IRC log for #koha, 2006-02-23

← Previous day | Today | Next day → | Search | Index

All times shown according to UTC.

Time Nick Message
11:58 kados morning stephen
11:59 hi all
11:59 shedges morning kados
12:58 morning owen
12:58 owen Hi
12:58 thd are you feeling better kados?
13:03 owen: the default intranet template has support for moving subfields up and down within the field in the record editor for  rel_2_2.  The intranet template provides no such support now.
13:04 owen Do you mean the NPL template provides no such support now?
13:04 thd owen: yes that is what I mean
13:04 owen You're talking about addbiblio, right?
13:05 thd owen: Yes.
13:05 owen Okay, I'll check it out
13:05 shedges thd:  has that feature been improved?  It used to be that you could keep moving subfields up and up, into the previous tag, and the previous tag...
13:05 (not good!)
13:05 paul no, it hasn't yet.
13:05 i'll do it for 2.2.6
13:05 shedges hi paul!
13:05 paul hello everybody. Still stuck with utf8 on this side of atlantic :-(
13:06 shedges (we all use Windows on this side of the atlantic)
13:06 thd shedges: Oh yes, up and up but that still allows changing the order of the subfields.
13:08 paul: Will you also add changing the order of entire repeated fields?
13:09 paul nope. This hack was just a dirty hack before Koha 3.0 that will be shipped with a new marc editor
13:11 thd paul: Changing field order is important for sequencing linked fields and also for altering primary, secondary, etc. importance in repeated fields.
13:14 paul: This hack is actually sufficiently functional to provide for the possibility of creating standards compliant records in the Koha record editor that had been impossible for many common cases previously.
13:21 paul: Why was .the inconspicuous "..." chosen to hide the best features of Koha instead of something more visible?  Especially in the OPAC, it must be missed by most patrons who would not have special information about Koha features.
13:23 paul: Should '...' not be some symbol or text that signifies its function clearly to any new user.
13:23 ?
13:26 hdl: you had been asking about shipping costs divided by the number of items received and associated with the item in normal acquisitions.
13:26 hdl thd : No.
13:27 I asked about the freight being multiplied for each order line :)
13:28 thd: that seems quite different ;) But I wait and listen to you.
13:28 thd hdl: You mean multiplied improperly and not based on the number of copies or total items ordered?
13:28 hdl thd: And sent a mail on koha-devel about that matter
13:29 thd: I mean
13:29 To my mind freight is like "frais de port" in french.
13:29 You pay it for the whole parcel because of its weight.
13:29 owen thd: a "best feature" for a librarian isn't necessarily the best feature for a patron
13:30 hdl So it should not be multiplied for each order line but should be added to total.
13:31 And maybe, but I am not an expert, after that the VAT (GST) is calculated.
13:32 Moreover, some ppl told me that VAT in france depended upon the books or the booksellers. (But Koha is great. It manages that for a parcel.)
13:35 (except that it is a value and not a percentile)
13:46 kados hehe
13:46 one of my clients was confused and thought that
13:46 'debarr' meant 'not barred'
13:46 :-)
13:47 so they were 'debarring' every patron they added :-)
13:47 causing no end of confusion until I realized how they were using the word :-)
14:04 thd hdl: There is often a difference between how freight is assessed on shipments to some types of customers such as bookshops and libraries as opposed to ordinary retail customers.
14:05 hdl thd:  tell me
14:06 thd hdl: In an ordinary retail transaction you usually know the shipping charge in advance.
14:07 hdl kados : are they speaking french ? Som of our customers would also ask the question the same way.
14:07 thd: still listening.
14:08 kados hdl: maybe we should think up a new term for 3.0
14:08 hdl: that's not so confusing :-)
14:08 thd hdl: The vendor in an ordinary retail transaction will set a charge based on the vendor's needs but the real fee for shipping may be hidden in the profit on the price of the books themselves.
14:10 hdl: In the ordinary retail transaction the vendor will seldom raise the price originally set for shipping even if there are many partial shipments to fulfil the order.
14:10 paul a quick question :
14:10 how to dump the content of a variable in perl ?
14:10 kados use Data::Dumper'
14:11 paul data::dumper, thanks
14:11 kados ; even
14:11 warn Dumper($variable);
14:12 thd hdl: For customers such as libraries and bookshops the business relationship that the vendor has may be rather different.
14:14 paul kados : it's almost time to leave for me.
14:14 BUT :
14:14 a GOOD news, that has to be confirmed in 2 days (tomorrow i won't be here)
14:14 it seems that i have a correct solution to our utf8 problem.
14:14 it relies on mysqlPP package.
14:15 that is a 100% Pure Perl mySQL connector.
14:15 it is 100% pure perl. It also requires only 1 line change in Context.pm
14:15 AND :
14:15 I have included Encode::decode_utf8() at the right place, and it seems to work fine !
14:16 I changed nothing except this in most scripts, and that work fine.
14:18 thd hdl: In every case their may be a vendor set basis of a fixed minimum fee for the presumed package plus something for each item shipped.  The vendor's own calculation may be based on weight and size or more usually a set fee per item that is averaged in the vendor's own calculations between the expected total volume of heavy and large and light and small,
14:19 s/their/there/
14:22 hdl: For any customers there may be an itemised invoice or packing slip attributing a shipping price to each item shipped.
