* PHP-5.5:
fix#65070, bgcolor does not use the same format as the input image with imagerotate
fix#65070, bgcolor does not use the same format as the input image with imagerotate
Fix the spelling of the php_cli_server_http_response_status_code_pair typedef.
Change the search in get_status_string() to correctly handle unknown codes.
Update git rules (5.5 is stable, 5.3 sec only)
skip test for bug #64936 when tokenizer not built
Bump version
Update credits
Merge NEWS from PHP 5.4 branch
Reorder NEWS for PHP 5.5.0 final
Conflicts:
configure.in
ext/gd/libgd/gd_interpolation.c
main/php_version.h
* PHP-5.5:
Fix the spelling of the php_cli_server_http_response_status_code_pair typedef.
Change the search in get_status_string() to correctly handle unknown codes.
* PHP-5.4:
Fix the spelling of the php_cli_server_http_response_status_code_pair typedef.
Change the search in get_status_string() to correctly handle unknown codes.
This previously used a buggy implementation of binary search that would loop
infinitely for unknown codes when searching in reason arrays of particular
sizes (such as the one we have at the moment). Since C provides bsearch(),
we'll just use that instead, since libc authors hopefully get this right.
There was also an additional bug that was masked by the first one: the design
was that an unknown code would result in get_status_string() returning NULL,
which would then result in a segfault in append_http_status_line(), since it
assumed that it would always receive a valid string pointer that could be
handed off to smart_str_appends_ex(). We'll now return a placeholder in that
case.
Fixes bug #65066 (Cli server not responsive when responding with 422 http
status code).
nNumOfElements was incremented after the pDestructor code, so any
code in the dtor would get a wrong number of elements.
Right now the bucket deletion code is replicated in four places,
it should probably be moved off into one function (or rather,
zend_hash_apply_deleter should be used everywhere). The codes are
subtly different though in that the HANDLE_UNBLOCK_INTERRUPTIONS()
happens in different places. In particular it seems odd that in
some cases interruptions stay blocked during the destructor call.
