l -multi -all/-table ; same with tb

This commit is contained in:
aarne
2006-05-30 22:02:25 +00:00
parent 6f57b5b530
commit 993f2b3e1e
19 changed files with 112 additions and 64 deletions

View File

@@ -12,6 +12,9 @@ Changes in functionality since May 17, 2005, release of GF Version 2.2
</center>
30/5 (AR) It is now possible to use the flags <tt>-all, -table, -record</tt>
in combination with <tt>l -multi</tt>, and also with <tt>tb</tt>.
<p>
18/5 (AR) Introduced a wordlist format <tt>gfwl</tt> for

View File

@@ -1910,7 +1910,7 @@ of such a theory, represented as an ``abstract`` module in GF.
A resource grammar is a grammar built on linguistic grounds,
to describe a language rather than a domain.
The GF resource grammar library contains resource grammars for
The GF resource grammar library, which contains resource grammars for
10 languages, is described more closely in the following
documents:
- [Resource library API documentation ../../lib/resource-1.0/doc/]:

View File

@@ -5,7 +5,7 @@ concrete IdiomDan of Idiom = CatDan **
lin
ImpersCl vp = mkClause "det" (agrP3 neutrum Sg) vp ;
GenericCl vp = mkClause "man" (agrP3 neutrum Sg) vp ;
GenericCl vp = mkClause "man" (agrP3 utrum Sg) vp ;
ExistNP np =
mkClause "det" (agrP3 neutrum Sg) (insertObj

View File

@@ -12,7 +12,7 @@ lin
airplane_N = mk2N "fly" "flyet" ;
answer_V2S = mkV2S (regV "svare") "til" ;
apartment_N = mk2N "værelse" "værelsen" ;
apple_N = mk2N "æble" "æblet" ;
apple_N = mk3N "æble" "æblet" "æbler" ;
art_N = mk2N "kunst" "kunsten" ;
ask_V2Q = mkV2Q spørge_V [] ;
baby_N = mk2N "baby" "babyen" ; ---- babyen
@@ -252,7 +252,7 @@ lin
cloud_N = mk2N "sky" "skyen" ;
day_N = mk3N "dag" "dagen" "dage" ;
dust_N = mk2N "støv" "støvet" ;
ear_N = mk2N "øre" "øret" ;
ear_N = mk3N "øre" "øret" "ører" ;
earth_N = mk2N "jord" "jorden" ;
egg_N = mk2N "æg" "æget" ;
eye_N = mk3N "øje" "øjet" "øjne" ;
@@ -269,7 +269,7 @@ lin
hair_N = mk2N "hår" "håret" ;
hand_N = mk2N "hånd" "hånden" ;
head_N = mk2N "hoved" "hovedet" ;
heart_N = mk2N "hjerte" "hjertet" ;
heart_N = mk3N "hjerte" "hjertet" "hjerter" ;
horn_N = mk2N "horn" "hornet" ;
husband_N = mk2N "ægtefælle" "ægtefællen" ; ----
ice_N = mk2N "is" "isen" ;

View File

@@ -226,6 +226,8 @@ oper
vaereV : V -> V ;
--3 Verbs with a particle
--
-- The particle, such as in "switch on", is given as a string.

View File

@@ -79,7 +79,7 @@ concrete StructuralDan of Structural = CatDan **
there7to_Adv = ss "dit" ;
there7from_Adv = ss "derfra" ;
therefore_PConj = ss "derfor" ;
these_NP = regNP ["de her"] ["det hers"] Plg ;
these_NP = regNP ["disse"] ["disses"] Plg ;
they_Pron = MorphoDan.mkNP "de" "dem" "deres" "deres" "deres" Plg P1 ;
this_Quant =
{s = table {
@@ -88,7 +88,7 @@ concrete StructuralDan of Structural = CatDan **
} ;
det = DDef Indef
} ;
this_NP = regNP ["det her"] ["det hers"] SgNeutr ;
this_NP = regNP ["dette"] ["dettes"] SgNeutr ;
those_NP = regNP ["de der"] ["de ders"] Plg ;
through_Prep = ss "gennem" ;
too_AdA = ss "for" ;