14:22 paul mmm... seems to work imperfectly.
14:22 but at least works a little !
14:22 hdl thd: Would you have some bills to send me as examples.
14:24 thd hdl: Unfortunately, I was not the record keeper for my own business and no longer have such documentation.
14:25 hdl: However, you must be able to ask some of your libraries to send copies of some examples that show varied practices.
14:27 hdl: Otherwise, if you know any small shopkeeper well enough to ask any business would have such records.
14:27 hdl thd: Sure, but would certainly be quite awkward ;)
14:28 kados, owen, chris, rach, shedges, or russ  : could you send some different orders and reception bill exemples ?
14:29 thd hdl: Packing slips are also useful to see what accompanies a shipment when the bill or invoice is usually separate.
14:30 hdl send those too :)
14:30 I hope bills are not too localized but doubt it.
14:31 thd hdl: Often packing slips will be a form similar to the invoice except that the important numbers apart from quantity may be greyed out.
14:34 hdl: For an ongoing business relationship the vendor is likely to charge for each partial shipment and because there is likely to be a minimum charge for each package the total may be more than if everything could have been sent in one package.
14:36 hdl: Therefore, the original expected shipping price at the time the order was placed may be understood as only an estimate.
14:37 kados paul: do you know of a problem with the MARC editor where ISBN cannot be deleted?
14:38 paul mmm... not that I remember atm
14:38 hdl thd: If you have some time, try Acquisition an parcel management and you will see the core of my question.
14:38 kados one of my clients is unable to delete isbns or save the record without the isbn value filled
14:38 might be a template issue ... I'll look into it
14:39 hdl kados: It could be a framework parameter also :)
14:39 kados ahh
14:39 but it's default framework
14:39 hdl default framework can have been a little modified.
14:40 thd hdl: The original order being based on estimated or provisional prices where even the price of the books might change by the time that the order was fulfilled.
14:40 kados hdl: if it's set to 'mandatory' it will behave that way?
14:40 hdl: 020$a is set to:
14:41 hdl: Tab:0, | Koha field: biblioitems.isbn, Not repeatable, Mandatory
14:41 thd kados: Almost certainly 020 $a has been set to mandatory.
14:41 hdl kados: When set to mandatory : value is needed.
14:41 Unless provided a value you cannot input data.
14:41 kados thd: but in some cases it's not there ... why would it be mandatory?
14:42 hdl Is that not the strict norm ? ;)
14:42 And norms must get adapted to uses.... :D
14:43 thd kados: If and only if you were in a library where you expected all books catalogued using the new books framework to have ISBNs then in that case it probably should be set to mandatory.
14:44 hdl thd Thx.
14:44 kados thanks guys, that really helps
14:44 thd kados: That should not be a generic default for all cases but would certainly be an understandable one.
14:44 hdl Well folks,
14:44 kados should isbn be repeatable?
14:44 I think yes right?
14:45 thd good evening hdl_away
14:45 kados well ... the tag 020 anyway, how about the subfield $a?
14:45 thd kados: absolutely
14:45 kados both?
14:45 thd kados: $a is unique within 020
14:46 kados I'm getting some framework suggestions from a cataloger who works in a prospective client's library
14:46 so hopefully soon I'll have several framework options on the liblime demos
14:46 she was describing 'materials designation' as the most important distinction to use when deciding what framework to use
14:46 245$h and a number of other fields ...
14:46 thd: that sound familiar?
14:48 thd kados: I have not finished the compile and fully validated default framework.  I was delayed starting and continuing by odd template behaviour.
14:48 s/compile/complete/
14:49 kados: 245 $h is not in all records and cannot be used alone as the basis.
14:50 kados: 000/06-07 is the best starting point for framework selection.
14:52 kados: 000/06-07 are absolutely required but libraries using Koha with an incomplete framework that lost their original records may have no 000.
14:52 kados thd: so the first job for me is to get a nice looking default framework on the demo
14:52 thd: so that libraries won't be scared off :-)
14:54 thd kados: I was distracted by appearance in the MARC display that was obviously scaring off libraries independent of having even a perfect framework.
14:54 kados thd: right ... but that's cleared up now right?
14:54 thd: or at least we have a fix
14:55 thd: I can make the modif to LibLime's demo
14:55 thd: if you think that would be good
14:56 thd kados: I have a fix for one issue for both the intranet and the OPAC but we need a new preference to commit to rel_2_2.
14:56 kados thd: I can do the pref and the fix
14:57 thd: just let me know what the fix is
14:57 thd kados: paul had made the MARC display less verbose and making it appear as if something is wrong underneath even when everything is right.
14:58 kados thd: I understand that
14:58 thd: we just need to put a simple 'if' statement before some lines right?
14:58 thd: can you think of a good name for the new syspref?
14:59 thd kados: That was a complaint of some of his customers so less verbose and and more correct need a system preference.
15:01 kados: MARC Semantic View Verbosity: standard | economical
15:02 kados hehe
15:02 too verbose I'm afraid
15:02 :-)
15:02 thd kados: There should also be a very traditional MARC view that is very compact with no semantic labels.
15:04 kados: MARCSVV: 0 | 1 :)
15:05 kados: that is much less verbose but also unclear.
