- Koha - lists.katipo.co.nz

Version 1.07 database update
by Tonnesen Steve 21 Dec '00

21 Dec '00

The database update script failed for me. The itemnumber field was already in the 1.06 table, so the itemtype field never got added. Also, I noticed that the new 1.07 koha.mysql data had a key on the datetime field. I had to modify the statistics datetime field so that it was NOT NULL and then add the key. Steve.

2 3

1.07 is here
by Chris Cormack 20 Dec '00

20 Dec '00

Hi guys Ive finally tarred up and put 1.07 up availiable for download. Ive also this time, created a patch file and an update script to patch the perl scripts and update the database http://www.koha.org/download/update1.06-1.07.tar.gz The full installation is http://www.koha.org/download/koha-1.07.tar.gz Changelog is at http://www.koha.org/download/Changelog Im also in the process of setting up koha on sourceforge http://sourceforge.net/projects/koha/ I will put the files up on soureforge for download also, unfortunately at the moment they are updating things, and releasing files isnt working properly. Ill put them up there as soon as the sourceforge ppl have finished their updates. I will be setting up cvs there, so if ppl would like to work on the project please contact me, and ill enable cvs access for you. Of course anonymous cvs updates will be available for all, but I want to set it up so developers can also commit their changes to the cvs tree. Much thanks to Glen Stewart, and Steve Tonnesen for their help. I have included Glen's data import script in the tarball as well. Hope you all have a good Christmas Chris -- Chris Cormack Programmer 025 500 789 Katipo Communications Ltd chris(a)katipo.co.nz www.katipo.co.nz

1 0

More installation log:
by Tonnesen Steve 15 Dec '00

15 Dec '00

Some more issues (mostly resolved) that I have come up with in testing Koha: --- 1 --- In circ program: Error message when returning from ISSUES or RETURNS scripts messed up the main menu. Added a 'system("tput clear");' to clear the screen before redrawing the main menu to fix this. --- 2 --- Reading record only shows books currently signed out. --- 3 --- When viewing an item's record and clicking on the cardnumber beside "On loan to", it always goes to borrower number 2. Found problem. Was always going to previous borrower, instead of current. Changed 'borrower0' to 'borrower' in moredetail.pl, line 133. Also had to modify Search.pm so that $data->{'borrower'} was set to current borrower. --- 4 --- When viewing items in the moredetail.pl script, the header at the top is blank (just '()' where the title, author and dewey number should be) and biblio/biblioitem info is missing in the box at left. Found a fix for this. Changed "$data=bibitemdata($bi)" to "$data=bibdata($bi)" in moredetail.pl. Probably would have been better to find out what was wrong with bibitemdata subroutine? --- 5 ---  INCLUDE lines in htdocs files are not working for me. I'm assuming that I'm missing something in my apache configuration... Yes. Needed to enable Server Side Includes. Modifed apache config and added: Options +Includes AddHandler server-parsed .html to VirtualHost sections for intranet and opac virtual hosts. Note that apache web site discourages using .html for parsed files in favour of using .shtml. But if all your flat text files for Koha are parsed it doesn't make any difference. --- 6 --- I have also started to write a simpler interface for adding books to the database. Since I have schools that have no existing library automation in place, all existing books will have to be entered by hand (hoping MARC records will come in handy here if I can find a CD to borrow) and the acquisitions module is giving me problems, and is more tedious than I need. My script asks for an ISBN, if a biblioitem already exists, it just asks for the item info. If a biblioitem does not exist, it asks for the biblio info (title, author, copyright, etc.). If there already appears to be a biblio with this info it uses that and then asks the user to enter the biblioitem info and item info. Once the existing holdings are entered, they can revert to using the standard acquisitions module, or just keep entering books using my simpler script. Steve.

1 0

Re: MARC Records
by glen_stewart＠associate.com 15 Dec '00

15 Dec '00

Hi Folks, I recently sent Chris a script developed for our small company library, which ran GLAS (a very yucky Windows software) until yesterday. It takes what GLAS exports and acts as a SQL generator. GLAS thinks it's exporting MARC records, but I wasn't so sure. Regardless, it may be trivial to modify it for use with MARC data. Because I couldn't make sense of the MARC output GLAS provided, I used the subfield option instead of MARC tags to identify each record item (a-z line prefixes). You may be able to plug the MARC tags back into the code and put it to good use. When the script is done running, the admin types "mysql Koha < koha.sql" and all the records are brought into Koha. GLAS uses the concept of "book ID's" to identify each book as unique in the database. I used these same ID's in Koha, as biblionumber and barcode number. Our library uses these numbers on book labels as well as Dewey numbers. I should note that we still use physical cards to track overdue items. The librarians want to switch to Koha to do that, as soon as it supports automatic member number assignment. I'd love to get one of those free Radio Shack scanners and have Koha check out items using that. (-: Anyway, Chris has the script and will hopefully get it out with 1.07 or sooner. Enjoy, Glen

1 0

KOHA : Please unsubscribe
by Laurence Zerafa 15 Dec '00

15 Dec '00

Presently not using service. Please unsubscribe till further notice. laurencezerafa(a)hotmail.com Laurence _________________________________________________________________________ Get Your Private, Free E-mail from MSN Hotmail at http://www.hotmail.com.

