On NixOS we run `make` & `make check` inside `ext/soap` which broke the test
like this:
001+ Fatal error: Uncaught SoapFault exception: [WSDL] SOAP-ERROR: Parsing WSDL: Couldn't load from 'ext/soap/tests/gh15711.wsdl' : failed to load "ext/soap/tests/gh15711.wsdl": No such file or directory
002+ in /build/php-8.3.13/ext/soap/tests/gh15711.php:29
003+ Stack trace:
004+ #0 /build/php-8.3.13/ext/soap/tests/gh15711.php(29): SoapClient->__construct('ext/soap/tests/...', Array)
005+ #1 {main}
006+ thrown in /build/php-8.3.13/ext/soap/tests/gh15711.php on line 29
Fix is to make the path dependant on `__DIR__` as it's the case in other
testcases including WSDLs.
Closes GH-16733.
If get_iterator() fails, we should not destroy the object.
Also changes the check to a NULL check to be more defensive, and to
match the VM.
Closes GH-16441.
Bisect points to 94ee4f9, however this only reveals the problem.
Cloning an object on a lower branch and trying to call its methods
crashes as well. Cloning the object shouldn't be possible in the first
place because there's an engine constraint that when we have a new
object handler we should also have a clone handler. This constraint is
not fulfilled here.
Closes GH-16245.
These failures are caused by the fix for GHSA-p99j-rfp4-xqvq. Since
the two bug*.phpt tests don't need the "wsdl" query string, and don't
even need php-cgi, we just remove the `--GET--` section. The two
server*.phpt tests are harder to fix, since during evaluation of the
`--SKIPIF--` section, the soap extension can be loaded, but it may not
during evaluation of the `--FILE--` section. So for now, we skip these
tests on Windows altogether.
Co-authored-by: Niels Dossche <7771979+nielsdos@users.noreply.github.com>
Closes GH-16084.
Apparently example.org now rejects POST requests, so we would need to
adjust the test expectation ("Method not allowed"). However, there is
no need for an online test; instead we're just using the CLI test
server. The serialization is a bit fiddly, but as long as there are
no quotes in `PHP_CLI_SERVER_ADDRESS` we're fine.
Closes GH-16063.
The one error message indeed had a wrong namespace, and in general they
weren't very descriptive, this also makes them more descriptive.
Furthermore, two additional bugs were fixed:
- Persistent memory leak of `location`.
- UAF issues when printing the error message.
Closes GH-15830.
This code is modelled after how `http_fopen_wrapper.c` does things,
which apparently is just looping over the array and handling each string
the same way as if we passed a header string directly.
Also fixes a potential crash in `php_sdl.c` but without adding support
for header arrays there (yet) because the code is untested.
Closes GH-15817.
libxml2 2.13 has different formatting behaviour: it outputs `<faultcode/>`
instead of `<faultcode></faultcode>`, and similarly for `env:Value`.
Normalize the output.
Closes GH-15801.
Two issues:
1) We should not modify the object when we pass invalid values
2) We should reset the properties to their default value otherwise we
get a UAF.
Regressed in df219ccf9d
Closes GH-15248.
RFC 2617 and 7616 describe that for the "Authorization" header we should
not put the qop nor nc value inside quotes. This differs from the
WWW-Authenticate header, which may have been the source of the confusion
in the implementation. While the version with quotes seems to work fine
in some cases, clearly not all servers accept the non-standard form.
To fix the issue, simply removing the quotes of those two header fields
of the client request to be in line with the RFC suffices.
I refer further to example 3.5 in RFC 2617 and example 3.9.1 in
RFC 7616.
RFC 2617: https://datatracker.ietf.org/doc/html/rfc2617
RFC 7616: https://datatracker.ietf.org/doc/html/rfc7616
Closes GH-14328.
There's a hash table that maps type names to class name, but names with
a leading backslash are not supported. The engine has logic to strip
away the leading backslash that we should replicate here.
It works by checking if we need to make an actual copy in case an
unexpected (e.g. invalid data or leading backslash) situations are
detected. Upon making a copy we normalize the data in the table.
Furthermore, previously the code assumed that the key was always valid
and that the structure was a non-packed hash table. This isn't
necessarily the case. The new code fixes this as well.
Closes GH-14398.
The naming of the userland functions is terrible and confused me.
gzdecode() is actually the function to decompress a gzip stream, and
gzuncompress() is the one to decompress a deflate stream...
See zlib.c to see the internal function -> type mapping.
The incorrect functions are being called to deal with incoming
compressed data.
gzip/x-gzip corresponds to gzuncompress(), while deflate corresponds to
gzinflate().
The existing code for gzip compression also plays with removing the
first 10 bytes (i.e. the gzip header) to pass it to the inflate
implementation but that doesn't always work properly due to trailer
data. Get rid of that entirely by using the correct functions.
Closes GH-14321.
Setting the stream context via php_stream_context_to_zval() will
increase the reference count. So if the new context is created, then it
will end up with a reference count of 2 while it should be 1.
Credits to cmb for the analysis. I arrived at the same patch as he did.
Closes GH-12523.
There are two issues:
- UAF because the hashmap resized while being iterated over, yet the local
variables used internally in the macros are not updated.
- The hashmap being iterated over is modified: entries are deleted after
other entries have been added. This causes the deletion to fail sometimes
because indices of buckets have shifted.
Fix it by using a while loop iteration and HashPosition position tracker
instead.
Issue exists on PHP 8.1 too, but is much harder to trigger.
The test file reproduces the issue reliably on PHP 8.2 and up.
Closes GH-12409.
If you build soap as a shared object, then these tests fail on
non-Windows, or when the PHP install hasn't been make install-ed yet,
but is executed from the development directory.
Closes GH-11211.