15:05 kados yep
15:05 how about
15:06 Standard MARC Layout
15:06 yes/no
15:08 thd kados: There should be some attempt at distinguishing the traditional codes only MARC view from the one with semantic labels otherwise the user will expect something that Koha does not have at the moment but which could be added with fairly trivial work.
15:09 kados thd: er? what's that?
15:09 thd kados: also I have no clue about what I would expect when choosing no in your example
15:09 kados you mean MARC view with no labels at all?
15:10 syspref MARC Labels yes/no
15:10 thd: that's above and beyond what we're talking about though
15:11 thd kados: The usual one you see everywhere with only codes is something that is missing from Koha and probably expected by MARC fanatics and is certainly beyond what we are talking about.
15:12 kados: However, that is in the context of meeting librarians' expectations so they do not run in the other direction after a brief look at Koha before they understand its fantastic benefits.
15:14 kados: Labled MARC View:  standard | economical
15:15 kados: That is less verbose and perfectly clear.
15:17 kados: It also reserves a distinction for the coded but otherwise unlabelled MARC view that should be there and could easily be added.
15:18 kados thd: right ...
15:18 thd: ok ... I'll make it so
15:18 thd kados: I did stay up much of last night working on the default framework for MARC 21 but I had forgotten how much was missing and incorrect.
15:18 kados thd: cool
15:19 thd kados: I still have much more to do.
15:20 kados: I hope there are no more additional places for libraries to use the wrong mislabeled field to store ISBN numbers.
15:21 kados: I wonder how many inexperienced part time cataloguers may have mucked up new records with that alone.
15:23 kados: actually, that field would not have appeared previously in the record editor so it would have gone unnoticed.
15:23 kados: I have been using the latest rel_2_2 but I do not see the system preferences that you added.
15:24 kados: Did you commit the extra system preferences to rel_2_2?
15:25 kados: are you still there?
15:27 kados: or have you succumbed to your fever from last night?
15:28 kados I'm here
15:28 was just handling a support issue
15:28 done now
15:28 thd: which sys prefs?
15:28 thd: do you mean?
15:29 thd kados: You had added ones for Amazon and one for linking to other resources outside Koha.
15:30 kados: I do not see them in the latest rel_2_2 that I am running.
15:31 kados did you updatedatabase?
15:31 thd kados: As far as I know nothing updated when I ran the database updater.
15:32 kados well they're there
15:32 try exporting KOHA_CONF and running it again
15:34 thd kados: I should try updating the database again because I found that sudo had lost PERL5LIB even though root and I still had it.
15:35 kados: Do have any idea what could clobber PERL5LIB for sudo on a Debian testing system?
15:35 s/Do/Do you/
15:38 maybe root had actually unset it and it was restored with a new shell as it should have been when I tested.
15:40 kados I'm not sure
15:40 just run updatedatabase from within the CVS koha dir
15:40 and it'll automatically put C4 in PERL5LIB
15:41 thd kados: Actually, thinking carefully there is a comment in updatedatabase that seems to suppose no value in reading koha.conf but that may be a forgotten ancient historic artifact.
15:43 kados: I may not have used cd to the directory first.  Does that make a difference?
15:44 kados: Does updatedatabase need the working directory context?
15:44 kados thd: updatedatabase is updated
15:44 thd: with the new syspref
15:45 thd: can you tell me which lines to enclose in the syspref for opac-MARC detail et all
15:45 thd: if you know
15:45 thd: before I ever run any script I always set KOHA_CONF and PERL5LIB
15:45 thd: because so many things can go wrong if you don't
15:45 thd: it's a good rule of thumb
15:46 thd kados: I have the environment variables in my bash profile so I assumed that I was safe.
15:51 kados: I could commit the two files to rel_2_2 now and you can update them with the needed nested if else clauses for using the system preference.  I have not studied hdl's instructions to you or closely examined an isolated example to do that yet.
15:51 kados thd: when I get back, let me know which lines I should look at in the MARC detail pages
15:51 thd: just tell me the lines, I'll commit the change
15:52 thd kados: I have to go to the dentist.  Wil you be around in several hours?
15:52 s/several/a few/
16:05 kados yep
16:07 thd kados: I will provide all information in commented files when I return from having a filling replaced.  ouch
16:30 kados ouch is right :/
16:57 thd-away: LabelMARCView support is now committed into CVS and seems to be working on LibLime's demo
16:58 thd-away: once you confirm it's working as expected I'll post a note to koha-devel
16:59 thd-away: we may also want to formulate an answer to Stephen about it, explaining that it was nothing but a display issue
17:55 shedges kados:  so there was no possibility of saving scrambled tags?
17:56 kados shedges: saving scrambled tags?
17:56 shedges: do you mean for libraries like NPL?
17:57 btw: remember way back when when i was trying to get us to look at the leader and fixed fields rather than a local use tag for itemtype?
17:57 turns out I was right :-)
17:58 shedges is the leader flexible enough to store user-defined item types?
17:58 kados no
17:58 you have to follow AACR2
17:59 the unfortunate bit is that NPL lost all the 00X fields on import
17:59 shedges ah, right
18:00 kados because paul's original bulkmarcimport discarded those fields
18:00 it doesn't strictly affect any performance ...
18:01 it just means that NPL won't ever have script MARC-compliant records
18:01 unless we can come up with a clever way to intuit that info from the data
18:02 shedges ...or just goes to all XML and let's MARC die...
