7828 Commits

This Branch

This Branch

All Branches

Author	SHA1	Message	Date
aarne	1c3d7ea22d	inheriting some translator Extensions in NDTrans. Many of them are not needed, since they are now covered by Pred.	2014-02-08 20:18:27 +00:00
aarne	e9af74ae09	moved development-phrase modules to subdir alternatives	2014-02-08 19:55:59 +00:00
aarne	2ff1587382	factored out fun's for lifting RGL functions to Pred, and added some	2014-02-08 19:53:11 +00:00
aarne	3c97dc4ed3	nondependent versions of Pred grammars, implemented by a functor over the dependent version so that no new concrete syntax code is needed. This is at the moment the grammar of choice for translation: NDTransEng is several times faster than TransEng and even faster than the old TranslateEng. The next step is to make sure that all coverage is preserved; some new coverage is already there.	2014-02-08 14:28:58 +00:00
aarne	b1aec70571	predication Test: top-level grammar with Lexicon and Structural instead of the big Dictionary	2014-02-08 10:45:17 +00:00
aarne	2e7190c55f	functorization of Pred and Trans now includes both Eng and Swe	2014-02-08 09:10:57 +00:00
aarne	b502c26e83	qualifications in finnish/stemmed/WNKotus to help overloading	2014-02-07 22:54:54 +00:00
aarne	1da64de767	PredEng now works with a functor. Some decisions about name clashes to do.	2014-02-07 17:59:37 +00:00
hallgren	5edf45c2a0	lib/src/Makefile: TranslateChi needs a bigger stack ... Note: TranslateFin fails to compile because of an overloading problem.	2014-02-07 17:00:16 +00:00
hallgren	01b90b95f8	gf.cabal: pgf-shell needs containers now	2014-02-07 16:42:51 +00:00
hallgren	96594e405d	pgf-shell: show the names of the concrete syntaxes after loading the grammar	2014-02-07 16:40:13 +00:00
inari	ea49a7b388	haskell-bind: add function languages	2014-02-07 15:56:12 +00:00
hallgren	d1ab3072db	pgf-shell: enable -rtsopts, call performGC between commands ... This is to make it easier to find the cause of space leaks.	2014-02-07 13:57:13 +00:00
aarne	ed45089b83	completed lifting Eng verbs categories to Pred.V	2014-02-07 09:53:38 +00:00
john.j.camilleri	74d6cd6d85	Remove that annoying <li> before Punjabi in Synopsis	2014-02-07 08:57:52 +00:00
aarne	dfb00211a8	refinements in VVType	2014-02-06 20:46:11 +00:00
inari	13e7d3e518	fixed some memory leaks	2014-02-06 10:50:59 +00:00
aarne	2bc7f1049a	added information to V in Predication	2014-02-06 10:49:55 +00:00
kr.angelov	4b8c1db6d2	bugfix in the recognizer for literals in the C runtime	2014-02-05 20:13:52 +00:00
aarne	935e8ee4ff	predication-based TransSwe delivering the first sentences	2014-02-05 20:03:51 +00:00
aarne	b11977a152	started TransSwe with the new Predication	2014-02-05 15:33:35 +00:00
aarne	4fe3b482d8	qualification of type PN in Heb.mkPN to resolve overloading	2014-02-05 12:39:01 +00:00
aarne	ddc40ee2fd	Predication: factored out Prep as Adv/NP; added linref's to test with chunking	2014-02-05 09:31:23 +00:00
aarne	bda94ebb95	some more rules included in experimental/TransEng	2014-02-04 21:35:09 +00:00
aarne	56bda6bb09	first experiments running Predication with the big lexicon and the other RGL modules	2014-02-04 20:11:03 +00:00
aarne	9e8be757d8	started embedding Predication in the big translation grammar	2014-02-04 16:12:22 +00:00
aarne	3953ac3371	PredicationEng complete with all agreement forms and questions; questions with contracted auxiliaries still to do	2014-02-04 15:24:06 +00:00
hallgren	abfd4b9c96	PGF.Optimize: one more fix for unsafeFreeze	2014-02-04 13:47:37 +00:00
hallgren	8a5ade8897	PGF.Optimize: import unsafeFreeze from Data.Array.Unsafe for compatibility with ghc-7.8	2014-02-04 13:44:07 +00:00
hallgren	c8310cf17c	GFC.hs: encodeFile_ is not available in the standard Data.Binary	2014-02-04 13:36:16 +00:00
aarne	263ae29551	contractions of auxiliaries and negations in PredicationEng	2014-02-04 07:46:43 +00:00
aarne	caf3ed1053	PredicationEng except auxiliary in questions and agreement of copula. Swe is so much simpler...	2014-02-03 18:41:32 +00:00
