|
5db5a704
|
2023-08-09T18:39:14
|
|
html: Fix UAF in htmlCurrentChar
Short-lived regression found by OSS-Fuzz.
|
|
|
b973ceaf
|
2023-08-09T18:37:20
|
|
parser: Fix mistake in xmlDetectEncoding
Short-lived regression.
|
|
|
cb717d7e
|
2023-08-09T16:52:02
|
|
parser: Update line number after coalescing text nodes
This should make the line number of text nodes deterministic. Before,
it depended on the callback sequence which depends on the size of chunks
fed to the parser.
|
|
|
855818bd
|
2023-08-08T15:21:37
|
|
parser: Check for truncated multi-byte sequences
When decoding input data, check whether the "raw" buffer is empty after
parsing the document. Otherwise, the input ends with a truncated
multi-byte sequence which shouldn't be silently ignored.
|
|
|
95e81a36
|
2023-08-08T15:21:31
|
|
parser: Decode all data in xmlCharEncInput
Even with flush set to true, xmlCharEncInput didn't guarantee to decode
all data. This complicated the push parser.
Remove the flush flag and always decode all available data.
Also fix ICU code where the flush flag has a different meaning. Always
set flush to false and retry even with empty input buffers.
|
|
|
834b8123
|
2023-08-08T15:21:28
|
|
parser: Stream data when reading from memory
Don't create a copy of the whole input buffer. Read the data chunk by
chunk to save memory.
Historically, it was probably envisioned to read data from memory
without additional copying. This doesn't work reliably with the current
design of the XML parser which requires a terminating null byte at the
end of input buffers. This lead to xmlReadMemory interfaces, which
expect pointer and size arguments, being changed to make a
zero-terminated copy of the input buffer. Interfaces based on
xmlReadDoc, which actually expect a zero-terminated string and
would make zero-copy operation work, were then simplified to rely on
xmlReadMemoryi, resulting in an unnecessary copy.
To avoid copying (possibly gigabytes) of memory temporarily, we now
stream in-memory input just like content read from files in a
chunk-by-chunk fashion (using a somewhat outdated INPUT_CHUNK size of
250 bytes). As a side effect, we also avoid another copy of the whole
input when handling non-UTF-8 data which was made possible by some
earlier commits.
Interfaces expecting zero-terminated strings now make use of strnlen
which unfortunately isn't part of the standard C library and only
mandated since POSIX 2008.
|
|
|
5aff27ae
|
2023-08-08T15:21:25
|
|
parser: Optimize xmlLoadEntityContent
Load entity content via xmlParserInputBufferGrow, avoiding a copy.
This also fixes an entity size accounting error.
|
|
|
facc2a06
|
2023-08-08T15:21:21
|
|
parser: Don't overwrite EOF parser state
|
|
|
59fa0bb3
|
2023-08-08T15:21:14
|
|
parser: Simplify input pointer updates
The base member always points to the beginning of the buffer.
|
|
|
c88ab7e3
|
2023-08-08T15:19:54
|
|
parser: Don't reinitialize parser input members
The parser input struct should already be initialized.
|
|
|
4ee08155
|
2023-08-08T15:19:51
|
|
encoding: Move rawconsumed accounting to xmlCharEncInput
|
|
|
a0462e2d
|
2023-08-08T15:19:49
|
|
test: Add push parser test with overridden encoding
After recent changes, it should work to call xmlSwitchEncoding to
override the encoding for the push parser. This was never properly
supported, so Chromium and WebKit added a hack to reset the encoding in
the startDocument SAX handler.
|
|
|
ec7be506
|
2023-08-08T15:19:46
|
|
parser: Rework encoding detection
Introduce XML_INPUT_HAS_ENCODING flag for xmlParserInput which is set
when xmlSwitchEncoding is called. The parser can use the flag to
reliably detect whether an encoding was already set via user override,
BOM or other auto-detection. In this case, the encoding declaration
won't be used to switch the encoding.
