Daniel, Tonights daily has been applied, and the new Morfologic speller is running.
First results are promising, though it is missing some compounds apparently; but I can easily make a procedure for those. Proper Names are hit mostly; but that is natural. An exception is still the single quote: 'plomp is apperantly sent to the spellchecker from the sentence: Correct gespeld: 'plomp' komt voor in de Woordenlijst Nederlandse Taal van de Taalunie en in de spellingwoordenlijst van OpenTaal. In that case, both quotes should have been stripped off by the preparing mechanisms. It look like there is a bug in the Dutch tokenizer (specially adjusted for quote in a word by Marcin: .... 'plomp['plomp/null,] '['/null,] .... Maybe one of the Java specialists could have a look at this? The ' behaves like this in Dutch: When it is between 2 letters (without a space), it is part of the word: z'n, m'n, opa's, rock'n-roll When it is after a whitespace (or beginning) and forms 's , 't, 'm ,'n (or any other single letter), it is part of the token. When the word starts like 's- , and an uppercase letter follows, it is part of some placenames. In most other cases it is a token by itself. Ruud ------------------------------------------------------------------------------ Want excitement? Manually upgrade your production database. When you want reliability, choose Perforce Perforce version control. Predictably reliable. http://pubads.g.doubleclick.net/gampad/clk?id=157508191&iu=/4140/ostg.clktrk _______________________________________________ Languagetool-devel mailing list Languagetool-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/languagetool-devel
