new_offset = lseek(mine->fd, skip, SEEK_CUR)) >= 0))
-		return (new_offset - old_offset);
+	if ((old_offset = lseek(mine->fd, 0, SEEK_CUR)) >= 0) {
+		if (old_offset >= mine->size ||
+		    skip > mine->size - old_offset) {
+			/* Do not seek past end of file. */
+			errno = ESPIPE;
+		} else if ((new_offset = lseek(mine->fd, skip, SEEK_CUR)) >= 0)
+			return (new_offset - old_offset);
+	}

 	/* If seek failed once, it will probably fail again. */
 	mine->use_lseek = 0;
@@ -178,11 +185,24 @@ static int64_t
 file_seek(struct archive *a, void *client_data, int64_t request, int whence)
 {
 	struct read_fd_data *mine = (struct read_fd_data *)client_data;
+	off_t seek = (off_t)request;
 	int64_t r;
+	int seek_bits = sizeof(seek) * 8 - 1;  /* off_t is a signed type. */

 	/* We use off_t here because lseek() is declared that way. */
-	/* See above for notes about when off_t is less than 64 bits. */
-	r = lseek(mine->fd, request, whence);
+
+	/* Reduce a request that would overflow the 'seek' variable. */
+	if (sizeof(request) > sizeof(seek)) {
+		const int64_t max_seek =
+		    (((int64_t)1 << (seek_bits - 1)) - 1) * 2 + 1;
+		const int64_t min_seek = ~max_seek;
+		if (request > max_seek)
+			seek = (off_t)max_seek;
+		else if (request < min_seek)
+			seek = (off_t)min_seek;
+	}
+
+	r = lseek(mine->fd, seek, whence);
 	if (r >= 0)
 		return r;

@@ -52,13 +52,15 @@
 struct read_FILE_data {
 	FILE    *f;
 	size_t	 block_size;
+	int64_t	 size;
 	void	*buffer;
 	char	 can_skip;
 };

-static int	file_close(struct archive *, void *);
-static ssize_t	file_read(struct archive *, void *, const void **buff);
-static int64_t	file_skip(struct archive *, void *, int64_t request);
+static int	FILE_close(struct archive *, void *);
+static ssize_t	FILE_read(struct archive *, void *, const void **buff);
+static int64_t	FILE_seek(struct archive *, void *, int64_t, int);
+static int64_t	FILE_skip(struct archive *, void *, int64_t);

 int
 archive_read_open_FILE(struct archive *a, FILE *f)
@@ -69,7 +71,7 @@ archive_read_open_FILE(struct archive *a, FILE *f)
 	void *b;

 	archive_clear_error(a);
-	mine = malloc(sizeof(*mine));
+	mine = calloc(1, sizeof(*mine));
 	b = malloc(block_size);
 	if (mine == NULL || b == NULL) {
 		archive_set_error(a, ENOMEM, "No memory");
@@ -90,22 +92,23 @@ archive_read_open_FILE(struct archive *a, FILE *f)
 		archive_read_extract_set_skip_file(a, st.st_dev, st.st_ino);
 		/* Enable the seek optimization only for regular files. */
 		mine->can_skip = 1;
-	} else
-		mine->can_skip = 0;
+		mine->size = st.st_size;
+	}

 #if defined(__CYGWIN__) || defined(_WIN32)
 	setmode(fileno(mine->f), O_BINARY);
 #endif

-	archive_read_set_read_callback(a, file_read);
-	archive_read_set_skip_callback(a, file_skip);
-	archive_read_set_close_callback(a, file_close);
+	archive_read_set_read_callback(a, FILE_read);
+	archive_read_set_skip_callback(a, FILE_skip);
+	archive_read_set_seek_callback(a, FILE_seek);
+	archive_read_set_close_callback(a, FILE_close);
 	archive_read_set_callback_data(a, mine);
 	return (archive_read_open1(a));
 }

 static ssize_t
-file_read(struct archive *a, void *client_data, const void **buff)
+FILE_read(struct archive *a, void *client_data, const void **buff)
 {
 	struct read_FILE_data *mine = (struct read_FILE_data *)client_data;
 	size_t bytes_read;
@@ -119,16 +122,17 @@ file_read(struct archive *a, void *client_data, const void **buff)
 }

 static int64_t
-file_skip(struct archive *a, void *client_data, int64_t request)
+FILE_skip(struct archive *a, void *client_data, int64_t request)
 {
 	struct read_FILE_data *mine = (struct read_FILE_data *)client_data;
-#if HAVE_FSEEKO
-	off_t skip = (off_t)request;
-#elif HAVE__FSEEKI64
+#if HAVE__FSEEKI64
 	int64_t skip = request;
+#elif HAVE_FSEEKO
+	off_t skip = (off_t)request;
 #else
 	long skip = (long)request;
 #endif
+	int64_t old_offset, new_offset = -1;
 	int skip_bits = sizeof(skip) * 8 - 1;

 	(void)a; /* UNUSED */
@@ -144,7 +148,7 @@ file_skip(struct archive *a, void *client_data, int64_t request)

 	/* If request is too big for a long or an off_t, reduce it. */
 	if (sizeof(request) > sizeof(skip)) {
-		int64_t max_skip =
+		const int64_t max_skip =
 		    (((int64_t)1 << (skip_bits - 1)) - 1) * 2 + 1;
 		if (request > max_skip)
 			skip = max_skip;
@@ -152,23 +156,92 @@ file_skip(struct archive *a, void *client_data, int64_t request)

