strcoll: Remove incorrect STRDIFF-based optimization (Bug 18589).

The optimization introduced in commit
f13c2a8dff2329c6692a80176262ceaaf8a6f74e, causes regressions in
sorting for languages that have digraphs that change sort order, like
cs_CZ which sorts ch between h and i.

My analysis shows the fast-forwarding optimization in STRCOLL advances
through a digraph while possibly stopping in the middle which results
in a subsequent skipping of the digraph and incorrect sorting. The
optimization is incorrect as implemented and because of that I'm
removing it for 2.23, and I will also commit this fix for 2.22 where
it was originally introduced.

This patch reverts the optimization, introduces a new bug-strcoll2.c
regression test that tests both cs_CZ.UTF-8 and da_DK.ISO-8859-1 and
ensures they sort one digraph each correctly. The optimization can't be
applied without regressing this test.

Checked on x86_64, bug-strcoll2.c fails without this patch and passes
after. This will also get a fix on 2.22 which has the same bug.

5 files changed, 1 insertions, 22 deletions

diff --git a/locale/C-collate.c b/locale/C-collate.c
index d7f3c55..06dfdfa 100644
--- a/locale/C-collate.c
+++ b/locale/C-collate.c
@@ -144,8 +144,6 @@ const struct __locale_data _nl_C_LC_COLLATE attribute_hidden =
     /* _NL_COLLATE_COLLSEQWC */
     { .string = (const char *) collseqwc },
     /* _NL_COLLATE_CODESET */
-    { .string = _nl_C_codeset },
-    /* _NL_COLLATE_ENCODING_TYPE */
-    { .word = __cet_8bit }
+    { .string = _nl_C_codeset }
   }
 };
diff --git a/locale/categories.def b/locale/categories.def
index 045489d..a8dda53 100644
--- a/locale/categories.def
+++ b/locale/categories.def
@@ -58,7 +58,6 @@ DEFINE_CATEGORY
   DEFINE_ELEMENT (_NL_COLLATE_COLLSEQMB,        "collate-collseqmb",        std, wstring)
   DEFINE_ELEMENT (_NL_COLLATE_COLLSEQWC,        "collate-collseqwc",        std, wstring)
   DEFINE_ELEMENT (_NL_COLLATE_CODESET,		"collate-codeset",	    std, string)
-  DEFINE_ELEMENT (_NL_COLLATE_ENCODING_TYPE,   "collate-encoding-type",    std, word)
   ), NO_POSTLOAD)
 
 
diff --git a/locale/langinfo.h b/locale/langinfo.h
index ffc5c7f..a565d9d 100644
--- a/locale/langinfo.h
+++ b/locale/langinfo.h
@@ -255,7 +255,6 @@ enum
   _NL_COLLATE_COLLSEQMB,
   _NL_COLLATE_COLLSEQWC,
   _NL_COLLATE_CODESET,
-  _NL_COLLATE_ENCODING_TYPE,
   _NL_NUM_LC_COLLATE,
 
   /* LC_CTYPE category: character classification.
diff --git a/locale/localeinfo.h b/locale/localeinfo.h
index 434b839..789da44 100644
--- a/locale/localeinfo.h
+++ b/locale/localeinfo.h
@@ -110,14 +110,6 @@ enum coll_sort_rule
   sort_mask
 };
 
-/* Collation encoding type.  */
-enum collation_encoding_type
-{
-  __cet_other,
-  __cet_8bit,
-  __cet_utf8
-};
-
 /* We can map the types of the entries into a few categories.  */
 enum value_type
 {
diff --git a/locale/programs/ld-collate.c b/locale/programs/ld-collate.c
index a39a94f..dc0fe30 100644
--- a/locale/programs/ld-collate.c
+++ b/locale/programs/ld-collate.c
@@ -32,7 +32,6 @@
 #include "linereader.h"
 #include "locfile.h"
 #include "elem-hash.h"
-#include "../localeinfo.h"
 
 /* Uncomment the following line in the production version.  */
 /* #define NDEBUG 1 */
@@ -2131,8 +2130,6 @@ collate_output (struct localedef_t *locale, const struct charmap_t *charmap,
 	  /* The words have to be handled specially.  */
 	  if (idx == _NL_ITEM_INDEX (_NL_COLLATE_SYMB_HASH_SIZEMB))
 	    add_locale_uint32 (&file, 0);
-	  else if (idx == _NL_ITEM_INDEX (_NL_COLLATE_ENCODING_TYPE))
-	    add_locale_uint32 (&file, __cet_other);
 	  else
 	    add_locale_empty (&file);
 	}
@@ -2496,12 +2493,6 @@ collate_output (struct localedef_t *locale, const struct charmap_t *charmap,
   add_locale_raw_data (&file, collate->mbseqorder, 256);
   add_locale_collseq_table (&file, &collate->wcseqorder);
   add_locale_string (&file, charmap->code_set_name);
-  if (strcmp (charmap->code_set_name, "UTF-8") == 0)
-    add_locale_uint32 (&file, __cet_utf8);
-  else if (charmap->mb_cur_max == 1)
-    add_locale_uint32 (&file, __cet_8bit);
-  else
-    add_locale_uint32 (&file, __cet_other);
   write_locale_data (output_path, LC_COLLATE, "LC_COLLATE", &file);
 
   obstack_free (&weightpool, NULL);