http://sourceware.org/ml/gdb-patches/2011-04/msg00546.html Subject: [obv] Fix completer pre-expansion for case insensitive lookups Hi, due to the pending patches there happened a semantic collision. Both pathsets were right on their own but together it did not work well. There could be a testcase for it but I do not provide it, it seems obvious enought to me. The two patches were: Re: [patch 3/3] case insensitive: the fix [rediff] http://sourceware.org/ml/gdb-patches/2011-04/msg00418.html - see the change in completion_list_add_name and [patch][+7.3] Fix physname completion regression http://sourceware.org/ml/gdb-patches/2011-04/msg00140.html - see expand_partial_symbol_name which just called completion_list_add_name before but now it has explicit strncmp. Another problem is this code is still wrong. That is the subject of: CU expansion problem for parameters http://sourceware.org/bugzilla/show_bug.cgi?id=12708 But I did not want to mix two unrelated fixes into a single patch No regressions on {x86_64,x86_64-m32,i686}-fedora15-linux-gnu. Checked it in already to not get lost in the inter-patch dependencies again. Sorry, Jan http://sourceware.org/ml/gdb-cvs/2011-04/msg00186.html ### src/gdb/ChangeLog 2011/04/28 15:52:53 1.12964 ### src/gdb/ChangeLog 2011/04/28 17:37:02 1.12965 ## -1,3 +1,8 @@ +2011-04-28 Jan Kratochvil + + * symtab.c (expand_partial_symbol_name): New variable NCMP. Support + case insensitive comparison. + 2011-04-28 Ulrich Weigand * infrun.c (proceed): Revert previous change. --- src/gdb/symtab.c 2011/04/27 20:03:03 1.269 +++ src/gdb/symtab.c 2011/04/28 17:37:06 1.270 @@ -3707,8 +3707,11 @@ expand_partial_symbol_name (const char *name, void *user_data) { struct add_name_data *datum = (struct add_name_data *) user_data; + int (*ncmp) (const char *, const char *, size_t); - return strncmp (name, datum->sym_text, datum->sym_text_len) == 0; + ncmp = (case_sensitivity == case_sensitive_on ? strncmp : strncasecmp); + + return ncmp (name, datum->sym_text, datum->sym_text_len) == 0; } char **