- Koha-devel - lists.koha-community.org

Database size with InnoDB
by Nahuel ANGELINETTI 16 Jan '09

16 Jan '09

Hi people, I don't know if some people have a big experience from InnoDB and MySQL. We just see in mysql documentation innoDB does not reallow the disk space of deleted datas. An example : the disk used by one of our clients, who have only 50 000 notices, is 4GB and is still growing. The explication is innodb don't delete data from disk after a "drop" or a "delete". http://dev.mysql.com/doc/refman/5.0/en/adding-and-removing.html "Currently, you cannot remove a data file from the tablespace. To decrease the size of your tablespace, use this procedure: " The solution to decrease the size of data file, is to dump, and restore the database... A really bad solution for production installations. Does someone have experimented the same problem? What are your solution about this problem ? bests, -- Nahuel ANGELINETTI

3 3

Another UTF-8 encoding question
by Chadwick, John, DCA 15 Jan '09

15 Jan '09

Play time is over and I am now building a couple of servers using Ubuntu 8.04 and the latest git version of Koha. So far the installations have been clean and easy. I have just one more UTF-8 encoding question. Being in the southwestern U.S. we are trying to deal with some titles that have Spanish character sets to go along with our primarily English character sets. In looking over Koha Wiki entry for encoding and character sets, http://wiki.koha.org/doku.php?id=encodingscratchpad., I have a question about the section on combining characters and collations. The search entry may not necessarily have the special character, but we need to return records with special characters. For those who have had to deal with this, which would you recommend, utf8_unicode_ci or utf8_general_ci for the collation collection? The word Univerzalitás is a unicode combining form. When you copy/paste it into a text editor or use a keyboard to type it, it is most likely going to be the non-combining form: Univerzalitás. (in the non-combining form, the hex for the accented a is: Hex 0301; for the non-combining form it’s: Hex 61, Hex 00e1). Non-combining form: http://www.fileformat.info/info/unicode/char/00e1/index.htm Combining form: http://www.fileformat.info/info/unicode/char/61/index.htm http://www.fileformat.info/info/unicode/char/0301/index.htm Univerzalitás Univerzalitás It seems that the utf8_general_ci collation doesn’t support equality for those two forms. However, utf8_unicode_ci seems to work. If you have combining characters in your data, you may want to go with statements like: ALTER TABLE marc_word MODIFY word VARCHAR(255) CHARACTER SET utf8 COLLATE utf8_unicode_ci; and be sure to add init-connect = ‘SET collation_connection = utf8_unicode_ci’ to your my.cnf Thanks, John +----------------------------------------------------------------------------+ John Chadwick, Ed.D. Information Technology Manager New Mexico State Library 1209 Camino Carlos Rey Santa Fe, NM 87507 Phone: 505-476-9740 Cell: 505-629-8116 Fax: 505-476-9761 john.chadwick(a)state.nm.us http://www.nmstatelibrary.org Confidentiality Notice: This e-mail, including all attachments is for the sole use of the intended recipient(s) and may contain confidential and privileged information. Any unauthorized review, use, disclosure or distribution is prohibited unless specifically provided under the New Mexico Inspection of Public Records Act. If you are not the intended recipient, please contact the sender and destroy all copies of this message. -- This email has been scanned by the Sybari - Antigen Email System.

1 0

Fixing holding info after an import
by Jason Macki 13 Jan '09

13 Jan '09

Hello, I'm doing an import from another library system to Koha version 3 on Linux. I used the bulkmarcimport.pl script to import the bibliographic information, populating the "biblio" and "biblioitems" tables. Then I ran a program to fill in the holding information by directly inserting rows into the "items" table. That worked, except that the "biblioitems" table has a field called "marcxml", which includes the holding information for that book. Now, if I use the Koha management application to edit and save a holding item, it fills in the holding information in the "marcxml" field of the corresponding "biblioitems" record, and the Koha features which depend on the holding information in the "marcxml" field work properly. My question is, is there a script in Koha 3 that will automatically put the holding information contained in the "items" table into the "marcxml" field of the "biblioitems" table? Thanks very much, Jason

