diff --git a/.travis.yml b/.travis.yml index 7f2c3d43..166b6815 100644 --- a/.travis.yml +++ b/.travis.yml @@ -1,14 +1,27 @@ sudo: required - language: c -services: - - docker +os: + - linux + - osx + - windows + +addons: + apt: + packages: + - ghc before_install: - - docker pull odanoburu/haskell-gf:3.9 - - mkdir rgl + - if [[ "$TRAVIS_OS_NAME" == "osx" ]]; then brew update && brew install ghc@8.2 && export PATH="/usr/local/opt/ghc@8.2/bin:$PATH" ; fi + - if [[ "$TRAVIS_OS_NAME" == "osx" ]]; then curl http://www.grammaticalframework.org/download/gf-3.9-bin-intel-mac.tar.gz > gf.tar.gz && sudo tar --no-same-owner --no-same-permissions -C /usr/local -zxf gf.tar.gz && rm gf.tar.gz; fi + - if [[ "$TRAVIS_OS_NAME" == "linux" ]]; then curl http://www.grammaticalframework.org/download/gf_3.9.1-1_amd64-trusty.deb > gf.deb && sudo dpkg -i gf.deb && rm gf.deb ; fi + - if [[ "$TRAVIS_OS_NAME" == "windows" ]]; then choco install ghc --version=8.4.4 && export PATH="/c/ProgramData/chocolatey/lib/ghc/tools/ghc-8.4.4/bin:$PATH"; fi + - if [[ "$TRAVIS_OS_NAME" == "windows" ]]; then curl http://www.grammaticalframework.org/download/gf-3.9-bin-windows.zip > gf.zip && unzip gf.zip && rm gf.zip && export PATH="$TRAVIS_BUILD_DIR/gf-3.9/bin:$PATH"; fi + - if [[ "$TRAVIS_OS_NAME" == "windows" ]]; then alias gf='gf.exe' && alias runghc='runghc.exe' ; fi script: - - docker run --mount src="$(pwd)",target=/home,type=bind odanoburu/haskell-gf:3.9 /bin/bash -c "cd /home/; export GF_LIB_PATH=/home/rgl ; runghc Make.hs build prelude all --verbose ;" - - docker run --mount src="$(pwd)",target=/home,type=bind odanoburu/haskell-gf:3.9 /bin/bash -c "cd /home/; export GF_LIB_PATH=/home/rgl; bash Make.sh --dest=rgl --verbose ;" + - runghc Make.hs build prelude all --verbose + - rm -rf dist + - mkdir dist-bash ; bash Make.sh --dest=dist-bash --verbose + - rm -rf dist ; mkdir dist-bat + - if [[ "$TRAVIS_OS_NAME" == "windows" ]]; then cmd //c Make.bat --dest=dist-bat --verbose ; fi diff --git a/Config.hs b/Config.hs new file mode 100644 index 00000000..b671eacb --- /dev/null +++ b/Config.hs @@ -0,0 +1,70 @@ +-- | Reading language config file +module Config ( + LangInfo (..), + loadLangs, loadLangsFrom, configFile + ) where + +import Data.List (unfoldr) +import System.IO (hPutStrLn,stderr) +import System.Exit (exitFailure) + +-- | Path to language config file +configFile :: FilePath +configFile = "languages.csv" + +-- | Information about a language +data LangInfo = LangInfo + { langCode :: String -- ^ 3-letter ISO 639-2/B code + , langName :: String -- ^ language name + , langDir :: String -- ^ directory name + , langFunctor :: Maybe String -- ^ functor (not used) + , langUnlexer :: Maybe String -- ^ decoding for postprocessing linearizations + , langPresent :: Bool + , langAll :: Bool + , langTry :: Bool + , langSymbolic :: Bool + , langCompatibility :: Bool + , langSynopsis :: Bool -- ^ include in RGL synopsis + } deriving (Show,Eq) + +-- | Load language information from default config file +loadLangs :: IO [LangInfo] +loadLangs = loadLangsFrom configFile + +-- | Load language information from specified config file +loadLangsFrom:: FilePath -> IO [LangInfo] +loadLangsFrom configFile = do + lns <- readFile configFile >>= return . lines + mapM mkLangInfo (tail lns) + where + maybeBit bits n = if length bits >= (n+1) && length (bits !! n) > 0 then Just (bits !! n) else Nothing + boolBit bits n def = if length bits >= (n+1) && length (bits !! n) > 0 then (if def then bits !! n /= "n" else bits !! n == "y") else def + mkLangInfo s = + let bits = separateBy ',' s in + if length bits < 2 + then die $ "Invalid entry in " ++ configFile ++ ": " ++ s + else return $ LangInfo + { langCode = bits !! 0 + , langName = bits !! 1 + , langDir = bits !! 