 #ifdef __ANDROID__
        /* fileno() isn't safe on all platforms ... see above. */
-	if (lseek(fileno(mine->f), skip, SEEK_CUR) < 0)
+	old_offset = lseek(fileno(mine->f), 0, SEEK_CUR);
 #elif HAVE__FSEEKI64
-	if (_fseeki64(mine->f, skip, SEEK_CUR) != 0)
+	old_offset = _ftelli64(mine->f);
 #elif HAVE_FSEEKO
-	if (fseeko(mine->f, skip, SEEK_CUR) != 0)
+	old_offset = ftello(mine->f);
 #else
-	if (fseek(mine->f, skip, SEEK_CUR) != 0)
+	old_offset = ftell(mine->f);
 #endif
-	{
-		mine->can_skip = 0;
-		return (0);
+	if (old_offset >= 0) {
+		if (old_offset < mine->size &&
+		    skip <= mine->size - old_offset) {
+#ifdef __ANDROID__
+			new_offset = lseek(fileno(mine->f), skip, SEEK_CUR);
+#elif HAVE__FSEEKI64
+			if (_fseeki64(mine->f, skip, SEEK_CUR) == 0)
+				new_offset = _ftelli64(mine->f);
+#elif HAVE_FSEEKO
+			if (fseeko(mine->f, skip, SEEK_CUR) == 0)
+				new_offset = ftello(mine->f);
+#else
+			if (fseek(mine->f, skip, SEEK_CUR) == 0)
+				new_offset = ftell(mine->f);
+#endif
+			if (new_offset >= 0)
+				return (new_offset - old_offset);
+		}
 	}
-	return (request);
+
+	mine->can_skip = 0;
+	return (0);
+}
+
+/*
+ * TODO: Store the offset and use it in the read callback.
+ */
+static int64_t
+FILE_seek(struct archive *a, void *client_data, int64_t request, int whence)
+{
+	struct read_FILE_data *mine = (struct read_FILE_data *)client_data;
+#if HAVE__FSEEKI64
+	int64_t seek = request;
+#elif HAVE_FSEEKO
+	off_t seek = (off_t)request;
+#else
+	long seek = (long)request;
+#endif
+	int seek_bits = sizeof(seek) * 8 - 1;
+	(void)a; /* UNUSED */
+
+	/* Reduce a request that would overflow the 'seek' variable. */
+	if (sizeof(request) > sizeof(seek)) {
+		const int64_t max_seek =
+		    (((int64_t)1 << (seek_bits - 1)) - 1) * 2 + 1;
+		const int64_t min_seek = ~max_seek;
+		if (request > max_seek)
+			seek = max_seek;
+		else if (request < min_seek)
+			seek = min_seek;
+	}
+
+#ifdef __ANDROID__
+	/* Newer Android versions have fseeko...to meditate. */
+	int64_t ret = lseek(fileno(mine->f), seek, whence);
+	if (ret >= 0) {
+		return ret;
+	}
+#elif HAVE__FSEEKI64
+	if (_fseeki64(mine->f, seek, whence) == 0) {
+		return _ftelli64(mine->f);
+	}
+#elif HAVE_FSEEKO
+	if (fseeko(mine->f, seek, whence) == 0) {
+		return ftello(mine->f);
+	}
+#else
+	if (fseek(mine->f, seek, whence) == 0) {
+		return ftell(mine->f);
+	}
+#endif
+	/* If we arrive here, the input is corrupted or truncated so fail. */
+	archive_set_error(a, errno, "Error seeking in FILE* pointer");
+	return (ARCHIVE_FATAL);
 }

 static int
-file_close(struct archive *a, void *client_data)
+FILE_close(struct archive *a, void *client_data)
 {
 	struct read_FILE_data *mine = (struct read_FILE_data *)client_data;

@@ -74,6 +74,7 @@ struct read_file_data {
 	size_t	 block_size;
 	void	*buffer;
 	mode_t	 st_mode;  /* Mode bits for opened file. */
+	int64_t	 size;
 	char	 use_lseek;
 	enum fnt_e { FNT_STDIN, FNT_MBS, FNT_WCS } filename_type;
 	union {
@@ -303,7 +304,7 @@ file_open(struct archive *a, void *client_data)
 		}
 		if (fd < 0) {
 			archive_set_error(a, errno,
-			    "Failed to open '%S'", wfilename);
+			    "Failed to open '%ls'", wfilename);
 			return (ARCHIVE_FATAL);
 		}
 #else
@@ -315,7 +316,7 @@ file_open(struct archive *a, void *client_data)
 	if (fstat(fd, &st) != 0) {
 #if defined(_WIN32) && !defined(__CYGWIN__)
 		if (mine->filename_type == FNT_WCS)
-			archive_set_error(a, errno, "Can't stat '%S'",
+			archive_set_error(a, errno, "Can't stat '%ls'",
 			    wfilename);
 		else
 #endif
@@ -400,8 +401,10 @@ file_open(struct archive *a, void *client_data)
 	mine->st_mode = st.st_mode;

 	/* Disk-like inputs can use lseek(). */
-	if (is_disk_like)
+	if (is_disk_like) {
 		mine->use_lseek = 1;
+		mine->size = st.st_size;
+	}

 	return (ARCHIVE_OK);
 fail:
@@ -447,7 +450,7 @@ file_read(struct archive *a, void *client_data, const void **buff)
 				    "Error reading '%s'", mine->filename.m);
 			else
 				archive_set_error(a, errno,
-				    "Error reading '%S'", mine->filename.w);
+				    "Error reading '%ls'", mine->filename.w);
 		}
 		return (bytes_read);
 	}
@@ -479,21 +482,30 @@ file_skip_lseek(struct archive *a, void *client_data, int64_t request)
 	struct read_file_data *mine = (struct read_file_data *)client_data;
 #if defined(_WIN32) && !defined(__CYGWIN__)
 	/* We use _lseeki64() on Windows. */
-	int64_t old_offset, new_offset;
+	int64_t old_offset, new_offset, skip = request;
 #else
-	off_t old_offset, new_offset;
+	off_t old_offset, new_offset, skip = (off_t)request;
 #endif
+	int skip_bits = sizeof(skip) * 8 - 1;

 	/* We use off_t here because lseek() is declared that way. */

-	/* TODO: Deal with case where off_t isn't 64 bits.
-	 * This shouldn't be a problem on Linux or other POSIX
-	 * systems, since the configuration logic for libarchive
-	 * tries to obtain a 64-bit off_t.
-	 */
-	if ((old_offset = lseek(mine->fd, 0, SEEK_CUR)) >= 0 &&
-	    (new_offset = lseek(mine->fd, request, SEEK_CUR)) >= 0)
-		return (new_offset - old_offset);
+	/* Reduce a request that would overflow the 'skip' variable. */
+	if (sizeof(request) > sizeof(skip)) {
+		const int64_t max_skip =
+		    (((int64_t)1 << (skip_bits - 1)) - 1) * 2 + 1;
+		if (request > max_skip)
+			skip = max_skip;
+	}
+
+	if ((old_offset = lseek(mine->fd, 0, SEEK_CUR)) >= 0) {
+		if (old_offset >= mine->size ||
+		    skip > mine->size - old_offset) {
+			/* Do not seek past end of file. */
+			errno = ESPIPE;
+		} else if ((new_offset = lseek(mine->fd, skip, SEEK_CUR)) >= 0)
+			return (new_offset - old_offset);
+	}

