Polerio writes: "pyCatalog is a Python, MySQL, wxPython, Reportlab application specifically usable in library and information centers. It simply produces book catalog and card catalog in pdf format rendered using reportlab. The program takes MARC file as its source data. Catalog output is one of the core output needed in library software today. Because open source system for libraries as of this date didn't have this kind of feature as most other proprietary library software have today. This includes Koha, Obiblio, and PhpMyLibrary. By making this generalized solution, those open source system for libraries mentioned will be able to take advantage from this application because the software is not centered on one library system. The application simply needs a MARC file in order to output the desired book and card catalog. The application was written in Python to make it platform independent. Thus it will run in Linux, Mac, and Windows environment. The application require MySQL database to manipulate the order of the catalog. Card catalog is a 3x5 page size output that comprises one book bibliographic entry. Book catalog is a full page output that comprise 1 or more bibliograrphic entry. It was named pycatalog, simply because it is written entirely in python, and catalog wholly describes the purpose of this application. I wanted to make this application open source because I believe open source developers will put this application above all expensive proprietary software for libraries. And will make this application fully-functional, I may not have that complete resource to create a world-accepted catalog generator but I think ideas and contributions from the developers around the world would help improve this application. I brought this idea to the public because of the necessity I felt during my project. I have searched all day long at Google to find a catalog generator but I find nothing. Hope this application would be able to help small to large scale libraries around the world."
Ed writes: "Net::OAI::Harvester is a new Perl module available from CPAN for easily interacting with data providers that implement the Open Archives Initiative Protocol for Metadata Harvesting. It provides an object-oriented client interface to OAI-PMH in the same way that LWP::UserAgent does for HTTP. Net::OAI::Harvester is basically a toolkit which allows you to quickly write programs that retrieve data from OAI-PMH archives without having to get your hands dirty generating HTTP requests, parsing large amounts of XML, and representing lots of data in memory.
Net::OAI::Harvester uses XML::SAX internally, so potentially large documents are handled as streams (no large DOMs in memory). An OAI DublinCore module is included for representing Dublin Core metadata; and there is an extension mechanism so that you can plug in your own metadata handler to handle other flavors of metadata. For more information see: http://search.cpan.org/author/ESUMMERS/OAI-Harvester-0.2/lib/Net/OAI/Harvester.pm
If you are interested in contributing at all to the project please express your interest at on the perl4lib list."
Bas writes: "This new version supports full character set conversion between MARC-8 and UCS/Unicode. For UNIMARC character set conversions are supported between ISO5426 and UCS/Unicode and ISO6937 and UCS/Unicode. This new version also fixes some bugs. The goal of MARC4J is to provide an easy to use library for working with MARC records in Java. MARC4J adopts MARC related standards and is tightly integrated with SAX2, a high performance XML interface, to enable robust conversion routines between MARC and XML. MARC4J is written in Java and published under the terms of the GNU Lesser General Public License as published by the Free Software Foundation."
Jeremy writes: "The initial release of the metadata3 application is now available. You can download the .gz file from the project's website. This is version 0.1 of the md3 application, and is a developer's release. The metadata3 (also known as md3) application is a peer-to-peer client/server that promotes the sharing and automatic translation of bibliographic metadata. The current version of the application focuses on the basic peer-to-peer functionality; the next version will add in basic metadata translation capabilities."
Ed writes: "MARC::Charset 0.4 now supports MARC-8 character translations to Unicode (UTF-8) for the East Asian character sets. More information is available at MARC/Perl. Now that there is good support for MARC-8 character conversions to Unicode it is possible to update MARC::XML to do a good solid job of converting to LC's MARC schema."
Bas writes: "This is the first MARC4J release and it includes a lot of new features, including support for LC's MARC XML. Currently it's the only released tool that supports this new format. The MARC4J project is a continuation of James (Java Marc Events)." [note: the last James release was beta 5.]
Mark writes: "The MARC Template Library is a C++ API for reading MARC records using C++ templates and the Standard Template Library. It features an event parser inspired by Bas Peters' James (Java MARC Events), althought I have taken a very different approach by using templates and other features not found in Java. The parser is very fast - able to read over 30,000 records a second on my 450 mhz AMD k3, that is if you don't include the time to load the records from disk. In real world situations the speed is about 5,000 to 10,000 per second. It is still in the very early stages of development - so parts of it are not really usable yet and there's quite a few bugs."
Kevin writes: "A new version of YAZ has been released. YAZ is a C/C++ programmer's toolkit supporting the development of Z39.50v3 clients and servers. A sample client, a sample server, and documentation are included in the distribution. YAZ also includes experimental support for the industry standard ZOOM API for Z39.50. There are several changes from the last version (mostly bug fixes)."
Bas writes: "New in this release is a version of MarcXmlWriter that supports XSLT processing using any JAXP compliant XSLT processor (like SAXON or Xalan) and a RecordHandler interface to support complex conversion routines using MARC record objects.
Added to the site is a tutorial about writing MARC to XML conversion programs using the James API and SAX2 events.
The James API provides a sequential model to access a collection of MARC bibliographic records in tape format. The goal of James is to provide a generic Application Programming Interface (API) for records that conform to the MARC standard.
James (Java MARC Events) is written in Java and published under the terms of the GNU Lesser General Public License as published by the Free Software Foundation. Downloads and additional information, including the tutorial and Javadoc documentation, can be found at www.bpeters.com."