Recently I have seen people giving advice how to use CPAN to install the perl modules that Koha depends on. Now for non .deb based systems (Redhat, Centos, etc) this advice is ok. But if you are use a Debian based distro, like Debian, Ubuntu, Mint, etc please use the packages.
If it is not in your distro proper, you can get the packages from debian.koha-community.org. As new dependencies are added, we package them and submit them to Debian, so they end up in the next release, but you can always get them from debian.koha-community.org until then.
A really easy way to get them is of course to just install the koha-common package from there. You can read good instructions at http://wiki.koha-community.org/wiki/Debian or there is the koha-perldeps package which only pulls in the dependencies.
Why not use CPAN? Because then it is up to you to make sure you keep all your installed modules up to date, and with the latest security patches. Or you could let the package maintainers worry about that instead.
If you do bump into a module that isn’t packaged, let us know and we will fix that right away.
So for no real reason, other than to see if I could, I figured out how to get the top 10 changed (well it could be any number) files in the Koha code base
git rev-list --objects --all | awk '$2' | sort -k2 | uniq -cf1 | sort -rn |
while read frequency sample path
[ "blob" == "$(git cat-file -t $sample)" ] && echo -e "$frequency\t$path";
done | head
This told me (where the number is the number of commits)
- 3006 installer/data/mysql/updatedatabase.pl
- 1220 C4/Biblio.pm
- 1141 C4/Search.pm
- 1137 kohaversion.pl
- 997 C4/Circulation.pm
- 996 installer/data/mysql/kohastructure.sql
- 914 C4/Auth.pm
- 796 opac/opac-search.pl
- 752 circ/circulation.pl
- 710 opac/opac-detail.pl
You could of course leave the head off to get all or do head -20 to get the top twenty etc.
Like I said, useless statistic, but kinda interesting nonetheless
This year we had 6 students working on the Koha project as part of their project work for the Catalyst Open Source Academy and a very productive 3.5 days it was too.
In total we submitted 29 new patches, tested and signed off on 9 more, and even rescued one patch from languishing in the ‘patch does not apply’ status.
So far most of the patches have been signed off, and some have passed QA, and a few have even been committed into the codebase. In fact the upcoming 3.18.3 Koha release will have 5 patches from 3 of the students from 2015, and 1 patch from a former Academy student who is now an intern at Catalyst.
After all that hard work, I figured they deserved a cupcake, so I asked the wonderful Sweet Release Cake and Treats to make some, and we ended up with these, which everyone loved.
A++ would academy again.
2014 was another big year for Koha, there are now 2847 libraries in Libwebcats listed as using Koha. That doesn’t include 1037 public libraries in Argentina, the 1106 public libraries in Turkey, and the multitude more in many other countries.
On top of the ever growing number of users, many more features were added, and bugs fixed, here are some statistics for the year:
- There were 2225 commits to the master branch of the Koha codebase
- The busiest month was May with 278 commits
- 1570 Patch sets were signed off
- 1299 Patch sets Passed QA
- 673 Patch sets Failed QA (almost all were fixed and went on to pass in the end)
- Katrin Fischer signed off on 47, and QAed 815
- Jonathan Druart – was author of the year with 325 patches, he also signed off on 74 and QAed 336
- 87 People had patches in Koha
- There were 34 people who got their first patch in Koha this year
Yesterday Kahurangi came to work with me, as he likes to do when he is on school holidays, but this day was a little bit different. Normally Kahu spends most of the day reading, and playing on the tablet, but yesterday he decided he wanted to do some work.
One of the bonuses of working on Free Software is that if your child says “Can I do some work with you?” you can say “Yes, lets fix a bug together” knowing that there is a good chance you can get your patch submitted.
