alarm-clock ftbfs in maverick

Bug #626184 reported by Bhavani Shankar
10
This bug affects 1 person
Affects Status Importance Assigned to Milestone
alarm-clock (Ubuntu)
Fix Released
Medium
Unassigned
Tags: patch
Bhavani Shankar (bhavi)
Changed in alarm-clock (Ubuntu):
assignee: nobody → Bhavani Shankar (bhavi)
status: New → In Progress
Revision history for this message
Bhavani Shankar (bhavi) wrote :
Changed in alarm-clock (Ubuntu):
status: In Progress → Confirmed
assignee: Bhavani Shankar (bhavi) → nobody
Revision history for this message
Stefano Rivera (stefanor) wrote :

Does that patch really work? It seems wrong to change a function prototype in a header but not the function and its callers.

When adding quilt patches, it really helps the reviewers (and future developers) if you add nice patch headers: http://dep.debian.net/deps/dep3/

You will forward these patches upstream?

Revision history for this message
Bhavani Shankar (bhavi) wrote :

Hello stefano,

sorry for the late reply,

afaik this error arises when the namespace already has the variable under view defined. In essence it conflicts with alarm function declaration in unistd.h here

from the buildlog:

//usr/include/unistd.h:429: note: previous declaration of 'alarm' was here

so it arises an ambiguity wrt declarations when compiling in the compiler

so as a workaround you rename the function in view such that ambiguity doesn't arise

and it works as it builds fine on my PPA

https://edge.launchpad.net/~bhavi/+archive/crickinfo/+build/1939060

https://edge.launchpad.net/~bhavi/+archive/crickinfo/+build/1939061

(I named it as -2 in my PPA as i fixed the mime bug too)

Here is the attached patch with the required dep 3 headers

regards

tags: added: patch
Revision history for this message
Stefano Rivera (stefanor) wrote :

I understand what you are doing (renaming). But I don't understand why you rename the prototype of run_alarm but not run_alarm itself or any of the callers. I also don't understand why run_alarm_struct is a good name for run_alarm.

As to the alarm GKeyFile, I can't see anywhere where it's used.

Btw, a patch labelled "fix-ftbfs" is pretty misleading.

Revision history for this message
Bhavani Shankar (bhavi) wrote : Re: [Bug 626184] Re: alarm-clock ftbfs in maverick

Sure stefano, but as I am down with a sprained neck please feel free
to do the required modification and upload the same

regards

On Wed, Sep 8, 2010 at 5:12 AM, Stefano Rivera <email address hidden> wrote:
> I understand what you are doing (renaming). But I don't understand why
> you rename the prototype of run_alarm but not run_alarm itself or any of
> the callers. I also don't understand why run_alarm_struct is a good name
> for run_alarm.
>
> As to the alarm GKeyFile, I can't see anywhere where it's used.
>
> Btw, a patch labelled "fix-ftbfs" is pretty misleading.
>
> --
> alarm-clock ftbfs in maverick
> https://bugs.launchpad.net/bugs/626184
> You received this bug notification because you are a direct subscriber
> of the bug.
>
> Status in “alarm-clock” package in Ubuntu: Confirmed
>
> Bug description:
> Binary package hint: alarm-clock
>
> Buildlogs:
>
> http://people.ubuntuwire.org/~lucas/ubuntu-nbs/32/alarm-clock_1.2.5-1_lubuntu32.buildlog
>
> http://people.ubuntuwire.org/~lucas/ubuntu-nbs/64/alarm-clock_1.2.5-1_lubuntu64.buildlog
>
> To unsubscribe from this bug, go to:
> https://bugs.launchpad.net/ubuntu/+source/alarm-clock/+bug/626184/+subscribe
>

--

Bhavani Shankar.R
https://launchpad.net/~bhavi, a proud ubuntu community  member.
What matters in life is application of mind!,
It makes great sense to have some common sense..!

Revision history for this message
Bhavani Shankar (bhavi) wrote :

Hello stefano,

here is the updated patch

regards

Revision history for this message
Bhavani Shankar (bhavi) wrote :

Hello stefano,

here is the updated patch

regards

Revision history for this message
Bhavani Shankar (bhavi) wrote :

Sorry for the double post!

Revision history for this message
Stefano Rivera (stefanor) wrote :

OK, I'm happier, but I still don't get why we are renaming a variable that's never used. Why not just comment it out?

Revision history for this message
Bhavani Shankar (bhavi) wrote :

Alright stefano,

Its done :)

attaching the modified diff

regards

Revision history for this message
Stefano Rivera (stefanor) wrote :

Right. The changelog entry is still a little misleading. I'd say something like:

    + comment out unused "alarm" variable in src/alarm_runner.h which clashes
      with the alarm() function declared in unistd.h (LP: #626184)

Is MimeType application/octet-stream really correct? Shouldn't we just delete that line? With it, my desktop tries to open random binary files with alarm-clock with is totally incorrect.

Revision history for this message
Bhavani Shankar (bhavi) wrote :

Alright stefano,

done

attaching revised diff

regards

Revision history for this message
Stefano Rivera (stefanor) wrote :

Yay, thanks for being patient so we could produce a good quality fix.

Bug auto-closing is currently broken, doing it manually:

alarm-clock (1.2.5-1ubuntu1) maverick; urgency=low

  * debian/patches/fix-desktop-file.patch:
    + remove MIME type declaration to avoid alarm-clock associating with
      text files (LP: #502871)
  * debian/patches/fix-ftbfs-with-gcc4.5.patch:
    + comment out unused "alarm" variable in src/alarm_runner.h which clashes
      with the alarm() function declared in unistd.h (LP: #626184)

Thank you for your contribution to Ubuntu.

Changed in alarm-clock (Ubuntu):
importance: Undecided → Medium
status: Confirmed → Fix Released
To post a comment you must log in.
This report contains Public information  
Everyone can see this information.

Other bug subscribers

Remote bug watches

Bug watches keep track of this bug in other bug trackers.