hallgren
dd78657191
testsuite: Use Cabal's new test suite interface
...
* The old way: a user hook in Setup.hs
* The new way: specify it in gf.cabal
* The test suite is now called gf-tests, and it runs testsuite/run.hs.
* You can run it manually with 'runhaskell testsuite/run.hs'. It also runs,
together with rgl-tests, when you do 'cabal test'
* Currently only 9 of 34 tests pass. Many failures have silly causes:
- Error messages that look slightly different
- Same output but in a different order
- Absolute paths in output
2013-12-03 17:13:39 +00:00
hallgren
9118f72fda
testsuite: use forward slashes in paths
2013-12-03 16:59:33 +00:00
aarne
938587c159
top-level access to contracted auxiliaries in ExtraEng. Introduced them as variants in ParseEng.
2013-12-03 14:52:12 +00:00
gregoire.detrez
23f13d5b55
Fix bescherelle tests broken (True/False order)
2013-12-03 14:29:36 +00:00
aarne
302e2d3ebd
introduced contracted auxiliaries in English. So far only available internally in the tables created from Cl. TODO: make them accessible on top level in Extra; use them as variants in Parse. Also the use in QCl needs some more work: we can say "he's here", "I'm here", "where's he" but not "'s he here", "where'm I".
2013-12-03 13:54:57 +00:00
aarne
e0ded7e7e8
changed the order of Prelude.Bool to False | True. This is how it is in Haskell to create expected Ord and Enum. Now it also gives a better default for English clause linearizations, as False = uncontracted.
2013-12-03 13:52:52 +00:00
aarne
1778cd7c19
removed the unlines-lines wrapper from Lexing.unlexer to prevent empty lines when an unlexer (such as -bind or -unchars) is used as an option in linearization. Don't know really why the input had been broken into lines in the first place. You can see the effect by importing LangEng and running "gr -cat=Cl | l -table -bind" before and after recompiling GF.
2013-12-03 13:27:22 +00:00
kr.angelov
98e0f79324
fix in inflection_en.xml
2013-12-03 12:15:06 +00:00
kr.angelov
5d93c1d4de
split teacher_N into two genders
2013-12-03 10:39:45 +00:00
kr.angelov
ff5dcf0f6a
FeatN and FeatN2 in English now show the gender if it is masculine or feminine
2013-12-03 09:05:20 +00:00
kr.angelov
41642e8d30
in DictEng all hyphenated words are again merged into a single token without spaces. This is necessary because there is no lexer anymore
2013-12-03 08:31:04 +00:00
kr.angelov
63c779f844
forgoten entry in DictEngAbs.gf
2013-12-03 08:28:09 +00:00
john.j.camilleri
f7e91e06c1
Maltese RG: Fix apposition. Add ExtAdvNP & ExtAdvVP. Fix adjective order.
2013-12-03 08:19:38 +00:00
kr.angelov
2356aa7e65
one big patch with several changes in the different dictionaries
...
1. the duplicated variants in the French entries are merged
2. ~150 abstract words are split into masculine and feminine entries.
The actual defininitions for English, Bulgarian, French and German differ.
For the other languages they are just copied.
3. fan_N, intelligency_N, character_N and letter_N are split into
different senses. The entries for Bulgarian, Swedish, German and French
are sense disambiguated. For English they are just copied since they
are ambiguous. For the other languages they are copied since I don't
know how to translate them.
2013-12-03 08:14:05 +00:00
inari
299630e95e
fixed good_A in DictEngFre
2013-12-02 15:58:54 +00:00
aarne
385d5f45fc
ui/android/.../Translate.java uses Parse8 (change the name and language list if you need something else)
2013-12-02 13:34:51 +00:00
aarne
4fab9c1bb8
mkInterj in some languages; 'make Parse' in lib/src builds Parse8.pgf
2013-12-02 13:34:07 +00:00
aarne
45787ff1e6
interjections in DictEng Fin and Ger
2013-12-02 11:09:28 +00:00
hallgren
25b3d5adc0
examples/phrasebook/Makefile2: Romanian needs a larger stack
2013-11-29 17:34:09 +00:00
hallgren
7934c0a88b
Commment code and options relating to the old partial evaluator
...
This means that the -old-comp and -new-comp flags are not recognized anymore.
The only functional difference is that printnames were still normalized with
the old partial evaluator. Now that is done with the new partial evaluator.
