php/archived-php-src
1
0
Fork 0
mirror of https://github.com/php/php-src.git synced 2026-04-23 16:08:35 +02:00
Code Issues Packages Projects Releases Wiki Activity
124,652 Commits 549 Branches 1,485 Tags
22d700b66a29c149afa4bc6969937089fecaefdc
Commit Graph

124652 Commits

This Branch
This Branch
All Branches
Author SHA1 Message Date
Christoph M. Becker 22d700b66a Merge branch 'PHP-8.0' 
* PHP-8.0:
  Fix test in case fallocate(1) is available, but fails
 2021-06-17 19:08:58 +02:00
Christoph M. Becker 3374adcd14 Merge branch 'PHP-7.4' into PHP-8.0 
* PHP-7.4:
  Fix test in case fallocate(1) is available, but fails
 2021-06-17 19:06:18 +02:00
Christoph M. Becker 99e7c5cea8 Fix test in case fallocate(1) is available, but fails 
That happens on Travis s390x for whatever reasons.  Thus, instead of
checking for `fallocate -h`, we attempt the real allocation and skip if
that fails.
 2021-06-17 19:05:07 +02:00
Dmitry Stogov d72c320ff6 JIT/ARM64: Fixed possible incorrect register allocation 2021-06-17 18:43:15 +03:00
Dmitry Stogov c02d7c4de8 Merge branch 'PHP-8.0' 
* PHP-8.0:
  JIT/x86: Fixed possible incorrect register allocation
 2021-06-17 18:42:57 +03:00
Dmitry Stogov ced8e88438 JIT/x86: Fixed possible incorrect register allocation 2021-06-17 18:42:11 +03:00
Joe Watkins 6318040df2 remove specialized printing from phpdbg (#7156) 2021-06-17 14:22:33 +02:00
Dmitry Stogov 82bcc9b497 JIT/ARM64: Fixed "may be used uninitialized" compilation warning 2021-06-17 15:08:14 +03:00
Dmitry Stogov 6442cbfbc4 Merge branch 'PHP-8.0' 
* PHP-8.0:
  JIT: Fixed failures of tracing JIT with CALL VM
 2021-06-17 14:26:36 +03:00
Dmitry Stogov 1082669e24 JIT: Fixed failures of tracing JIT with CALL VM 
This fixes:
- tests/lang/bug28800.phpt
- Zend/tests/settype_resource.phpt
- Zend/tests/type_declarations/scalar_return_basic_64bit.phpt
 2021-06-17 14:21:40 +03:00
Christoph M. Becker 83517340f6 Merge branch 'PHP-8.0' 
* PHP-8.0:
  Fix #81145: copy() and stream_copy_to_stream() fail for +4GB files
 2021-06-17 13:18:59 +02:00
Christoph M. Becker 8daed6d0f2 Merge branch 'PHP-7.4' into PHP-8.0 
* PHP-7.4:
  Fix #81145: copy() and stream_copy_to_stream() fail for +4GB files
 2021-06-17 13:16:30 +02:00
Christoph M. Becker 2555efadbc Fix #81145: copy() and stream_copy_to_stream() fail for +4GB files 
When mapping the file, we need to pass the proper `dwFileOffsetHigh`
instead of `0`.

Co-authored-by: Nikita Popov <nikita.ppv@gmail.com>

Closes GH-7158.
 2021-06-17 13:13:47 +02:00
Alex Dowad b1ab76f742 Minor formatting tweaks in mbfilter_euc_kr.c 2021-06-17 13:12:40 +02:00
Alex Dowad 958ef47d2b When flushing CP5022x conversion filter, also flush next filter in chain 
All the mbstring encoding conversion filters do this. I missed it when
adding a flush function for CP5022x.
 2021-06-17 13:12:40 +02:00
Alex Dowad caeaa662ab Strict conversion of UHC text to Unicode 
Previously, mbstring would accept a lot of things which were not valid
UHC text. No more.