1 0

please unsubscribe
by michael collins 15 Dec '00

15 Dec '00

mcollins(a)ihug.co.nz

1 0

MARC Records
by Jonathan Moore 14 Dec '00

14 Dec '00

-------------------------------------------------------- ------------------------ >MARC records have an astonishing amount of detail, both >in terms of data ... >possibility of adding a field to bibliosubject to > >specify a subject > heading system used? The system we are using seems to use the redirection entries that someone mentioned. I think we must have a whole table of entries that say to place a particular marc entry tag into a particular area of the programs database. It also seems to have a similar table for exporting. It would be nice to program this flexibility into Koha. I think it would help down the road in supporting other input formats such as the differences between USMARC and UKMARC. I agree with the idea of storing the marc record en mass in the database. There is a lot of use for being able to export it out at a later time, plus I think a lot of librarians are going to want to be able to modify their collection data from a marc oriented interface. >I also have a request. Could people send me some >samples of their MARC >records so I can do some sanity checking? A lot of my >MARC records are >not from "formal" sources, but were hand entered by >the librarian here. I have posted a couple of marc record files to our ftp server. I am working on a data conversion to get our data out of the system we are currently using to something new, possibly Koha. The old system doesn't place everything in the correct place so keep that in mind. Some of the information is being put in the 900+ range instead of the "correct" location. For example the one file with bar code information has it in the 945 range and it should be a sub field of the 852 tag. One of the files is quite large 30 megs it contains marc records for just about every item in our high school collection. There should be some good examples of multiple records that should when processed have a single shared marc bib record. ftp://usd465.com/pub/marc-samples/ I'm also curious to know how librarians typically use commercial CDs of MARC records. Do they import all of the records into their library system at once, or just import records for books they hold? How are the MARC records typically stored on a CD? Is it just one big file full of MARC records? The way our librarians want to use MARC is done by an add on to the Follet system. It is a yearly cd subscription of common titles/all titles you are likely to have. There are two main uses 1. Run it through your existing holdings data and it will insert correct mark records for the items. 2. Use it when you get new titles. The marc cd is searchable. Search using whatever field you desire: title, author, isbn, etc. Once you find the title click on add copy and put in copy specific data. You are down. I think we order books pre bar coded, so I think they look up the Marc record, click create copy, scan the bar code and enter the price. Some thoughts on determining when to create a new marc bib record in the system. I think in most cases that if you have items with the same title and author then they are probably different copies of the same intellectual property and hence should be link together in the system in some we whether that is by a common marc record or some other method. What if we had 2 configurable threshholds? I list of one or more boolean conditions that must be true for automatic combining and one for queried combining. For example If author and title and publisher are all the same that the items are automatically combine in some manor or cross referenced. If the author and title are the same and nothing else then query the user if he/she wants to combine or cross reference. What are you using to do the Marc processing? I have been looking at the MARC.pm code with the same idea in mind, but I haven't done any coding yet. I have just been looking at the documentation to get an idea of what it can do. It seems pretty powerful and flexible. The website for MARC.pm is http://marcpm.sourceforge.net/ PS I am not sure how this works but I remember reading on the Library of Congress site on marc record format that there are four kinds of marc records 1. Bibliographic 2. Authority 3. Holdings 4. ??? These could be all wrong I am just going by memory. I didn't see any obvious way that these are flagged anywhere. For example Follet seems to treat any marc record without a bar code as a bib record and anything with a bar code can end up creating both a bib record and a holding record that is tied to it. Sorry to go on so long, but while we are talking about MARC record cds, how hard would it be and what legal issues are their on creating a shared marc record library for Koha? Either something that could be accessed live from remote, or perhaps a distributed network where groups of libraries inter publish marc records. Does this already exist somewhere from a non commercial source? Jonathan Moore District Technology Coordinator Winfield Public Schools 316.221.5100 Technology Office 316.221.0508 Office Fax --------------------------------------------------------------- Visit Winfield Public Schools' website at http://www.usd465.com

2 1

Re: [Koha] MARC Records
by kaye.guidetti＠wesley.com.au 13 Dec '00

13 Dec '00

As a librarian my suggestion would be to combine all subject headings into one field and allow some sort of check against the authority file (ie. do you want to accept it or not if it does not match an existing subject heading). This would allow the librarian to reject sears headings if you use LC etc. Alternatively allow something in set up that allows you to decide at the start if you will accept only the fields which contain sears headings or only the fields that contain LC headings and ignore the others. I know nothing about programming so I don't know how easy this is to do - or if it fits in with what you are trying to do Tonnesen Steve <tonnesen(a)cmsd.bc.ca>@lists.katipo.co.nz on 13/12/2000 07:19:19 Sent by: koha-admin(a)lists.katipo.co.nz To: Koha Open Source Library System <koha(a)lists.katipo.co.nz> cc: Fax to: Subject: [Koha] MARC Records MARC records have an astonishing amount of detail, both in terms of data about the book, as well as ways in which the data can be described. I'm wondering how I should handle the first difficulty: subject headings. MARC records have several different entries for subjects: 650 Subject added entry -- Topical term Subjects like Mammals, Nuclear Warfare, Law, Alcohol, Child Welfare, etc. Appears to be the most common type of subject entry, at least with my data. 651 Subject added entry -- Geographic name Subjects like Canada, Rome, Moon, China, British Columbia, etc. 600 Subject added entry -- Personal name Subjects like Plato, Helen Keller, Charlemagne, etc. In addition each subject entry also specifies a source "system" such as Library of Congress Subject Headings or Sears Subject Headings. Obviously Koha doesn't have this complexity of subject headings, and probably doesn't need it, but I'm suspecting that we probably don't want to mix subject headings from different cataloging systems. Any possibility of adding a field to bibliosubject to specify a subject heading system used? Also, should I be adding any subject headings gathered from MARC records to the catalogueentry table (which I assume is used to deter librarians from adding new, conflicting subject headings like "Cows" and "Bovines"). I also have a request. Could people send me some samples of their MARC records so I can do some sanity checking? A lot of my MARC records are not from "formal" sources, but were hand entered by the librarian here. I'm also curious to know how librarians typically use commercial CDs of MARC records. Do they import all of the records into their library system at once, or just import records for books they hold? How are the MARC records typically stored on a CD? Is it just one big file full of MARC records? Steve. _______________________________________________ Koha mailing list Koha(a)lists.katipo.co.nz http://lists.katipo.co.nz/mailman/listinfo/koha =========================================== This email message has been swept by MIMESweeper for The Wesley Group of Hospitals- Australia Although this email has been scanned for any known virus, please ensure that due care is taken and correct virus management procedures are maintained. If you have any questions contact the Help Desk on x7029 or via email at help(a)wesley.com.au ===========================================

3 2

Re: [Koha] biblio vs. biblioitems
by Tonnesen Steve 13 Dec '00

13 Dec '00

On Wed, 13 Dec 2000, Rosalie Blake wrote: > Chris has already answered, but I'll just add the rationale - which > was that biblio describes the intellectual property. So author title, > copyright date. > > Then any copies we have of this piece of copyright, whether a large > print edition, or a hardback or paperback is attached to the biblio > with the ISBN, publishing info, location in the biblioitem record. If > we've got two large print copies, they'll share one Biblioitem record, > and each copy will have its own item record. That makes sense to me. It looks to me like MARC records don't make that distinction, though. If there is a normal and large print edition, they will have seperate MARC records. I guess the title,author,copyright date combination is always unique and can be used as a key to determine if the two records actually refer to the same "intellectual property" then? If I wanted to import MARC records, and I had a MARC record for a regular David Copperfield book and a MARC record for a David Copperfield video, would they really have the same copyright date? In any case, I would have to be able to recognize somehow when entering the video into the Koha database that a biblio already exists for the item (because the book was entered previously, for example) and then link the biblioitem entry for the video back to the original biblio entry for the book. Not impossible, but I sense some possibility for errors if bibliographic data is being entered automatically from MARC records. Steve.

2 1

Re: [Koha] biblio vs. biblioitems
by Simon Blake 13 Dec '00

13 Dec '00

On Wed, Dec 13, 2000 at 12:13:02PM +1300, Rosalie Blake said: > Hi again > > Chris has already answered, but I'll just add the rationale - which > was that biblio describes the intellectual property. So author title, > copyright date. > > Then any copies we have of this piece of copyright, whether a large > print edition, or a hardback or paperback is attached to the biblio > with the ISBN, publishing info, location in the biblioitem record. If > we've got two large print copies, they'll share one Biblioitem record, > and each copy will have its own item record. At some stage after the table name was well buried in lines of code, we decided that a better name for biblioitem would be "formats". And you never know, sometime in the future we may get around to a search and replace and change it in the code. Cheers Si -- Simon Blake simon(a)katipo.co.nz Katipo Communications +64 25 300 825

1 0