Hi list !
I am still trying to solve bug 79714 (See:
https://bugs.php.net/bug.php?id=79714)
I am no expert at C and/or the PHP Core at all so I need a little advice
from you guys.
Seeing lines 3238-3314 in /ext/phar/phar.c (from php-src repo):
static zend_op_array *phar_compile_file(zend_file_handle *file_handle,
int type) /* {{{ */
{
zend_op_array *res;
char *name = NULL;
int failed;
phar_archive_data *phar;
if (!file_handle || !file_handle->filename) {
return phar_orig_compile_file(file_handle, type);
}
if (strstr(file_handle->filename, ".phar") &&
!strstr(file_handle->filename, "://")) {
if (SUCCESS ==
phar_open_from_filename((char*)file_handle->filename,
strlen(file_handle->filename), NULL, 0, 0, &phar, NULL)) {
if (phar->is_zip || phar->is_tar) {
zend_file_handle f = *file_handle;
/* zip or tar-based phar */
spprintf(&name, 4096, "phar://%s/%s",
file_handle->filename, ".phar/stub.php");
if (SUCCESS == phar_orig_zend_open((const char *)name,
&f)) {
efree(name);
name = NULL;
f.filename = file_handle->filename;
if (f.opened_path) {
efree(f.opened_path);
}
f.opened_path = file_handle->opened_path;
f.free_filename = file_handle->free_filename;
switch (file_handle->type) {
case ZEND_HANDLE_STREAM:
if (file_handle->handle.stream.closer &&
file_handle->handle.stream.handle) {
file_handle->handle.stream.closer(file_handle->handle.stream.handle);
}
file_handle->handle.stream.handle = NULL;
break;
default:
break;
}
*file_handle = f;
}
} else if (phar->flags & PHAR_FILE_COMPRESSION_MASK) {
zend_file_handle_dtor(file_handle);
/* compressed phar */
file_handle->type = ZEND_HANDLE_STREAM;
/* we do our own reading directly from the phar, don't
change the next line */
file_handle->handle.stream.handle = phar;
file_handle->handle.stream.reader =
phar_zend_stream_reader;
file_handle->handle.stream.closer = NULL;
file_handle->handle.stream.fsizer =
phar_zend_stream_fsizer;
file_handle->handle.stream.isatty = 0;
phar->is_persistent ?
php_stream_rewind(PHAR_G(cached_fp)[phar->phar_pos].fp) :
php_stream_rewind(phar->fp);
}
}
}
zend_try {
failed = 0;
CG(zend_lineno) = 0;
res = phar_orig_compile_file(file_handle, type);
} zend_catch {
failed = 1;
res = NULL;
} zend_end_try();
if (name) {
efree(name);
}
if (failed) {
zend_bailout();
}
return res;
}
I *think* that what maybe would fix this bug is if line 3299 (res =
phar_orig_compile_file(file_handle, type);)
was - if BCGEN extension is loaded - is affected so that
phar_orig_compile_file was changed to point to another function.
Let's imagine that phar_orig_compile_file pointed to
bcgen_phar_compile_file() which was then implemented by BCGEN:
ZendAccelerator.h (of BCGEN): (Link:
https://github.com/vjardin/bcgen/blob/a863b8d596cebb698a491f7475903f5aa3664719/ZendAccelerator.h)
zend_op_array *bcgen_phar_compile_file(zend_file_handle *file_handle,
int type);
ZendAccelerator.c (of BCGEN): (Link:
https://github.com/vjardin/bcgen/blob/a863b8d596cebb698a491f7475903f5aa3664719/ZendAccelerator.c)
zend_op_array *bcgen_phar_compile_file(zend_file_handle *file_handle,
int type) {
if (!file_handle->filename || !ZCG(enabled) || !accel_startup_ok) {
/* The Accelerator is disabled, act as if without BCGEN */
return zend_compile(file_handle, type);
} else {
/* Use the zend_compile replacement of BCGEN for this file */
return persistent_load_file(file_handle, type);
}
}
Now, for my question:
How can I make sure that phar_orig_compile_file() points to
bcgen_phar_compile_file() when line 3299 of /ext/phar/phar.c is hit
*WITHOUT* changing line 3299 if - and only if BCGEN module is present ?
Would a #define do it ?
(something like:
#define phar_orig_compile_file bcgen_phar_compile_file
in ZendAccelerator.h (of BCGEN) (Link:
https://github.com/vjardin/bcgen/blob/a863b8d596cebb698a491f7475903f5aa3664719/ZendAccelerator.h)
NOTE: /ext/phar/phar.c already has this on line 3230:
zend_op_array *(*phar_orig_compile_file)(zend_file_handle *file_handle,
int type);
which would conflict with my - proposed - define/solution or ?
Please advice !
If it is inappropriate to ask for coding advice on this list, I do
apologize for this mail !
Thank you in advance !
Best regards,
Henrik Skov
Secuno A/S