 	/* If lseek() fails, don't bother trying again. */
 	mine->use_lseek = 0;
@@ -509,7 +521,7 @@ file_skip_lseek(struct archive *a, void *client_data, int64_t request)
 		archive_set_error(a, errno, "Error seeking in '%s'",
 		    mine->filename.m);
 	else
-		archive_set_error(a, errno, "Error seeking in '%S'",
+		archive_set_error(a, errno, "Error seeking in '%ls'",
 		    mine->filename.w);
 	return (-1);
 }
@@ -540,11 +552,24 @@ static int64_t
 file_seek(struct archive *a, void *client_data, int64_t request, int whence)
 {
 	struct read_file_data *mine = (struct read_file_data *)client_data;
+	off_t seek = (off_t)request;
 	int64_t r;
+	int seek_bits = sizeof(seek) * 8 - 1;

 	/* We use off_t here because lseek() is declared that way. */
-	/* See above for notes about when off_t is less than 64 bits. */
-	r = lseek(mine->fd, request, whence);
+
+	/* Reduce a request that would overflow the 'seek' variable. */
+	if (sizeof(request) > sizeof(seek)) {
+		const int64_t max_seek =
+		    (((int64_t)1 << (seek_bits - 1)) - 1) * 2 + 1;
+		const int64_t min_seek = ~max_seek;
+		if (request > max_seek)
+			seek = (off_t)max_seek;
+		else if (request < min_seek)
+			seek = (off_t)min_seek;
+	}
+
+	r = lseek(mine->fd, seek, whence);
 	if (r >= 0)
 		return r;

@@ -555,7 +580,7 @@ file_seek(struct archive *a, void *client_data, int64_t request, int whence)
 		archive_set_error(a, errno, "Error seeking in '%s'",
 		    mine->filename.m);
 	else
-		archive_set_error(a, errno, "Error seeking in '%S'",
+		archive_set_error(a, errno, "Error seeking in '%ls'",
 		    mine->filename.w);
 	return (ARCHIVE_FATAL);
 }
@@ -37,46 +37,32 @@ archive_read_support_filter_by_code(struct archive *a, int filter_code)
 	switch (filter_code) {
 	case ARCHIVE_FILTER_NONE:
 		return archive_read_support_filter_none(a);
-		break;
 	case ARCHIVE_FILTER_GZIP:
 		return archive_read_support_filter_gzip(a);
-		break;
 	case ARCHIVE_FILTER_BZIP2:
 		return archive_read_support_filter_bzip2(a);
-		break;
 	case ARCHIVE_FILTER_COMPRESS:
 		return archive_read_support_filter_compress(a);
-		break;
 	case ARCHIVE_FILTER_LZMA:
 		return archive_read_support_filter_lzma(a);
-		break;
 	case ARCHIVE_FILTER_XZ:
 		return archive_read_support_filter_xz(a);
-		break;
 	case ARCHIVE_FILTER_UU:
 		return archive_read_support_filter_uu(a);
-		break;
 	case ARCHIVE_FILTER_RPM:
 		return archive_read_support_filter_rpm(a);
-		break;
 	case ARCHIVE_FILTER_LZIP:
 		return archive_read_support_filter_lzip(a);
-		break;
 	case ARCHIVE_FILTER_LRZIP:
 		return archive_read_support_filter_lrzip(a);
-		break;
 	case ARCHIVE_FILTER_LZOP:
 		return archive_read_support_filter_lzop(a);
-		break;
 	case ARCHIVE_FILTER_GRZIP:
 		return archive_read_support_filter_grzip(a);
-		break;
 	case ARCHIVE_FILTER_LZ4:
 		return archive_read_support_filter_lz4(a);
-		break;
 	case ARCHIVE_FILTER_ZSTD:
 		return archive_read_support_filter_zstd(a);
-		break;
 	}
 	return (ARCHIVE_FATAL);
 }
@@ -328,6 +328,7 @@ next_code(struct archive_read_filter *self)
 	static int debug_buff[1024];
 	static unsigned debug_index;

+again:
 	code = newcode = getbits(self, state->bits);
 	if (code < 0)
 		return (code);
@@ -360,7 +361,7 @@ next_code(struct archive_read_filter *self)
 		state->section_end_code = (1 << state->bits) - 1;
 		state->free_ent = 257;
 		state->oldcode = -1;
-		return (next_code(self));
+		goto again;
 	}

 	if (code > state->free_ent
@@ -323,7 +323,6 @@ lz4_filter_read(struct archive_read_filter *self, const void **p)
 		archive_set_error(&self->archive->archive,
 		    ARCHIVE_ERRNO_MISC, "Program error.");
 		return (ARCHIVE_FATAL);
-		break;
 	}

 	while (state->stage == SELECT_STREAM) {
@@ -28,6 +28,9 @@
 #ifdef HAVE_ERRNO_H
 #include <errno.h>
 #endif
+#if HAVE_STDINT_H
+#include <stdint.h>
+#endif
 #ifdef HAVE_STDLIB_H
 #include <stdlib.h>
 #endif
@@ -50,6 +53,7 @@
 #include "archive_ppmd7_private.h"
 #include "archive_private.h"
 #include "archive_read_private.h"
+#include "archive_time_private.h"
 #include "archive_endian.h"

 #ifndef HAVE_ZLIB_H
@@ -59,7 +63,26 @@
 #define _7ZIP_SIGNATURE	"7z\xBC\xAF\x27\x1C"
 #define SFX_MIN_ADDR	0x27000
 #define SFX_MAX_ADDR	0x60000
+#define SFX_MAX_OFFSET	(SFX_MAX_ADDR - SFX_MIN_ADDR)