18:03 kados yep
18:03 that would be truly innovative
18:04 shedges it would certainly make life in the Web much easier!
18:06 kados I think for that we'll have to wait until 3.0's finished and we've got the new holdings forest in place
18:06 then we can start working on some kind of XML-based profile for library records
18:27 yay logbot
22:03 thd kados: I see that you committed the code to opac-MARCdetail.pl to rel_2_2.
22:04 kados thd: yep
22:04 thd: but it's a syspref
22:04 thd: so we should be safe :-)
22:04 thd kados: I can commit the corresponding changes for the intranet and also to HEAD.
22:05 kados already did
22:06 wait ... didin't to head
22:06 thd kados: That does not cure all about which Steven was commenting.  That is however, where, he was observing that issue.  Other issues that he had were not mere appearance.
22:08 kados: I saw no commit to MARCdetail.pl only to opac-MARCdetail.pl.
22:09 kados i commited both MARCdetail and opac-MARCdetail in rel_2_2 IIRC
22:09 thd: are those a matter of a better default framework?
22:09 thd kados: Also paul had cited an additional if clause for MARCdetail.pl that seems not to be present in opac-MARCdetail.pl.
22:10 kados: yes, a better default framework.
22:11 kados thd: which you're working on right?
22:11 thd kados: yes
22:12 kados: back to that just after I succeed with updatedatabase
22:12 kados cool
22:14 thd: do this exactly:
22:14 thd kados: NPL has an external record store that it can use for recovering 00X and anything else that may need recovering from the SQL void in Koha does it not?
22:14 kados cd /path/to/koha/cvs/repo
22:14 ie, the 'koha' directory
22:15 export KOHA_CONF=/path/to/koha.conf
22:15 updater/updatedatabase
22:15 then you'll be done :-)
22:15 and you can get to work on frameworks :-)
22:17 thd kados: I will add that to my grabkoha.sh script that is doing all the repetitious tasks now.
22:17 kados: Dose NPL not have the data preserved for recovering what was lost inside Koha?
22:19 kados thd: pass
22:19 thd kados: pass?
22:20 kados thd: they don't
22:21 thd kados: was it not your job to look after their data for them? :)
22:21 kados thd: I tried ... did you see my earlier message to stephen?
22:23 thd kados: I saw some log from today but I must have missed part of the discussion and logbot was down.
22:23 kados sorry then :-)
22:27 thd kados: I have from this point on and seem to have had an uninterrupted connection for much of today..
22:27 shedges kados:  so there was no possibility of saving scrambled tags?
22:27 kados shedges: saving scrambled tags?
22:28 kados thd: in fact, it doesn't matter for NPL at all
22:28 thd: having valid marc would give them zippo ... nadda
22:28 thd kados: There must have been something before that which was not a public message.
22:28 kados nope
22:29 thd kados: Why would invalid MARC help NPL?  :-)
22:29 kados um ... the point is valid marc would do nothing _for_ them
22:29 I didn't say invalid would help :-)
22:29 just that valid wouldn't help :-)
22:30 in fact, there would be absolutely no advantage for NPL
22:30 thd kados: You merely mean that Koha ignores that data now?
22:30 kados no ...
22:30 I mean that even if Koha didn't, it wouldn't matter to them
22:30 they don't have professional catalogers even
22:31 thd kados: That is not an uncommon problem even where there are simply not enough professional cataloguers.
22:32 kados there's nothing that valid MARC would give them that they don't already have
22:32 due largely to their current cataloging practices
22:33 thd kados: After I finish the framework, I will send you the long delayed message about how that can be remedied for every library.
22:33 kados thd: I may have finished my framework before you finish yours :-)
22:34 thd kados: Too much work required to finish tonight.
22:34 kados I got a specification from a potential client
22:34 several actually
22:35 thd kados: Do you mean the have given you a copy of the SQL already?
22:35 s/the/they/
22:35 kados thd: of course not ... just the tags/subfields they need to be in certain frameworks and how to fill the fix fields, etc.
22:35 materials designation, etc.
22:36 thd kados: one default framework will be insufficient for meeting needs.
22:36 kados of course ...
22:37 that's why they sent me several :-)
22:37 thd kados: Some plugins need to be written as well but nothing major for minimal compliance.
22:38 kados yep, and I've already written one of those :-)
22:41 thd kados: I am specifying minimal requirements for all subfields.  After that is done and carefully checked we can discuss what other standard frameworks should be created.  I am carefully noting the salient differences for national framework which would be a modest modification.
22:42 kados thd: I'm starting to think we're going to need several frameworks
22:42 a general one where the user is prompted to enter values for all relevant fields
22:42 then one for each materials designation
22:42 minimally:
22:42 Book (monograph)
22:42 Computer File
22:42 thd kados: There should be a national level one just to prove that Koha can be used for national level cataloguing.
22:43 kados Video (film, filmstrip, transparency)
22:43 Picture, study Print, Photograph, Chart
22:43 Kit (Book and nonmusical Casette)
22:43 Manuscript Lanuage Material
22:43 Maps Globes and Atlases
22:43 Mixed Material
22:43 Music (Printed)
22:43 thd kados: UNIMARC Koha has those, however, they are full of mistakes and the libraries that paul has do not care.
22:43 kados Three dimensional Artifact, Object
22:44 Sound Recordings (nonmusical)
22:44 Serial (Printed Periodical, etc.)