View File

@@ -1,3 +1,11 @@
30/5/2006
LangFre
231m, 48s, *.gfc 718k
--------------
11/5/2006
LangFre

View File

@@ -17,15 +17,16 @@ instance DiffNor of DiffScand = open CommonScand, Prelude in {
_ => Plg
} ;
detDef : Species = Indef ;
detDef : Species = Def ;
Verb : Type = {
s : VForm => Str ;
part : Str ;
vtype : VType
vtype : VType ;
isVaere : Bool
} ;
hasAuxBe _ = False ;
hasAuxBe v = v.isVaere ;
-- Strings.
@@ -44,13 +45,13 @@ instance DiffNor of DiffScand = open CommonScand, Prelude in {
verbHave =
mkVerb "ha" "har" "ha" "hadde" "hatt" nonExist nonExist nonExist
** noPart ;
** {part = [] ; isVaere = False} ;
verbBe =
mkVerb "være" "er" "var" "var" "vært" "væren" "været" "værne"
** noPart ;
** {part = [] ; isVaere = False} ;
verbBecome =
mkVerb "bli" "blir" "bli" "ble" "blitt" "bliven" "blivet" "blivne"
** noPart ;
** {part = [] ; isVaere = True} ;
-- auxiliary
noPart = {part = []} ;

View File

@@ -6,7 +6,7 @@ concrete IdiomNor of Idiom = CatNor **
lin
ImpersCl vp = mkClause "det" (agrP3 neutrum Sg) vp ;
GenericCl vp = mkClause "man" (agrP3 neutrum Sg) vp ;
GenericCl vp = mkClause "man" (agrP3 utrum Sg) vp ;
ExistNP np =
mkClause "det" (agrP3 neutrum Sg) (insertObj