+/*
+ * PE format
+ */
+#define PE_DOS_HDR_LEN				0x40
+#define PE_DOS_HDR_ELFANEW_OFFSET	0x3c
+#define PE_COFF_HDR_LEN				0x18
+#define PE_COFF_HDR_SEC_CNT_OFFSET	0x6
+#define PE_COFF_HDR_OPT_SZ_OFFSET	0x14
+#define PE_SEC_HDR_LEN 				0x28
+#define PE_SEC_HDR_RAW_ADDR_OFFSET	0x14
+#define PE_SEC_HDR_RAW_SZ_OFFSET	0x10
+
+/*
+ * ELF format
+ */
+#define ELF_HDR_MIN_LEN 0x34
+#define ELF_HDR_EI_CLASS_OFFSET 0x04
+#define ELF_HDR_EI_DATA_OFFSET 0x05

 /*
 * Codec ID
@@ -83,6 +106,7 @@
 #define _7Z_ARM		0x03030501
 #define _7Z_ARMTHUMB	0x03030701
 #define _7Z_ARM64	0xa
+#define _7Z_RISCV	0xb
 #define _7Z_SPARC	0x03030805

 #define _7Z_ZSTD	0x4F71101 /* Copied from https://github.com/mcmilk/7-Zip-zstd.git */
@@ -144,7 +168,6 @@ struct _7z_digests {
 	uint32_t	*digests;
 };

-
 struct _7z_folder {
 	uint64_t		 numCoders;
 	struct _7z_coder {
@@ -224,13 +247,13 @@ struct _7zip_entry {
 #define CRC32_IS_SET	(1<<3)
 #define HAS_STREAM	(1<<4)

-	time_t			 mtime;
-	time_t			 atime;
-	time_t			 ctime;
-	long			 mtime_ns;
-	long			 atime_ns;
-	long			 ctime_ns;
-	uint32_t		 mode;
+	int64_t			 mtime;
+	int64_t			 atime;
+	int64_t			 ctime;
+	uint32_t		 mtime_ns;
+	uint32_t		 atime_ns;
+	uint32_t		 ctime_ns;
+	__LA_MODE_T		 mode;
 	uint32_t		 attr;
 };

@@ -388,7 +411,6 @@ static int	decode_encoded_header_info(struct archive_read *,
 static int	decompress(struct archive_read *, struct _7zip *,
 		    void *, size_t *, const void *, size_t *);
 static ssize_t	extract_pack_stream(struct archive_read *, size_t);
-static void	fileTimeToUtc(uint64_t, time_t *, long *);
 static uint64_t folder_uncompressed_size(struct _7z_folder *);
 static void	free_CodersInfo(struct _7z_coders_info *);
 static void	free_Digest(struct _7z_digests *);
@@ -424,7 +446,9 @@ static ssize_t	read_stream(struct archive_read *, const void **, size_t,
 		    size_t);
 static int	seek_pack(struct archive_read *);
 static int64_t	skip_stream(struct archive_read *, size_t);
-static int	skip_sfx(struct archive_read *, ssize_t);
+static int	skip_sfx(struct archive_read *, const ssize_t);
+static ssize_t	find_pe_overlay(struct archive_read *);
+static ssize_t	find_elf_data_sec(struct archive_read *);
 static int	slurp_central_directory(struct archive_read *, struct _7zip *,
 		    struct _7z_header_info *);
 static int	setup_decode_folder(struct archive_read *, struct _7z_folder *,
@@ -435,6 +459,8 @@ static void	arm_Init(struct _7zip *);
 static size_t	arm_Convert(struct _7zip *, uint8_t *, size_t);
 static size_t	arm64_Convert(struct _7zip *, uint8_t *, size_t);
 static ssize_t		Bcj2_Decode(struct _7zip *, uint8_t *, size_t);
+static size_t	sparc_Convert(struct _7zip *, uint8_t *, size_t);
+static size_t	powerpc_Convert(struct _7zip *, uint8_t *, size_t);


 int
@@ -522,15 +548,17 @@ archive_read_format_7zip_bid(struct archive_read *a, int best_bid)
 	 * It may a 7-Zip SFX archive file. If first two bytes are
 	 * 'M' and 'Z' available on Windows or first four bytes are
 	 * "\x7F\x45LF" available on posix like system, seek the 7-Zip
-	 * signature. Although we will perform a seek when reading
-	 * a header, what we do not use __archive_read_seek() here is
-	 * due to a bidding performance.
+	 * signature. While find_pe_overlay can be performed without
+	 * performing a seek, find_elf_data_sec requires one,
+	 * thus a performance difference between the two is expected. 
 	 */
 	if ((p[0] == 'M' && p[1] == 'Z') || memcmp(p, "\x7F\x45LF", 4) == 0) {
-		ssize_t offset = SFX_MIN_ADDR;
+		const ssize_t min_addr = p[0] == 'M' ? find_pe_overlay(a) :
+						       find_elf_data_sec(a);
+		ssize_t offset = min_addr;
 		ssize_t window = 4096;
 		ssize_t bytes_avail;
-		while (offset + window <= (SFX_MAX_ADDR)) {
+		while (offset + window <= (min_addr + SFX_MAX_OFFSET)) {
 			const char *buff = __archive_read_ahead(a,
 					offset + window, &bytes_avail);
 			if (buff == NULL) {
@@ -580,21 +608,14 @@ check_7zip_header_in_sfx(const char *p)
 }

 static int
-skip_sfx(struct archive_read *a, ssize_t bytes_avail)
+skip_sfx(struct archive_read *a, const ssize_t min_addr)
 {
 	const void *h;
 	const char *p, *q;
 	size_t skip, offset;
 	ssize_t bytes, window;

-	/*
-	 * If bytes_avail > SFX_MIN_ADDR we do not have to call
-	 * __archive_read_seek() at this time since we have
-	 * already had enough data.
-	 */
-	if (bytes_avail > SFX_MIN_ADDR)
-		__archive_read_consume(a, SFX_MIN_ADDR);
-	else if (__archive_read_seek(a, SFX_MIN_ADDR, SEEK_SET) < 0)
+	if (__archive_read_seek(a, min_addr, SEEK_SET) < 0)
 		return (ARCHIVE_FATAL);

 	offset = 0;
@@ -627,7 +648,7 @@ skip_sfx(struct archive_read *a, ssize_t bytes_avail)
 				    (struct _7zip *)a->format->data;
 				skip = p - (const char *)h;
 				__archive_read_consume(a, skip);
-				zip->seek_base = SFX_MIN_ADDR + offset + skip;
+				zip->seek_base = min_addr + offset + skip;
 				return (ARCHIVE_OK);
 			}
 			p += step;
@@ -644,6 +665,207 @@ fatal:
 	return (ARCHIVE_FATAL);
 }

