Re: [PATCH][TEST-CASE][DOC] Implementation of pipe to pass GDB's command output to a shell command.

[Abhijit Halder <abhijit.k.halder@gmail.com>]

On Mon, Sep 5, 2011 at 5:23 PM, Abhijit Halder
<abhijit.k.halder@gmail.com> wrote:
> On Sun, Sep 4, 2011 at 8:35 PM, Abhijit Halder
> <abhijit.k.halder@gmail.com> wrote:
>> On Sun, Sep 4, 2011 at 8:22 PM, Jan Kratochvil
>> <jan.kratochvil@redhat.com> wrote:
>>> On Sat, 03 Sep 2011 11:21:19 +0200, Abhijit Halder wrote:
>>>> >> +    regsub -all {\$[0-9]+} $fdata {} pattern
>>>> >> +    if ![string match $pattern " = 120 'x'\n"] then {
>>>> >
>>>> > Instead of regsub + string match just:
>>>> >    if ![string match "* = 120 'x'\n" $fdata] then {
>>>> >
>>>> The intention was to eliminate only numeric patterns here.
>>>> Instead if I used "if ![string match "* = 120 'x'\n" $fdata] then {"
>>>> following pattern will also match:
>>>> $junk = 120 'x'
>>>> that I did not want.
>>>
>>> It is currently a common practice to match this kind of pattern just from
>>> ` = ' upwards such as:
>>> gdb_test "p callme ()" " = 42"
>>> etc. everywhere in the testsuite.
>>>
>>> But if you want to match the initial $number part such as in
>>> gdb_test "print test1.test" "\\$\[0-9\]* = true" "simple object, const bool"
>>> then it is more simple by:
>>>    if ![regexp {^\$[0-9]+ = 120 'x'\n$} $fdata] then {
>>>
>> Got it. Correcting the same in the next patch.
>>>
>>> Thanks,
>>> Jan
>>>
>>
>
>
>
>
> *********************************************************************************
> Following invalid c-statements are getting parsed successfully:
> 1. int(*)
> 2. int*(*)
> 3. int(*)()()
> 4. int*(*)()()
> 5. int(*)()()()
> 6. int*(*)()()()
> 7. int*(*)()[]()
> ....etc.
>
> Following valid c-statement failed to be parsed:
> 1. int(**)()
>
> Probably a few more may appear.........
> I am putting all this as the test-cases.
>
> Thanks,
> Abhijit Halder
>
>
> 2. int*(**)()
> 3.
>

Sorry for using this thread to post this. Anyway the above statements
were not correct. The c-statement told as "invalid" are actually
valid. In gdb some of then are just not being displayed properly. Once
again sorry for this confusion.


Regards,
Abhijit Halder