3 2

strange recent indexing problems on koha 3.0.x branch, anyone?
by Mason James 13 Jan '09

13 Jan '09

Hi All The basic issue seems to be that the latest koha 3.0.x branch has some strange indexing (and encoding?) errors (for me at least) i have spotted these 'Invalid XML encoding name' errors when running the zebraqueue_daemon.pl / koha-zebraqueue-ctl.sh scripts - after adding a new bib via... * a z39-pull * web-gui batch-import * cmdline bulkmarcimport.pl script i can replicate this on 2 separate debian-etch boxes rolling my repo back from the latest 3.0.x commit, to the 'v3.00.00' tag fixes the issue (which is good, but i havent clicked to what the issue actually is yet ;/ ) the best way to replicate the indexing error for me, was this... fetch, and rewind the repo $ git-fetch $ git-fetch --tags #needed too ?? $ git rebase origin/3.0.x $ git reset --hard v3.00.00 $ ./misc/migration_tools/bulkmarcimport.pl -d -v -fk -n 10 -file / tmp/sample.mrc deleting biblios .......... 10 MARC records done in 0.157172918319702 seconds $ ./misc/bin/zebraqueue_daemon.pl -f ~/etc/koha-conf.xml 22:52:52-05/01 zebrasrv(17) [session] Session - OK 17 unix:/home/ mason/koha/ashs/var/run/zebradb/bibliosocket 26837 22:52:52-05/01 zebrasrv(17) [request] Auth idPass kohauser - 22:52:52-05/01 zebrasrv(17) [request] Init OK - ID:81/81 Name:ZOOM-C/ YAZ Version:1.116/2.1.54 22:52:52-05/01 zebrasrv(17) [log] function: 1 22:52:52-05/01 zebrasrv(17) [log] Waitaction: 2 22:52:52-05/01 zebrasrv(17) [log] Received DB Update 22:52:52-05/01 zebrasrv(17) [log] action 22:52:52-05/01 zebrasrv(17) [log] specialUpdate 22:52:52-05/01 zebrasrv(17) [log] database: biblios 22:52:52-05/01 zebrasrv(17) [log] enabling shadow spec=/home/mason/ koha/ashs/var/lib/zebradb/biblios/shadow:4G 22:52:52-05/01 zebrasrv(17) [log] cache_fname = /home/mason/koha/ashs/ var/lib/zebradb/biblios/shadow/cache 22:52:52-05/01 zebrasrv(17) [log] record 0 type XML 22:52:52-05/01 zebrasrv(17) [log] 2095 bytes: <?xml version="1.0" encoding="UTF-8"?> <record xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance" xsi:schemaLocation="http://www.loc.gov/MARC21/slim http://w ... 