View File

@@ -17,9 +17,9 @@ lin
bad_A = regADeg "dårlig" ; ----
bank_N = mk2N "bank" "banken" ;
beautiful_A = mk3ADeg "vakker" "vakkert" "vakra" ;
become_VA = mkVA bli_V ;
become_VA = mkVA (vaereV bli_V) ;
beer_N = regGenN "øl" neutrum ;
beg_V2V = mkV2V be_V [] "att" ;
beg_V2V = mkV2V be_V [] "at" ;
big_A = irregADeg "stor" "større" "størst";
bike_N = mkN "sykkel" "sykkelen" "sykler" "syklene" ;
bird_N = mk2N "fugl" "fuglen" ;
@@ -53,19 +53,19 @@ lin
close_V2 = dirV2 (mk2V "lukke" "lukket") ;
coat_N = regGenN "frakk" masculine ;
cold_A = regADeg "kald" ;
come_V = komme_V ;
come_V = vaereV komme_V ;
computer_N = mk2N "datamaskin" "datamaskinen" ;
country_N = mk2N "land" "landet" ;
cousin_N = mk2N "fetter" "fetteren" ; ----
cow_N = mkN "ku" "kua" "kyr" "kyrne" ; ----
die_V = dø_V ;
die_V = vaereV dø_V ;
dirty_A = mk3ADeg "skitten" "skittent" "skitne" ; ----
distance_N3 = mkN3 (regGenN "avstand" masculine) "fra" "til" ;
doctor_N = mk2N "lege" "legen" ;
dog_N = regGenN "hund" masculine ;
door_N = regGenN "dør" feminine ;
drink_V2 = dirV2 drikke_V ;
easy_A2V = mkA2V (regA "grei") "før" ;
easy_A2V = mkA2V (regA "grei") "for" ;
eat_V2 = dirV2 (mk2V "spise" "spiste") ;
empty_A = mkADeg "tom" "tomt" "tomme" "tommere" "tommest" ;
enemy_N = regGenN "fiende" masculine ;
@@ -85,7 +85,7 @@ lin
glove_N = regGenN "hanske" masculine ;
gold_N = regGenN "gull" neutrum ;
good_A = mkADeg "god" "godt" "gode" "bedre" "best" ;
go_V = gå_V ;
go_V = vaereV gå_V ;
green_A = mk2ADeg "grønn" "grønt" ;
harbour_N = regGenN "havn" feminine;
hate_V2 = dirV2 (regV "hate") ;
@@ -150,7 +150,7 @@ lin
rock_N = regGenN "stein" masculine ;
roof_N = regGenN "tak" neutrum ;
rubber_N = mk2N "gummi" "gummien" ;
run_V = springe_V ;
run_V = vaereV springe_V ;
say_VS = mkVS si_V ;
school_N = regGenN "skole" feminine;
science_N = mk2N "vitenskap" "vitenskapen" ;
@@ -189,7 +189,7 @@ lin
thick_A = mk2ADeg "tykk" "tykt" ;
thin_A = mk2ADeg "tynn" "tynt" ;
train_N = regGenN "tog" neutrum ;
travel_V = mk2V "reise" "reiste" ;
travel_V = vaereV (mk2V "reise" "reiste") ;
tree_N = mkN "tre" "treet" "trær" "træne" ;
---- trousers_N = regGenN "trousers" ; ---- pl t !
ugly_A = mk2ADeg "stygg" "stygt" ;
@@ -197,7 +197,7 @@ lin
university_N = regGenN "universitet" neutrum ;
village_N = mk2N "grend" "grenda" ;
wait_V2 = mkV2 (regV "vente") "på" ;
walk_V = (irregV "" "gikk" "gått") ;
walk_V = vaereV gå_V ;
warm_A = regADeg "varm" ;
war_N = regGenN "krig" masculine ;
watch_V2 = mkV2 se_V "på" ;
@@ -220,7 +220,7 @@ lin
add_V3 = mkV3 (partV (irregV "legge" "la" "lagt") "til") [] "til" ;
number_N = mk2N "nummer" "nummeret" ;
put_V2 = mkV2 (irregV "sette" "satte" "satt") [] ;
stop_V = regV "stanse" ;
stop_V = vaereV (regV "stanse") ;
jump_V = regV "hoppe" ;
left_Ord = {s = "venstre" ; isDet = True} ;
@@ -308,12 +308,12 @@ lin
count_V2 = dirV2 (regV "regne") ;
cut_V2 = dirV2 (skjære_V) ;
dig_V = mk2V "grave" "gravde" ;
fall_V = falle_V ;
fall_V = vaereV falle_V ;
fear_V2 = dirV2 (regV "frykte") ;
fight_V2 = dirV2 (slåss_V) ;
float_V = flyte_V ;
flow_V = renne_V ;
fly_V = fly_V ;
fly_V = vaereV fly_V ;
freeze_V = fryse_V ;
give_V3 = dirdirV3 gi_V ;
hit_V2 = dirV2 (slå_V) ;
@@ -335,7 +335,7 @@ lin
split_V2 = dirV2 (mk2V "kløyve" "kløyvde") ;
squeeze_V2 = dirV2 (mk2V "klemme" "klemte") ;
stab_V2 = dirV2 (stikke_V) ;
stand_V = stå_V ;
stand_V = vaereV stå_V ;
suck_V2 = dirV2 (suge_V) ;
swell_V = partV (regV "hovne") "opp" ;
swim_V = regV "simme" ;