+static ssize_t
+find_pe_overlay(struct archive_read *a)
+{
+	const char *h;
+	ssize_t bytes, max_offset, offset, sec_end;
+	ssize_t opt_hdr_sz, sec_cnt;
+
+	for (;;) {
+		/*
+		 * Read Dos header to find e_lfanew
+		 */
+		h = __archive_read_ahead(a, PE_DOS_HDR_LEN, &bytes);
+		if (h == NULL || h[0] != 'M' || h[1] != 'Z') {
+			break;
+		}
+		offset = archive_le32dec(h + PE_DOS_HDR_ELFANEW_OFFSET);
+
+		/*
+		 * Read COFF header to find opt header size and sec cnt
+		 */
+		if (bytes < offset + PE_COFF_HDR_LEN) {
+			h = __archive_read_ahead(a, offset + PE_COFF_HDR_LEN,
+			    &bytes);
+			if (h == NULL || h[offset] != 'P' ||
+			    h[offset + 1] != 'E') {
+				break;
+			}
+		}
+		sec_cnt = archive_le16dec(
+		    h + offset + PE_COFF_HDR_SEC_CNT_OFFSET);
+		opt_hdr_sz = archive_le16dec(
+		    h + offset + PE_COFF_HDR_OPT_SZ_OFFSET);
+
+		/*
+		 * Skip optional header
+		 */
+		if (opt_hdr_sz != 0) {
+			offset += PE_COFF_HDR_LEN + opt_hdr_sz;
+		} else {
+			break;
+		}
+
+		/*
+		 * Traverse sec table to find max raw offset (i.e., overlay)
+		 */
+		if (bytes < offset + sec_cnt * PE_SEC_HDR_LEN) {
+			h = __archive_read_ahead(a,
+			    offset + sec_cnt * PE_SEC_HDR_LEN, NULL);
+			if (h == NULL) {
+				break;
+			}
+		}
+		max_offset = offset;
+		while (sec_cnt > 0) {
+			sec_end = archive_le32dec(
+				      h + offset + PE_SEC_HDR_RAW_SZ_OFFSET) +
+			    archive_le32dec(
+				h + offset + PE_SEC_HDR_RAW_ADDR_OFFSET);
+			if (sec_end > max_offset) {
+				max_offset = sec_end;
+			}
+			offset += PE_SEC_HDR_LEN;
+			sec_cnt--;
+		}
+		return (max_offset);
+	}
+
+	/*
+	 * If encounter any weirdness, revert to old brute-force style search
+	 */
+	return (SFX_MIN_ADDR);
+}
+
+static ssize_t
+find_elf_data_sec(struct archive_read *a)
+{
+	const char *h;
+	char big_endian, format_64;
+	ssize_t bytes, min_addr = SFX_MIN_ADDR;
+	uint64_t e_shoff, strtab_offset, strtab_size;
+	uint16_t e_shentsize, e_shnum, e_shstrndx;
+	uint16_t (*dec16)(const void *);
+	uint32_t (*dec32)(const void *);
+	uint64_t (*dec64)(const void *);
+
+	for (;;) {
+		/*
+		 * Read Elf header to find bitness & endianness
+		 */
+		h = __archive_read_ahead(a, ELF_HDR_MIN_LEN, &bytes);
+		if (h == NULL || memcmp(h, "\x7F\x45LF", 4) != 0) {
+			break;
+		}
+		format_64 = h[ELF_HDR_EI_CLASS_OFFSET] == 0x2;
+		big_endian = h[ELF_HDR_EI_DATA_OFFSET] == 0x2;
+		if (big_endian) {
+			dec16 = &archive_be16dec;
+			dec32 = &archive_be32dec;
+			dec64 = &archive_be64dec;
+		} else {
+			dec16 = &archive_le16dec;
+			dec32 = &archive_le32dec;
+			dec64 = &archive_le64dec;
+		}
+
+		/*
+		 * Read section header table info
+		 */
+		if (format_64) {
+			e_shoff = (*dec64)(h + 0x28);
+			e_shentsize = (*dec16)(h + 0x3A);
+			e_shnum = (*dec16)(h + 0x3C);
+			e_shstrndx = (*dec16)(h + 0x3E);
+			if (e_shnum < e_shstrndx || e_shentsize < 0x28)
+				break;
+
+		} else {
+			e_shoff = (*dec32)(h + 0x20);
+			e_shentsize = (*dec16)(h + 0x2E);
+			e_shnum = (*dec16)(h + 0x30);
+			e_shstrndx = (*dec16)(h + 0x32);
+			if (e_shnum < e_shstrndx || e_shentsize < 0x18)
+				break;
+		}
+
+		/*
+		 * Reading the section table to find strtab section
+		 */
+		if (__archive_read_seek(a, e_shoff, SEEK_SET) < 0) {
+			break;
+		}
+		h = __archive_read_ahead(a, (size_t)e_shnum * (size_t)e_shentsize, NULL);
+		if (h == NULL) {
+			break;
+		}
+		if (format_64) {
+			strtab_offset = (*dec64)(
+			    h + e_shstrndx * e_shentsize + 0x18);
+			strtab_size = (*dec64)(
+			    h + e_shstrndx * e_shentsize + 0x20);
+		} else {
+			strtab_offset = (*dec32)(
+			    h + e_shstrndx * e_shentsize + 0x10);
+			strtab_size = (*dec32)(
+			    h + e_shstrndx * e_shentsize + 0x14);
+		}
+
+		/*
+		 * Read the STRTAB section to find the .data offset
+		 */
+		if (__archive_read_seek(a, strtab_offset, SEEK_SET) < 0) {
+			break;
+		}
+		h = __archive_read_ahead(a, strtab_size, NULL);
+		if (h == NULL) {
+			break;
+		}
+		ssize_t data_sym_offset = -1;
+		for (size_t offset = 0; offset < strtab_size - 6; offset++) {
+			if (memcmp(h + offset, ".data\00", 6) == 0) {
+				data_sym_offset = offset;
+				break;
+			}
+		}
+		if (data_sym_offset == -1) {
+			break;
+		}
+
+		/*
+		 * Find the section with the .data name
+		 */
+		if (__archive_read_seek(a, e_shoff, SEEK_SET) < 0) {
+			break;
+		}
+		h = __archive_read_ahead(a, (size_t)e_shnum * (size_t)e_shentsize, NULL);
+		if (h == NULL) {
+			break;
+		}
+		ssize_t sec_tbl_offset = 0, name_offset;
+		while (e_shnum > 0) {
+			name_offset = (*dec32)(h + sec_tbl_offset);
+			if (name_offset == data_sym_offset) {
+				if (format_64) {
+					min_addr = (*dec64)(
+					    h + sec_tbl_offset + 0x18);
+				} else {
+					min_addr = (*dec32)(
+					    h + sec_tbl_offset + 0x10);
+				}
+				break;
+			}
+			sec_tbl_offset += e_shentsize;
+			e_shnum--;
+		}
+		break;
+	}
+
+	__archive_read_seek(a, 0, SEEK_SET);
+	return (min_addr);
+}
+
 static int
 archive_read_format_7zip_read_header(struct archive_read *a,
 	struct archive_entry *entry)
@@ -770,7 +992,7 @@ archive_read_format_7zip_read_header(struct archive_read *a,
 		/* allocate for ",rdonly,hidden,system" */
 		fflags_text = malloc(22 * sizeof(*fflags_text));
 		if (fflags_text != NULL) {
-			ptr = fflags_text; 
+			ptr = fflags_text;
 			if (zip_entry->attr & FILE_ATTRIBUTE_READONLY) {
 				strcpy(ptr, ",rdonly");
 				ptr = ptr + 7;
@@ -1109,7 +1331,9 @@ init_decompression(struct archive_read *a, struct _7zip *zip,
 			if (coder2->codec != _7Z_X86 &&
 			    coder2->codec != _7Z_X86_BCJ2 &&
 			    coder2->codec != _7Z_ARM &&
-			    coder2->codec != _7Z_ARM64) {
+			    coder2->codec != _7Z_ARM64 &&
+			    coder2->codec != _7Z_POWERPC &&
+			    coder2->codec != _7Z_SPARC) {
 				archive_set_error(&a->archive,
 				    ARCHIVE_ERRNO_MISC,
 				    "Unsupported filter %lx for %lx",
@@ -1223,6 +1447,12 @@ init_decompression(struct archive_read *a, struct _7zip *zip,
 				filters[fi].id = LZMA_FILTER_ARM64;
 				fi++;
 				break;
+#endif
+#ifdef LZMA_FILTER_RISCV
+			case _7Z_RISCV:
+				filters[fi].id = LZMA_FILTER_RISCV;
+				fi++;
+				break;
 #endif
 			case _7Z_SPARC:
 				filters[fi].id = LZMA_FILTER_SPARC;
@@ -1396,6 +1626,7 @@ init_decompression(struct archive_read *a, struct _7zip *zip,
 	case _7Z_ARM:
 	case _7Z_ARMTHUMB:
 	case _7Z_ARM64:
+	case _7Z_RISCV:
 	case _7Z_SPARC:
 	case _7Z_DELTA:
 		archive_set_error(&a->archive, ARCHIVE_ERRNO_MISC,
@@ -1708,6 +1939,10 @@ decompress(struct archive_read *a, struct _7zip *zip,
 			*outbytes = arm_Convert(zip, buff, *outbytes);
 		} else if (zip->codec2 == _7Z_ARM64) {
 			*outbytes = arm64_Convert(zip, buff, *outbytes);
+		} else if (zip->codec2 == _7Z_SPARC) {
+			*outbytes = sparc_Convert(zip, buff, *outbytes);
+		} else if (zip->codec2 == _7Z_POWERPC) {
+			*outbytes = powerpc_Convert(zip, buff, *outbytes);
 		}
 	}