- Don't allow single-byte control characters to appear where the 2nd
  byte of a multi-byte character should be.
- Validate that the 2nd byte of a multi-byte character is in the
  expected range.
- Treat it as an error if a multi-byte character is truncated.

Also add a test suite to confirm that UHC conversion (both to and from
Unicode) works according to spec.
 2021-06-17 13:12:40 +02:00
Alex Dowad 4550036d96 Minor formatting tweaks in mbfilter_uhc.c 2021-06-17 13:12:40 +02:00
Alex Dowad 9868c17368 Mark CP932 and CP51932 encoding tests as 'slow tests' 2021-06-17 13:12:40 +02:00
Alex Dowad e2459857af Remove duplicate implementation of CP932 from mbstring 
Sigh. Double sigh. After fruitlessly searching the Internet for information on
this mysterious text encoding called "SJIS-open", I wrote a script to try
converting every Unicode codepoint from 0-0xFFFF and compare the results from
different variants of Shift-JIS, to see which one "SJIS-open" would be most
similar to.

The result? It's just CP932. There is no difference at all. So why do we have
two implementations of CP932 in mbstring?

In case somebody, somewhere is using "SJIS-open" (or its aliases "SJIS-win" or
"SJIS-ms"), add these as aliases to CP932 so existing code will continue to
work.
 2021-06-17 13:12:40 +02:00
Alex Dowad 7502c86342 Add test suite for UTF-{7,8,16,32} 
Also fix a couple small problems with UTF-32 and UTF-8 support:

- UTF-32 would pass very large codepoints (>= 0x80000000), which are
  not valid.
- UTF-8 would sometimes emit two error marker characters for a single
  bad input byte.
 2021-06-17 13:12:40 +02:00
Dmitry Stogov cc0aac3f73 Merge branch 'PHP-8.0' 
* PHP-8.0:
  Fixed incorrect type inference for "(array)$null".
 2021-06-17 13:05:45 +03:00
Dmitry Stogov df16fd149b Fixed incorrect type inference for "(array)$null". 2021-06-17 13:03:09 +03:00
Nikita Popov deb7955bf4 Fix #81150 Add UPGRADING note for float to int deprecation 
[ci skip]
 2021-06-17 09:52:44 +02:00
Alex Dowad 5d37cd97e4 Improve documentation for contributors 2021-06-17 06:29:38 +02:00
Nikita Popov 0d6358f2cf Drop support for printf p modifier 
To be conservative, cause a fatal error if the p modifier is
encountered, in case this is still used by some extension.
 2021-06-16 17:59:00 +02:00
Nikita Popov 7eec57c263 Test ARM64 JIT on Travis (#7157) 
Run tests with tracing JIT, function JIT and tracing JIT under --repeat 2.

As the overall build with three test runs only takes ~15m, I'm enabling this unconditionally, rather than just for nightly builds.
 2021-06-16 17:13:46 +02:00
Nikita Popov c492c90a6e Fix mysqlnd debug formats on 32-bit 2021-06-16 17:08:20 +02:00
Nikita Popov 57f06b6058 Fix more mysqlnd debug printf formats 
These didn't show up on my system...
 2021-06-16 16:23:30 +02:00
Nikita Popov 3acdab864d Fix printf formats in mysql debug logging 
Enable printf format verification and fix the reported errors.
 2021-06-16 16:09:06 +02:00
Dmitry Stogov 3a78259525 JIT: Avoid too aggressive loop unrolling 
This fixes tests/func/010.phpt failure with tracing JIT on ARM64.
 2021-06-16 16:44:27 +03:00
Nikita Popov 591dcdbdb0 Use different error condition in ACCEL_LOG_FATAL test 
This test fails on s390x with opcache. Presumably the large allocation
works fine there as long as it's not used.