22:44 Sound Recording(Musical)
22:44 phew :-)
22:45 thd: do you know where I can find out what the leader should be for each of those materials?
22:45 thd kados: UNIMARC Koha only has some of that more complete list.  They have no realia framework for cataloguing cannonballs and paintings.
22:46 kados: That same table for 000/06-07
22:51 kados: I had given you the related 008 table before but see the table at the bottom of http://www.itsmarc.com/crs/Bib0021.htm
22:51 kados shit ... I don't want to have to decode that :-)
22:52 i need a valid string for 'BOOK(MONOGRAPH)' :-)
22:52 thd kados: That is easy, I have the Perl code already.
22:52 kados yea?
22:53 for now could you just tell me what the value should be for a book?
22:53 '00428nam 2200145u' 4500 look right to you?
22:53 oops ...
22:53 thd kados: yes, I started it months ago for a message to koha-devel until too many additional fields were involved for a simple message.
22:54 kados '00428nam 2200145u 4500'
22:55 thd kados: where did n come from?
22:55 kados no offense, but you've got to just send those messages even if they are not 100% complete
22:56 I got that from one of my client's catalogs
22:56 Code n (Special instructional material) is obsolete. Defined 1975-1983; use code k (Two-dimensional nonprojectable graphic) and code r (Three-dimensional artifacts and naturally occurring objects).
22:57 thd kados: I did send the message I just did not attach incomplete code to it to prove an obvious point.
22:57 kados how's this:
22:57 '00994pam 2200301 a 4500'
23:00 thd: did you write the 'marc21leader.pl'?
23:00 thd: the plugin?
23:01 thd kados: paul wrote that,  I suggested good default values for the most likely use in Koha.
23:03 kados: your last example is the monograph component part of a mixed material.
23:03 kados can you tell me what the correct code for a simple Book (Monograph) should be?
23:04 thd kados: your previous example is the monograph component part of a nonstandard (not in the MARC standard) record type.
23:05 kados: maybe the first part is a typo.
23:06 kados: plugins may not cure typos but can certainly reduce them and at least leave you with a valid record of some sort.
23:08 kados hmmm ... I added a 000 to my 'Book' Framewokr
23:08 but it still doesn't show up when I try to edit a MARC record
23:09 do I need to add it as @000 ?
23:09 thd kados: 000/06 is not defined for 'n'.  There is a remote possibility that it is an historical usage which should have been converted long ago.
23:11 kados: you need to add it to your '@' subfield for 000 in the bibliographic framework editor.
23:11 kados not enough characters to fit it :(
23:11 the template is preventing it :-)
23:12 thd kados: What do you mean?  Not enough characters to fit what?
23:13 s/what/what where/
23:16 kados I mean that I can't enter in more than 3 chars in the 'tag' input box
23:17 thd kados: you mean the add field box?
23:18 sorry retract last post
23:18 kados found the prob
23:18 have to commit it to cvs tonight
23:19 thd kados: It is working on my rel_2_2 system so it must already be in CVS.
23:19 kados so when I add it as @000
23:19 it doesn't show up :-)
23:20 thd kados: Add field 000.
23:20 kados er?
23:20 I thought you said to do it as @000
23:20 thd kados: create subfield '@'
23:20 kados ahh
23:21 I'm a dumbass :-)
23:21 thd kados: actually if you leave the subfiled name blank the default templates will create it as '@'.
23:21 kados ahh
23:21 doesn't on mine
23:22 thd kados: try the default template.
23:22 kados: default only for intranet in MARC.
23:23 kados shit ... npl template doesn't even save the @ subfield
23:23 wtf
23:25 thd kados: The default templates are now only non-MARC in the OPAC for some mistake that happened during 2.2.4.
23:25 kados doesn't work in default either
23:26 thd kados: Some translations are tied to the default OPAC templates so there is now a problem for users of those languages.
23:26 If hey are using MARC Koha.
23:26 s/hey/they/
23:27 kados no, I'm not even able to add a @ subfield in the 000 tag
23:27 it won't 'stick'
23:27 thd kados: update to the latest rel_2_2 in CVS.  I only had 2.2.5 for a couple of days.
23:28 kados: I run a single script to update everything now.
23:28 kados it _is_ latest cvs
23:29 thd kados: Is there some legacy problem in your bibliographic framework table?
23:30 kados could be
23:30 that's the thing about updatedatabase
23:31 it doesn't update if the table exists :(
23:31 well ... table and column that is
23:31 ok ... I'll delete the framework table and run updatedatabase
23:31 thd kados: owen told me that no one would touch the frameworks at NPL because of some problem from versions ago would give permanent headaches that no one else had ever been able to reproduce.
23:32 kados: That problem with updatedatabase should be cured.
23:33 kados: You must be able to add columns to an existing table or nothing would hardly ever update without being dropped first.
23:35 kados actually, I'm not even sure which table the framework stuff is in
23:38 ok ... deleted biblio_framework and readded it with updatedatabase
23:39 ok that must not be it
23:41 I'm guessing it's probably marc_subfield_structure and marc_tag_structure
23:41 maybe even others
23:41 this sucks
23:42 thd kados: careful not to delete the wrong table :)
23:44 kados I may have :-)
23:44 yea, looks like it's toast :-)
23:45 totally
23:45 shit
23:49 thd kados: you built everything from an SQL dump right?
