Hi Oliver,
Thanks for the contribution! I've taken a look over the code and I have
a couple of questions:
1) Why is the preparsing step required ? Is there something too complex
for antlr to handle in there ?
2) If the preparsing step is really needed, could we use jakarta-regexp
as the regular expression library instead of jakarta-oro. It is just
that jakarta-regexp is already present in the TM4J distribution and I
would prefer not to have two regular expression libraries in there!
Great work BTW :-)
Cheers,
Kal
Oliver Leimig wrote:
>Hi there,
>
>I'm contributing the AsTMa= importer 4 TM4J, see attachment.
>
>-Imports AsTMa= topic maps (.atm) to TM4J
>-the included jakarta-oro jar is required, J2SE1.3 compliant
>-creates Topic Maps in TMAPI and also returns Topic Maps in TM4J-API
>-resides in org/tm4j/topicmap/utils (ATM*)
>-test cases are provided in org/tm4j/topicmaps/utils/test
>-further AsTMa= topic maps can be obtained from
> http://astma.it.bond.edu.au/authoring.xsp e.g. the sumo topic map
>-Usage: org.tm4j.topicmap.TopicMap t = provider.addTopicMap(new
>SerializedTopicMapSource(filename.atm));
>-ATMPreLexer.java (temporarily?) has a main()-method for easy usage test
>-Call sequence: YourClass, ATMBuilder, ATMPreLexer, ATMLexer and ATMParser
>-ASTMAImporter.java is the integration of the AsTMa= importer in
>tmapi-utils(.sf.net),
> tmapi-utils usage: org.tmapi.core.TopicMap tMap =
>astmaImporter.parse(filename.atm);
>-ATMParser.java, ATMLexer.java and ATMParserTokenTypes.java can be generated
>from
> the antlr grammar file ATM.g by executing "java antlr.Tool ATM.g"
>-tested with tm4j 0.9.6 and 0.9.7
>-should work (tm)
>
>
>Oliver Leimig
>
>
|
