mirror of
https://sourceware.org/git/glibc.git
synced 2024-12-28 05:21:13 +00:00
0f9e585480
[Modified from the original email by Siddhesh Poyarekar] This patch solves bug #16009 by implementing an additional path in strxfrm that does not depend on caching the weight and rule indices. In detail the following changed: * The old main loop was factored out of strxfrm_l into the function do_xfrm_cached to be able to alternativly use the non-caching version do_xfrm. * strxfrm_l allocates a a fixed size array on the stack. If this is not sufficiant to store the weight and rule indices, the non-caching path is taken. As the cache size is not dependent on the input there can be no problems with integer overflows or stack allocations greater than __MAX_ALLOCA_CUTOFF. Note that malloc-ing is not possible because the definition of strxfrm does not allow an oom errorhandling. * The uncached path determines the weight and rule index for every char and for every pass again. * Passing all the locale data array by array resulted in very long parameter lists, so I introduced a structure that holds them. * Checking for zero src string has been moved a bit upwards, it is before the locale data initialization now. * To verify that the non-caching path works correct I added a test run to localedata/sort-test.sh & localedata/xfrm-test.c where all strings are patched up with spaces so that they are too large for the caching path. |
||
---|---|---|
.. | ||
charmaps | ||
locales | ||
tests | ||
tests-mbwc | ||
tst-fmon-locales | ||
bug-iconv-trans.c | ||
bug-setlocale1-static.c | ||
bug-setlocale1.c | ||
bug-usesetlocale.c | ||
ChangeLog | ||
collate-test.c | ||
cs_CZ.in | ||
da_DK.in | ||
de_DE.in | ||
Depend | ||
dump-ctype.c | ||
en_US.in | ||
fr_FR.in | ||
gen-locale.sh | ||
gen-unicode-ctype.c | ||
hr_HR.in | ||
Makefile | ||
README | ||
show-ucs-data.c | ||
si_LK.in | ||
sort-test.sh | ||
SUPPORTED | ||
sv_SE.in | ||
th_TH.in | ||
tr_TR.in | ||
tst-ctype-de_DE.ISO-8859-1.in | ||
tst-ctype.c | ||
tst-ctype.sh | ||
tst-digits.c | ||
tst-fmon.c | ||
tst-fmon.data | ||
tst-fmon.sh | ||
tst-langinfo-static.c | ||
tst-langinfo.c | ||
tst-langinfo.sh | ||
tst-leaks.c | ||
tst-locale.sh | ||
tst-mbswcs1.c | ||
tst-mbswcs2.c | ||
tst-mbswcs3.c | ||
tst-mbswcs4.c | ||
tst-mbswcs5.c | ||
tst-mbswcs6.c | ||
tst-numeric.c | ||
tst-numeric.data | ||
tst-numeric.sh | ||
tst-rpmatch.c | ||
tst-rpmatch.sh | ||
tst-setlocale2.c | ||
tst-setlocale3.c | ||
tst-setlocale.c | ||
tst-sscanf.c | ||
tst-strfmon1.c | ||
tst-trans.c | ||
tst-trans.sh | ||
tst-wctype.c | ||
tst-wctype.input | ||
tst-xlocale1.c | ||
tst-xlocale2.c | ||
xfrm-test.c |
POSIX locale descriptions and POSIX character set descriptions Ulrich Drepper Time-stamp: <2004/11/27 13:06:54 drepper> drepper@redhat.com This directory contains the data needed to build the locale data files to use the internationalization features of the GNU libc. POSIX.2 describes the `localedef' utility which is part of the GNU libc. You need this program to "compile" the locale description in a form suitable for fast access by the GNU libc functions. Any compilation is based on a given character set. Once you run `make install' for the GNU libc the data files are automatically installed in the right place, ready for use by the `localedef' program. To compile the locale data files you simply have to decide which locale (based on the location and the language) and which character set you use. E.g., French speaking Canadians would use the locale `fr_CA' and the character set `ISO_8859-1,1987'. Calling `localedef' to get the desired data should happen like this: localedef -i fr_CA -f ISO-8859-1 fr_CA This will place the 6 output files in the appropriate directory where the GNU libc functions can find them. Please note that you need permission to write to this directory ($(prefix)/share/locale, where $(prefix) is the value you specified while configuring GNU libc). If you do not have the necessary permissions, you can write the files into an arbitrary directory by giving a path including a '/' character instead of `fr_CA'. E.g., to put the new files in a subdirectory of the current directory simply use localedef -i fr_CA -f ISO-8859-1 ./fr_CA How to use these data files is described in the GNU libc manual, especially in the section describing the `setlocale' function. All problems should be reported using http://sourceware.org/bugzilla/ One more note: the `POSIX' locale definition is not meant to be used as an input file for `localedef'. It is rather there to show the values with are built in the libc binaries as default values when no legal locale is found or the "C" or "POSIX" locale is selected. The collation test suite ######################## This package also contains a (beginning of a) test suite for the collation functions in the GNU libc. The files are provided sorted. The test program shuffles the lines and sort them afterwards. Some of the files are provided in 8bit form, i.e., not only ASCII characters. So the tools you use to process the files should be 8bit clean. To run the test program the appropriate locale information must be installed. Therefore the localedef program is used to generate this data used the locale and charmap description files contained here. Since we cannot run the localedef program in case of cross-compilation no tests at all are performed. ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ Local Variables: mode:text eval:(load-library "time-stamp") eval:(make-local-variable 'write-file-hooks) eval:(add-hook 'write-file-hooks 'time-stamp) eval:(setq time-stamp-format '(time-stamp-yyyy/mm/dd time-stamp-hh:mm:ss user-login-name)) End: