libxkbcommon/test/buffercomp.c

152 lines
5.9 KiB
C
Raw Normal View History

keymap: add xkb_keymap_new_from_buffer() The current API doesn't allow the caller to create keymaps from mmap()'ed files. The problem is, xkb_keymap_new_from_string() requires a terminating 0 byte. However, there is no way to guarantee that when using mmap() so a user currently has to copy the whole file just to get the terminating zero byte (assuming they cannot use xkb_keymap_new_from_file()). This adds a new entry xkb_keymap_new_from_buffer() which takes a memory location and the buffer size in bytes. Internally, we depend on yy_scan_{string,byte}() helpers. According to flex documentation these already copy the input string because they are wrappers around yy_scan_buffer(). yy_scan_buffer() on the other hand has some insane requirements. The buffer must be writeable and the last two bytes must be ASCII-NUL. But the buffer may contain other 0 bytes just fine. Because we don't want these constraints in our public API, xkb_keymap_new_from_buffer() needs to create a copy of the input memory. But it then calls yy_scan_buffer() directly. Hence, we have the same number of buffer-copies as with *_from_string() but without the terminating 0 requirement. The explicit yy_scan_buffer() call is preferred over yy_scan_byte() so the buffer-copy operation is not hidden somewhere in flex. Maybe some day we no longer depend on flex and can have a zero-copy API. A user could mmap() a file and it would get parsed right from this buffer. But until then, we shouldn't expose this limitation in the API but instead provide an API that some day can work with zero-copy. Signed-off-by: David Herrmann <dh.herrmann@gmail.com> [ran: rebased on top of my branch] Conflicts: Makefile.am src/xkbcomp/xkbcomp.c
2013-03-11 05:53:39 -06:00
/*
* Copyright © 2009 Dan Nicholson
*
* Permission is hereby granted, free of charge, to any person obtaining a
* copy of this software and associated documentation files (the "Software"),
* to deal in the Software without restriction, including without limitation
* the rights to use, copy, modify, merge, publish, distribute, sublicense,
* and/or sell copies of the Software, and to permit persons to whom the
* Software is furnished to do so, subject to the following conditions:
*
* The above copyright notice and this permission notice (including the next
* paragraph) shall be included in all copies or substantial portions of the
* Software.
*
* THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
* IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
* FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL
* THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
* LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING
* FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER
* DEALINGS IN THE SOFTWARE.
*/
#include "config.h"
keymap: add xkb_keymap_new_from_buffer() The current API doesn't allow the caller to create keymaps from mmap()'ed files. The problem is, xkb_keymap_new_from_string() requires a terminating 0 byte. However, there is no way to guarantee that when using mmap() so a user currently has to copy the whole file just to get the terminating zero byte (assuming they cannot use xkb_keymap_new_from_file()). This adds a new entry xkb_keymap_new_from_buffer() which takes a memory location and the buffer size in bytes. Internally, we depend on yy_scan_{string,byte}() helpers. According to flex documentation these already copy the input string because they are wrappers around yy_scan_buffer(). yy_scan_buffer() on the other hand has some insane requirements. The buffer must be writeable and the last two bytes must be ASCII-NUL. But the buffer may contain other 0 bytes just fine. Because we don't want these constraints in our public API, xkb_keymap_new_from_buffer() needs to create a copy of the input memory. But it then calls yy_scan_buffer() directly. Hence, we have the same number of buffer-copies as with *_from_string() but without the terminating 0 requirement. The explicit yy_scan_buffer() call is preferred over yy_scan_byte() so the buffer-copy operation is not hidden somewhere in flex. Maybe some day we no longer depend on flex and can have a zero-copy API. A user could mmap() a file and it would get parsed right from this buffer. But until then, we shouldn't expose this limitation in the API but instead provide an API that some day can work with zero-copy. Signed-off-by: David Herrmann <dh.herrmann@gmail.com> [ran: rebased on top of my branch] Conflicts: Makefile.am src/xkbcomp/xkbcomp.c