Before, an inscrutable mix of ctxt->charset, ctxt->input->encoding
and ctxt->input->buf->encoder was used.
Introduce private helper functions to switch encodings used by both the
XML and HTML parser:
- xmlDetectEncoding which skips over the BOM, allowing to remove the
BOM checks from other encoding functions.
- xmlSetDeclaredEncoding, replacing htmlCheckEncodingDirect, which warns
about encoding mismatches.
If users override the encoding, store the declared instead of the actual
encoding in xmlDoc. In this case, the actual encoding is known and the
raw value from the doc is more useful.
Also use the input flags to store the ISO-8859-1 fallback state.
Restrict the fallback to cases where no encoding was specified. (The
fallback is only useful in recovery mode and these days broken UTF-8 is
probably more likely than ISO-8859-1, so it might eventually be removed
completely.)
The 'charset' member of xmlParserCtxt is now unused. The 'encoding'
member of xmlParserInput is now unused.
The 'standalone' member of xmlParserInput is renamed to 'flags'.
A new parser state XML_PARSER_XML_DECL is added for the push parser.
|
|
|
d38e73f9
|
2023-08-08T15:19:44
|
|
parser: Always create UTF-8 in xmlParseReference
It seems that this code path could only be triggered after an encoding
error in recovery mode. Creating char-ref nodes is unnecessary and
typically unexpected.
|
|
|
131d0dc0
|
2023-08-08T15:19:39
|
|
parser: Don't use 'standalone' member of xmlParserInput
The standalone declaration is only parsed in the main input stream.
|
|
|
d9ec182b
|
2023-08-08T15:19:36
|
|
parser: Don't detect encoding in xmlCtxtResetPush
The encoding will be detected in xmlParseTryOrFinish.
|
|
|
3a64f394
|
2023-08-08T15:19:25
|
|
html: Remove some debugging code in htmlParseTryOrFinish
|
|
|
58de9d31
|
2023-08-03T12:00:55
|
|
valid: Fix c1->parent pointer in xmlCopyDocElementContent
Fixes #572.
|
|
|
75693281
|
2023-07-21T14:50:30
|
|
malloc-fail: Fix memory leak in xmlCompileAttributeTest
Found by OSS-Fuzz, see #344.
|
|
|
90bcbcfc
|
2023-07-20T21:08:01
|
|
parser: Fix potential use-after-free in xmlParseCharDataInternal
Return immediately if a SAX handler stops the parser.
Fixes #569.
|
|
|
88447447
|
2023-06-23T23:04:30
|
|
parser: Fix typo in previous commit
|
|
|
9d0541dd
|
2023-06-22T18:06:53
|
|
parser: Make xmlSwitchEncoding always skip the BOM
Chromium calls xmlSwitchEncoding from the start document handler and
relies on this function to skip the BOM. Commit 98840d40 changed the
behavior when switching to UTF-16 since inspecting the input buffer at
this point is fragile.
Revert part of the commit to also skip a potential (decoded UTF-8) BOM
when switching to UTF-16. Make sure that we do this only at the start of
an input stream to avoid U-FEFF characters being lost.
BOM handling should ultimately be moved to the parsing code to avoid
such bugs.
See https://bugs.chromium.org/p/chromium/issues/detail?id=1451026
|
|
2473b485
|
2023-06-21T14:15:02
|
|
autotools: fix Python module file ext for cygwin/msys2
both use .dll, not .pyd
|
|
5f54bac9
|
2023-06-10T10:50:02
|
|
testapi: test_xmlSAXDefaultVersion() leaves xmlSAX2DefaultVersionValue set to 1 with LIBXML_SAX1_ENABLED
Add code to save and to restore the default value of
xmlSAX2DefaultVersionValue.
Fixes #554.
|
|
|
b236b7a5
|
2023-06-08T21:53:05
|
|
parser: Halt parser when growing buffer results in OOM
Fix short-lived regression from previous commit.