23:55 kados yea, restoring now
00:16 I decided to completely start over with a fresh install
00:16 to grab the frameworks stuff
00:19 now updatedatabase has run
00:37 thd: do you know this syspref:
00:37 itemcallnumber  676a  The MARC field/subfield that is used to calculate the itemcallnumber (in UNIMARC
00:37 it cuts off
00:40 thd kados: That preference is supposed to aid in filling an item call number from a standard location.
00:40 kados does it apply to usmarc/marc21 too?
00:40 thd kados: It is not UNIMARC dependent but I could never get it to work
00:40 kados thd: also, even now, my marc structure doesn't display a 000 tag
00:40 thd: or does the new default still not have it?
00:41 thd kados: Default is very poor and is missing much.
00:43 kados: Inaccurate repeatability was added to the original work but nothing more.
00:45 kados ARRRG
00:45 thd kados: The librarian who helped paul originally prepare the specifications for the frameworks did not have the compulsion for validity that people may loose their data if this is incomplete or inaccurate.
00:45 kados just added 000
00:45 then attempted to addd subfield @
00:45 and it didn't work!
00:45 thd kados: what template are you using?
00:45 kados npl
00:45 switching to default now
00:46 thd: using default I get the same thing!
00:47 yea, it's definitely not working!
00:47 this is SO frustrating
00:47 thd: koha.liblime.com/cgi-bin/koha/admin/marc_subfiel​ds_structure.pl?tagfield=000&frameworkcode=BOOK
00:48 give it a shot yourself
00:48 see if you can make it dance
00:51 thd: are you trying it?
01:08 thd: is it possible that the non-default frameworks can't have their own values for LEADER -- only the default can?
01:09 nope :(
01:11 thd kados: I succeeded
01:11 kados thd: in what?
01:12 thd kados: There were no plugins listed though.
01:12 kados I still dont' see the @ subfield listed
01:12 if you do, please show me where
01:13 thd kados: http://koha.liblime.com/cgi-bi[…]rameworkcode=BOOK
01:13 kados hey!
01:13 what was I missing then?
01:14 Text for librarian i bet
01:14 thd kados: maybe the templates reject the attempt to add the subfield if you do not provide text for the librarian or OPAC
01:14 kados shoot ... that was a complete waste of time :/
01:14 thd: I was gonna use a 'Auth value' not a plugin
01:14 thd: because I thought the BOOK framework would only need one value for leader
01:14 thd: am I wrong?
01:15 thd kados: You cannot use an authorised value for the leader and have an accurate leader.
01:15 kados why not?
01:16 thd kados: The leader is supposed to include information about the record size etc. that should be filled automatically.
01:17 kados: I am not certain that paul's plugin calculates the record size when you choose the submit button but it could eventually.
01:18 kados: If I recall correctly, as long as the plugins directory is in the correct place all plugins will be listed.
01:19 kados: The apache user has to be able to read and execute them of course. :)
01:21 kados they are of course executable by apache
01:21 thd kados: You should merely alter the leader plugin with other copies set to different correct defaults for different bibliographic media.
01:21 kados thd: what is in your 'scripts' directory?
01:22 thd: and, from the sound of it, calculate the length of the record :-)
01:24 thd kados: they are in the value builder directory for CVS.
01:24 kados I know where they are
01:24 I just don't know why my Koha's not finding them
01:24 are they also in your 'scripts' directory?
01:26 thd: we're back to not having the '000' $@ again now that I switched back to npl templates
01:26 thd kados: not that I see.
01:27 kados both the @ and the plugins
01:27 thd: what was the value for 'Text for librarian'?
01:27 thd kados: I will run the find command
01:28 kados: I used 'control field' in '@' for 000
01:29 kados wow ... plugin's just magically started working!
01:30 hey, looks like it's even working !
01:30 yay!
01:30 http://koha.liblime.com/cgi-bi[…]mple/addbiblio.pl
01:31 well ... kinda
01:31 the values don't really look right to me
01:32 interesting ... you can even specify if there should be links
01:34 thd kados: koha/intranet/cgi-bin/value_builder
01:34 kados yea, I have no idea why they were't showing up before
01:35 thd: I didn't realize that you would use a plugin as paul's marc21leader.pl works
01:35 thd kados: links work now in both MARC view and the default non-MARC detail view
01:35 kados: what else would you use?
01:36 kados thd: it looks like perhaps it is meant to cover all possible cases
01:36 thd: would you simply specify defaults if you were going to make a marc21.books.leader.pl?
01:36 or would you change the actual way the script works?
01:37 thd kados: I would mak a copy of the script and change the defaults
01:37 kados what should the values be for the types I listed above ... do you know off hand?
01:42 thd kados: Book (monograph) 000/06 a 000/07 m
01:43 kados thd: I don't know what that means
01:44 thd kados: Computer File: 000/06 m, 000/07 whatever applies
01:45 kados: leader positions six and seven
01:46 kados thd: are those the only two that need to be changed?