2013-03-11 05:53:39 -06:00
#include <assert.h>
#include <stdio.h>
#include <stdlib.h>
#include "test.h"
#define DATA_PATH "keymaps/stringcomp.data"
static bool
test_encodings(struct xkb_context *ctx)
{
struct xkb_keymap *keymap;
/* Accept UTF-8 encoded BOM (U+FEFF) */
const char utf8_with_bom[] =
"\xef\xbb\xbfxkb_keymap {"
" xkb_keycodes { include \"evdev\" };"
" xkb_types { include \"complete\" };"
" xkb_compat { include \"complete\" };"
" xkb_symbols { include \"pc\" };"
"};";
keymap = test_compile_buffer(ctx, utf8_with_bom, sizeof(utf8_with_bom));
assert(keymap);
xkb_keymap_unref(keymap);
/* Reject UTF-16LE encoded string */
const char utf16_le[] =
"x\0k\0b\0_\0k\0e\0y\0m\0a\0p\0 \0{\0\n\0"
" \0 \0x\0k\0b\0_\0k\0e\0y\0c\0o\0d\0e\0s\0 \0{\0 \0i\0n\0c\0l\0u\0d\0e\0 \0\"\0e\0v\0d\0e\0v\0\"\0 \0}\0;\0\n\0"
" \0 \0x\0k\0b\0_\0t\0y\0p\0e\0s\0 \0{\0 \0i\0n\0c\0l\0u\0d\0e\0 \0\"\0c\0o\0m\0p\0l\0e\0t\0e\0\"\0 \0}\0;\0\n\0"
" \0 \0x\0k\0b\0_\0c\0o\0m\0p\0a\0t\0 \0{\0 \0i\0n\0c\0l\0u\0d\0e\0 \0\"\0c\0o\0m\0p\0l\0e\0t\0e\0\"\0 \0}\0;\0\n\0"
" \0 \0x\0k\0b\0_\0s\0y\0m\0b\0o\0l\0s\0 \0{\0 \0i\0n\0c\0l\0u\0d\0e\0 \0\"\0p\0c\0\"\0 \0}\0;\0\n\0"
"}\0;\0";
keymap = test_compile_buffer(ctx, utf16_le, sizeof(utf16_le));
assert(!keymap);
/* Reject UTF-16LE with BOM encoded string */
const char utf16_le_with_bom[] =
"\xff\xfex\0k\0b\0_\0k\0e\0y\0m\0a\0p\0 \0{\0\n\0"
" \0 \0x\0k\0b\0_\0k\0e\0y\0c\0o\0d\0e\0s\0 \0{\0 \0i\0n\0c\0l\0u\0d\0e\0 \0\"\0e\0v\0d\0e\0v\0\"\0 \0}\0;\0\n\0"
" \0 \0x\0k\0b\0_\0t\0y\0p\0e\0s\0 \0{\0 \0i\0n\0c\0l\0u\0d\0e\0 \0\"\0c\0o\0m\0p\0l\0e\0t\0e\0\"\0 \0}\0;\0\n\0"
" \0 \0x\0k\0b\0_\0c\0o\0m\0p\0a\0t\0 \0{\0 \0i\0n\0c\0l\0u\0d\0e\0 \0\"\0c\0o\0m\0p\0l\0e\0t\0e\0\"\0 \0}\0;\0\n\0"
" \0 \0x\0k\0b\0_\0s\0y\0m\0b\0o\0l\0s\0 \0{\0 \0i\0n\0c\0l\0u\0d\0e\0 \0\"\0p\0c\0\"\0 \0}\0;\0\n\0"
"}\0;\0";
keymap = test_compile_buffer(ctx, utf16_le_with_bom, sizeof(utf16_le_with_bom));
assert(!keymap);
/* Reject UTF-16BE encoded string */
const char utf16_be[] =
"\0x\0k\0b\0_\0k\0e\0y\0m\0a\0p\0 \0{\0\n\0"
" \0 \0x\0k\0b\0_\0k\0e\0y\0c\0o\0d\0e\0s\0 \0{\0 \0i\0n\0c\0l\0u\0d\0e\0 \0\"\0e\0v\0d\0e\0v\0\"\0 \0}\0;\0\n\0"
" \0 \0x\0k\0b\0_\0t\0y\0p\0e\0s\0 \0{\0 \0i\0n\0c\0l\0u\0d\0e\0 \0\"\0c\0o\0m\0p\0l\0e\0t\0e\0\"\0 \0}\0;\0\n\0"
" \0 \0x\0k\0b\0_\0c\0o\0m\0p\0a\0t\0 \0{\0 \0i\0n\0c\0l\0u\0d\0e\0 \0\"\0c\0o\0m\0p\0l\0e\0t\0e\0\"\0 \0}\0;\0\n\0"
" \0 \0x\0k\0b\0_\0s\0y\0m\0b\0o\0l\0s\0 \0{\0 \0i\0n\0c\0l\0u\0d\0e\0 \0\"\0p\0c\0\"\0 \0}\0;\0\n\0"
"}\0;";
keymap = test_compile_buffer(ctx, utf16_be, sizeof(utf16_be));
assert(!keymap);
return true;
}
keymap: add xkb_keymap_new_from_buffer() The current API doesn't allow the caller to create keymaps from mmap()'ed files. The problem is, xkb_keymap_new_from_string() requires a terminating 0 byte. However, there is no way to guarantee that when using mmap() so a user currently has to copy the whole file just to get the terminating zero byte (assuming they cannot use xkb_keymap_new_from_file()). This adds a new entry xkb_keymap_new_from_buffer() which takes a memory location and the buffer size in bytes. Internally, we depend on yy_scan_{string,byte}() helpers. According to flex documentation these already copy the input string because they are wrappers around yy_scan_buffer(). yy_scan_buffer() on the other hand has some insane requirements. The buffer must be writeable and the last two bytes must be ASCII-NUL. But the buffer may contain other 0 bytes just fine. Because we don't want these constraints in our public API, xkb_keymap_new_from_buffer() needs to create a copy of the input memory. But it then calls yy_scan_buffer() directly. Hence, we have the same number of buffer-copies as with *_from_string() but without the terminating 0 requirement. The explicit yy_scan_buffer() call is preferred over yy_scan_byte() so the buffer-copy operation is not hidden somewhere in flex. Maybe some day we no longer depend on flex and can have a zero-copy API. A user could mmap() a file and it would get parsed right from this buffer. But until then, we shouldn't expose this limitation in the API but instead provide an API that some day can work with zero-copy. Signed-off-by: David Herrmann <dh.herrmann@gmail.com> [ran: rebased on top of my branch] Conflicts: Makefile.am src/xkbcomp/xkbcomp.c