It might be safer to make xmlBufSetInputBaseCur use the original buffer
even in case of errors.
Found by OSS-Fuzz.
|
|
|
20f5c734
|
2023-06-07T14:05:34
|
|
parser: Recover more input from encoding errors
Don't halt the parser in xmlParserGrow to allow more input to be
recovered in case of encoding errors.
Fixes #543.
|
|
|
db21cd5d
|
2023-06-06T14:25:30
|
|
malloc-fail: Handle malloc failures in xmlAddEncodingAlias
Avoid memory errors if an allocation fails.
See #344. Fixes #553.
|
|
|
305a75cc
|
2023-06-06T13:15:46
|
|
malloc-fail: Fix null-deref with xmllint --copy
See #344. Fixes #552.
|
|
|
6273df6c
|
2023-05-30T12:30:27
|
|
xpath: Ignore entity ref nodes when computing node hash
XPath queries only work reliably if entities are substituted.
Nevertheless, it's possible to query a document with entity reference
nodes. xmllint even deletes entities when the `--dropdtd` option is
passed, resulting in dangling pointers, so it's best to skip entity
reference nodes to avoid a use-after-free.
Fixes #550.
|
|
|
e2f21c22
|
2023-05-25T13:01:48
|
|
win32: Deprecate old Windows build system
|
|
|
1e8ab697
|
2023-05-25T03:03:33
|
|
gitlab-ci: Lower _XOPEN_SOURCE value
|
|
|
cb8ccb10
|
2023-05-25T03:07:57
|
|
testapi: Don't set http_proxy environment variable
We already disable network access, so this has no effect.
|
|
|
9fd57df8
|
2023-05-25T02:37:57
|
|
autotools: Improve iconv check
Use a custom test program which includes iconv.h, so we can check
whether the possibly redefined symbols in this header file match the
symbols in the iconv library.
Should fix #547.
|
|
|
c3c6cc62
|
2023-05-24T20:08:33
|
|
runtest: Fix compilation without LIBXML_HTML_ENABLED
Fixes #545.
|
|
|
981093ab
|
2023-05-18T19:23:58
|
|
test: Add push parser tests for split UTF-8 sequences
|
|
|
e0f3016f
|
2023-05-18T17:31:44
|
|
parser: Fix regression when push parsing UTF-8 sequences
Partial UTF-8 sequences are allowed when push parsing.
Fixes #542.
|
|
|
687a2b71
|
2023-05-08T17:05:13
|
|
xinclude: Lower initial table size when fuzzing
We don't have test cases with many documents, so set the initial table
size to 1 when fuzzing, so there is a chance to detect reallocation
issues.
|
|
|
c40cbf07
|
2023-05-08T17:03:00
|
|
malloc-fail: Fix null deref after xmlXIncludeNewRef
See #344.
|
|
|
105ce73d
|
2023-05-08T16:45:28
|
|
xinclude: Fix false positives in inclusion loop detection
xmlXIncludeRecurseDoc can realloc the cache.
|
|
|
bdb5667a
|
2023-05-10T18:13:47
|
|
autotools: Fix ICU detection
Fixes #540.
|
|
|
9dae389c
|
2023-05-09T13:28:06
|
|
parser: Fix "huge input lookup" error with push parser
Fix parsing of larger documents without XML_PARSE_HUGE.
Should fix #538.
|
|
|
b8961df6
|
2023-05-09T03:25:24
|
|
SAX: Always validate xml:ids
The behavior shouldn't depend on mostly random configuration options.
|
|
|
f24ffddb
|
2023-05-08T23:33:04
|
|
Stop using sprintf
Switch remaining users to snprintf.
|
|
|
01723fc6
|
2023-05-08T23:12:33
|
|
xpath: Fix build without LIBXML_XPATH_ENABLED
Move static function declaration into XPATH block. Also move comparison
functions.
