IRC log for #koha, 2005-07-12

All times shown according to UTC.

Time S Nick Message
15:58 thd kados: pinging kados
16:15 jo Chris: are you about
17:04 rach he is in dunedin today
17:05 rach so I wouldn't expect so
17:28 jo hi Rach
17:29 jo I need our budget reset and I guess he is the only one who can do it ...
17:31 rach sorry I'm not sure - drop him an e-mail, he's likely to get it
17:35 jo thanks Rach. I have sent him a msg so he may be able to . otherwise we will wait until tomorow.
17:36 rach how's your leg?
18:26 jo good - a bit sore
18:26 jo but really good.
20:18 Genji hiya. the items.location field, what does it do? whats its purpose?
20:24 mason genji, im not too sure, but i think its a descriptive field. probably used for comments by library staff?
20:24 mason chris will know for sure, but i think hes a bit busy now with a problem ;(
20:25 rach he's in dunedin
20:25 rach so not really around at all
20:26 rach but I know that paul put in some extra fields, so it might be one that identifes the "sub location"
20:26 mason thats what i initially thought, but...
20:26 rach we just had branches, but people wanted to identify rooms in the branches
20:27 rach so I'm not sure which of those it might be
20:27 Genji right, okay. so im adding an extra field to items, for the physical shelf number... that looks up to a virtual shelf.
20:27 rach paul has it set up so you can say  an item is in freds office
20:28 mason ah, makes good sense.
20:28 mason ta rach.
20:29 Genji right. and yet, after its been in freds office, with my physical shelf number, it can be returned to the correct shelf... complementry.
20:38 rach um no it's not issued to freds office
20:38 rach that
20:38 rach that's where it lives I think
20:38 rach but you might do best to check this with paul
20:40 Genji Oh. okay. hmm... then my physical shelf tiein to virtual shelf system could replace his system?
20:42 rach I really don't know sorry
20:43 rach if you're changing something that much, you might want to drop a line to koha-devl explaining what you're trying to do, and what the problem is you're trying to solve
20:43 rach so that paul (who knows what it currently does) and joshua at least can comment - and others as well
20:44 rach because you might be able to extend what paul has already done rather than put in a new thing
20:45 Genji But ill continue devel on this physicalshelf thing.
01:47 hdl hi
01:48 thd hdl: hello
01:50 hdl how are you ?
01:50 thd still awake somehow
01:53 thd hdl: Is paul away on holiday?
01:54 hdl yes, until the end of this week.
01:54 hdl Do you have problems ?
01:56 thd hdl: No special problems.  I just wanted to ask when he might commit his fix for the data corruption aspect of bug 997 backported to the current stable release.  He said he had that fixed on his desktop.
02:05 thd happy koha to one and all
02:06 hdl ;)
02:08 hdl thd : Has Paul sent you an email about 997 issue or was it on the chan ?
02:09 hdl It seems to me not such a trivial problem... Would certainly need an order field. or an inidcator management, which, atm is quite poorly done.
02:10 thd hdl: paul reported that he had the corruption aspect fixed on his desktop over #koha
02:11 thd hdl: I think he made a change to store the order in a datase field without addressing the cataloguing issue for 997
02:13 hdl But wasnot ceommted for two possible reasons :
02:14 thd hdl: The cataloguing issue could wait until zebra and I offered to start working seriously on a better cataloguing interface starting in about 3 weeks or after some OpenURL work :)
02:14 hdl 1) requires a database structure modification, not allowed in 2.2 branch.
02:16 hdl 2) needs many modifications in addbiblio..
02:16 hdl thd : he hasnot told me.
02:18 thd hdl: Is the 2.4 branch working?
02:18 hdl I don't know for 2.4. I am working on HEAD.
02:19 hdl But 2.4 may be called 3.0. It is not clearly defined.
02:19 thd hdl: paul had mentioned something on koha-dev about head being broken around the time he announced 2.2.3RC
02:20 hdl I think that ALL HEAD branch is not broken. But some scripts are...
02:20 thd hdl: Which parts of head are broken?
02:22 hdl Some people would commit things without :
02:22 hdl 1) Testing it works on their machine.
02:22 hdl 2) Mentioning THEY ARE working on that... And shall be fixed soon.
02:23 hdl For Instance, I noticed that search was broken last week.
02:23 hdl I sent a fix.
02:25 thd hdl: What did kados mean about making sure code at least compiles before being committed?  Run without throughing an untrapped error?
02:25 thd s/throughing/throwing
02:26 hdl for instance ;)
02:26 hdl When you commit things. Try to make a complete commit.
02:27 hdl THat is to say... If you modified base, send updatedatabase also.
02:27 hdl if you modified many files, send them at once.
02:27 thd hdl: I have no particular for instance :)  I just did not follow the usage of 'compile' with Perl.
02:28 hdl test your modifications on your machine and get it work before sending.
02:29 thd hdl: would anyone do other than commit a tested modification? :)
02:30 hdl Some did, since we had to send this e-mail ;)
02:31 thd hdl: and is there no process for ensuring commits have been tested and approved appart from the formal release?
02:34 hdl ATM, we can only count on the consciousness of ppl that commit code.
02:34 hdl That's why e-mails are important.
02:36 thd hdl: Shoudn't there be a validated part of the source tree that is upstream from the stable release that requires special privileges for committing?
02:38 hdl asa you are an acknowledged developper, you can commit. That's the point.
02:42 thd hdl: shouldn't the protection of a level of the development source be stronger than an admonishment to please play nicely?
02:45 hdl Shouldn't the persone which contribute to Koha and rely on it and are willing to develop and contribute to it be sensible enough to keep it safe and secure ?
02:45 hdl philosophy, when you catch us....
02:45 hdl ;)
02:45 hdl ;))))
02:50 thd hdl: I meant shouldn't there be a hierarchy of development privileges to protect the careful from the careless?  But of course, everyone should always act in his best interest, and therefore, no one could ever commit a mistake as a logical impossibility of acting against himself :)
02:52 hdl ATM, as I said, it is not so.
02:52 hdl For the future, why not sending an e-mail on koha-devel ?
02:56 thd hdl: I was merely trying to clarify the present situation before considering an email if one might be warranted.  Perhaps there are not enough developers for this to be any significant issue.
03:00 thd hdl: also I wanted to know how easily I might break something for everyone if I became a possible culprit :)
03:06 thd hdl: Accidental incomplete commits would seem to be the greatest potential problem if everyone is playing nicely.  I wish it were actually a logical impossibility :)
03:08 hdl thd : accidents may happen. But when one spoils the milk, this one can also take a sweep ;)))
03:08 hdl ASA he realizes it of course ;)
03:14 thd I would keep tidy :)
03:16 hdl Anyway, this should only happen in HEAD version. There are too few programmers on stable version to brak things badly.
03:16 hdl Even on HEAD, few persons commit things. One answer could be to keep a safe personal backup...
03:16 hdl Then when trying to compile sthg and failing to, you could copy the old code, modify what you want, test and then try and commit.
03:16 hdl My question would be about CVS management of that kind of action....
03:16 hdl Anyway, keeping tidy is the best way.
07:34 Genji hiya all.
07:39 hdl hi
11:48 kados msg owen looks like some stuff has changed there
11:48 kados oops ;-)
11:49 owen So maybe it wasn't our fault? ;)
11:50 kados right

| Channels | #koha index | Today | | Search | Google Search | Plain-Text | plain, newest first | summary