mirror of
https://github.com/GrammaticalFramework/gf-core.git
synced 2026-04-12 06:19:33 -06:00
39 lines
1.1 KiB
Plaintext
39 lines
1.1 KiB
Plaintext
The French resource API implementation, based on romance.
|
|
|
|
The Verbes* modules were translated from an XML resource created
|
|
by Sylvain Pogodalla.
|
|
|
|
WARNING: The subcategorization information is not very reliable
|
|
(the extraction method is the main thing to blame).
|
|
|
|
The biggest test grammar is TestVerbesFre.
|
|
|
|
We recommend using GF with extra stack and heap:
|
|
|
|
gf +RTS -K10M -H200M
|
|
|
|
We recommend reading grammars from source and not emitting gfc code,
|
|
since, the size of gfc files becomes enormous. Thus:
|
|
|
|
i -noemit -src -cflexer TestVerbesFre.gf
|
|
|
|
So it seems we'll have to work on effectivizing GF!
|
|
|
|
Useful tests:
|
|
|
|
l -table choirVN -- create infl table
|
|
|
|
ma "cours cherra" -- analyse a couple of forms
|
|
|
|
gr -cat=Cl | l -table -- generate random sentence
|
|
|
|
gr PredVG HeNP (PredV2 (UseVN2 ?) SheNP) | l -table -- idem, of form "il la X"
|
|
|
|
rf labible.txt | ma | wf labible.tagged -- tag the bible (first get it!)
|
|
|
|
mq -cat=VN2 -- run a morphology quiz
|
|
|
|
Aarne Ranta 30/11/2004
|
|
|
|
|