Fixes #537.
|
|
|
235b15a5
|
2023-05-08T17:58:02
|
|
SAX: Always initialize SAX1 element handlers
Follow-up to commit d0c3f01e. A parser context will be initialized to
SAX version 2, but this can be overridden with XML_PARSE_SAX1 later,
so we must initialize the SAX1 element handlers as well.
Change the check in xmlDetectSAX2 to only look for XML_SAX2_MAGIC, so
we don't switch to SAX1 if the SAX2 element handlers are NULL.
|
|
34630630
|
2023-05-05T17:34:57
|
|
autoconf: fix iconv library paths
and pass cflags when building executables
See 0f77167f for prior related work
|
|
|
d0c3f01e
|
2023-05-06T17:47:37
|
|
parser: Fix old SAX1 parser with custom callbacks
For some reason, xmlCtxtUseOptionsInternal set the start and end element
SAX handlers to the internal DOM builder functions when XML_PARSE_SAX1
was specified. This means that custom SAX handlers could never work with
that flag because these functions would receive the wrong user data
argument and crash immediately.
Fixes #535.
|
|
|
06a2c251
|
2023-05-06T15:28:13
|
|
hash: Fix possible startup crash with old libxslt versions
Call xmlInitParser in xmlHashCreate to make it work if the library
wasn't initialized yet.
Otherwise, exsltRegisterAll from libxslt 1.1.24 or older might cause
a crash.
See #534.
|
|
|
a800b7e0
|
2023-05-04T12:47:00
|
|
regexp: Fix null deref in xmlFAFinishReduceEpsilonTransitions
Short-lived regression found by OSS-Fuzz.
|
|
|
8d5e33ef
|
2023-05-03T20:42:10
|
|
Fix compiler warning on GCC < 8
-Wcast-function-type is only available since GCC 8.
|
|
|
d6882f64
|
2023-05-03T18:33:20
|
|
threads: Fix startup crash with weak symbol hack
Fix another issue when running with older libc, threads and libpthread
not linked in.
|
|
|
7f3f3f11
|
2023-05-03T03:20:14
|
|
dict: Raise MAX_DICT_HASH limit
This fixes quadratic behavior with large dictionaries.
Also rework testdict.c to support tests with larger dictionaries.
|
|
|
11a95279
|
2023-05-02T13:32:24
|
|
win32: Don't depend on removed .def file
Fixes broken build after 21cec82b.
Fixes #532.
|
|
|
c613ab14
|
2023-05-02T00:32:50
|
|
regexp: Fix mistake in previous commit
The `ret = 0` line should have been deleted.
Fixes #531.
|
|
|
a06eaa61
|
2023-03-09T06:58:24
|
|
regexp: Fix determinism checks
Swap arguments in initial call to xmlFARecurseDeterminism.
Fix the check whether we revisit the initial state in
xmlFARecurseDeterminism.
If there are transitions with equal atoms and targets but different
counters, treat the regex as deterministic but mark the transitions as
non-deterministic internally.
Don't overwrite zero return value of xmlFAComputesDeterminism
with non-zero value from xmlFARecurseDeterminism.
Most of these errors lead to non-deterministic regexes not being
detected which typically isn't an issue. The improved code may break
users who relied on buggy behavior or cause other bugs to become
visible.
Fixes #469.
|
|
|
e301865e
|
2023-03-09T05:34:38
|
|
regexp: Fix checks for eliminated transitions
'to' can be set to -1 or -2 when eliminating transitions, so check for
all negative values.
|
|
|
90759c59
|
2023-03-09T16:34:11
|
|
regexp: Simplify xmlFAReduceEpsilonTransitions
|
|
|
9f7b1142
|
2023-03-09T05:25:09
|
|
regexp: Fix cycle check in xmlFAReduceEpsilonTransitions
The visited flag must only be reset after the first call to
xmlFAReduceEpsilonTransitions has finished. Visiting states multiple
times could lead to unnecessary processing of duplicate transitions.