View File

@@ -217,6 +217,13 @@ oper
irregV : (drikke, drakk, drukket : Str) -> V ;
--3 Verbs with "være" as auxiliary
--
-- By default, the auxiliary is "have". This function changes it to "være".
vaereV : V -> V ;
--3 Verbs with a particle.
--
-- The particle, such as in "switch on", is given as a string.
@@ -357,14 +364,14 @@ oper
mkPreposition p = p ;
mkV a b c d e f = mkVerb6 a b c d e f **
{part = [] ; vtype = VAct ; lock_V = <>} ;
{part = [] ; vtype = VAct ; isVaere = False ; lock_V = <>} ;
regV a = case last a of {
"e" => vHusk (init a) ;
_ => vBo a
} ** {part = [] ; vtype = VAct ; lock_V = <>} ;
} ** {part = [] ; vtype = VAct ; isVaere = False ; lock_V = <>} ;
mk2V a b = regVerb a b ** {part = [] ; vtype = VAct ; lock_V = <>} ;
mk2V a b = regVerb a b ** {part = [] ; vtype = VAct ; isVaere = False ; lock_V = <>} ;
irregV =
\drikke,drakk,drukket ->
@@ -380,10 +387,17 @@ oper
in
mkV drikke drikker (drikke + "s") drakk drukket drikk ;
vaereV v = {
s = v.s ;
part = [] ;
vtype = v.vtype ;
isVaere = True ;
lock_V = <>
} ;
partV v p = {s = v.s ; part = p ; vtype = v.vtype ; lock_V = <>} ;
depV v = {s = v.s ; part = v.part ; vtype = VPass ; lock_V = <>} ;
reflV v = {s = v.s ; part = v.part ; vtype = VRefl ; lock_V = <>} ;
partV v p = {s = v.s ; part = p ; vtype = v.vtype ; isVaere = v.isVaere ; lock_V = <>} ;
depV v = {s = v.s ; part = v.part ; vtype = VPass ; isVaere = False ; lock_V = <>} ;
reflV v = {s = v.s ; part = v.part ; vtype = VRefl ; isVaere = False ; lock_V = <>} ;
mkV2 v p = v ** {c2 = p ; lock_V2 = <>} ;
dirV2 v = mkV2 v [] ;

View File

@@ -79,16 +79,16 @@ concrete StructuralNor of Structural = CatNor **
there7to_Adv = ss "dit" ;
there7from_Adv = ss "derfra" ;
therefore_PConj = ss "derfor" ;
these_NP = regNP ["de her"] ["det hers"] Plg ;
these_NP = regNP ["disse"] ["disses"] Plg ;
they_Pron = MorphoNor.mkNP "de" "dem" "deres" "deres" "deres" Plg P1 ;
this_Quant =
{s = table {
Sg => \\_ => genderForms ["den her"] ["det her"] ;
Pl => \\_,_ => ["de her"]
Sg => \\_ => genderForms ["denne"] ["dette"] ;
Pl => \\_,_ => ["disse"]
} ;
det = DDef Indef
det = DDef Def
} ;
this_NP = regNP ["det her"] ["det hers"] SgNeutr ;
this_NP = regNP ["dette"] ["dettes"] SgNeutr ;
those_NP = regNP ["de der"] ["de ders"] Plg ;
through_Prep = ss "gjennom" ;
too_AdA = ss "for" ;

View File

@@ -1,6 +1,8 @@
interface DiffScand = open CommonScand, Prelude in {
-- Parameters.
--1 Differences between Scandinavian languages
-- Norway has three genders, Danish and Swedish have two.
param
Gender ;
@@ -14,13 +16,14 @@ interface DiffScand = open CommonScand, Prelude in {
detDef : Species ;
-- Danish verbs have a marking for compound-tense auxiliary ("have" or "være").
-- Danish and Norwegian verbs, but not Swedish verbs,
-- have two possible compound-tense auxiliaries ("have" or "være").
Verb : Type ;
hasAuxBe : Verb -> Bool ;
-- Strings.
-- The rest of the parameters are function words used in the syntax modules.
conjThat : Str ;
conjThan : Str ;

View File

@@ -5,7 +5,7 @@ concrete IdiomSwe of Idiom = CatSwe **
lin
ImpersCl vp = mkClause "det" (agrP3 neutrum Sg) vp ;
GenericCl vp = mkClause "man" (agrP3 neutrum Sg) vp ;
GenericCl vp = mkClause "man" (agrP3 utrum Sg) vp ;
ExistNP np =
mkClause "det" (agrP3 neutrum Sg) (insertObj
(\\_ => np.s ! accusative) (predV (depV finna_V))) ;

