Bug #88995 for File-Tail-Dir: missing private method

Fri Sep 27 01:21:37 2013 spragues [...] gmail.com - Ticket created

Subject:	missing private method
Date:	Thu, 26 Sep 2013 22:21:04 -0700
To:	bug-File-Tail-Dir [...] rt.cpan.org
From:	Stephen Sprague <spragues [...] gmail.com>

Hi Jon, i'm really curious if this ever finds its way to you. been using File::Tail::Dir for a couple of days now and i really like it. thanks so much for making it public! one thing. it seems when i set $tailer->runnning(0) the module will eventually end its loop and call _stop_watchdog. it appears that method is missing. figured you might want to know about it. i've put an eval {} around the watch_files() and just exit when the error is caught. I'm thinking there isn't much of any cleanup to do since the state was saved prior to the _stop_watchdog call. anyway. that's it. and thanks again for the module. cheers, stephen sprague spragues@gmail.com

Fri Sep 27 02:05:33 2013 JJSCHUTZ [...] cpan.org - Correspondence added

Hi Stephen, Funny how such issues can live undetected for so long. Fixed in release 0.16. Regards, Jon Schutz On Fri Sep 27 01:21:37 2013, spragues@gmail.com wrote: Show quoted text

> Hi Jon, > > i'm really curious if this ever finds its way to you. been using > File::Tail::Dir for a couple of days now and i really like it. thanks so > much for making it public! > > one thing. it seems when i set $tailer->runnning(0) the module will > eventually end its loop and call _stop_watchdog. it appears that method is > missing. figured you might want to know about it. > > i've put an eval {} around the watch_files() and just exit when the error > is caught. I'm thinking there isn't much of any cleanup to do since the > state was saved prior to the _stop_watchdog call. > > anyway. that's it. and thanks again for the module. > > cheers, > stephen sprague > spragues@gmail.com

Fri Sep 27 02:05:33 2013 The RT System itself - Status changed from 'new' to 'open'

Fri Sep 27 02:05:33 2013 JJSCHUTZ [...] cpan.org - Status changed from 'open' to 'resolved'

Fri Sep 27 11:52:46 2013 spragues [...] gmail.com - Correspondence added

Subject:	Re: [rt.cpan.org #88995] missing private method
Date:	Fri, 27 Sep 2013 08:52:12 -0700
To:	bug-file-tail-dir <bug-File-Tail-Dir [...] rt.cpan.org>
From:	Stephen Sprague <spragues [...] gmail.com>

Thanks so much Jon. The way you're storing state is really a bang up job and pretty hard-core. Its looks like you meant business when you started off on this! :) I'm using it to feed Kafka (LinkedIn's opensource pub/sub app) and so far things look great! Thanks again. Cheers, stephen sprague spragues@gmail.com On Thu, Sep 26, 2013 at 11:05 PM, Jon Schutz via RT < bug-File-Tail-Dir@rt.cpan.org> wrote: Show quoted text

> <URL: https://rt.cpan.org/Ticket/Display.html?id=88995 > > > Hi Stephen, > > Funny how such issues can live undetected for so long. Fixed in release > 0.16. > > Regards, > > Jon Schutz > > On Fri Sep 27 01:21:37 2013, spragues@gmail.com wrote:

> > Hi Jon, > > > > i'm really curious if this ever finds its way to you. been using > > File::Tail::Dir for a couple of days now and i really like it. thanks so > > much for making it public! > > > > one thing. it seems when i set $tailer->runnning(0) the module will > > eventually end its loop and call _stop_watchdog. it appears that method

> is

> > missing. figured you might want to know about it. > > > > i've put an eval {} around the watch_files() and just exit when the error > > is caught. I'm thinking there isn't much of any cleanup to do since the > > state was saved prior to the _stop_watchdog call. > > > > anyway. that's it. and thanks again for the module. > > > > cheers, > > stephen sprague > > spragues@gmail.com

> > > >

Mon Sep 30 17:12:12 2013 spragues [...] gmail.com - Correspondence added

Subject:	Re: [rt.cpan.org #88995] missing private method
Date:	Mon, 30 Sep 2013 14:11:36 -0700
To:	bug-file-tail-dir <bug-File-Tail-Dir [...] rt.cpan.org>
From:	Stephen Sprague <spragues [...] gmail.com>

Hi Jon, This isn't a bug but it would still be great to get your feedback. Here's my current dilemma. How can i exit the tailing process, watch_files(), programmatically from my code? I know about your running(0) method but its kinda hard to call it when my process_lines method isn't invoked. The simple case is an empty file. My callback is never called so i can't do anything to stop your module's event loop - from what i can tell anyway. In my case, if i don't get any records for 1 minute i can assume no more records will ever be written to the file i'm tailing - so i'd like to exit gracefully. I was thinking maybe a heartbeat type of callback could be employed in your module that you could call when in a wait mode. Then execution comes back to my code. just a thought. Anyhoo, what would you recommend in this case? thanks, stephen sprague spragues@gmail.com On Fri, Sep 27, 2013 at 8:52 AM, Stephen Sprague <spragues@gmail.com> wrote: Show quoted text