Similar to 68eadabd.
|
|
|
550eaac6
|
2023-04-30T19:40:43
|
|
writer: Add error check in xmlTextWriterEndDocument
|
|
|
2f12e3a9
|
2023-04-30T18:46:05
|
|
encoding: Stop calling xmlEncodingErr
This invokes the global error handler which should be avoided.
|
|
|
b230861d
|
2023-04-30T18:38:16
|
|
xmlIO: Remove some calls to xmlIOErr
The xmlIOErr functions use the global error handler and should be
avoided if possible.
|
|
|
4f49017e
|
2023-04-30T21:26:55
|
|
tests: Test streaming schema validation
|
|
|
d88763cc
|
2023-04-30T21:26:03
|
|
schemas: Fix filename in xmlSchemaValidateFile
Make sure that filename appears in error messages.
|
|
|
165f3436
|
2023-04-30T21:24:50
|
|
schemas: Fix line numbers in streaming validation
|
|
|
57d88da6
|
2023-04-30T21:30:21
|
|
schemas: Fix memory leak in xmlSchemaValidateStream
Regressed in 9a82b94a.
Fixes #530.
|
|
|
0ffc2d82
|
2023-04-30T20:28:47
|
|
runtest: Skip element name in schema error messages
This makes sure that memory and streaming tests will report the same
messages.
|
|
|
320f5084
|
2023-04-30T18:25:09
|
|
parser: Improve handling of encoding and IO errors
Make sure that xmlCharEncInput, xmlParserInputBufferPush and
xmlParserInputBufferGrow set the correct error code in the
xmlParserInputBuffer. Handle errors when calling these functions.
|
|
|
fc69cf56
|
2023-04-30T17:51:29
|
|
parser: Move xmlFatalErr to parserInternals.c
|
|
|
3ff6abbf
|
2023-02-22T17:11:20
|
|
encoding: Rework error codes
Use an enum instead of magic numbers. Fix a few error codes. Simplify
handling of "space" and "partial" errors.
See #506.
|
|
|
b463b38b
|
2023-04-30T16:19:28
|
|
.gitignore: Split up and rearrange .gitignore files
|
|
|
0260de55
|
2023-04-30T16:00:44
|
|
.gitignore: Add runsuite.log
|
|
|
886bf4e6
|
2023-04-30T15:35:47
|
|
Stop calling xmlMemoryDump
This was used to check for memory leaks but could potentially create a
.memdump file. These days, there are better ways to check for memory
leaks.
|
|
|
fc119e32
|
2023-04-30T15:28:12
|
|
examples: Don't call xmlCleanupParser and xmlMemoryDump
xmlCleanupParser is dangerous and shouldn't be called in most cases.
Being part of the examples led many people to use it incorrectly.
xmlMemoryDump is an obsolete way to test for memory leaks.
|
|
|
3da09318
|
2023-04-30T13:16:12
|
|
libxml.m4: Fix "possibly undefined macro" error
See #527.
|
|
fdc9cfaa
|
2023-04-30T13:02:31
|
|
cmake: Fix va_copy detection
va_copy is defined by the standard as a macro, so check_function_exists
will not detect it; check_symbol_exists will.
Fixes #528.
|
|
|
fa993130
|
2023-04-30T12:57:09
|
|
xpath: Remove remaining references to valueFrame
Fixes #529.
|
|
|
a5bc4605
|
2023-04-29T20:43:54
|
|
cmake: Link with --undefined-version
|
|
|
a8fe4900
|
2023-04-29T20:17:38
|
|
autoconf: Only use --undefined-version with version scripts
|
|
|
85d4aba4
|
2023-04-29T19:48:50
|
|
libxml.m4: Fix quoting
Should fix #527.
|
|
|
d8f6e760
|
2023-04-29T19:12:23
|
|
Link with --undefined-version
lld 16 defaults to --no-undefined-version but the version script can
contain symbols disabled by configuration options.