2013-03-11 05:53:39 -06:00
int
main(int argc, char *argv[])
{
struct xkb_context *ctx = test_get_context(0);
struct xkb_keymap *keymap;
char *original, *dump;
assert(ctx);
/* Load in a prebuilt keymap, make sure we can compile it from memory,
* then compare it to make sure we get the same result when dumping it
* to a string. */
original = test_read_file(DATA_PATH);
assert(original);
/* Load a prebuild keymap, once without, once with the trailing \0 */
for (int i = 0; i <= 1; i++) {
keymap = test_compile_buffer(ctx, original, strlen(original) + i);
assert(keymap);
keymap: add xkb_keymap_new_from_buffer() The current API doesn't allow the caller to create keymaps from mmap()'ed files. The problem is, xkb_keymap_new_from_string() requires a terminating 0 byte. However, there is no way to guarantee that when using mmap() so a user currently has to copy the whole file just to get the terminating zero byte (assuming they cannot use xkb_keymap_new_from_file()). This adds a new entry xkb_keymap_new_from_buffer() which takes a memory location and the buffer size in bytes. Internally, we depend on yy_scan_{string,byte}() helpers. According to flex documentation these already copy the input string because they are wrappers around yy_scan_buffer(). yy_scan_buffer() on the other hand has some insane requirements. The buffer must be writeable and the last two bytes must be ASCII-NUL. But the buffer may contain other 0 bytes just fine. Because we don't want these constraints in our public API, xkb_keymap_new_from_buffer() needs to create a copy of the input memory. But it then calls yy_scan_buffer() directly. Hence, we have the same number of buffer-copies as with *_from_string() but without the terminating 0 requirement. The explicit yy_scan_buffer() call is preferred over yy_scan_byte() so the buffer-copy operation is not hidden somewhere in flex. Maybe some day we no longer depend on flex and can have a zero-copy API. A user could mmap() a file and it would get parsed right from this buffer. But until then, we shouldn't expose this limitation in the API but instead provide an API that some day can work with zero-copy. Signed-off-by: David Herrmann <dh.herrmann@gmail.com> [ran: rebased on top of my branch] Conflicts: Makefile.am src/xkbcomp/xkbcomp.c
2013-03-11 05:53:39 -06:00
dump = xkb_keymap_get_as_string(keymap, XKB_KEYMAP_USE_ORIGINAL_FORMAT);
assert(dump);
keymap: add xkb_keymap_new_from_buffer() The current API doesn't allow the caller to create keymaps from mmap()'ed files. The problem is, xkb_keymap_new_from_string() requires a terminating 0 byte. However, there is no way to guarantee that when using mmap() so a user currently has to copy the whole file just to get the terminating zero byte (assuming they cannot use xkb_keymap_new_from_file()). This adds a new entry xkb_keymap_new_from_buffer() which takes a memory location and the buffer size in bytes. Internally, we depend on yy_scan_{string,byte}() helpers. According to flex documentation these already copy the input string because they are wrappers around yy_scan_buffer(). yy_scan_buffer() on the other hand has some insane requirements. The buffer must be writeable and the last two bytes must be ASCII-NUL. But the buffer may contain other 0 bytes just fine. Because we don't want these constraints in our public API, xkb_keymap_new_from_buffer() needs to create a copy of the input memory. But it then calls yy_scan_buffer() directly. Hence, we have the same number of buffer-copies as with *_from_string() but without the terminating 0 requirement. The explicit yy_scan_buffer() call is preferred over yy_scan_byte() so the buffer-copy operation is not hidden somewhere in flex. Maybe some day we no longer depend on flex and can have a zero-copy API. A user could mmap() a file and it would get parsed right from this buffer. But until then, we shouldn't expose this limitation in the API but instead provide an API that some day can work with zero-copy. Signed-off-by: David Herrmann <dh.herrmann@gmail.com> [ran: rebased on top of my branch] Conflicts: Makefile.am src/xkbcomp/xkbcomp.c
2013-03-11 05:53:39 -06:00
if (!streq(original, dump)) {
fprintf(stderr,
"round-trip test failed: dumped map differs from original\n");
fprintf(stderr, "path to original file: %s\n",