> Thanks so much Jon. The way you're storing state is really a bang up job > and pretty hard-core. Its looks like you meant business when you started > off on this! :) > > I'm using it to feed Kafka (LinkedIn's opensource pub/sub app) and so far > things look great! > > Thanks again. > > Cheers, > stephen sprague > spragues@gmail.com > > > On Thu, Sep 26, 2013 at 11:05 PM, Jon Schutz via RT < > bug-File-Tail-Dir@rt.cpan.org> wrote: >

>> <URL: https://rt.cpan.org/Ticket/Display.html?id=88995 > >> >> Hi Stephen, >> >> Funny how such issues can live undetected for so long. Fixed in release >> 0.16. >> >> Regards, >> >> Jon Schutz >> >> On Fri Sep 27 01:21:37 2013, spragues@gmail.com wrote:

>> > Hi Jon, >> > >> > i'm really curious if this ever finds its way to you. been using >> > File::Tail::Dir for a couple of days now and i really like it. thanks so >> > much for making it public! >> > >> > one thing. it seems when i set $tailer->runnning(0) the module will >> > eventually end its loop and call _stop_watchdog. it appears that

>> method is

>> > missing. figured you might want to know about it. >> > >> > i've put an eval {} around the watch_files() and just exit when the

>> error

>> > is caught. I'm thinking there isn't much of any cleanup to do since the >> > state was saved prior to the _stop_watchdog call. >> > >> > anyway. that's it. and thanks again for the module. >> > >> > cheers, >> > stephen sprague >> > spragues@gmail.com

>> >> >> >>

Mon Sep 30 20:28:23 2013 jon.schutz [...] youramigo.com - Correspondence added

Subject:	Re: [rt.cpan.org #88995] missing private method
Date:	Tue, 01 Oct 2013 09:57:56 +0930
To:	bug-File-Tail-Dir [...] rt.cpan.org
From:	Jon Schutz <jon.schutz [...] youramigo.com>

Hi Stephen, The built-in watchdog will cater for this situation. If you set autostate_delay to 60 and enable autostate, the watchdog will cause the watch_files loop to be interrupted at least every 60 seconds and running(0) will be recognised. It will also save state. Arguably the autostate and loop checking functions should be separated - but that's the way it works at the moment. Regards, -- Jon Schutz CTO, YourAmigo Ltd 53 Gilbert St Adelaide SA 5000 Ph: +61 8 82119211 Fax: +61 8 8211 6356 http://www.youramigo.com On 01/10/13 06:42, Stephen Sprague via RT wrote: Show quoted text

> Queue: File-Tail-Dir > Ticket <URL: https://rt.cpan.org/Ticket/Display.html?id=88995 > > > Hi Jon, > This isn't a bug but it would still be great to get your feedback. > > Here's my current dilemma. How can i exit the tailing process, > watch_files(), programmatically from my code? I know about your running(0) > method but its kinda hard to call it when my process_lines method isn't > invoked. The simple case is an empty file. My callback is never called so > i can't do anything to stop your module's event loop - from what i can tell > anyway. > > In my case, if i don't get any records for 1 minute i can assume no more > records will ever be written to the file i'm tailing - so i'd like to exit > gracefully. I was thinking maybe a heartbeat type of callback could be > employed in your module that you could call when in a wait mode. Then > execution comes back to my code. just a thought. > > Anyhoo, what would you recommend in this case? > > thanks, > stephen sprague > spragues@gmail.com > > > On Fri, Sep 27, 2013 at 8:52 AM, Stephen Sprague <spragues@gmail.com> wrote: >

>> Thanks so much Jon. The way you're storing state is really a bang up job >> and pretty hard-core. Its looks like you meant business when you started >> off on this! :) >> >> I'm using it to feed Kafka (LinkedIn's opensource pub/sub app) and so far >> things look great! >> >> Thanks again. >> >> Cheers, >> stephen sprague >> spragues@gmail.com >> >> >> On Thu, Sep 26, 2013 at 11:05 PM, Jon Schutz via RT < >> bug-File-Tail-Dir@rt.cpan.org> wrote: >>

>>> <URL: https://rt.cpan.org/Ticket/Display.html?id=88995 > >>> >>> Hi Stephen, >>> >>> Funny how such issues can live undetected for so long. Fixed in release >>> 0.16. >>> >>> Regards, >>> >>> Jon Schutz >>> >>> On Fri Sep 27 01:21:37 2013, spragues@gmail.com wrote:

>>>> Hi Jon, >>>> >>>> i'm really curious if this ever finds its way to you. been using >>>> File::Tail::Dir for a couple of days now and i really like it. thanks so >>>> much for making it public! >>>> >>>> one thing. it seems when i set $tailer->runnning(0) the module will >>>> eventually end its loop and call _stop_watchdog. it appears that

>>> method is

>>>> missing. figured you might want to know about it. >>>> >>>> i've put an eval {} around the watch_files() and just exit when the

>>> error

>>>> is caught. I'm thinking there isn't much of any cleanup to do since the >>>> state was saved prior to the _stop_watchdog call. >>>> >>>> anyway. that's it. and thanks again for the module. >>>> >>>> cheers, >>>> stephen sprague >>>> spragues@gmail.com

>>> >>> >>>