This fixes incorrect type conversion and subsequent check for Windows
where returned socket is not an int.
It should be noted that this is not really an issue as previous int
would get negative so the check should still work. The issue actually
happens only in master (PHP 8.5) where refactoring has been done and the
type changed.
Closes GH-19881
On Linux, these two character devices are exceptions in that they can be
seeked. Check their major/minor device number.
Co-authored-by: divinity76 <hans@loltek.net>
Change the reproducer code in `bug35916.phpt` from `stream_bucket_append` to
`stream_bucket_prepend` and you have the same bug.
Furthermore, even in the append case the check is incorrect because the bucket
can already be in the brigade at a position other than the tail.
To solve this properly, unlink the brigade first and also use that as a
condition to manage the refcount.
Closes GH-18973.
Or on Windows it is going to use either FormatMessageW or strerror_s
for compatibility with previous error messages.
It also needs to accomodate for GNU and BSD versions of strerror_r
returning different type.
Closes GH-19251
* uri: Rename `uri_recomposition_mode_t` to `php_uri_recomposition_mode`
* uri: Align the names of the `php_uri_recomposition_mode` values
* uri: Rename `uri_component_read_mode_t` to `php_uri_component_read_mode`
* uri: Align the names of the `php_uri_component_read_mode` values
* uri: Rename `uri_property_name_t` to `php_uri_property_name`
* uri: Align the names of the `php_uri_property_name` values
* uri: Rename `uri_property_handler_t` to `php_uri_property_handler`
* uri: Rename `uri_(read|write)_t` to `php_uri_property_handler_(read|write)`
* uri: Rename `php_uri_property_handler`’s `(read|write)_func` to `read|write`
The `_func` is implied by the data type and the name of the struct.
* uri: Rename `uri_parser_t` to `php_uri_parser`
* uri: Shorten the names of `php_uri_parser` fields
The `_uri` suffix is implied, because this is an URI parser.
PSFS_FEED_ME is supposed to be returned when the filter did not receive
enough data and did not generate buckets for the output brigade.
The test generates buckets anyway on the output brigade, and the stream
layer did not handle that case causing a memory leak.
To solve this, discard any such buckets as it would conflict with the
status code returned by the filter. This keeps BC and solves the leak.
Closes GH-18972.
The string is converted twice for some reason.
This is pointless, and furthermore, this is observable in userspace code
when dealing with Stringable objects.
Closes GH-19713.
This one is not initialized. This is not hittable from userspace code
because all locations within first-party php-src code have a valid
`option` argument.
Closes GH-19714.
Use the PHP_STREAM_FLAG_NO_FCLOSE flag to prevent closing a stream while
a handler is running. We already do this in some other places as well.
Only handlers that do something with the stream afterwards need changes.
Closes GH-18797.
This reverts commit ca4a841921.
We like the error message change, but not the downgrade to notice
at this time in the release cycle.
@bukka will come back around
* Remove include "sanity check" to get a better error message
Instead of rejecting directories / non-regular files early with
a generic error, we should just accept them and error later when a
read is attempted. This is more general and will generate a better
error message on Linux. On Windows, the error remains the same as
before.
* Update error message to include include_path
fix format for include path
---------
Co-authored-by: Joe Watkins <krakjoe@php.net>
Jakub Zelenka