Re: [PATCH] Don't truncate the history file when history size is unlimited

[Pedro Alves <palves at redhat dot com>]

On 06/15/2015 05:00 PM, Patrick Palka wrote:
> On Mon, Jun 15, 2015 at 11:25 AM, Pedro Alves <palves@redhat.com> wrote:
>> On 06/09/2015 08:27 PM, Patrick Palka wrote:
>>> We still do not handle "set history size unlimited" correctly.  In
>>> particular, after writing to the history file, we truncate the history
>>> even if it is unlimited.
>>
>> Whoops.
>>
>>>
>>> This patch makes sure that we do not call history_truncate_file() if the
>>> history is not stifled (i.e. if it's unlimited).  This bug causes the
>>> history file to be truncated to zero on exit when one has "set history
>>> size unlimited" in their gdbinit file.  Although this code exists in GDB
>>> 7.8 it is masked by a pre-existing bug that's been only fixed in GDB 7.9
>>> (PR gdb/17820).
>>>
>>> I tried to make a test to check that the history does not get truncated
>>> on exit when the history size is unlimited, but I could not get the test
>>> to work properly.
>>
>> How so?
> 
> I could not observe that the history file gets truncated with this
> patch reverted.  Maybe I simply have forgotten to revert the patch or
> something stupid like that.  I'll take another stab at it.
> 
>>
>>> Also I could not figure out a good way to create a
>>> temporary file (to act as the history file) in tcl versions earlier than
>>> 8.6.  I am not sure if it's worth the effort to add a test.
>>
>> Not sure I follow.  Why do you need a temporary file?
>> You can leave the file in the build/test dir.  In fact, that's
>> encouraged, to make it easier to debug the test on failures.
>> You can instead delete a stale file from a previous run at the
>> start of the (new) run.
> 
> Good point... What's the right way to refer to the
> $buildroot/gdb/testsuite/gdb.base directory?

Use [standard_output_file $some_file].

E.g., take a look at gdb.trace/tfile.exp, gdb.xml/tdesc-arch.exp,
gdb.xml/tdesc-regs.exp and others.

Thanks,
Pedro Alves