Fix #75273: php_zlib_inflate_filter() may not update bytes_consumed
Whenever we return with `PSFS_PASS_ON`, we need to update
`bytes_consumed` to not mislead the caller. Instead of fixing the
respective `if` clauses, we eschew the early bail-outs to simplify the
code a bit.
Fix #75696: posix_getgrnam fails to print details of group
According to the POSIX specification of `getgrnam_r()` the result of
`sysconf(_SC_GETGR_R_SIZE_MAX)` is an initial value suggested for the
size of the buffer, and `ERANGE` signals that insufficient storage was
supplied. So if we get `ERANGE`, we try again with a buffer twice as
big, and so on, instead of failing.
Anatol Belski [Mon, 6 Aug 2018 20:35:11 +0000 (22:35 +0200)]
Fix stack underflow in phar
The checks can issue reads below and above the temporary buffer. A read
itself doesn't seem dangerous, but the condition result can be
arbitrary. Such reads have to be avoided. Likely this patch should be
backported.
Fix #68180: iconv_mime_decode can return extra characters in a header
Basically, the algorithm to append a converted string to an existing
`smart_str` works by increasing the `smart_str` buffer, to let `iconv`
convert characters until there is no more space, to set the new length
of the `smart_str` and to repeat until there is no more input.
Formerly, the new length calculation has been wrong, though, since we
would have to take the old `out_len` into account (`buf_growth -
old_out_len - out_len`). However, since there is no need to take the
old `out_len` into account when increasing the `smart_str` buffer, we
can simplify the fix, avoiding an additional variable.
Fix #60494: iconv_mime_decode does ignore special characters
We must not ignore erroneous characters in mime headers, but rather let
iconv_mime_decode() fail in this case, issuing the usual notice
regarding illegal characters.
Fix #63839: iconv_mime_decode_headers function is skipping headers
We have to cater to the possibility that `=?` is not the start of an
encoded-word, but rather a literal `=?`. If a line break is found
while we're still looking for the charset, we can safely assume that
it's a literal `=?`, and act accordingly.
Fix #55146: iconv_mime_decode_headers() skips some headers
If we're expecting the start of an encoded word (`=?`), but instead of
the question mark get a line break (CR or LF), we must not append it to
the `pretval`.
There's no need to actually try to trigger an out-of-memory condition
to proof the leak; instead we can simply rely on the Zend MM to report
the memory leaks in debug mode (at least on Linux). Therefore we
simplify the regression test, which also makes it run much faster.
Kevin Abel [Mon, 9 Jul 2018 01:37:21 +0000 (20:37 -0500)]
Fix bug 76595: Update phpdbg man page
Send phpdbg.1 man page through configure replacements
Update phpdbg.1 man page to include all options
Fixes formatting to be more consistent with php.1
Fix paragraph whitespace and ignore phpdbg.1
Fix #52974: jewish.c: compile error under Windows with GBK charset
jewish.c includes ISO-8859-8 encoded Hebrew Hebrew month names, which
may cause compile errors, and is generally confusing. We replace the
literal month names with appropriate escape sequences.
* PHP-7.0:
Fixed bug #76459 windows linkinfo lacks openbasedir check
Add NEWS
Fixed bug #76459 windows linkinfo lacks openbasedir check
Fix bug #76557: heap-buffer-overflow (READ of size 48) while reading exif data
Fix bug #76423 - Int Overflow lead to Heap OverFlow in exif_thumbnail_extract of exif.c
* PHP-5.6:
Add NEWS
Fixed bug #76459 windows linkinfo lacks openbasedir check
Fix bug #76557: heap-buffer-overflow (READ of size 48) while reading exif data
Fix bug #76423 - Int Overflow lead to Heap OverFlow in exif_thumbnail_extract of exif.c
Fixed RecursiveDirectoryIterator with long path or with edge case length
The search path needs to be appended with the wild card. Till now, an
edge case existed, so then if a path is 259 bytes long, which is smaller
_MAX_PATH, the suffix would cause the final search path to become longer
than _MAX_PATH. It is an edge case, when the starting path happens to
have a specific length. If the starting path was longer than _MAX_PATH
or the addition of "\\*" would not exceed _MAX_PATH, the function was
correct. Except for rewind, which was broken in the case of the long
path.
Fix C++11 and up compatibility for zend_finite and more
C++11 puts isfinite, isinf, isnan and a lot of other stuff into the
std namespace. Thus, if a C++11 or newer source is compiled, these
symbols won't be available. A good solution would be to include cmath,
but depending on a particular compiler that might remove even more
stuff from the global namespace, so such a fix should only target master.
For now, just keep these defines same for C++11 and upper, as the actual
C++ code should use symbols from the std namespace anyway. This
especially concerns older GCC versions like at least 4 and 5, which are
used by default in the LTS Linux distros.
The log header can be saved in the globals on startup. At the same
time, the log header can be changed per request. In case that
happened, wrong pointer will be free'd on shutdown. It can happen at
any point when zend_error() or similar is called at startup, like for
example in the case of the ini deprecation warnings. Thus, ZMM cannot
be used here.
Validate that "C" serialization payload is followed by "}" prior to
calling the unserialize() handler. This mitigates issues caused by
unserialize() not correctly handling strings that are not NUL
terminated. Making sure that there is a "}" at the end avoids the
problem.
projects / php / log