aarne	cf576d245d	PredicationEng now almost a functor with Swe ; details to be fixed on e.g. aux/non-aux V distinction	2014-02-03 14:04:04 +00:00
hallgren	a81d7db70b	gf.cabal: add default-language to silence warning	2014-02-03 12:24:42 +00:00
aarne	26cee4374d	mkV2V second preposition mistakenly omitted in Dan and Nor, now fixed	2014-02-03 08:51:03 +00:00
aarne	9c415d6be6	Lat.VocNP issue resolved	2014-02-03 08:49:22 +00:00
aarne	cf196cb0dd	new overload resolution, taking care of ad hoc overloading.	2014-02-01 13:17:52 +00:00
aarne	2552b859fd	ad hoc overloading: a new way to deal with name resolution conflicts. Previously the renamer warned if there was e.g. an unqualified reference to mkAdv, which could come from either Syntax or Paradigms. The renamer picked randomly one of the alternatives, which then often failed in type checking. Now, all candidates are collected into a new structure AdHocOverload [Term], which is accessed by the type checker to make the choice based on the type of the constant. This eliminates some of the warnings and some of the error due to wrong choices. In some rare cases, the inherited constants have the same type, which cannot be resolved by overloading. In such cases, the type checker does the same as the renamer did before: pick the "first" option (i.e. the one that happens to be the first in the list returned by the renamer) and issues a warning. In this patch, only a couple of lines are changed. The typechecker (RConcrete) has more substantial changes, and will be recorded as the next patch.	2014-02-01 13:10:36 +00:00
aarne	4cc72722a6	missing import in CatLat causing an eval error with "no reason given"; commented out PhraseLat.VocNP that gives a mysterious PMCFG generation error	2014-02-01 13:03:08 +00:00
aarne	1a71260608	duplicated param type definition in TenseScand removed	2014-02-01 12:51:54 +00:00
aarne	1df85aa1fe	fixed an ambiguous reference in CatAmh.	2014-02-01 12:47:08 +00:00
aarne	4404e9c899	another type error revealed by record extension checking, in Est.	2014-02-01 12:30:40 +00:00
aarne	c727f7c85c	typo in Concrete type checker: when type checking <e : t>, return annotated e and not e itself. Fixed in RConcrete, which should soon replace Concrete; and hopefully will be replaced by some cleaner code soon, such as ConcreteNew which has been under construction for quite some time.	2014-01-31 21:16:07 +00:00
aarne	e97f31c175	found one type error in the whole RGL with the new (experimental) type checker, and fixed it.	2014-01-31 21:15:21 +00:00
aarne	4d4327ad8a	PredicationSwe now uses record updates; PredicationSwO is old-style	2014-01-31 20:44:34 +00:00
aarne	e19a28f46c	experimental type checker module GF.Compile.TypeCheck.RConcrete.hs with record updates enabled by ** expressions. Can be changed back to Concrete.hs by just changing one import command in GF.Compile.CheckGrammar.hs. The worst thing that *should* happen with the new type checker is that some old code is detected to be invalid, which happens if it contains a type-incompatible record extension, e.g. {x = "foo"} ** {x = 1}. Previously such errors were silently ignored. A set of test runs detected one such error in the RGL, which was corrected. On the positive side, the new type checker now enables record updates in the natural way: R ** {x = 1} will give value x = 1 overshadowing any value of x in R (provided the expected type of x is Int). lib/src/experimental/PredicationSwe.gf illustrates this, as opposed to PredicationSwO.gf which has to use old-style copying of even the unchanged record fields.	2014-01-31 20:28:25 +00:00
hallgren	82fd8c9f94	gf-server.cabal: flag to include C run-time support in the FastCGI PGF service	2014-01-31 16:11:41 +00:00
kr.angelov	f7b978f063	fix in the bracketedLinearize in Haskell	2014-01-31 08:52:32 +00:00
aarne	8b1b17c33b	abstracting complement and extension insertion in the test file PredicationSwR.gf	2014-01-31 08:13:28 +00:00
aarne	4e126a1356	an example of how grammars look with record updates: PredicationSwR (as compared with PredicationSwe)	2014-01-30 06:50:57 +00:00