[4/5] Fix latent bug in test_gdb_complete_cmd_multiple
Commit Message
A following patch will add the following to a testcase:
test_gdb_completion_offers_commands "| "
And that tripped on a latent testsuite bug:
(gdb) | PASS: gdb.base/shell.exp: tab complete "| "
^CQuit
(gdb) complete |
| !
| +
PASS: gdb.base/shell.exp: cmd complete "| "
| *** List may be truncated, max-completions reached. ***
(gdb) FAIL: gdb.base/shell.exp: set max-completions 200
set max-completions 200
The issue is that "|" ends up as part of a regexp, and "|" in regexps
has a special meaning...
Fix this with string_to_regexp.
gdb/testsuite/ChangeLog:
yyyy-mm-dd Pedro Alves <palves@redhat.com>
* lib/completion-support.exp (test_gdb_complete_cmd_multiple): Use
string_to_regexp.
---
gdb/testsuite/lib/completion-support.exp | 3 ++-
1 file changed, 2 insertions(+), 1 deletion(-)
Comments
>>>>> "Pedro" == Pedro Alves <palves@redhat.com> writes:
Pedro> if {$max_completions} {
Pedro> + set cmd_prefix_re [string_to_regexp $cmd_prefix]
Pedro> append expected_re \
Pedro> - "$cmd_prefix \\*\\*\\* List may be truncated, max-completions reached\\. \\*\\*\\*.*\r\n"
Pedro> + "$cmd_prefix_re \\*\\*\\* List may be truncated, max-completions reached\\. \\*\\*\\*.*\r\n"
Pedro> }
Pedro> set cmd_re [string_to_regexp "complete $cmd_prefix$completion_word"]
It looks like this is going to end up running string_to_regexp twice --
once in the "if" and then again here, which is probably not desirable.
Tom
On 6/28/19 3:53 PM, Tom Tromey wrote:
>>>>>> "Pedro" == Pedro Alves <palves@redhat.com> writes:
>
> Pedro> if {$max_completions} {
> Pedro> + set cmd_prefix_re [string_to_regexp $cmd_prefix]
> Pedro> append expected_re \
> Pedro> - "$cmd_prefix \\*\\*\\* List may be truncated, max-completions reached\\. \\*\\*\\*.*\r\n"
> Pedro> + "$cmd_prefix_re \\*\\*\\* List may be truncated, max-completions reached\\. \\*\\*\\*.*\r\n"
> Pedro> }
>
> Pedro> set cmd_re [string_to_regexp "complete $cmd_prefix$completion_word"]
>
> It looks like this is going to end up running string_to_regexp twice --
> once in the "if" and then again here, which is probably not desirable.
It won't because the second time is still referring to $cmd_prefix,
while the new code introduces a new $cmd_prefix_re variable.
Thanks,
Pedro Alves
>>>>> "Pedro" == Pedro Alves <palves@redhat.com> writes:
Pedro> On 6/28/19 3:53 PM, Tom Tromey wrote:
>>>>>>> "Pedro" == Pedro Alves <palves@redhat.com> writes:
>>
Pedro> if {$max_completions} {
Pedro> + set cmd_prefix_re [string_to_regexp $cmd_prefix]
Pedro> append expected_re \
Pedro> - "$cmd_prefix \\*\\*\\* List may be truncated, max-completions reached\\. \\*\\*\\*.*\r\n"
Pedro> + "$cmd_prefix_re \\*\\*\\* List may be truncated, max-completions reached\\. \\*\\*\\*.*\r\n"
Pedro> }
>>
Pedro> set cmd_re [string_to_regexp "complete $cmd_prefix$completion_word"]
>>
>> It looks like this is going to end up running string_to_regexp twice --
>> once in the "if" and then again here, which is probably not desirable.
Pedro> It won't because the second time is still referring to $cmd_prefix,
Pedro> while the new code introduces a new $cmd_prefix_re variable.
Indeed, I misread that. Thanks.
Tom
@@ -200,8 +200,9 @@ proc test_gdb_complete_cmd_multiple { cmd_prefix completion_word completion_list
set expected_re [make_cmd_completion_list_re $cmd_prefix $completion_list $start_quote_char $end_quote_char]
if {$max_completions} {
+ set cmd_prefix_re [string_to_regexp $cmd_prefix]
append expected_re \
- "$cmd_prefix \\*\\*\\* List may be truncated, max-completions reached\\. \\*\\*\\*.*\r\n"
+ "$cmd_prefix_re \\*\\*\\* List may be truncated, max-completions reached\\. \\*\\*\\*.*\r\n"
}
set cmd_re [string_to_regexp "complete $cmd_prefix$completion_word"]