test_get_path(DATA_PATH));
fprintf(stderr, "length: dumped %lu, original %lu\n",
(unsigned long) strlen(dump),
(unsigned long) strlen(original));
fprintf(stderr, "dumped map:\n");
fprintf(stderr, "%s\n", dump);
fflush(stderr);
assert(0);
}
free(dump);
xkb_keymap_unref(keymap);
keymap: add xkb_keymap_new_from_buffer() The current API doesn't allow the caller to create keymaps from mmap()'ed files. The problem is, xkb_keymap_new_from_string() requires a terminating 0 byte. However, there is no way to guarantee that when using mmap() so a user currently has to copy the whole file just to get the terminating zero byte (assuming they cannot use xkb_keymap_new_from_file()). This adds a new entry xkb_keymap_new_from_buffer() which takes a memory location and the buffer size in bytes. Internally, we depend on yy_scan_{string,byte}() helpers. According to flex documentation these already copy the input string because they are wrappers around yy_scan_buffer(). yy_scan_buffer() on the other hand has some insane requirements. The buffer must be writeable and the last two bytes must be ASCII-NUL. But the buffer may contain other 0 bytes just fine. Because we don't want these constraints in our public API, xkb_keymap_new_from_buffer() needs to create a copy of the input memory. But it then calls yy_scan_buffer() directly. Hence, we have the same number of buffer-copies as with *_from_string() but without the terminating 0 requirement. The explicit yy_scan_buffer() call is preferred over yy_scan_byte() so the buffer-copy operation is not hidden somewhere in flex. Maybe some day we no longer depend on flex and can have a zero-copy API. A user could mmap() a file and it would get parsed right from this buffer. But until then, we shouldn't expose this limitation in the API but instead provide an API that some day can work with zero-copy. Signed-off-by: David Herrmann <dh.herrmann@gmail.com> [ran: rebased on top of my branch] Conflicts: Makefile.am src/xkbcomp/xkbcomp.c
2013-03-11 05:53:39 -06:00
}
free(original);
/* Make sure we can't (falsely claim to) compile an empty string. */
keymap = test_compile_buffer(ctx, "", 0);
assert(!keymap);
assert(test_encodings(ctx));
keymap: add xkb_keymap_new_from_buffer() The current API doesn't allow the caller to create keymaps from mmap()'ed files. The problem is, xkb_keymap_new_from_string() requires a terminating 0 byte. However, there is no way to guarantee that when using mmap() so a user currently has to copy the whole file just to get the terminating zero byte (assuming they cannot use xkb_keymap_new_from_file()). This adds a new entry xkb_keymap_new_from_buffer() which takes a memory location and the buffer size in bytes. Internally, we depend on yy_scan_{string,byte}() helpers. According to flex documentation these already copy the input string because they are wrappers around yy_scan_buffer(). yy_scan_buffer() on the other hand has some insane requirements. The buffer must be writeable and the last two bytes must be ASCII-NUL. But the buffer may contain other 0 bytes just fine. Because we don't want these constraints in our public API, xkb_keymap_new_from_buffer() needs to create a copy of the input memory. But it then calls yy_scan_buffer() directly. Hence, we have the same number of buffer-copies as with *_from_string() but without the terminating 0 requirement. The explicit yy_scan_buffer() call is preferred over yy_scan_byte() so the buffer-copy operation is not hidden somewhere in flex. Maybe some day we no longer depend on flex and can have a zero-copy API. A user could mmap() a file and it would get parsed right from this buffer. But until then, we shouldn't expose this limitation in the API but instead provide an API that some day can work with zero-copy. Signed-off-by: David Herrmann <dh.herrmann@gmail.com> [ran: rebased on top of my branch] Conflicts: Makefile.am src/xkbcomp/xkbcomp.c
2013-03-11 05:53:39 -06:00
/* Make sure we can recompile our output for a normal keymap from rules. */
keymap = test_compile_rules(ctx, NULL, NULL,
"ru,ca,de,us", ",multix,neo,intl", NULL);
assert(keymap);
dump = xkb_keymap_get_as_string(keymap, XKB_KEYMAP_USE_ORIGINAL_FORMAT);
assert(dump);
xkb_keymap_unref(keymap);
keymap = test_compile_buffer(ctx, dump, strlen(dump));
assert(keymap);
xkb_keymap_unref(keymap);
free(dump);
xkb_context_unref(ctx);
return 0;
}