View File

@@ -231,8 +231,11 @@ execC co@(comm, opts0) sa@(sh@(st,(h,_,_,_)),a) = checkOptions st co >> case com
CLinearize []
| oElem showMulti opts ->
changeArg (opTS2CommandArg (unlines. linearizeToAll
(allStateGrammars st)) . s2t) sa
changeArg (opTS2CommandArg (
unlines .
(\t -> [optLinearizeTreeVal opts gr t | gr <- allStateGrammars st])) . s2t) sa
| otherwise -> changeArg (opTS2CommandArg (optLinearizeTreeVal opts gro) . s2t) sa
---- CLinearize m -> changeArg (opTS2CommandArg (optLinearizeArgForm opts gro m)) sa

View File

@@ -185,11 +185,11 @@ txtHelpFile =
"\n HINT: see GF language specification for the syntax of Pattern and Term." ++
"\n You can also copy and past parsing results." ++
"\n options: " ++
"\n -table show parameters" ++
"\n -struct bracketed form" ++
"\n -record record, i.e. explicit GF concrete syntax term" ++
"\n -all show all forms and variants" ++
"\n -multi linearize to all languages (the other options don't work)" ++
"\n -table show parameters (not compatible with -record, -all)" ++
"\n -record record, i.e. explicit GF concrete syntax term (not compatible with -table, -all)" ++
"\n -all show all forms and variants (not compatible with -record, -table)" ++
"\n -multi linearize to all languages (can be combined with the other options)" ++
"\n flags:" ++
"\n -lang linearize in this grammar" ++
"\n -number give this number of forms at most" ++
@@ -243,6 +243,9 @@ txtHelpFile =
"\n options:" ++
"\n -c compare to existing xml-formatted treebank" ++
"\n -trees return the trees of the treebank" ++
"\n -all show all linearization alternatives (branches and variants)" ++
"\n -table show tables of linearizations with parameters" ++
"\n -record show linearization records" ++
"\n -xml wrap the treebank (or comparison results) with XML tags" ++
"\n -mem write the treebank in memory instead of a file TODO" ++
"\n examples:" ++

View File

@@ -186,7 +186,7 @@ optionsOfCommand co = case co of
CGenerateRandom -> both "cf prob" "cat lang number depth atoms noexpand doexpand"
CGenerateTrees -> both "metas" "atoms depth alts cat lang number noexpand doexpand"
CPutTerm -> flags "transform number"
CTreeBank -> opts "c xml trees"
CTreeBank -> opts "c xml trees all table record"
CLookupTreebank -> both "assocs raw strings trees" "treebank"
CWrapTerm _ -> opts "c"
CApplyTransfer _ -> flags "lang transfer"

