 |
 |
*** Cross-posted to code4lib, digipres, digital-preservation, diglib, dlf-announce, jhove-users, padiforum-l *** To the JHOVE/JHOVE2 user community: An alpha prototype version of JHOVE2 is now available for download and evaluation (v. 0.5.2, 2009-08-05). Distribution packages (in zip and tar.gz form) are available on the JHOVE2 public wiki at (http://confluence.ucop.edu/display/JHOVE2Info/Download). The new JHOVE2 architecture reflected in this prototype is described in the attached architectural overview (also available at http://confluence.ucop.edu/display/JHOVE2Info/Architecture). The distribution package contains two driver scripts in the JHOVE2 home directory: a DOS shell script (jhove2.bat) for Windows and a Bourne shell script (jhove2.sh) for Unix/Linux. These scripts MUST be modified to define the "JHOVE2_HOME" and "JAVA" variables to point to the JHOVE2 installation directory and the "java" command. jhove2.bat (as distributed): JHOVE2_HOME="C:\Program Files\jhove2" JAVA="C:\Program Files\java\jre1.6.0_13\bin\java" jhove2.sh (as distributed): JHOVE2_HOME=/usr/jhove2 JAVA=/usr/bin/java You can verify the installation with the command (for Unix): % ./jhove2.sh test which should produce results similar to (http://confluence.ucop.edu/display/JHOVE2Info/Sample+Results). The prototype supports the following features: * Appropriate recursive processing of directories and Zip files. * High performance buffered I/O using the Java nio package. * Message digesting for the following algorithms: Adler-32, CRC-32, MD2, MD5, SHA-1, SHA-256, SHA-384, SHA-512 * Results formatted as JSON, text (name/value pairs), and XML. * Use of the Spring Inversion-of-Control container for flexible module configuration. * A complete UTF-8 module. * An minimally functional Shapefile module. Please be aware of the following limitations and and caveats: * This prototype is being made available now to provide you with an early look at the new JHOVE2 architecture and APIs. While the full processing model is demonstrated, there is limited format support at this time. * DROID is not yet integrated into the identification module. The identification module has stub code that can recognize a Zip file; all other files are assumed to be UTF-8. * The aggregate-level identification module (i.e. the "aggrefier" module) has stub code that can recognize a Shapefile formed by the files "abc.shp", "abc.shx", and "abc.dbf". * There is no assessment module available for review at this time. * Memory mapped I/O buffers are not yet fully supported. * JHOVE2 requires a 1.6 JRE. The project team is now working on additional format modules. These will be added to the public distribution as they become available. We would very much like to receive your feedback on the new code. While the current state of the code is the product of much internal review and refactoring, your evaluations and suggestions, based on a wide diversity of experience and needs, will be welcome as we continue to move forward with our work. Please direct your comments and suggestions to the "JHOVE2-TechTalk-L" mailing list <http://confluence.ucop.edu/display/JHOVE2Info/Mailing+Lists> for community discussion. Thank you, Stephen Abrams / California Digital Library Tom Cramer / Stanford Univesity Evan Owens / Portico on behalf of the JHOVE2 project team