diff options
author | Tom Tromey <tromey@adacore.com> | 2025-02-24 13:18:27 -0700 |
---|---|---|
committer | Tom Tromey <tromey@adacore.com> | 2025-03-06 07:08:14 -0700 |
commit | 5393cfd07abcafad6f3a84dcd263f7248943ace6 (patch) | |
tree | 536cbfdd04bb796ef250235990b7da9074b87895 /gdb/cli/cli-utils.h | |
parent | 37b0b22f16c73903ad382fb8435bdd4b56464001 (diff) | |
download | binutils-5393cfd07abcafad6f3a84dcd263f7248943ace6.zip binutils-5393cfd07abcafad6f3a84dcd263f7248943ace6.tar.gz binutils-5393cfd07abcafad6f3a84dcd263f7248943ace6.tar.bz2 |
Avoid double-decoding in ada_add_global_exceptions
I noticed that ada_add_global_exceptions calls ada_decode on
'search_name' -- and then passes this to name_matches_regex, which
also calls ada_decode.
name_matches_regex is also used later, where the result of
'natural_name ()' is passed to it -- but natural_name also calls
ada_decode.
So, I think the call to ada_decode in name_matches_regex is redundant.
This patch removes it, and turns name_matches_regex into an inner
function to avoid propagating its use.
Note that, right now, the DWARF implementation of
expand_symtabs_matching does not in fact pass an encoded name to this
callback. So, this code remains slightly (but currently harmlessly)
wrong. expand_symtabs_matching is fixed by another pending series of
mine.
Diffstat (limited to 'gdb/cli/cli-utils.h')
0 files changed, 0 insertions, 0 deletions