Bug #85500 for Parse-Syslog-Line: datetime_raw became date

Tue May 21 13:57:17 2013 abraxxa [...] cpan.org - Ticket created

Subject:

datetime_raw became date_raw

It seems the patch you've applied has changed the contents of the returned hashref. It now contains a date_raw key instead of the documented datetime_raw. Please either fix the code, which is preferred because Message::Passing::Input::Syslog depends on it and currently fails its tests, or the synopsis docs that show the returned hashref format. Thanks!

Tue May 21 14:22:30 2013 brad [...] divisionbyzero.net - Correspondence added

On Tue May 21 13:57:17 2013, ABRAXXA wrote: Show quoted text

> It seems the patch you've applied has changed the contents of the > returned hashref. It now contains a date_raw key instead of the > documented datetime_raw. > Please either fix the code, which is preferred because > Message::Passing::Input::Syslog depends on it and currently fails its > tests, or the synopsis docs that show the returned hashref format. > Thanks!

Good catch, didn't see that. Fixed in 1.4, just pushed to the CPAN.

Tue May 21 14:22:30 2013 The RT System itself - Status changed from 'new' to 'open'

Tue May 21 14:23:13 2013 brad [...] divisionbyzero.net - TimeWorked changed from (no value) to '15'

Tue May 21 14:23:13 2013 brad [...] divisionbyzero.net - Status changed from 'open' to 'patched'

Tue May 21 14:23:14 2013 brad [...] divisionbyzero.net - Taken

Tue May 21 14:23:14 2013 brad [...] divisionbyzero.net - Severity Important added

Tue May 21 14:33:41 2013 abraxxa [...] cpan.org - Correspondence added

Thanks for the quick fix! Does it make sense to have the same value twice in the hash? I'm asking because I'm using it in a performance sensitive environment and don't want it unnecessarily degrade.

Tue May 21 14:46:20 2013 brad [...] divisionbyzero.net - Correspondence added

On Tue May 21 14:33:41 2013, ABRAXXA wrote: Show quoted text

> Thanks for the quick fix! > Does it make sense to have the same value twice in the hash? > I'm asking because I'm using it in a performance sensitive environment > and don't want it unnecessarily degrade.

There's barely any overhead for this. It did not impact my benchmark scores, and only adds a ~16 bytes to the size of each hash.

Thu May 23 15:13:03 2013 brad [...] divisionbyzero.net - Fixed in 1.4 added

Fri Aug 09 07:27:02 2013 abraxxa [...] cpan.org - Correspondence added

Am Di 21. Mai 2013, 14:46:20, BLHOTSKY schrieb: Show quoted text

> There's barely any overhead for this. It did not impact my benchmark > scores, and only adds a ~16 bytes to the size of each hash.

We're indexing syslog messages in ElasticSearch and every byte counts. I had to write a filter that removes all key/value pair that we don't need or which are duplicate. If you still insist on having the same data duplicate in the returned hashref you might close this bug report.

Sat Aug 10 06:05:09 2013 brad [...] divisionbyzero.net - Correspondence added

There's a few modules depending on this module, some of which started after this incompatibly was introduced. I'm hesitant to break compatibility. I'm using this for the same purpose, massive amounts of log data to ElasticSearch. Ping me if you want to compare notes on that stuff as well :) On Fri Aug 09 07:27:02 2013, ABRAXXA wrote: Show quoted text

> Am Di 21. Mai 2013, 14:46:20, BLHOTSKY schrieb:

> > There's barely any overhead for this. It did not impact my benchmark > > scores, and only adds a ~16 bytes to the size of each hash.

> > > We're indexing syslog messages in ElasticSearch and every byte counts. > I had to write a filter that removes all key/value pair that we don't > need or which are duplicate. > If you still insist on having the same data duplicate in the returned > hashref you might close this bug report.

Sat Aug 10 06:05:09 2013 brad [...] divisionbyzero.net - Status changed from 'patched' to 'resolved'

Bug #85500 for Parse-Syslog-Line: datetime_raw became date_raw