aboutsummaryrefslogtreecommitdiff
path: root/locale/programs
diff options
context:
space:
mode:
authorUlrich Drepper <drepper@redhat.com>2009-09-07 02:32:29 -0700
committerUlrich Drepper <drepper@redhat.com>2009-09-07 02:32:29 -0700
commit7a7e49c020125d444fec7fded51bd5f82bfc8c49 (patch)
tree25c434b7088fb319dbda653fb9c2e3490837fcf4 /locale/programs
parentd76da20f7f77e9dfc7e81ebaf0f9902699a873e8 (diff)
downloadglibc-7a7e49c020125d444fec7fded51bd5f82bfc8c49.zip
glibc-7a7e49c020125d444fec7fded51bd5f82bfc8c49.tar.gz
glibc-7a7e49c020125d444fec7fded51bd5f82bfc8c49.tar.bz2
Fix endless loop in localedef.
localedef got into an endless loop in case order_start was used for the unnamed_section twice and the first use didn't actually result into any definition.
Diffstat (limited to 'locale/programs')
-rw-r--r--locale/programs/ld-collate.c59
1 files changed, 31 insertions, 28 deletions
diff --git a/locale/programs/ld-collate.c b/locale/programs/ld-collate.c
index bf50e77..11bd7ea 100644
--- a/locale/programs/ld-collate.c
+++ b/locale/programs/ld-collate.c
@@ -1,4 +1,4 @@
-/* Copyright (C) 1995-2003, 2005-2007, 2008 Free Software Foundation, Inc.
+/* Copyright (C) 1995-2003, 2005-2008, 2009 Free Software Foundation, Inc.
This file is part of the GNU C Library.
Contributed by Ulrich Drepper <drepper@gnu.org>, 1995.
@@ -203,6 +203,8 @@ struct locale_collate_t
struct section_list *current_section;
/* There always can be an unnamed section. */
struct section_list unnamed_section;
+ /* Flag whether the unnamed section has been defined. */
+ bool unnamed_section_defined;
/* To make handling of errors easier we have another section. */
struct section_list error_section;
/* Sometimes we are defining the values for collating symbols before
@@ -634,7 +636,7 @@ find_element (struct linereader *ldfile, struct locale_collate_t *collate,
if (find_entry (&collate->seq_table, str, len, &result) != 0)
{
/* Nope, not define yet. So we see whether it is a
- collation symbol. */
+ collation symbol. */
void *ptr;
if (find_entry (&collate->sym_table, str, len, &ptr) == 0)
@@ -788,7 +790,7 @@ insert_weights (struct linereader *ldfile, struct element_t *elem,
if (*cp == '<')
{
/* Ahh, it's a bsymbol or an UCS4 value. If it's
- the latter we have to unify the name. */
+ the latter we have to unify the name. */
const char *startp = ++cp;
size_t len;
@@ -1302,8 +1304,8 @@ order for `%.*s' already defined at %s:%Zu"),
else
{
/* Determine the range. To do so we have to determine the
- common prefix of the both names and then the numeric
- values of both ends. */
+ common prefix of the both names and then the numeric
+ values of both ends. */
size_t lenfrom = strlen (startp->name);
size_t lento = strlen (endp->name);
char buf[lento + 1];
@@ -2222,14 +2224,14 @@ collate_output (struct localedef_t *locale, const struct charmap_t *charmap,
else
{
/* The entries in the list are sorted by length and then
- alphabetically. This is the order in which we will add the
- elements to the collation table. This allows simply walking
+ alphabetically. This is the order in which we will add the
+ elements to the collation table. This allows simply walking
the table in sequence and stopping at the first matching
- entry. Since the longer sequences are coming first in the
- list they have the possibility to match first, just as it
- has to be. In the worst case we are walking to the end of
- the list where we put, if no singlebyte sequence is defined
- in the locale definition, the weights for UNDEFINED.
+ entry. Since the longer sequences are coming first in the
+ list they have the possibility to match first, just as it
+ has to be. In the worst case we are walking to the end of
+ the list where we put, if no singlebyte sequence is defined
+ in the locale definition, the weights for UNDEFINED.
To reduce the length of the search list we compress them a bit.
