On 18 July 2012 13:46, Mikel Artetxe <[email protected]> wrote: >> Looks good, but ticking "Mark ambiguity" causes it to crash. :-( >> >> Using the following test phrase: >> "I want to head off to the beach now. When will we go next?" > > > Thank you for reporting it. > > I've discovered that it is lttoolbox-java who crashes and not Apertium > Caffeine itself. The problematic word is "will" in the es-en language pair > when passing the "-a" flag. In other words, the following causes exactly the > same crash: > > echo "will" | java -jar lttoolbox.jar apertium -a -d > /usr/local/share/apertium/ en-es > > Unfortunately, org.apertium.lttoolbox.process.FSTProcessor, which causes the > crash, is super complex (more than 2000 lines of code!), and my work so far > hasn't been related to it... Perhaps somebody that worked on it can help us? >
:D I think this goes to show how little the '-a' option is used, because neither implementation of generation in lttoolbox handles it. The word 'will' disappears because the tagger is picking the auxiliary form, which has a null translation in the bidix, but the generation error appears with other words: $ echo can |apertium -a en-es =#poder $ echo can |apertium en-es Puede The solution for you is simply to remove the option - it does nothing that's useful for an end user. -- <Sefam> Are any of the mentors around? <jimregan> yes, they're the ones trolling you ------------------------------------------------------------------------------ Live Security Virtual Conference Exclusive live event will cover all the ways today's security and threat landscape has changed and how IT managers can respond. Discussions will include endpoint security, mobile security and the latest in malware threats. http://www.accelacomm.com/jaw/sfrnl04242012/114/50122263/ _______________________________________________ Apertium-stuff mailing list [email protected] https://lists.sourceforge.net/lists/listinfo/apertium-stuff
