> Something here is broken for me (for transcode).. Sending SIG_ANYTHING to
> the process leader after making the recomended changes... (in util.popen3)
> doesn't kill all childs.. I can only think now I must hack my way around
> this...
>
> Bruno, in cd_burn, cleanup() calls kill on -pid not p
> -Original Message-
> From: [EMAIL PROTECTED]
> [mailto:[EMAIL PROTECTED] Behalf Of Bruno
> Tavares
> Sent: Monday, 16 August 2004 8:33 PM
> To: [EMAIL PROTECTED]
> Subject: Re: [Freevo-devel] Re: utils.popen3 and mmpython
>
>
> On Mon, 2004-08-16 at 12:5
> Now you got me. What are process groups? My knowledge says: when a
> child dies, the kernel sends SIGCHILD to the parent. The parent than
> should 'wait' for it to remove the zombie process. If a process dies,
> all children are now children of process 1 (init). So how can you get
> information w
On Mon, 2004-08-16 at 12:50, Dirk Meyer wrote:
> Now you got me. What are process groups? My knowledge says: when a
> child dies, the kernel sends SIGCHILD to the parent. The parent than
> should 'wait' for it to remove the zombie process. If a process dies,
> all children are now children of proc
Bruno Tavares wrote:
> Mick wrote:
>
>>>But how should Freevo be able to control a child of a child?
>>>
>>>
> Freevo just has to worry on killing all the spwaned childs and it's
> sub-childs. Since most of the helper processes do not notify anything
> about spwaning sub-childs the only way i see
Mick wrote:
But how should Freevo be able to control a child of a child?
Freevo just has to worry on killing all the spwaned childs and it's
sub-childs. Since most of the helper processes do not notify anything
about spwaning sub-childs the only way i see to make this work is to
assign a p
> But how should Freevo be able to control a child of a child?
>
I wish I knew.. This could more easily be broken into how can any python
know the pid of a child of a child.
I've been searching through python docs and mailing lists and I can't find a
solution to this. Any ideas are welcome.. The
Bruno Tavares wrote:
> On Wed, 2004-08-11 at 15:50, Dirk Meyer wrote:
>
>> It's in plugins/idlebar/__init__.py. The plugin is called something
>> with cd.
>>
>>
>> Dischi
>
> Found, but from what i can understand, the icon changes according to
> media types reported in config.REMOVABLE_MEDIA. The
Bruno Tavares wrote:
> On Wed, 2004-08-11 at 14:10, Dirk Meyer wrote:
>
>>
>> No, what is the problem you have now? You have full control over the
>> child.
> Currently none, just think ahead. It's probable that a child spawns a
> child and with the current code you loose control of that.
> Since
On Wed, 2004-08-11 at 15:50, Dirk Meyer wrote:
> It's in plugins/idlebar/__init__.py. The plugin is called something
> with cd.
>
>
> Dischi
Found, but from what i can understand, the icon changes according to
media types reported in config.REMOVABLE_MEDIA. The cd part of the
plugins don't list
Bruno Tavares wrote:
> On Wed, 2004-08-11 at 02:34, Mick wrote:
>
>> We should be able to do this by sending burn start and stop events, and
>> modifying the plugin that handles that image to also diaply something
>> when a burn is active..
>
> Ok, do you know what is the plugin that manages/handle
On Wed, 2004-08-11 at 02:34, Mick wrote:
> We should be able to do this by sending burn start and stop events, and
> modifying the plugin that handles that image to also diaply something
> when a burn is active..
Ok, do you know what is the plugin that manages/handles the cd image?
---
On Wed, 2004-08-11 at 14:10, Dirk Meyer wrote:
>
> No, what is the problem you have now? You have full control over the
> child.
Currently none, just think ahead. It's probable that a child spawns a
child and with the current code you loose control of that.
Since it doesn't break anything i think
"Mick" wrote:
>> Bad example because the recordserver works different. You can use
>> rc.post_event for sending events (grep the source for it). But you
>> still need someone catching it and do something.