22:52:52-05/01 zebrasrv(17) [log] update grs.xml <no file> 0 22:52:52-05/01 zebrasrv(17) [log] Iterations: isam/dict 8/2 22:52:52-05/01 zebrasrv(17) [log] Dict: inserts/updates/deletions: 1/0/1 22:52:52-05/01 zebrasrv(17) [log] Records: 1 i/u/d 0/1/0 22:52:52-05/01 zebrasrv(17) [request] Extended Service: Update (done) 22:52:52-05/01 zebrasrv(17) [log] function: 1 22:52:52-05/01 zebrasrv(17) [log] Waitaction: 2 22:52:52-05/01 zebrasrv(17) [log] adm request database biblios 22:52:52-05/01 zebrasrv(17) [log] adm-commit ... SUCCESS!!!!!!!!!!!!!!!!!!!! ---------------------------------------------------- roll the branch forward... $ git rebase origin/3.0.x $ ~/git/stable/misc/migration_tools/bulkmarcimport.pl -d -v -fk -n 10 -file /tmp/sample.mrc deleting biblios .......... 10 MARC records done in 0.157172918319702 seconds $ ./misc/bin/zebraqueue_daemon.pl -f ~/koha/ashs/etc/koha-conf.xml & $ sudo tail /var/log/syslog Jan 5 23:58:35 fugu Koha Zebraqueue [27085]: Starting Zebraqueue log at Mon Jan 5 23:58:35 2009 Jan 5 23:58:35 fugu Koha Zebraqueue [27085]: Mon Jan 5 23:58:35 2009 POE Session Jan 5 23:58:36 fugu Koha Zebraqueue [27085]: 10 operations waiting to be run Jan 5 23:58:36 fugu Koha Zebraqueue [27085]: Executing zebra operations Jan 5 23:58:36 fugu Koha Zebraqueue [27085]: biblioserver record 10 is malformed: Entity: line 1: parser error : Invalid XML encoding name Jan 5 23:58:36 fugu Koha Zebraqueue [27085]: biblioserver record 9 is malformed: Entity: line 1: parser error : Invalid XML encoding name Jan 5 23:58:36 fugu Koha Zebraqueue [27085]: biblioserver record 8 is malformed: Entity: line 1: parser error : Invalid XML encoding name Jan 5 23:58:36 fugu Koha Zebraqueue [27085]: biblioserver record 7 is malformed: Entity: line 1: parser error : Invalid XML encoding name Jan 5 23:58:36 fugu Koha Zebraqueue [27085]: biblioserver record 6 is malformed: Entity: line 1: parser error : Invalid XML encoding name Jan 5 23:58:36 fugu Koha Zebraqueue [27085]: biblioserver record 5 is malformed: Entity: line 1: parser error : Invalid XML encoding name Jan 5 23:58:36 fugu Koha Zebraqueue [27085]: biblioserver record 4 is malformed: Entity: line 1: parser error : Invalid XML encoding name Jan 5 23:58:36 fugu Koha Zebraqueue [27085]: biblioserver record 3 is malformed: Entity: line 1: parser error : Invalid XML encoding name Jan 5 23:58:36 fugu Koha Zebraqueue [27085]: biblioserver record 2 is malformed: Entity: line 1: parser error : Invalid XML encoding name Jan 5 23:58:36 fugu Koha Zebraqueue [27085]: biblioserver record 1 is malformed: Entity: line 1: parser error : Invalid XML encoding name FAIL !!!!!!!!!!!!!!!!!!!! any ideas, can anyone else replicate this, or confirm otherwise? Cheers, Mason

