gt -noexpand

This commit is contained in:
aarne
2006-03-31 13:20:40 +00:00
parent 2dd2b1e11e
commit c45100b056
6 changed files with 54 additions and 26 deletions

View File

@@ -234,6 +234,7 @@ optLinearizeTree opts0 gr t = case getOptVal opts transferFun of
| mk == markOptJava -> lin markXMLjgf
| mk == markOptStruct -> lin markBracket
| mk == markOptFocus -> lin markFocus
| mk == "metacat" -> lin metaCatMark
| otherwise -> lin noMark
_ -> lin noMark

View File

@@ -50,6 +50,13 @@ noMark :: Marker
noMark = defTMarker mk where
mk _ _ _ = ("","")
-- | mark metas with their categories
metaCatMark :: Marker
metaCatMark = defTMarker mk where
mk nod _ _ = case nod of
V.N (_,V.AtM _,val,_,_) -> ("", '+':prt val)
_ -> ("","")
-- | for vanilla brackets, focus, and position, use
markBracket :: Marker
markBracket = defTMarker mk where

View File

@@ -335,18 +335,20 @@ txtHelpFile =
"\n command completes the Tree with values to the metavariables in" ++
"\n the tree." ++
"\n options:" ++
"\n -metas also return trees that include metavariables" ++
"\n -metas also return trees that include metavariables" ++
"\n flags:" ++
"\n -depth generate to this depth (default 3)" ++
"\n -atoms take this number of atomic rules of each category (default unlimited)" ++
"\n -alts take this number of alternatives at each branch (default unlimited)" ++
"\n -cat generate in this category" ++
"\n -lang use the abstract syntax of this grammar" ++
"\n -number generate (at most) this number of trees" ++
"\n -depth generate to this depth (default 3)" ++
"\n -atoms take this number of atomic rules of each category (default unlimited)" ++
"\n -alts take this number of alternatives at each branch (default unlimited)" ++
"\n -cat generate in this category" ++
"\n -lang use the abstract syntax of this grammar" ++
"\n -number generate (at most) this number of trees" ++
"\n -noexpand don't expand these categories (comma-separated, e.g. -noexpand=V,CN)" ++
"\n examples:" ++
"\n gt -depth=10 -cat=NP -- generate all NP's to depth 10 " ++
"\n gt (PredVP ? (NegVG ?)) -- generate all trees of this form" ++
"\n gt -cat=S -tr | l -- gererate and linearize" ++
"\n gt -depth=10 -cat=NP -- generate all NP's to depth 10 " ++
"\n gt (PredVP ? (NegVG ?)) -- generate all trees of this form" ++
"\n gt -cat=S -tr | l -- generate and linearize" ++
"\n gt -noexpand=NP | l -mark=metacat -- the only NP is meta, linearized \"?0 +NP\"" ++
"\n" ++
"\nma, morphologically_analyse: ma String" ++
"\n Runs morphological analysis on each word in String and displays" ++
@@ -525,6 +527,7 @@ txtHelpFile =
"\n options" ++
"\n -all show the whole help file" ++
"\n -defs show user-defined commands and terms" ++
"\n -FLAG show the values of FLAG (works for grammar-independent flags)" ++
"\n examples:" ++
"\n h print_grammar -- show all information on the pg command" ++
"\n" ++
@@ -674,6 +677,11 @@ txtHelpFile =
"\n -unlexer=concat remove all spaces" ++
"\n -unlexer=bind like identity, but bind at \"&+\"" ++
"\n" ++
"\n-mark, marking of parts of tree in linearization. The default is none." ++
"\n -mark=metacat append \"+CAT\" to every metavariable, showing its category" ++
"\n -mark=struct show tree structure with brackets" ++
"\n -mark=java show tree structure with XML tags (used in gfeditor)" ++
"\n" ++
"\n-coding, Some grammars are in UTF-8, some in isolatin-1." ++
"\n If the letters ä (a-umlaut) and ö (u-umlaut) look strange, either" ++
"\n change your terminal to isolatin-1, or rewrite the grammar with" ++

View File

@@ -179,12 +179,12 @@ optionsOfCommand co = case co of
CStripState -> none
CTransformGrammar _ -> flags "printer"
CConvertLatex _ -> none
CLinearize _ -> both "utf8 table struct record all multi" "lang number unlexer"
CLinearize _ -> both "utf8 table struct record all multi" "lang number unlexer mark"
CParse -> both "cut new newer cfg mcfg n ign raw v lines all prob"
"cat lang lexer parser number rawtrees"
CTranslate _ _ -> opts "cat lexer parser"
CGenerateRandom -> both "cf prob" "cat lang number depth"
CGenerateTrees -> both "metas" "atoms depth alts cat lang number"
CGenerateTrees -> both "metas" "atoms depth alts cat lang number noexpand"
CPutTerm -> flags "transform number"
CTreeBank -> opts "c xml trees"
CLookupTreebank -> both "assocs raw strings trees" "treebank"
@@ -216,7 +216,7 @@ optionsOfCommand co = case co of
CPrintMultiGrammar -> both "utf8 utf8id" "printer"
CPrintSourceGrammar -> both "utf8" "printer"
CHelp _ -> opts "all alts atoms coding defs filter length lexer unlexer printer probs transform depth number"
CHelp _ -> opts "all alts atoms coding defs filter length lexer unlexer printer probs transform depth number cat"
CImpure ICEditSession -> both "f" "file"
CImpure ICTranslateSession -> both "f langs" "cat"

View File

@@ -43,17 +43,18 @@ import Data.List
generateTrees :: Options -> GFCGrammar -> Cat -> Int -> Maybe Int -> Maybe Tree -> [Exp]
generateTrees opts gr cat n mn mt = map str2tr $ generate gr' ifm cat' n mn mt'
where
gr' = gr2sgr ats gr
gr' = gr2sgr noe ats gr
cat' = prt $ snd cat
mt' = maybe Nothing (return . tr2str) mt
ifm = oElem withMetas opts
ats = getOptInt opts (aOpt "atoms")
noe = maybe [] (chunks ',') $ getOptVal opts (aOpt "noexpand")
------------------------------------------
-- translate grammar to simpler form and generated trees back
gr2sgr :: Maybe Int -> GFCGrammar -> SGrammar
gr2sgr un gr = buildTree [(c,rs) | rs@((_,(_,c)):_) <- prune rules] where
gr2sgr :: [SIdent] -> Maybe Int -> GFCGrammar -> SGrammar
gr2sgr noe un gr = buildTree [(c,noexp c rs) | rs@((_,(_,c)):_) <- prune rules] where
rules =
groupBy (\x y -> scat x == scat y) $
sortBy (\x y -> compare (scat x) (scat y))
@@ -71,8 +72,11 @@ gr2sgr un gr = buildTree [(c,rs) | rs@((_,(_,c)):_) <- prune rules] where
in take n rs1 ++ rs2
atom = null . fst . snd
noexp c rs = if elem c noe then [('?':c,([],c))] else rs
-- str2tr :: STree -> Exp
str2tr t = case t of
SApp ('?':c,[]) -> mkMeta 0 -- from noexpand=c
SApp (f,ts) -> mkApp (trId f) (map str2tr ts)
SMeta _ -> mkMeta 0
---- SString s -> K s

View File

@@ -306,18 +306,20 @@ gt, generate_trees: gt Tree?
command completes the Tree with values to the metavariables in
the tree.
options:
-metas also return trees that include metavariables
-metas also return trees that include metavariables
flags:
-depth generate to this depth (default 3)
-atoms take this number of atomic rules of each category (default unlimited)
-alts take this number of alternatives at each branch (default unlimited)
-cat generate in this category
-lang use the abstract syntax of this grammar
-number generate (at most) this number of trees
-depth generate to this depth (default 3)
-atoms take this number of atomic rules of each category (default unlimited)
-alts take this number of alternatives at each branch (default unlimited)
-cat generate in this category
-lang use the abstract syntax of this grammar
-number generate (at most) this number of trees
-noexpand don't expand these categories (comma-separated, e.g. -noexpand=V,CN)
examples:
gt -depth=10 -cat=NP -- generate all NP's to depth 10
gt (PredVP ? (NegVG ?)) -- generate all trees of this form
gt -cat=S -tr | l -- gererate and linearize
gt -depth=10 -cat=NP -- generate all NP's to depth 10
gt (PredVP ? (NegVG ?)) -- generate all trees of this form
gt -cat=S -tr | l -- generate and linearize
gt -noexpand=NP | l -mark=metacat -- the only NP is meta, linearized "?0 +NP"
ma, morphologically_analyse: ma String
Runs morphological analysis on each word in String and displays
@@ -496,6 +498,7 @@ h, help: h Command?
options
-all show the whole help file
-defs show user-defined commands and terms
-FLAG show the values of FLAG (works for grammar-independent flags)
examples:
h print_grammar -- show all information on the pg command
@@ -645,6 +648,11 @@ q, quit: q
-unlexer=concat remove all spaces
-unlexer=bind like identity, but bind at "&+"
-mark, marking of parts of tree in linearization. The default is none.
-mark=metacat append "+CAT" to every metavariable, showing its category
-mark=struct show tree structure with brackets
-mark=java show tree structure with XML tags (used in gfeditor)
-coding, Some grammars are in UTF-8, some in isolatin-1.
If the letters ä (a-umlaut) and ö (u-umlaut) look strange, either
change your terminal to isolatin-1, or rewrite the grammar with