Re: PATCH: Debugger Redesign

Hi Dave,

Please find the patch to resolve few of the problems with the debuggers and
introduced during redesigning it.

On Tue, Apr 30, 2013 at 12:56 AM, Ashesh Vashi <
ashesh(dot)vashi(at)enterprisedb(dot)com> wrote:

> On Tue, Apr 30, 2013 at 12:48 AM, Dave Page <dpage(at)pgadmin(dot)org> wrote:
>
>> On Mon, Apr 29, 2013 at 8:10 PM, Ashesh Vashi <
>> ashesh(dot)vashi(at)enterprisedb(dot)com> wrote:
>>
>>> On Mon, Apr 29, 2013 at 10:30 PM, Dave Page <dpage(at)pgadmin(dot)org> wrote:
>>>
>>>> Thanks Ashesh - committed (with a few minor tweaks, mostly to
>>>> messages)! Good work :-)
>>>>
>>> Thanks
>>>
>>>>
>>>> I think we have the following issues to discuss and possibly resolve
>>>> during beta:
>>>>
>>>> 1) I find the progress dialogue annoying, and as discussed previously
>>>> think we should replace it with a progress indicator on the status bar.
>>>>
>>> Yeah - I am aware of that.
>>> I will work on it once I get some time.
>>>
>>>>
>>>> 2) I don't see RAISE DEBUG messages in the server message pane. Perhaps
>>>> we should explicitly set client_min_messages so the user can see their
>>>> debug messages?
>>>>
>>> Yup
>>>
>> Done.

>
>>>> 3) The stack pane shows functions as
>>>> "foo(integer)(param_name=1)@<line>". I think perhaps we should change it
>>>> to: "foo(integer param_name=1)@<line>"
>>>>
>>> +1
>>>
>>
>> OK, please make those tweaks when you work on the progress indicator (or
>> before, if you get a chance).
>>
>>
>>>
>>>> 4) When in-process debugging, if the calling process is terminated we
>>>> get a connection lost error (which I special-cased in the code, as it
>>>> displays an unsightly message from the plugin by default). Instead, I think
>>>> we should just automatically start listening for another calling process.
>>>>
>>> The problem is: this error message is coming from backend (debugger
>>> plugin).
>>> And hence, it may require us to make changes in the plugin.
>>>
>>
>> Shouldn't do - don't we just need to reset the breakpoint and wait for
>> it to be hit again? FYI, the old code does seem to do it successfully,
>> though it doesn't do a good job of ensuring the dialogues are re-displayed.
>>
> It was ignore any error in the debugging connection (even if it is a
> connection error or something else).
> I will come up with some other way to figure out the same. (I will also
> take care of this, when doing the modification for the progress bar. Hope -
> that's ok with everybody).
>
Done.
Now - we check for the session (backend) on which the target was running,
if it does not exist, we wait for another session to invoke the target.

Also - resolved couple of other problem related to the debugger:
- On cancellation of progress dialog, it was not closing the debugger.
- In direct debugging, the target was invoked without taking care about
quotes.

This patch also includes couple of other issues introduced by the earlier
patch.
1. Showing the messages multiple times in the message window (Query window)
2. Cancellation of query wasn't working on Query window.

Still working on the progress-bar.

>
>> Thanks!
>>
>> --
>> Dave Page
>> Blog: http://pgsnake.blogspot.com
>> Twitter: @pgsnake
>>
>> EnterpriseDB UK: http://www.enterprisedb.com
>> The Enterprise PostgreSQL Company
>>
>
>
>
> --
> --
>
> Thanks & Regards,
>
> Ashesh Vashi
> EnterpriseDB INDIA: Enterprise PostgreSQL Company<http://www.enterprisedb.com>
>
>
>
> *http://www.linkedin.com/in/asheshvashi*<http://www.linkedin.com/in/asheshvashi>
>
--

Thanks & Regards,

Ashesh Vashi
EnterpriseDB INDIA: Enterprise PostgreSQL Company<http://www.enterprisedb.com>

*http://www.linkedin.com/in/asheshvashi*<http://www.linkedin.com/in/asheshvashi>