@@ -2841,23 +3076,6 @@ read_Header(struct archive_read *a, struct _7z_header_info *h,
 	return (0);
 }

-#define EPOC_TIME ARCHIVE_LITERAL_ULL(116444736000000000)
-static void
-fileTimeToUtc(uint64_t fileTime, time_t *timep, long *ns)
-{
-
-	if (fileTime >= EPOC_TIME) {
-		fileTime -= EPOC_TIME;
-		/* milli seconds base */
-		*timep = (time_t)(fileTime / 10000000);
-		/* nano seconds base */
-		*ns = (long)(fileTime % 10000000) * 100;
-	} else {
-		*timep = 0;
-		*ns = 0;
-	}
-}
-
 static int
 read_Times(struct archive_read *a, struct _7z_header_info *h, int type)
 {
@@ -2900,19 +3118,19 @@ read_Times(struct archive_read *a, struct _7z_header_info *h, int type)
 			goto failed;
 		switch (type) {
 		case kCTime:
-			fileTimeToUtc(archive_le64dec(p),
+			ntfs_to_unix(archive_le64dec(p),
 			    &(entries[i].ctime),
 			    &(entries[i].ctime_ns));
 			entries[i].flg |= CTIME_IS_SET;
 			break;
 		case kATime:
-			fileTimeToUtc(archive_le64dec(p),
+			ntfs_to_unix(archive_le64dec(p),
 			    &(entries[i].atime),
 			    &(entries[i].atime_ns));
 			entries[i].flg |= ATIME_IS_SET;
 			break;
 		case kMTime:
-			fileTimeToUtc(archive_le64dec(p),
+			ntfs_to_unix(archive_le64dec(p),
 			    &(entries[i].mtime),
 			    &(entries[i].mtime_ns));
 			entries[i].flg |= MTIME_IS_SET;
@@ -3007,7 +3225,9 @@ slurp_central_directory(struct archive_read *a, struct _7zip *zip,

 	if ((p[0] == 'M' && p[1] == 'Z') || memcmp(p, "\x7F\x45LF", 4) == 0) {
 		/* This is an executable ? Must be self-extracting... */
-		r = skip_sfx(a, bytes_avail);
+		const ssize_t min_addr = p[0] == 'M' ? find_pe_overlay(a) :
+						       find_elf_data_sec(a);
+		r = skip_sfx(a, min_addr);
 		if (r < ARCHIVE_WARN)
 			return (r);
 		if ((p = __archive_read_ahead(a, 32, &bytes_avail)) == NULL)
@@ -4007,6 +4227,121 @@ arm64_Convert(struct _7zip *zip, uint8_t *buf, size_t size)
 	return i;
 }

