This broke in CI but not on my local machine because of the different
compiler version. This is because there was an issue in GCC [1] that
caused the attribute to not properly work on multidimensional arrays.
This has since been fixed in GCC 15.
Therefore, we guard the attribute with a version check.
[1] https://gcc.gnu.org/cgit/gcc/commit/?id=afb46540d3921e96c4cd7ba8fa2c8b0901759455
Introducing a new attribute macro in lower branches is an ABI break and
not allowed. However, we still need to fix the warnings such that
-Werror builds don't break. So we copy the macro from the master branch
to the C files in the appropriate places.
Starting with gcc 15 the warning `-Wunterminated-string-initialization`
is enabled by default. We make now use of the `nonstring` attribute to
silence the warning for the cases where this is intended.
Closes GH-18603.
These were introduced in 6747068c, but they don't seem to be in upstream
(anymore). For the entry in rpm it may have even been a mixup with the
two sections in the rpm file: there's a "10 string" entry but only in
the delta part.
Make data_file.c's generator output its array initialization
"by list of strings", instead of "by list of chars": that makes the compiler
happier.
Use strtr() with a precomputed map, instead of a loop over ord(),
to generate in 1/100th the time.
Avoids ambiguous 0x tokens (as specified in C standards), by using octal.
Closes GH-10422.
* Update libmagic to 5.45
This also cleans up magicdata.patch: changes that are already in upstream file
were removed from that patch file.
There are five (expected) test output changes.
All these were also checked with the file command.
- bug77961.phpt changes because there's now an early error-return in the
`if (ts == FILE_BADSIZE) {` branch.
- cve-2014-1943.phpt and cve-2014-1943-mb.phpt change because now the crafted
data is recognised as a simh file.
- bug71434.phpt now properly recognises it as a Python file.
- ext/fileinfo/tests/finfo_file_basic.phpt more specific mime type.
* Adjust memory requirement for s390x fileinfo run
The larger database causes a higher memory usage.
Similar to 962c082a5b.
* [ci skip] NEWS
The change reduces the input size on patterns using posix character
classes. It is still better than reverting to the patterns from the
older version, as the upstream data uses posix classes also in the later
versions. The input reduction speeds up the pattern matching in some
cases.
This patch is functionally almost same as upstream, but might show some diff
when the input is very long. While the magic data in the newer versions is
functionally an improvement, with jit=0 it might show a performance regression.
The slowdown is negligible in the normal usage and is still acceptable for the
malicious input. If some functional regressions show up, this patch should be
reverted and the tests timing should be adapted instead.
yet unfinished port to libmagic 5.28
catch with missing libmagic port pieces
regenerate data file with magic from 5.28
test magic files from 5.28
missing files
fix path
pure c99 is still not supported
move right to 5.29, yet some bugs present
more sync with orig lib
more ZMM usage
use unpatched data for now
partial revert according to bug #67705
Revert "more ZMM usage"
This reverts commit 5e3c9b851034b455942b219b18bdc49c8315c3ee.
several fixes, so it's now closer to the clean port