>>
> Almost off topic.. But can an event be posted with a variable? Like can
> CAPTURE_START
Bruno Tavares wrote:
> On Wed, 2004-08-11 at 09:20, Dirk Meyer wrote:
>
>>
>> What do you need from Freevo for that?
>>
>
> Not sure if i understand the question? what do i need to add to freevo
> to solve this?
> is that the question?
No, what is the problem you have now? You have full control
> Bad example because the recordserver works different. You can use
> rc.post_event for sending events (grep the source for it). But you
> still need someone catching it and do something.
>
Almost off topic.. But can an event be posted with a variable? Like can
CAPTURE_START / STOP event be sent w
On Wed, 2004-08-11 at 09:20, Dirk Meyer wrote:
>
> What do you need from Freevo for that?
>
Not sure if i understand the question? what do i need to add to freevo
to solve this?
is that the question?
>
> Dischi
Bruno
---
SF.Net email is s
Mick wrote:
> Check out tv/generic_record.py for an example on how to send events (not
> sure I understand it, but it sends them..
Bad example because the recordserver works different. You can use
rc.post_event for sending events (grep the source for it). But you
still need someone catching it and
Bruno Tavares wrote:
> On Tue, 2004-08-10 at 15:44, Dirk Meyer wrote:
>
>> Bad hack:
>> command = 'foo bar something'
>> popen3(command.split(' '))
>>
>> Better: (because filenames can conatin spaces)
>> command = [ 'my_prog', 'param1', 'a param with spaces' ]
>> popen3(command)
>>
>
> Got it, th
On Tue, 2004-08-10 at 23:10, Bruno Tavares wrote:
> On Tue, 2004-08-10 at 14:54, Mick wrote:
>
> > In the future it would be great if mmpython could identify blank media
> > somehow.. That way we can intigate some checks in cd_burn down the track..
> > Of course I have no idea how to do this ;-)
On Tue, 2004-08-10 at 15:44, Dirk Meyer wrote:
> Bad hack:
> command = 'foo bar something'
> popen3(command.split(' '))
>
> Better: (because filenames can conatin spaces)
> command = [ 'my_prog', 'param1', 'a param with spaces' ]
> popen3(command)
>
Got it, thanks.
Still wont solve the "childs
On Tue, 2004-08-10 at 14:54, Mick wrote:
> In the future it would be great if mmpython could identify blank media
> somehow.. That way we can intigate some checks in cd_burn down the track..
> Of course I have no idea how to do this ;-)
Neither do i, but another cool feature was to mmpython to de
Bruno Tavares wrote:
> On Tue, 2004-08-10 at 14:14, Dirk Meyer wrote:
>> Bruno Tavares wrote:
>> > I've two questions:
>> >
>> > first; and i know Dischi don't like threads, is it possible to patch
>> > utils.popen3.Popen4 to make it launch all the childs under a group
>> > process id.
>> > This be
> But the problem persists if the child_app launches another command :/
>
I think this is the biggest issue.. My record app lauches 2 other apps to
pipe stuff too.. Its the second app that grabs the v4l device so when
recodings start really close to sopt times the second app doesn't get killed
so
> > second, when i started to devel the cd_burn plugin i noticed that
> > mmpython crashes when analysing a blank disc in the drive.
> > Can it be fixed?
>
> I thought I made it working some time ago. But I saw a patch for this
> some days ago, I will add it to mmpython.
>
In the future it would be
On Tue, 2004-08-10 at 14:14, Dirk Meyer wrote:
> Bruno Tavares wrote:
> > I've two questions:
> >
> > first; and i know Dischi don't like threads, is it possible to patch
> > utils.popen3.Popen4 to make it launch all the childs under a group
> > process id.
> > This because, since Popen4 launches t
Bruno Tavares wrote:
> I've two questions:
>
> first; and i know Dischi don't like threads, is it possible to patch
> utils.popen3.Popen4 to make it launch all the childs under a group
> process id.
> This because, since Popen4 launches the command using sh, you loose
> control of all the child pro
27 matches
Mail list logo
