[patch/testsuite] Fix expected output for "help set follow-fork-mode"

[patch/testsuite] Fix expected output for "help set follow-fork-mode"

[Joel Brobecker]

[-- Attachment #1: Type: text/plain, Size: 1079 bytes --]

There has been a recent change to infrun (made on 2004-01-13) that
removed the choice "ask" to the possible values for "follow-fork-mode".
The help text for that variable was then modified, but the test
verifying that the help for this command exists has not been modified.

This patch updates the expected output to match the new help text.

2004-04-04  Joel Brobecker  <brobecker@gnat.com>

        * gdb.base/foll-fork.exp: Update the expected output for
        "help set follow-fork-mode", to match a change that was made
        to the help of this variable on 2004-01-13.

Checked in as obvious.

I don't really know the purpose of this test. If, as the comment
suggests:

        # Verify that help is available for "set follow-fork-mode".

Then I would recommend that we don't need to check for the entire
help text word for word. I would simply check for the first line,
or even maybe just check that we receive something, anything, before
we get the prompt back.

(BTW, Is it really necessary to use send_gdb + gdb_expect? Could gdb_test
work just as well?)

-- 
Joel

[-- Attachment #2: foll-fork.exp.diff --]
[-- Type: text/plain, Size: 890 bytes --]

Index: foll-fork.exp
===================================================================
RCS file: /cvs/src/src/gdb/testsuite/gdb.base/foll-fork.exp,v
retrieving revision 1.2
diff -u -p -r1.2 foll-fork.exp
--- foll-fork.exp	6 Mar 2001 08:21:50 -0000	1.2
+++ foll-fork.exp	4 Apr 2004 20:24:51 -0000
@@ -280,8 +280,7 @@ proc do_fork_tests {} {
 A fork or vfork creates a new process.  follow-fork-mode can be:.*
 .*parent  - the original process is debugged after a fork.*
 .*child   - the new process is debugged after a fork.*
-.*ask     - the debugger will ask for one of the above choices.*
-For \"parent\" or \"child\", the unfollowed process will run free..*
+The unfollowed process will continue to run..*
 By default, the debugger will follow the parent process..*$gdb_prompt $"\
                       { pass "help set follow" }
       -re "$gdb_prompt $" { fail "help set follow" }

Re: [patch/testsuite] Fix expected output for "help set follow-fork-mode"

[Daniel Jacobowitz]

On Sun, Apr 04, 2004 at 01:35:39PM -0700, Joel Brobecker wrote:
> There has been a recent change to infrun (made on 2004-01-13) that
> removed the choice "ask" to the possible values for "follow-fork-mode".
> The help text for that variable was then modified, but the test
> verifying that the help for this command exists has not been modified.
> 
> This patch updates the expected output to match the new help text.
> 
> 2004-04-04  Joel Brobecker  <brobecker@gnat.com>
> 
>         * gdb.base/foll-fork.exp: Update the expected output for
>         "help set follow-fork-mode", to match a change that was made
>         to the help of this variable on 2004-01-13.
> 
> Checked in as obvious.

Thanks.  I keep meaning to fix up this test to run, when possible, on
GNU/Linux; but there's some trickiness involved and I've failed to do
it adequately a few times now.

> 
> I don't really know the purpose of this test. If, as the comment
> suggests:
> 
>         # Verify that help is available for "set follow-fork-mode".
> 
> Then I would recommend that we don't need to check for the entire
> help text word for word. I would simply check for the first line,
> or even maybe just check that we receive something, anything, before
> we get the prompt back.

I agree, personally.

> (BTW, Is it really necessary to use send_gdb + gdb_expect? Could gdb_test
> work just as well?)

Yes, feel free.

-- 
Daniel Jacobowitz
MontaVista Software                         Debian GNU/Linux Developer