Apache only generates REDIRECT_STATUS, so explicitly check for that
if the server name is Apache, don't allow other variable names.
Furthermore, redirect.so and Netscape no longer exist, so
remove those entries as we can't check their server name anymore.
We now also check for the configuration override *first* such that it
always take precedence. This would allow for a mitigation path if
something like this happens in the future.
It's no use trying to work around whatever the operating system and Apache
do because we'll be fighting that until eternity.
Change the skip_getopt condition such that when we're running in
CGI or FastCGI mode we always skip the argument parsing.
This is a BC break, but this seems to be the only way to get rid of this
class of issues.
multipart/form-data boundaries larger than the read buffer result in erroneous
parsing, which violates data integrity.
Limit boundary size, as allowed by RFC 1521:
Encapsulation boundaries [...] must be no longer than 70 characters, not
counting the two leading hyphens.
We correctly parse payloads with boundaries of length up to
FILLUNIT-strlen("\r\n--") bytes, so allow this for BC.
We should not early-out with success status if we found an ipv6
hostname, we should keep checking the rest of the conditions.
Because integrating the if-check of the ipv6 hostname in the
"Validate domain" if-check made the code hard to read, I extracted the
condition out to a separate function. This also required to make
a few pointers const in order to have some clean code.
The original code is error-prone due to the "best fit mapping" that
happens with the argument parsing but not with the query string.
When we get a non-ASCII character, try to remap it and see if it becomes
a hyphen.
An alternative approach is to create a custom main `wmain` receiving
wide-character variations that does the ANSI transformation with the
best-fit mapping, but that's more error-prone and could cause unexpected
breakage.
Another alternative was just don't doing this check altogether and
always check for `cgi || fastcgi` instead, but that breaks real-world
use-cases.
The old code checked for suffixes but didn't take into account trailing
whitespace. Furthermore, there is peculiar behaviour with trailing dots
too. This all happens because of the special path-handling code inside
CreateProcessW.
By studying Wine's code, we can see that CreateProcessInternalW calls
get_file_name [1] in our case because we haven't provided an application
name. That code gets the first whitespace-delimited string into app_name
excluding the quotes. It's then passed to create_process_params [2]
where there is the path handling code that transforms the command line
argument to an image path [3]. Inside Wine, the extension check if
performed after these transformations [4]. By doing the same thing in
PHP we match the behaviour and can properly match the extension even in
the given edge cases.
[1] 166895ae3a/dlls/kernelbase/process.c (L542-L543)
[2] 166895ae3a/dlls/kernelbase/process.c (L565)
[3] 166895ae3a/dlls/kernelbase/process.c (L150-L151)
[4] 166895ae3a/dlls/kernelbase/process.c (L647-L654)
* fix segfault in `ZEND_BIND_STATIC`
In case a `ZEND_BIND_STATIC` is being executed, while the current chunk is full,
the `zend_array_dup()` call will trigger a OOM in ZendMM which will crash, as
the opline might be a dangling pointer.
* add missing test
* `assert()`ing seems easier than trying to make the compiler to not optimize
* moved from function call to INI setting, so we can use this in other places as well
* make `assert()` work no NDEBUG builds
* document magic number
* fix segfault in `ZEND_FUNC_GET_ARGS`
In case a `ZEND_FUNC_GET_ARGS` is being executed, while the current chunk is
full, the `zend_new_array()` call will trigger a OOM in ZendMM which will crash,
as the opline might be a dangling pointer.
---------
Co-authored-by: Florian Engelhardt <florian@engelhardt.tc>
php_strip_url_passwd modifies url in-place. We cannot assume from
php_message_handler_for_zend that data is a temporary, modifiable string.
Fixes oss-fuzz #64209
Closes GH-12733
This pulls only the changes made to CertificateGenerator.inc in the PHP-8.2
branch from commit 505e8d2a04.
Co-authored-by: Jakub Zelenka <bukka@php.net>
The namespace data is freed and set to NULL, but there remain references
to the namespace declaration nodes. This (rightfully) confuses libxml2
because its invariants are broken. We also have to remove all remaining
references from the subtree. This fixes the data corruption bug.
Closes GH-12681.
005_variation2.phpt creates files with special names, and
filesize_variation5.phpt checks for filesize of inexistent files with special
names. Create the files in a separate directory to avoid these tests clashing.
Closes GH-12692
Have to use file_put_contents() instead of --FILE-- because we have to
actually load it using the exec command, *and* have to make multiple
files, and note that we can only load files relative from the current
directory, so we can't rely on files being in the sapi/phpdbg/tests
folder.
Closes GH-12680.