View File

@@ -30,7 +30,7 @@ module GF.UseGrammar.Treebank (
) where
import GF.Compile.ShellState
import GF.UseGrammar.Linear (linTree2string)
import GF.UseGrammar.Linear -- (linTree2string)
import GF.UseGrammar.Custom
import GF.UseGrammar.GetTree (string2tree)
import GF.Grammar.TypeCheck (annotate)
@@ -38,7 +38,7 @@ import GF.Canon.CMacros (noMark)
import GF.Grammar.Grammar (Trm)
import GF.Grammar.MMacros (exp2tree)
import GF.Grammar.Macros (zIdent)
import GF.Grammar.PrGrammar (prt_)
import GF.Grammar.PrGrammar (prt_,prt)
import GF.Grammar.Values (tree2exp)
import GF.Data.Operations
import GF.Infra.Option
@@ -47,6 +47,7 @@ import GF.Infra.UseIO
import qualified GF.Grammar.Abstract as A
import qualified Data.Map as M
import qualified Data.Set as S
import Control.Monad (liftM)
-- Generate a treebank with a multilingual grammar. AR 8/2/2006
-- (c) Aarne Ranta 2006 under GNU GPL
@@ -95,7 +96,7 @@ mkMultiTreebank opts sh com trees = putInXML opts "treebank" comm (concatMap mkI
mkItem(t,i)= putInXML opts "item" (cat i) (mkTree t ++ concatMap (mkLin t) langs)
-- mkItem(t,i)= putInXML opts "item" (cat i) (mkTree t >>mapM_ (mkLin t) langs)
mkTree t = putInXML opts "tree" [] (puts $ showTree t)
mkLin t lg = putInXML opts "lin" (lang lg) (puts $ linearize sh lg t)
mkLin t lg = putInXML opts "lin" (lang lg) (puts $ linearize opts sh lg t)
langs = [prt_ l | l <- allLanguages sh]
comm = "" --- " command=" ++ show com +++ "abstract=" ++ show abstr
@@ -110,7 +111,7 @@ mkUniTreebank :: Options -> ShellState -> Language -> [A.Tree] -> Treebank
mkUniTreebank opts sh lg trees = M.fromListWith (++) [(lin t, [prt_ t]) | t <- trees]
where
lang = prt_ lg
lin t = linearize sh lang t
lin t = linearize opts sh lang t
-- reads a treebank and linearizes its trees again, printing all differences
testMultiTreebank :: Options -> ShellState -> String -> Res
@@ -120,7 +121,7 @@ testMultiTreebank opts sh = putInXML opts "testtreebank" [] .
where
testOne (e,lang,str0) = do
let tr = annot gr e
let str = linearize sh lang tr
let str = linearize opts sh lang tr
if str == str0 then ret else putInXML opts "diff" [] $ concat [
putInXML opts "tree" [] (puts $ showTree tr),
putInXML opts "old" (" lang=" ++ show (prt_ (zIdent lang))) $ puts str0,
@@ -198,11 +199,18 @@ tagXML s = "<" ++ s ++ ">"
--- these handy functions are borrowed from EmbedAPI
linearize mgr lang =
untok .
linTree2string noMark (canModules mgr) (zIdent lang)
where
sgr = stateGrammarOfLangOpt False mgr (zIdent lang)
untok = customOrDefault (stateOptions sgr) useUntokenizer customUntokenizer sgr
linearize opts mgr lang = lin where
sgr = stateGrammarOfLangOpt False mgr zlang
cgr = canModules mgr
zlang = zIdent lang
untok = customOrDefault (addOptions opts (stateOptions sgr)) useUntokenizer customUntokenizer sgr
lin
| oElem showRecord opts = err id id . liftM prt . linearizeNoMark cgr zlang
| oElem tableLin opts =
err id id . liftM (unlines . map untok . prLinTable True) . allLinTables cgr zlang
| oElem showAll opts =
err id id . liftM (unlines . map untok . prLinTable False) . allLinTables cgr zlang
| otherwise = untok . linTree2string noMark cgr zlang
showTree t = prt_ $ tree2exp t

View File

@@ -156,11 +156,11 @@ l, linearize: l PattList? Tree
HINT: see GF language specification for the syntax of Pattern and Term.
You can also copy and past parsing results.
options:
-table show parameters
-struct bracketed form
-record record, i.e. explicit GF concrete syntax term
-all show all forms and variants
-multi linearize to all languages (the other options don't work)
-table show parameters (not compatible with -record, -all)
-record record, i.e. explicit GF concrete syntax term (not compatible with -table, -all)
-all show all forms and variants (not compatible with -record, -table)
-multi linearize to all languages (can be combined with the other options)
flags:
-lang linearize in this grammar
-number give this number of forms at most
@@ -214,6 +214,9 @@ tb, tree_bank: tb
options:
-c compare to existing xml-formatted treebank
-trees return the trees of the treebank
-all show all linearization alternatives (branches and variants)
-table show tables of linearizations with parameters
-record show linearization records
-xml wrap the treebank (or comparison results) with XML tags
-mem write the treebank in memory instead of a file TODO
examples: