18/5 (AR) Introduced a wordlist format gfwl for diff --git a/doc/tutorial/gf-tutorial2.txt b/doc/tutorial/gf-tutorial2.txt index af6243056..c06fc8190 100644 --- a/doc/tutorial/gf-tutorial2.txt +++ b/doc/tutorial/gf-tutorial2.txt @@ -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/]: diff --git a/src/GF/Shell.hs b/src/GF/Shell.hs index 4b9660ced..3dcec9fbd 100644 --- a/src/GF/Shell.hs +++ b/src/GF/Shell.hs @@ -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 diff --git a/src/GF/Shell/HelpFile.hs b/src/GF/Shell/HelpFile.hs index 16603b3ff..29f25aef8 100644 --- a/src/GF/Shell/HelpFile.hs +++ b/src/GF/Shell/HelpFile.hs @@ -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:" ++ diff --git a/src/GF/Shell/ShellCommands.hs b/src/GF/Shell/ShellCommands.hs index 2740101a7..a3a0139ea 100644 --- a/src/GF/Shell/ShellCommands.hs +++ b/src/GF/Shell/ShellCommands.hs @@ -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" diff --git a/src/GF/UseGrammar/Treebank.hs b/src/GF/UseGrammar/Treebank.hs index 54ac8fb04..940ef188c 100644 --- a/src/GF/UseGrammar/Treebank.hs +++ b/src/GF/UseGrammar/Treebank.hs @@ -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 diff --git a/src/HelpFile b/src/HelpFile index df785bda2..df2a643d2 100644 --- a/src/HelpFile +++ b/src/HelpFile @@ -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: