Re: Semantic silent mode

Perfect, thanks. I'll get back to you if I ever find anything specific.

-----Original Message-----
From: Eric Ludlam [mailto:[email protected]] 
Sent: November-03-15 7:17 AM
To: Jan Goh <[email protected]>; [email protected]
Subject: Re: [cedet-semantic] Semantic silent mode

The first step would be to find the offending file.  Once you notice the problem, you can run:

M-x toggle-debug-on-error RET
M-x revert-buffer

and hopefully it will drop you into the offending bit of code.  That should point at the next step in debugging.

If you happen to identify a specific file it is breaking with from the stack, you can sometimes take that file, chop it in half, and whittle away the code till you are left with the offending function or line which may be sharable once you hack out the body.

Eric


On 11/02/2015 10:09 AM, Jan Goh wrote:
> Oops, sorry, meant to reply to this earlier.
>
> That said, I don't think I have any code that I can offer--it's all proprietary, so I'd get in trouble for sharing it. Is there any other way I can track down what the problem is so I can send it along?
>
> -----Original Message-----
> From: Eric Ludlam [mailto:[email protected]]
> Sent: October-18-15 11:46 AM
> To: Jan Goh <[email protected]>; 
> [email protected]
> Subject: Re: [cedet-semantic] Semantic silent mode
>
> On 10/16/2015 11:15 AM, Jan Goh wrote:
>> Is there a way to make Semantic silent? I want NO update messages of 
>> any kind. I don't want errors or anything. The messages interrupt 
>> other things that are going on, like when I have an auto-revert 
>> buffer working, semantic will complain about something and the 
>> auto-reverts stop until I manually revert. But in the end, most of 
>> the messages that pop up are simply things I don't care about.
>>
>> Additionally, is there anything that might automatically turn on 
>> global-semantic-idle-scheduler-mode? I keep turning it off, but it 
>> turns back on (and then complains about end-of-buffer problems that, 
>> again, I don't care about). Though I suppose if I could quash the 
>> messages, it wouldn't be as important that I'm able to turn off the 
>> idle-scheduler and rely on it staying off.
>
> It sounds like there is one message breaking auto-revert that is most annoying.  Auto-revert shouldn't break.  Do you have some repro-code that causes this to happen?
>
> Several of the messages use Emacs' built in progress reporting.   I
> assume that is supposed to restore any previous message.  The old version from CEDET used to do that.
>
> Thanks
> Eric
>

------------------------------------------------------------------------------