This happens by collecting sequences of consecutive byte
@@ -2297,7 +2299,7 @@ collate_output (struct localedef_t *locale, const struct charmap_t *charmap,
obstack_1grow_fast (&extrapool, curp->mbs[i]);
/* Now find the end of the consecutive sequence and
- add all the indeces in the indirect pool. */
+ add all the indeces in the indirect pool. */
do
{
weightidx = output_weight (&weightpool, collate, curp);
@@ -2312,7 +2314,7 @@ collate_output (struct localedef_t *locale, const struct charmap_t *charmap,
obstack_int32_grow (&indirectpool, weightidx);
/* And add the end byte sequence. Without length this
- time. */
+ time. */
for (i = 1; i < curp->nmbs; ++i)
obstack_1grow_fast (&extrapool, curp->mbs[i]);
}
@@ -2356,7 +2358,7 @@ collate_output (struct localedef_t *locale, const struct charmap_t *charmap,
& (__alignof__ (int32_t) - 1)) == 0);
/* If the final entry in the list is not a single character we
- add an UNDEFINED entry here. */
+ add an UNDEFINED entry here. */
if (lastp->nmbs != 1)
{
int added = ((sizeof (int32_t) + 1 + 1 + __alignof__ (int32_t) - 1)
@@ -3293,7 +3295,7 @@ error while adding equivalent collating symbol"));
else
{
/* One should not be allowed to open the same
- section twice. */
+ section twice. */
if (sp->first != NULL)
lr_error (ldfile, _("\
%s: multiple order definitions for section `%s'"),
@@ -3349,7 +3351,7 @@ error while adding equivalent collating symbol"));
section. */
collate->current_section = &collate->unnamed_section;
- if (collate->unnamed_section.first != NULL)
+ if (collate->unnamed_section_defined)
lr_error (ldfile, _("\
%s: multiple order definitions for unnamed section"),
"LC_COLLATE");
@@ -3359,6 +3361,7 @@ error while adding equivalent collating symbol"));
the collate->sections list. */
collate->unnamed_section.next = collate->sections;
collate->sections = &collate->unnamed_section;
+ collate->unnamed_section_defined = true;
}
}
@@ -3579,9 +3582,9 @@ error while adding equivalent collating symbol"));
else
{
/* This is bad. The section after which we have to
- reorder does not exist. Therefore we cannot
- process the whole rest of this reorder
- specification. */
+ reorder does not exist. Therefore we cannot
+ process the whole rest of this reorder
+ specification. */
lr_error (ldfile, _("%s: section `%.*s' not known"),
"LC_COLLATE", (int) arg->val.str.lenmb,
arg->val.str.startmb);
@@ -3657,9 +3660,9 @@ error while adding equivalent collating symbol"));
if (state == 0)
{
/* We are outside an `order_start' region. This means
- we must only accept definitions of values for
- collation symbols since these are purely abstract
- values and don't need directions associated. */
+ we must only accept definitions of values for
+ collation symbols since these are purely abstract
+ values and don't need directions associated. */
void *ptr;
if (find_entry (&collate->seq_table, symstr, symlen, &ptr) == 0)
@@ -3741,7 +3744,7 @@ error while adding equivalent collating symbol"));
seqp->next->last = seqp->last;
/* We also have to check whether this entry is the
- first or last of a section. */
+ first or last of a section. */
if (seqp->section->first == seqp)
{
if (seqp->section->first == seqp->section->last)
@@ -3798,7 +3801,7 @@ error while adding equivalent collating symbol"));
}
/* Process the rest of the line which might change
- the collation rules. */
+ the collation rules. */
arg = lr_token (ldfile, charmap, result, repertoire,
verbose);
if (arg->tok != tok_eof && arg->tok != tok_eol)
@@ -3810,8 +3813,8 @@ error while adding equivalent collating symbol"));
else if (was_ellipsis != tok_none)
{
/* Using the information in the `ellipsis_weight'
- element and this and the last value we have to handle
- the ellipsis now. */
+ element and this and the last value we have to handle
+ the ellipsis now. */
assert (state == 1);
handle_ellipsis (ldfile, symstr, symlen, was_ellipsis, charmap,
@@ -3871,7 +3874,7 @@ error while adding equivalent collating symbol"));
case tok_ellipsis3: /* absolute ellipsis */
case tok_ellipsis4: /* symbolic decimal ellipsis */
/* This is the symbolic (decimal or hexadecimal) or absolute
- ellipsis. */
+ ellipsis. */
if (was_ellipsis != tok_none)
goto err_label;