3 3

Re: [Koha-devel] [Koha] Issues while installig Koha 3.00 on redhat 5.0 linux
by Rick Welykochy 13 Jan '09

13 Jan '09

Mason James wrote: > On 2009/01/13, at 6:24 PM, Uma Mukkamala wrote: > >> I am not able to install CPAN also. Basically i run my internet on >> proxy i am giving the proxy details while installing the CPAN, but >> i have been structed at the following point >> >> Select your continent (or several nearby continents) [] 4 >> invalid items entered, try again >> >> i have entered both values like 'Asia' and 1 but i am not going >> forward from this point. Please suggest. > > > try this > > http://sial.org/howto/perl/life-with-cpan/ This quote from the above article is mainly for Koha developers: "Perl modules used in production must be saved locally, either under version control, a package system, or software depot. Doing so ensures the modules can be installed on new systems, or rebuilt on new platforms. This supports reproducible module installs over time, as CPAN modules may change or be removed without warning." i.e. CPAN is a network of perl modules, subject to change at any time. cheers rickw -- _________________________________ Rick Welykochy || Praxis Services All truth passes through three stages. First, it is ridiculed. Second, it is violently opposed. Third, it is accepted as being self-evident. -- Arthur Schopenhauer

2 1

Re: [Koha-devel] Git Repository Questions
by Michael Hafen 10 Jan '09

10 Jan '09

-------- Forwarded Message -------- From: Gerald Van Baren <vanbargw(a)gmail.com> To: Michael Hafen <mdhafen(a)tech.washk12.org> Subject: Re: [Koha-devel] Git Repository Questions Date: Fri, 9 Jan 2009 15:05:15 -0500 On Fri, Jan 9, 2009 at 12:06 PM, Michael Hafen <mdhafen(a)tech.washk12.org> wrote: > I don't know if it is possible to clone a git repo over http. There is > a web interface to browsing a repo, git-web, but I don't think that is > suitable for cloning. I think the standard protocols for cloning a repo > are git:// ssh:// and maybe ftp:// > > I'm sure someone on this list can correct me if I'm wrong. git:// is a rsync-based protocol that can run over ssh if the repository server is set up to do that - gitosis runs git:// over ssh, plain git-daemon runs git:// unencrypted. <http://scie.nti.st/2007/11/14/hosting-git-repositories-the-easy-and-secure-…> <http://www.kernel.org/pub/software/scm/git/docs/git-daemon.html> Git can clone over http:// (simply use http://), assuming the git repository server is set up to support this (I don't know what is involved, never having done it). Cloning/pulling over http:// is somewhat inefficient and is a *lot* more fragile than using the git:// protocol. <http://www.kernel.org/pub/software/scm/git/docs/howto/setup-git-server-over…> HTH, gvb

1 0

authorities question
by Rafael Antonio 09 Jan '09

09 Jan '09

Dear all, I am installing KOHA 3.0 with UNIMARC flavour. When I define authority types if I leave blank field Summary all the information is seen but the link just give ".../cgi-bin/koha/detail.pl?authid=" and I receive an error message because there is no argument. If I fill Summary with [200a] [200a] that information appears in one large link with authid=12 which allow to acess the authority record. Does someone can help? What is wrong? Where is sintax to use [200a] Regards, Rafael Antonio

1 0

Git Repository Questions
by Kyle Hall 09 Jan '09

09 Jan '09

Hello All, I'm trying to do something that I'm sure is doable with git. I would like to set a 'middle-man' git repository that I can push and pull from to manage our development and production servers for our switchover to koha 3 and future development. (git.koha.org) --> (MyKohaGit) <---> (KohaDevServer) \-------------> (KohaProductionServer) So MyKohaGit pulls changes from git.koha.org, I pull and push changes between MyKohaGit and KohaDevServer, and when I'm satisfied that the changes are production ready, I pull the changes from MyKohaGit to KohaProductionServer. I know I could just go straight between KohaDevServer and KohaProductionServer, but I'd like to make MyKohaGit public so others can track my updates that haven't made it into the official git repo yet. Is it really as simple as cloning git.koha.org and making it available via http/ssh? I've tried creating an empty repository and pushing to it from a clone of git.koha.org, and it populates it, but I don't seem to have a mechanism to update MyKohaGit directly from git.koha.org. Does any of this make sense? Thanks, Kyle http://www.kylehall.info Information Technology Crawford County Federated Library System ( http://www.ccfls.org )

3 4

how to start customization
by Zahoor 08 Jan '09

08 Jan '09

Hello every body, I have got the clone of koha using the Git after configuring the koha on Debian 4.0. Kindly let me know how can I set development environment to customize koha according to the requirement of our own library. Clone of koha on my local system has all the code of koha modules. Should I access those modules and make changes in required and after saving that locally should commit it using Git? Please help me as I am new entrant in Perl and koha. Have a nice time and thanks in advance.

2 1

Koha3 populate
by Matteo Romanello 08 Jan '09

08 Jan '09

Yes, the installation usually offers that chance. But in my case, for example, installing Koha3 I had to force some installation steps manually and as a result the DB testing data weren't correctly loaded. Like Rick I will find useful to have those data available as a stand- off download or such. Are there already? Best Matteo Romanello >> Anyone have a script or SQL file(s) to populate a Koha 3 database >> with data for testing and development? > > Isn't that an option offered during installation? > > -- Owen > > -- > Web Developer > Athens County Public Libraries > http://www.myacpl.org

2 1