01:46 thd kados: those are the only two that set material type in the leader
01:47 kados: their are other fields that need plugins, well all of them do. :)
01:47 s/their/there/
01:48 kados yea, but lets focus on the leader first
01:48 thd kados: 008 is a required field and has much to do with material type
01:48 kados: leader only for now
01:48 kados thanks :-)
01:49 looking over this template, I can't figure out how to get it to default to a specific value
01:49 thd kados: Video (film, filmstrip, transparency) 000/06 g, 000/07 whatever applies
01:50 kados nevermind
01:50 I see how
01:51 thd kados: It is very simple with just the leader
01:53 kados: book is tricky by requiring both 000/06 and 000/07
01:55 kados: proper material typing is complex because it involves several fields including ones that cannot be relied upon being present in a found record and many cataloguers have no time to create in an original record.
01:55 kados hang on
01:55 I still can't figure out how to make it default to the values we want
01:55 the leader that is
01:56 do you know?
01:56 thd s/fields/fields, fixed field positions, and subfields/
01:57 kados: I have not tried changing the default yet.
01:57 kados: It may use JavaScript along with Perl.
01:59 kados: there must be something different about the current defaults or there reference :)
01:59 s/there/their/
02:00 kados I get it now
02:00 ok ... we're in business
02:05 thd yes line 82 $result = "     nam         7a      " unless $result;
02:05 kados thd: can you check to see if you like what you see?
02:05 http://koha.liblime.com/cgi-bi[…]mple/addbiblio.pl
02:05 I think what we should do ... if you have time ...
02:06 is to go through the other fields up to 008
02:06 plus the materials designations (if you know what those should be)
02:06 and try to pick values for them that would apply for BOOK (MONOGRAPH)
02:06 or write plugins for them if the values are shifty
02:06 thd: are you up for it?
02:07 thd kados: I am well rested and have a new indestructible filling
02:07 kados heh
02:08 thd mostly indestructible but real tooth enamel is better
02:08 kados shoot ... not sure it's working right
02:09 thd kados: I grew up where fluoridated water was thought to be a communist plot.
02:09 kados arrrg
02:09 for the marc21_computerfile.pl I changed it to:
02:10 $result = "     nmm         7a      " unless $result;
02:10 but still doesn't work :(
02:10 thd Although I never had a cavity until I had been living a few years in New York.
02:11 kados: The existing default is right for books already as the most common thing to be catalogued.
02:11 kados: what is the problem?
02:12 kados: I think that is an illegal value
02:12 kados thd: did you visit the above link?
02:12 thd: and try to catalog a new Computer File
02:12 using the COMPUTER FILE framework?
02:12 it doesn't fill 000/07 with 'm'
02:13 like it should
02:13 well ... 000/06 I mean
02:14 I bet the problem is this:
02:14 defaultvalue=document.form​s[0].field_value[i].value;
02:15 where the heck is that stored?
02:17 document.forms[0] is the first form in the document
02:17 thd kados: perhaps there is even more to change for the default
02:18 kados field_value[i] is a variable for the value of the field within that form ... each one of them has a value
02:21 thd kados: there is a reference to unimarc field 100 in line 56
02:24 kados well ... we'll have to ask paul tomorrow
02:24 I can't figure it out
02:25 thd: lets move from the leader
02:25 thd: what's next in terms of importance ... 008?
02:25 thd kados: 001 is mandatory
02:25 kados ok ... so 001 then
02:25 it shouhhld be linked to biblionumber for now, right?
02:26 biblio.biblionumber ... right?
02:26 thd kados: 001 should be filled automatically from biblionumber yes
02:26 kados thd: what's the label?
02:27 thd kados: what label do you mean?
02:27 kados thd: what should I call it ?
02:27 thd kados: you mean subfield '@'?
02:27 kados Control Number
02:28 no, I meant 001 itself
02:28 thd CONTROL NUMBER
02:29 kados it's not repeatable
02:29 it has one subfield ... @ right
02:30 thd kados: all '@' subfields are non-repeatable
02:30 kados ok ... what's next?
02:30 ahh
02:30 thd kados: the fields may be repeatable but not in this case
02:30 kados right
02:30 what's next ... 002? :-)
02:30 I'm guessing 003
02:30 Control Number Identifier
02:31 thd kados: a plugin should move any old 001 value to 035
02:31 kados: there is no 002
02:31 kados the 003 can have an authorized value
02:32 so I'll set that up quickly
02:32 with LibLime's code
02:33 thd kados: that should be the branch doing the cataloguing read from wherever the cataloguing user information is located.
02:33 kados what's next?
02:34 005?
02:34 thd kados: That is supposed to be an authorised code assigned by LC for the institution
02:34 kados yep, we have one
02:34 LibLIme that is
02:34 OLvl I think
02:34 I can look it up later ... lets move on
02:34 005?  Date and Time of Latest Transaction  
02:34 should be filled with today's date right?
02:35 or last time's date?
02:35 thd kados: an existing 003 from a copy catalogued record should be moved to whatever field I said it should in my email from months ago
02:35 kados thd: I routinely do this for my clients
02:36 thd: when I bulk-import records
02:36 but we need a plugin that can handle an existing value
02:36 and put it in the right field right?
02:36 thd that would be the one I finally sent when I had finished the default framework this past weekend :)
02:36 kados right
02:37 I have a plugin script that can be changed to do this
02:37 should be fairly easy
02:37 thd: so ... 005 ... todays' date?
02:37 thd 005 has to be regenerated every time a record edit is submitted.
02:37 kados ok ... got a plugin for tha talready
02:38 thd kados: date and time in the required format
02:38 kados thd: which is what?
