IRC log for #koha, 2008-12-10

All times shown according to UTC.

Time S Nick Message
13:11 kyle hey all.
13:11 paul_p hello kyle
13:11 hdl hi kyle
13:12 kyle I'm having some trouble with git, are there any git experts around at the moment?
13:12 paul_p not sure i'm a git expert, but throw your question
13:14 kyle I suppose my first question isn't a big one. I made some patches, reverted to master, and are trying to apply these patches to master. When I do, I get a list of 'patch failed' and 'patch does not apply' errors. How do I fix those? I'm using the command 'git apply mycode.patch'
13:16 hdl kyle: it is because your patches are not based on the new code parts but on old ones.
13:17 hdl To fix this,
13:17 kyle please go on...
13:17 hdl revert to your previous point
13:17 hdl make a branch mywork
13:18 hdl apply your patches
13:18 hdl THEN rebase to origin
13:18 kyle I don't know what point I was at, is it hidden in the patch file anywhere?
13:19 hdl Do you know the version of the database you're at ?
13:19 hdl or you were at ?
13:20 kyle The patch I'm trying to apply is the Reserves System update that I posted on the 4th.
13:27 kyle I'm looking at the git log and if I can revert to just after the patch 'Minor logical cleanup.' I think I will be good to go. Any idea how I would to that?
13:27 kyle hi acmoore
13:29 hdl where you uptodate ?
13:29 hdl git reset --hard commitnumber
13:30 kyle Yes, I've done that before. I'll give that a shot.
13:30 kyle sucess!
13:31 kyle success! that is ; )
13:39 kyle I applied the patch and git gives me a list of files that 'needs update'. I checked them out, and they all look like they were patched just fine, no modifications needed. Any suggestions on what I should do next?
13:42 hdl then git branch mybranch
13:43 hdl git apply your patches
13:43 hdl git commit edited files
13:43 hdl git rebase origin
13:43 hdl should do the job
13:59 kyle I think I have a handle on it now. Thanks for all the help. It would be nice if the patch files contained the number of the previous commit.
14:49 ryan owen: around ?
14:49 owen Yes
14:50 ryan owen: have you played at all with inline table cell editing for yui ?
14:50 ryan[…]_cellediting.html
14:50 owen No I haven't. Someone was working on some AJAXy inline table editing though...for system prefs I think?
14:51 gmcharlt pianohacker; also working on adding AJAX to circ pages
14:53 ryan gmcharlt: were jesse's patches submitted ?  
14:54 gmcharlt he's still working on them
14:57 ryan gmcharlt, others:[…]w_bug.cgi?id=2761
14:57 ryan any consensus on the 'correct' column length?
14:57 ryan for item call number ?
14:57 ryan i was thinking 64, but i see nicole suggests 255.
14:58 gmcharlt 255 is too long, but anything less than 100 is is too short
14:58 ryan 128, then
14:58 gmcharlt (edge cases - you would think anything longer than 30 characters would defeat the purpose of having one in the first place, but...)
14:58 gmcharlt 128 is fine
14:59 ryan yes, it would seem to me that > 30 characters would be a misuse of the field.
14:59 ryan but may as well allow it
15:44 kados that has fairly substantial implications for spine label printing
15:45 paul_p a lot of ppl today here ;-). Hello kados, owen, gmcharlt, ryan
15:47 hdl hi
15:49 kados hi guys
15:49 danny hey everyone
16:00 danny I've been brainstorming about support for multiple cover art providers, basically the script will need to check to verify if the image exists if not go on to check until it finds a valid one and display it, or display owen's message that no cover art exists. Two options I have thought about are to use AJAX and go through the checks each time a cover art needs to be loaded, the other option would be to maybe have the script load once a day/week and dump into a data
16:00 kados danny: I'm in favor of doing that processing client-side FWIW
16:01 kados danny: doing it server-side could be pretty expensive
16:03 danny ok, it will still be partially server-side though because the only way to check Baker&Taylor is to do a Perl ua->get and check to see if the image is valid
16:04 danny that perl script would be called with ajax, but still a bit of backend work
16:08 gmcharlt some caching would be good, though
16:08 atz danny: really, ajax is the only choice since google's access methods require it
16:08 gmcharlt although not all providers will allow it IIRC
16:09 atz gmcharlt: yeah, i think we can defer the question of caching jacket images to a more robust proxy setup.... somebody already implemented this in the UK
16:10 danny ok
16:20 danny not 100% sure but I dont think you need JS to use google's method, if you can parse the JSON response in Perl it would still work, but not necessary if using ajax anyway
01:32 chris gah not again

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