2013-11-29 16:26:49 +00:00
hallgren
ac16928d87
Minor release/download/developer doc changes
2013-11-29 16:04:52 +00:00
hallgren
537caca039
Move typePredefined from GF.Compile.Compute.AppPredefined to GF.Compile.TypeCheck.Primitives
...
Also move the list of primitives
2013-11-29 15:47:26 +00:00
hallgren
6f590a3f2c
Move checkPredefError from GF.Compile.Compute.ConcreteLazy to GF.Grammar.Macros
...
Also simplified its type.
2013-11-29 14:43:01 +00:00
kr.angelov
d7f3aab994
-optimize-pgf should also apply to the linrefs
2013-11-29 14:25:23 +00:00
kr.angelov
724b2e04e1
fixes in the dictionaries
2013-11-29 13:56:28 +00:00
kr.angelov
938d97fad5
cleanup in LangBul
2013-11-29 13:55:46 +00:00
kr.angelov
4ae1a16a1e
added caching to be able to quickly generate the inflection table for Finnish
2013-11-28 22:06:55 +00:00
hallgren
134f6a200f
examples/foods: remove all -path flags
...
All languages n the Foods grammar compile fine with alltenses.
2013-11-29 13:03:19 +00:00
aarne
3d5f1afde6
added forms to Finnish xml table
2013-11-29 08:21:00 +00:00
aarne
e903f85a90
some missing forms in Fin (agent participle, Inf3 instructives); now also in stemmed version.
2013-11-29 08:05:55 +00:00
aarne
86937017cd
html templates for Finnish noun and verb inflection started
2013-11-28 22:04:30 +00:00
aarne
ece03aca51
added some esoteric verb forms to Fin (not used in resource API but nice to show): potential, passive imperatives, infinitives 2,4,5. Remains to carry out in stemmed/StemFin.
2013-11-28 22:01:38 +00:00
kr.angelov
945650f653
bugfix in the android app
2013-11-28 20:00:17 +00:00
kr.angelov
1c75721871
now the TTS is used only if there is actually a language model. this is better than using English for speaking Swedish
2013-11-28 19:59:03 +00:00
kr.angelov
442b5fb305
a bit of cleanup in the Android app
2013-11-28 19:57:45 +00:00
kr.angelov
953ae7e76d
extension in the template format for inflection tables. This makes it possible to use the templates with Finish
2013-11-28 19:52:31 +00:00
hallgren
dcf07aaad6
GFI: adapt to character encoding changes
...
GF.Compile.Coding is not used any more.
2013-11-28 13:55:08 +00:00
hallgren
4ef30ca450
SimpleEditor.Convert: adapt to recent character encoding changes
...
Things got simpler!
2013-11-28 13:27:45 +00:00
aarne
6c0498c479
changed annotations in DictEngFin: --UNCHK for unchecked entries, ---ARB for arbitrary choices that should be treated by separating senses; many other old annotations suppressed, but most informative ones preserved
2013-11-28 13:27:12 +00:00
kr.angelov
99ac055997
fixes in the dictionaries
2013-11-28 13:16:03 +00:00
kr.angelov
35580255c3
the Android App now show the valency patterns for verbs in English, Bulgarian and Swedish
2013-11-28 12:48:55 +00:00
aarne
ab90b3c7ef
added linref's to Ger Hin Romance
2013-11-28 08:05:58 +00:00
aarne
6401a37e21
Translate grammar for Ger and Hin added. Some functions in Ger phrasebook need to be commented out to compile. Now a system for 7 languages can be built, but including Fre as well is too heavy on my laptop although it can be compiled separately.
2013-11-28 08:03:50 +00:00
kr.angelov
f49a8df60c
changes in the parallel dictionaries
2013-11-27 22:18:22 +00:00
kr.angelov
9de4812482
better dictionary visualization in the Android App
2013-11-27 22:16:05 +00:00
aarne
848573afdc
555 new words checked in DictEngFin; latin1 flags in some esoteric files
2013-11-27 19:22:37 +00:00
hallgren
7215991e9b
Adding an alternate style for the GF home page
2013-11-27 15:32:25 +00:00
kr.angelov
38b4a88c2b
minor tweaks in the Android UI
2013-11-27 12:37:55 +00:00
kr.angelov
105071d11e
fix the character encoding problem in the web view in the Android UI
2013-11-27 12:37:15 +00:00
kr.angelov
3509a8d4cc
added API in the C runtime and the Java binding for checking whether a given function is linearizable in a given language. This is used in the Android UI for better vizualizations
2013-11-27 12:35:11 +00:00