Switch to a different error condition that produces a more reliable
failure.
 2021-06-16 13:08:54 +02:00
Christoph M. Becker 46a6d481d3 Merge branch 'PHP-8.0' 
* PHP-8.0:
  Fix #72809: Locale::lookup() wrong result with canonicalize option
 2021-06-16 10:38:14 +02:00
Christoph M. Becker 13f3999a77 Merge branch 'PHP-7.4' into PHP-8.0 
* PHP-7.4:
  Fix #72809: Locale::lookup() wrong result with canonicalize option
 2021-06-16 10:36:51 +02:00
Christoph M. Becker 0f1b17e378 Fix #72809: Locale::lookup() wrong result with canonicalize option 
Canonicalization converts the locale to ICU format[1].  However, the
lookup described in RFC 4647, section 3.4, is about POSIX format.  To
make that lookup work for ICU format, we also need to cater to keyword
separators.

The results are somewhat unexpected, but apparently canonical lookup is
explicitly supposed to return canonical language tags[2].

[1] <https://unicode-org.github.io/icu/userguide/locale/#canonicalization>
[2] <https://github.com/php/php-src/blob/php-7.4.20/ext/intl/locale/locale_methods.c#L1504>

Closes GH-7151.
 2021-06-16 10:33:59 +02:00
Dmitry Stogov 8ae06582c6 Fixed incorrect map_ptr slots counting 2021-06-16 11:33:43 +03:00
Derick Rethans e63ed9d0da Merge branch 'PHP-8.0' 2021-06-15 19:53:03 +01:00
Derick Rethans 77bfe96109 Merge branch 'PHP-7.4' into PHP-8.0 2021-06-15 19:52:14 +01:00
Derick Rethans 8370a084e4 Prepare for 7.4.22 2021-06-15 19:52:02 +01:00
Aaron Piotrowski 5d4f05527b Drop fiber block hooks 
Removes the ability for an extension to allow Fiber switching in blocked contexts.

See discussion on fdc22744a8.
 2021-06-15 08:58:07 -05:00
Nikita Popov b64213872a Fix missing error for single index unset on wrong type 
The error was correctly thrown for a nested index, but not for a
single index. Make sure both have the same behavior.
 2021-06-15 14:41:11 +02:00
Nikita Popov 34257e1cfc Remove unnecessary IS_UNUSED check 
This operand cannot be UNUSED.
 2021-06-15 14:27:41 +02:00
Nikita Popov 0eafc97d29 Fix prototype 2021-06-15 10:08:50 +02:00
Joe Watkins ba3c8027bc Adds zend_fiber_startup at the right stage, moves setting of switch hooks 
to correct stage

In a threaded environment, these ought to be set before any threads are
created, and they ought to be set once.
 2021-06-15 09:56:15 +02:00
Joe Watkins f46063495d reset blocking state on init 2021-06-15 09:41:36 +02:00
Gabriel Caruso b81c3b5b1c Prepare for PHP 8.0.9 2021-06-15 09:19:13 +02:00
Christoph M. Becker 139a73b5a4 Improve performance of AppVeyor test runs 
We only load a minimum set of extensions, and rely on dynamic loading
of others due to `--EXTENSION--` triggers.  We do not run the imap,
ldap and snmp test suites, because most of the tests would be skipped
after timeouts anyway.

Closes GH-7150.
 2021-06-14 23:36:17 +02:00
Aaron Piotrowski 859524c246 Remove fiber context embedding 2021-06-14 14:34:36 -05:00
Aaron Piotrowski fdc22744a8 Add API to prevent Fiber switch in select contexts 
Co-authored-by: Martin Schröder <m.schroeder2007@gmail.com>
 2021-06-14 14:19:00 -05:00
Nikita Popov 9871a624bf Clean up some more function_exists() checks 2021-06-14 15:47:08 +02:00
Nikita Popov 892421c9a8 Make some zend_test dependencies explicit 
Rather than checking that specific functions exist.
 2021-06-14 15:40:34 +02:00