+static size_t
+sparc_Convert(struct _7zip *zip, uint8_t *buf, size_t size)
+{
+	// This function was adapted from
+	// static size_t bcj_sparc(struct xz_dec_bcj *s, uint8_t *buf, size_t size)
+	// in https://git.tukaani.org/xz-embedded.git
+
+	/*
+	 * Branch/Call/Jump (BCJ) filter decoders
+	 *
+	 * Authors: Lasse Collin <lasse.collin@tukaani.org>
+	 *          Igor Pavlov <https://7-zip.org/>
+	 *
+	 * Copyright (C) The XZ Embedded authors and contributors
+	 *
+	 * Permission to use, copy, modify, and/or distribute this
+	 * software for any purpose with or without fee is hereby granted.
+	 *
+	 * THE SOFTWARE IS PROVIDED "AS IS" AND THE AUTHOR DISCLAIMS ALL
+	 * WARRANTIES WITH REGARD TO THIS SOFTWARE INCLUDING ALL IMPLIED
+	 * WARRANTIES OF MERCHANTABILITY AND FITNESS. IN NO EVENT SHALL
+	 * THE AUTHOR BE LIABLE FOR ANY SPECIAL, DIRECT, INDIRECT, OR
+	 * CONSEQUENTIAL DAMAGES OR ANY DAMAGES WHATSOEVER RESULTING FROM
+	 * LOSS OF USE, DATA OR PROFITS, WHETHER IN AN ACTION OF CONTRACT,
+	 * NEGLIGENCE OR OTHER TORTIOUS ACTION, ARISING OUT OF OR IN
+	 * CONNECTION WITH THE USE OR PERFORMANCE OF THIS SOFTWARE.
+	 */
+
+	size_t i;
+	uint32_t instr;
+
+	size &= ~(size_t)3;
+
+	for (i = 0; i < size; i += 4) {
+		instr = (uint32_t)(buf[i] << 24)
+			| ((uint32_t)buf[i+1] << 16)
+			| ((uint32_t)buf[i+2] << 8)
+			| (uint32_t)buf[i+3];
+
+		if ((instr >> 22) == 0x100 || (instr >> 22) == 0x1FF) {
+			instr <<= 2;
+			instr -= zip->bcj_ip + (uint32_t)i;
+			instr >>= 2;
+			instr = ((uint32_t)0x40000000 - (instr & 0x400000))
+			        | 0x40000000 | (instr & 0x3FFFFF);
+
+			buf[i] = (uint8_t)(instr >> 24);
+			buf[i+1] = (uint8_t)(instr >> 16);
+			buf[i+2] = (uint8_t)(instr >> 8);
+			buf[i+3] = (uint8_t)instr;
+		}
+	}
+
+	zip->bcj_ip += (uint32_t)i;
+
+	return i;
+}
+
+static size_t
+powerpc_Convert(struct _7zip *zip, uint8_t *buf, size_t size)
+{
+	// This function was adapted from
+	// static size_t powerpc_code(void *simple, uint32_t now_pos, bool is_encoder, uint8_t *buffer, size_t size)
+	// in https://git.tukaani.org/xz.git
+
+	/*
+	 * Filter for PowerPC (big endian) binaries
+	 *
+	 * Authors: Igor Pavlov
+	 *          Lasse Collin
+	 *
+	 * Copyright (C) The XZ Utils authors and contributors
+	 *
+	 * Permission to use, copy, modify, and/or distribute this
+	 * software for any purpose with or without fee is hereby granted.
+	 *
+	 * THE SOFTWARE IS PROVIDED "AS IS" AND THE AUTHOR DISCLAIMS ALL
+	 * WARRANTIES WITH REGARD TO THIS SOFTWARE INCLUDING ALL IMPLIED
+	 * WARRANTIES OF MERCHANTABILITY AND FITNESS. IN NO EVENT SHALL
+	 * THE AUTHOR BE LIABLE FOR ANY SPECIAL, DIRECT, INDIRECT, OR
+	 * CONSEQUENTIAL DAMAGES OR ANY DAMAGES WHATSOEVER RESULTING FROM
+	 * LOSS OF USE, DATA OR PROFITS, WHETHER IN AN ACTION OF CONTRACT,
+	 * NEGLIGENCE OR OTHER TORTIOUS ACTION, ARISING OUT OF OR IN
+	 * CONNECTION WITH THE USE OR PERFORMANCE OF THIS SOFTWARE.
+	 */
+
+	size &= ~(size_t)3;
+
+	size_t i;
+	for (i = 0; i < size; i += 4) {
+		// PowerPC branch 6(48) 24(Offset) 1(Abs) 1(Link)
+		if ((buf[i] >> 2) == 0x12
+			&& ((buf[i + 3] & 3) == 1)) {
+
+			const uint32_t src
+				= (((uint32_t)(buf[i + 0]) & 3) << 24)
+				| ((uint32_t)(buf[i + 1]) << 16)
+				| ((uint32_t)(buf[i + 2]) << 8)
+				| ((uint32_t)(buf[i + 3]) & ~UINT32_C(3));
+
+			uint32_t dest = src - (zip->bcj_ip + (uint32_t)(i));
+
+			buf[i + 0] = 0x48 | ((dest >> 24) &  0x03);
+			buf[i + 1] = (dest >> 16);
+			buf[i + 2] = (dest >> 8);
+			buf[i + 3] &= 0x03;
+			buf[i + 3] |= dest;
+		}
+	}
+
+	zip->bcj_ip += (uint32_t)i;
+
+	return i;
+}
+
 /*
 * Brought from LZMA SDK.
 *
@@ -4029,8 +4364,17 @@ arm64_Convert(struct _7zip *zip, uint8_t *buf, size_t size)

 #define RC_READ_BYTE (*buffer++)
 #define RC_TEST { if (buffer == bufferLim) return SZ_ERROR_DATA; }
-#define RC_INIT2 zip->bcj2_code = 0; zip->bcj2_range = 0xFFFFFFFF; \
-  { int ii; for (ii = 0; ii < 5; ii++) { RC_TEST; zip->bcj2_code = (zip->bcj2_code << 8) | RC_READ_BYTE; }}
+#define RC_INIT2 do {							\
+	zip->bcj2_code = 0;						\
+	zip->bcj2_range = 0xFFFFFFFF;					\
+	{								\
+		int ii;							\
+		for (ii = 0; ii < 5; ii++) {				\
+			RC_TEST;					\
+			zip->bcj2_code = (zip->bcj2_code << 8) | RC_READ_BYTE; \
+		}							\
+	}								\
+} while (0)

 #define NORMALIZE if (zip->bcj2_range < kTopValue) { RC_TEST; zip->bcj2_range <<= 8; zip->bcj2_code = (zip->bcj2_code << 8) | RC_READ_BYTE; }

@@ -4177,4 +4521,3 @@ Bcj2_Decode(struct _7zip *zip, uint8_t *outBuf, size_t outSize)

 	return ((ssize_t)outPos);
 }
-
@@ -41,49 +41,34 @@ archive_read_support_format_by_code(struct archive *a, int format_code)
 	switch (format_code & ARCHIVE_FORMAT_BASE_MASK) {
 	case ARCHIVE_FORMAT_7ZIP:
 		return archive_read_support_format_7zip(a);
-		break;
 	case ARCHIVE_FORMAT_AR:
 		return archive_read_support_format_ar(a);
-		break;
 	case ARCHIVE_FORMAT_CAB:
 		return archive_read_support_format_cab(a);
-		break;
 	case ARCHIVE_FORMAT_CPIO:
 		return archive_read_support_format_cpio(a);
-		break;
 	case ARCHIVE_FORMAT_EMPTY:
 		return archive_read_support_format_empty(a);
-		break;
 	case ARCHIVE_FORMAT_ISO9660:
 		return archive_read_support_format_iso9660(a);
-		break;
 	case ARCHIVE_FORMAT_LHA:
 		return archive_read_support_format_lha(a);
-		break;
 	case ARCHIVE_FORMAT_MTREE:
 		return archive_read_support_format_mtree(a);
-		break;
 	case ARCHIVE_FORMAT_RAR:
 		return archive_read_support_format_rar(a);
-		break;
 	case ARCHIVE_FORMAT_RAR_V5:
 		return archive_read_support_format_rar5(a);
-		break;
 	case ARCHIVE_FORMAT_RAW:
 		return archive_read_support_format_raw(a);
-		break;
 	case ARCHIVE_FORMAT_TAR:
 		return archive_read_support_format_tar(a);
-		break;
 	case ARCHIVE_FORMAT_WARC:
 		return archive_read_support_format_warc(a);
-		break;
 	case ARCHIVE_FORMAT_XAR:
 		return archive_read_support_format_xar(a);
-		break;
 	case ARCHIVE_FORMAT_ZIP:
 		return archive_read_support_format_zip(a);
-		break;
 	}
 	archive_set_error(a, ARCHIVE_ERRNO_PROGRAMMER,
 	    "Invalid format code specified");
@@ -363,7 +363,12 @@ archive_read_support_format_cab(struct archive *_a)
 		return (ARCHIVE_FATAL);
 	}
 	archive_string_init(&cab->ws);
-	archive_wstring_ensure(&cab->ws, 256);
+	if (archive_wstring_ensure(&cab->ws, 256) == NULL) {
+		archive_set_error(&a->archive, ENOMEM,
+		    "Can't allocate memory");
+		free(cab);
+		return (ARCHIVE_FATAL);
+	}

 	r = __archive_read_register_format(a,
 	    cab,
@@ -189,6 +189,7 @@ struct cpio {
 };

 static int64_t	atol16(const char *, unsigned);
+static uint64_t	atol16u(const char *, unsigned);
 static int64_t	atol8(const char *, unsigned);
 static int	archive_read_format_cpio_bid(struct archive_read *, int);
 static int	archive_read_format_cpio_options(struct archive_read *,
@@ -835,6 +836,7 @@ header_afiol(struct archive_read *a, struct cpio *cpio,
    struct archive_entry *entry, size_t *namelength, size_t *name_pad)
 {
 	int64_t t;
+	uint64_t u;
 	const void *h;
 	const char *header;

@@ -851,12 +853,12 @@ header_afiol(struct archive_read *a, struct cpio *cpio,

 	archive_entry_set_dev(entry, 
 		(dev_t)atol16(header + afiol_dev_offset, afiol_dev_size));
-	t = atol16(header + afiol_ino_offset, afiol_ino_size);
-	if (t < 0) {
-		archive_set_error(&a->archive, 0, "Nonsensical ino value");
-		return (ARCHIVE_FATAL);
-	}
-	archive_entry_set_ino(entry, t);
+	u = atol16u(header + afiol_ino_offset, afiol_ino_size);
+#if ARCHIVE_VERSION_NUMBER < 4000000
+	archive_entry_set_ino(entry, (int64_t)(u & INT64_MAX));
+#else
+	archive_entry_set_ino(entry, u);
+#endif
 	archive_entry_set_mode(entry,
 		(mode_t)atol8(header + afiol_mode_offset, afiol_mode_size));
 	archive_entry_set_uid(entry, atol16(header + afiol_uid_offset, afiol_uid_size));
@@ -1030,6 +1032,12 @@ atol8(const char *p, unsigned char_cnt)

 static int64_t
 atol16(const char *p, unsigned char_cnt)
+{
+	return ((int64_t)atol16u(p, char_cnt));
+}
+
+static uint64_t
+atol16u(const char *p, unsigned char_cnt)
 {
 	uint64_t l;
 	int digit;
@@ -1048,7 +1056,7 @@ atol16(const char *p, unsigned char_cnt)
 		l <<= 4;
 		l |= digit;
 	}
-	return ((int64_t)l);
+	return (l);
 }

 static int