Browse Prior Art Database

Method to identify processing for overridden file types

IP.com Disclosure Number: IPCOM000014199D
Original Publication Date: 2003-Jun-19
Included in the Prior Art Database: 2003-Jun-19
Document File: 1 page(s) / 38K

Publishing Venue

IBM

Abstract

Disclosed is a method to identify the types of processing intended for overridden file types. This is a problem in handling translations of human language and in building during product development. With the rise of multipurpose source type files (XML, Java are examples) the file type or extension is no longer dependable for determining the type of processing that the file is intended to undergo. This is a problem for automated systems such as build and translation systems.

This text was extracted from a PDF file.
This is the abbreviated version, containing approximately 88% of the total text.

Page 1 of 1

Method to identify processing for overridden file types

Build and make utilities determine the processing necessary from the file extension of the source file; mapping the type with an action to take. For example, Java* may be associated with 'javac' as the Java compiler. However the file may also need to be processed as a document source file.

Translation XML files may be the source for multiple output files; the source .xml file may be output as a xpg4 message catalogue and also as a Java resource bundle, both of which require different compilers to process them. This may be OK during compilation since the output files are mapped 1-1 by make to the compiler required; however when doing the acceptance testing of the xml source following translation the file may be in an acceptable format for zml and even for one of the types of generated source output but not for all. Thus the file may be accepted with flaws that will not be discovered until build time. Which will require the translator to fix, causing a build break that cannot be remedied quickly - and this negates the acceptance testing.

Since it is not possible to know the intended output formats of the files by their source alone it is necessary to have knowledge of the intended output formats and manually monitor the files.

The solution is to encode an 'intended output format' tag into the header of the file either as an xml meta tag or as a comment of other types of source files. This field can be read by...