So the two of us sat down to do some pair programming, Kahu at the keyboard and me beside him. It took us about 30 mins to get his dev environment set up, and check out a copy of the Koha source code. I knew of the perfect bug, while we were having our Koha team Xmas function (Bubbles and bugs, drink some bubbles and fix some bugs) we spotted that one of the developers names was missing from the about page. A perfect low barrier to entry bug to fix, but still one well worth fixing. It took us a little while to find the file that needed editing (got distracted showing him how to use locate) but once we found it, Kahu found the place that needed editing (the names are in alphabetical order so it wasn’t too hard) and we added the missing name. We then committed the fix to git, and created and attached a patch to the bug in bugzilla (bugs.koha-community.org)
Last night, Katrin (from Germany) checked and signed off the patch and this morning the Release manager (Tomás from Argentina) pushed it to the master branch. So now Kahurangi has a patch in Koha. http://git.koha-community.org/gitweb/?p=koha.git;a=commit;h=70de2cae53d93a47a7666015a79f35d7f458f4fe
I think we may be the first parent and child team to both have patches in Koha.
(And yes, that is a Cat Stevens lyric)
Paul presented about Coral which is an ERM (Electronic Resource Management) software. Paul told us it is well documented, with an extensive manual. It is software for Librarians to use, not end users. It is for managing all your subscriptions for digital resources. Large universities often have hundreds of subscriptions and tracking licenses and url, and acquisitions etc are very hard to track. Coral was developed to manage this.
It is not packaged, but is still pretty easy to set up. Setting up authentication is the hardest bit, but once you have done that, the rest is pretty easy. It has one big problem it is only available in English and it currently cannot be translated. Hopefully since it is OSS we can fix this issue, but it is a big task.
Coral is divided into modules
- Define all publishers, vendors, partners
- Define all licenses
- Highly flexible
- You can attach files like pdf
- This tracks the subscription
- You can define workflows
- store access details
Paul then demoed the software for us.
My battery on the laptop went flat, so Tom kindly took notes on this session for me so I could write this blog post
Alvet and Ricardo from EBSCO talked about EDS and the integration work they have done with Koha
To start they began by explaining why they had developed this plugin
- NIWA, a research institute in NZ, wanted a simple interface where their users could access discovery services
- An a interface where Koha was the Front end
They then explained what a discovery service is:
- A way to access all of the library’s full-text content (electronic and print) in a single search
- High quality metadata = high quality search results
- Relevancy ranking
- Match on subject heading
- Match on article titles
- Match on other keywords
- Match on keywords in abstracts
- Match on full text
Next the showed us what EDS looks like in its’ native interface so that they could show that it is quite similar in Koha using the plugin
Then they showed what EDS and Koha look like
- showed the NIWA search (Discovery) box, including the detailed list of field codes you can search on.
- Demoed the Catalyst work to use tabs instead of drop-downs to segregate the search options
- When the drop downs are changed a pop-up appears that notifies you of the search mode
- demoed the integration between the Koha catalog and the EDS resources, including patron services, accessing EBSCO resources through the koha interface, limiting the resources searched to what the library has available,
- reserving resources across Koha/Ebsco and cart functionality
- demoed guest access restrictions to show that limited resources will not be displayed unless authorized
- demoed login and authorization methods (userid/passwords, ip addr restrictions, etc)
- demoed search options (options per page, boolean search options, sort, etc.)
- demoed interaction between search results and checkboxes including the functionality that depends on those boxes (more details, etc)\
- demoed accessing EBSCO content from the Koha interface (html results, PDF, etc)
- showed that all EBSCO functionality is available through Koha interface in advanced search and that the Koha advanced search can be toggled through tabs/links including add/remove of search
They explained that support is only available in 3.12+ because integration has been implemented as a plug-in. Plugins allows for features to be added quickly through plug-ins in-between release Koha cycles.They then stepped us through how you install and configure the plugin
Improvements for the future
- Support for newer versions of Koha (3.16, 3.18) (ready at the end of October)
- Research Starters
- Multi-facet support
- Looking to enable default language support in Koha and the Ebsco plug-in
Where do I get it
If you google for EDS API Koha, you will find a github page that has the newest version available for download. EBSCO also provide a wiki, an Integration Kit as well as training and help