02:38 thd kados: mostly ISO
02:39 kados 2006-12-30 then?
02:39 thd The date requires 8 numeric characters in the pattern yyyymmdd. The time requires 8 numeric characters in the pattern hhmmss.f, expressed in terms of the 24-hour (00-23) clock.
02:40 there is a dot in between the date and the time
02:41 retract the dot is where the above sentence says
02:41 kados thd: should this be linked to a koha field?
02:42 biblioitems.timestamp maybe?
02:42 depending on how that's used
02:42 thd kados: maybe but no koha field stores the value in the exact required format unless I am mistaken
02:42 kados I'll go with it for now then I'll check to make sure I'm right later
02:43 k ... what's next?
02:43 006 I'm guessing
02:43 thd fractional second appear after the dot but could be fractional seconds '0'
02:44 kados thd: yep, I've got it noted ... for now it's just 2006-12-30
02:44 thd: but I'll definitely add that functionality as it's quite easy
02:44 thd: but I'm just eager to get a working framework in place
02:44 because I've got a webex demo on Thur :-)
02:44 thd kados: updates also affect a leader value
02:44 kados updates to what?
02:44 005?
02:45 thd kados: record updates
02:45 kados right
02:45 I know that already
02:45 we already support that
02:45 I don't see any need to create a separate framework for each material JUST for updates
02:45 lets move to 005
02:45 sorry ... 006
02:45 this looks a lot like the leader
02:46 in terms of what it's supposed to do
02:46 well ... look like
02:46 leader has 24 char positions
02:46 006 has 17
02:46 no sweat
02:46 thd kados: 006 s for additional material and is not needed unless you have a complex type of material to catalogue composed of multiple parts
02:47 kados so what's our default value supposed to be?
02:47 thd kados: That could of course be something as simple as a CD with a booklet insert though.
02:47 kados I guess I'll need to wait on that one since I still dont know how to adjust those leader scripts
02:47 I forgot
02:47 I'll ask paul about it tomorrow
02:48 but in the meatime, what shoudl the values be for a BOOK or for DEFAULT?
02:48 (so that I can fill them when i find out from paul tomorrow how to do so)
02:48 thd: do you know?
02:48 thd kados: 006 is not needed in most cases but would be for some types of material
02:48 kados ok ... should we move on then>?
02:49 thd kados: yes
02:49 kados 007 Physical Description Fixed Field
02:50 thd kados: 007 is not mandatory but is very nice for someone like me who wants to do a lot of work to make good cataloguing easy.
02:50 kados right
02:50 so should it be an authorized value?
02:50 thd kados: it needs a complex plugin
02:51 kados: it is like the leader only more complex
02:51 kados right I see that
02:51 ok ... 008 then
02:51 same deal there
02:51 thd kados: 008 is similar but mandatory
02:51 kados ok ... so I need to ask paul tomorrow morning about how to handle these
02:51 i just can't figure out how the plugin works in this case
02:52 thd: what's next?
02:52 thd kados: If we can understand how his plugins work it should merely be work to build more complex ones.
02:52 kados yep, shouldn't be too hard
02:53 thd kados: That is all the control fields
02:53 kados sweet ... not really that bad after all :-)
02:54 thd kados: other fields are mandatory and could use much support but 000-008 are the most difficult for humans
02:55 kados strange ...
02:55 003 has stopped showing up
02:55 ahh ... I bet I know why
02:56 thd: http://koha.liblime.com/cgi-bi[…]eworkcode=Default
02:56 thd: it's a start anyway
02:56 thd: thanks for your help
02:56 thd kados: paul's code may do nothing for record size etc. in the leader but he may envision treating that outside the editor.
02:57 kados yea ... I think MARC::Record already handles that actually
02:58 thd kados: it should not require pressing a JavaScript link to fill an initial default value.
02:59 kados: There should be some way of auto activating default values for plugins if the field is empty.
03:00 kados there is such a way
03:01 thd kados: also the three dots are difficult to spot.  There should be something more visible that signifies the function.
03:01 kados yep
03:02 thd kados: cataloguers can always be informed but there is a related function in the OPAC that must be missed by most users
03:03 kados: About the best feature of Koha is almost hidden.
03:04 kados: It should be shouted as a great advancement in user interfaces for library applications.
03:04 kados yep
03:09 thd: before you go
03:09 thd: http://koha.liblime.com/cgi-bi[…]mple/addbiblio.pl
03:10 check out the functionality of the 005
03:10 don't click on the ...
03:10 just put your cursor in the field
03:10 today's date should appear :-)
03:10 thd kados: I just thought of a cheap way to separate the OPAC MARC view from the fields specified for the cataloguer to use by default.
03:14 kados: the old value should appear until the record is submitted but a new value should not require any user action other than submitting the record.
03:15 above sentence for 005
03:15 kados ok ... I've got to get to bed
03:15 thd kados: of curse you need the right date time format still but you understood that.
03:16 kados or else I'll miss paul ... won't get up in time :-)
03:16 thd like me unless I stay up and then miss you
03:17 kados heh
03:32 thd kados: I had forgotten something important that confused me about your leader examples.
03:33 kados: position numbering starts with 00 so that 06 is in the seventh position.
05:31 osmoze hello

← Previous day | Today | Next day → | Search | Index

koha1