See #473.
|
|
|
131c2dec
|
2023-04-29T18:57:50
|
|
libxml2.syms: Revert removal of version information
Removing version information caused problems when relinking with shared
libraries depending on libxml2. It also broke the ABI on Android.
Revert libxml2.syms to the 2.10.0 version.
Fixes #526.
|
|
|
8979d739
|
2023-04-28T20:22:05
|
|
Clean up .gitignore
Remove stuff which probably shouldn't be ignored.
|
|
|
e07f765f
|
2023-04-28T14:51:06
|
|
Bump version to 2.12.0
|
|
|
f296934a
|
2023-04-26T22:11:09
|
|
Release v2.11.0
|
|
|
6e12409b
|
2023-04-26T21:21:31
|
|
Regenerate docs and testapi.c
|
|
|
2df2562e
|
2023-04-26T21:17:21
|
|
autoconf: Add AC_LANG_SOURCE to fix warning
|
|
|
107e00a5
|
2023-04-26T20:36:33
|
|
libxml2.syms: Remove LIBXML2_2.11.0 version node
This was never part of a public release and can be removed after commit
bbb2b8f1.
|
|
|
3ffcc03b
|
2023-03-13T19:38:41
|
|
parser: Deprecate more internal functions
|
|
|
b51b99ef
|
2023-04-21T02:56:10
|
|
testchar: Fix return value in testUserEncoding
|
|
|
db32b09d
|
2023-04-20T16:22:11
|
|
xmllint: Validate --maxmem integer option
Fixes #520.
|
|
|
250faf3c
|
2023-04-20T12:35:21
|
|
parser: Fix regression in xmlParserNodeInfo accounting
Commit 62150ed2 broke begin_pos and begin_line when extra node info was
recorded.
Fixes #523.
|
|
|
eca1116b
|
2023-04-20T13:52:16
|
|
testchar: Add test for memory pull parser with encoding
|
|
|
9282b084
|
2023-04-19T21:55:24
|
|
parser: Fix regression in memory pull parser with encoding
Revert another change from commit 98840d40.
Decode the whole buffer when reading from memory and switching to the
initial encoding. Add some comments about potential improvements.
|
|
|
1db3567c
|
2023-04-19T12:19:36
|
|
autoconf: Warn about outdated C compilers
|
|
|
33fb297b
|
2023-04-15T16:53:00
|
|
encoding: Fix compiler warning in ICU build
|
|
|
86105c04
|
2023-04-15T18:04:03
|
|
Fix use-after-free in xmlParseContentInternal()
* parser.c:
(xmlParseCharData):
- Check if the parser has stopped before advancing
`ctxt->input->cur`. This only occurs if a custom SAX error
handler calls xmlStopParser() on fatal errors.
Fixes #518.
|
|
|
a19fa11e
|
2023-04-13T15:11:47
|
|
parser: Fix regression when switching input encodings
Revert some changes from commit 98840d40.
WebKit/Chromium can actually switch from ISO-8859-1 to UTF-16 in the
middle of parsing. This is a bad idea, but we have to keep supporting
this use case.
|
|
|
b4d46cee
|
2023-04-12T15:10:01
|
|
parser: Remove first line handling in xmlParseChunk
After reworking EBCDIC detection, this isn't necessary.
|
|
|
921796b0
|
2023-04-12T13:43:28
|
|
parser: Don't grow push parser buffers
This should fix a short-lived regression when push parsing with
encodings.
|
|
|
547edbf1
|
2023-04-07T11:49:27
|
|
[CVE-2023-29469] Hashing of empty dict strings isn't deterministic
When hashing empty strings which aren't null-terminated,
xmlDictComputeFastKey could produce inconsistent results. This could
lead to various logic or memory errors, including double frees.
For consistency the seed is also taken into account, but this shouldn't
have an impact on security.
Found by OSS-Fuzz.
Fixes #510.
|