2 + , langFunctor = maybeBit bits 3 + , langUnlexer = maybeBit bits 4 + , langPresent = boolBit bits 5 False + , langAll = boolBit bits 6 True + , langTry = boolBit bits 7 True + , langSymbolic = boolBit bits 8 True + , langCompatibility = boolBit bits 9 False + , langSynopsis = boolBit bits 10 False + } + +-- | Separate a string on a character +-- Source: https://stackoverflow.com/a/4978733/98600 +separateBy :: Eq a => a -> [a] -> [[a]] +separateBy chr = unfoldr sep where + sep [] = Nothing + sep l = Just . fmap (drop 1) . break (== chr) $ l + +die :: String -> IO a +die s = do + hPutStrLn stderr s + exitFailure diff --git a/README.md b/README.md index 687257bc..ec0c1008 100644 --- a/README.md +++ b/README.md @@ -27,22 +27,25 @@ It will look for, in this order: ## Language config -A list of all languages and their properties is maintained centrally in `languages.csv`. +A list of all languages and their properties is maintained centrally in [`languages.csv`](languages.csv). This file should be kept up-to-date and all build methods should read this config file. **If you see something wrong, please report/fix it.** -Description of columns: -- Code, e,g, `Eng` -- Directory, e.g. `english` -- Functor (not used) -- Unlexer (not used) -- Present: languages that have `--# notpresent` marked -- All: languages for which to compile `All` -- Try: languages for which to compile `Try` -- Symbolic: languages for which to compile `Symbolic` -- Compatibility: languages for which to complile `Compatibility` +| # | Column | Description | Default | +|:---|:--------------|:-----------------------------------------|:-------:| +| 1 | Code | 3-letter language code, e.g. `Eng` | - | +| 2 | Name | language name in English, e.g. `English` | - | +| 3 | Directory | folder name under `src`, e.g. `english` | - | +| 4 | Functor | functor name (not used) | - | +| 5 | Unlexer | unlexer (not used) | - | +| 6 | Present | language is marked with `--# notpresent` | n | +| 7 | All | compile `All` module | y | +| 8 | Try | compile `Try` module | y | +| 9 | Symbolic | compile `Symbolic` module | y | +| 10 | Compatibility | complile `Compatibility` module | n | +| 11 | Synopsis | include language in the RGL synopsis | n | -Columns can be a string, just `y`'s (where nothing means `n`) or just (`n`'s where nothing means `y`). +If default is `y` then anything other than `n`, including the empty string, is treated as true (and vice versa when default is `n`). ## Haskell script: `Setup.hs` @@ -108,11 +111,9 @@ You can pass the following flags: ## Windows batch file: `Setup.bat` -**This script is still untested.** +This method is provided as an alternative for Windows users who don't have Haskell or Bash installed. -This method is provided as an alternative for Windows users who don't have Haskell installed. - -It is supposed to be a port of Setup.sh and works in largely the same way. +It is supposed to be a port of `Setup.sh` and works in largely the same way. In particular, it accepts the same flags (in the same format) as described above. However it currently tries to build all modules for all languages and doesn't consider the details of which modules should be compiled for each language (specified in `languages.csv`) diff --git a/Setup.bat b/Setup.bat index d7c77595..9dc085fb 100644 --- a/Setup.bat +++ b/Setup.bat @@ -12,35 +12,45 @@ set modules_langs=All Symbol Compatibility set modules_api=Try Symbolic REM Defaults (may be overridden by options) -set gf=gf-default +set gf=gf set dest= set verbose=false REM Check command line options -set arg_gf_next=false -set arg_dest_next=false -for %%i in (%*) do ( - if !arg_gf_next!==true ( - set gf=%%i - set arg_gf_next=false - ) - if !arg_dest_next!==true ( - set dest=%%i - set arg_dest_next=false - ) - if %%i==-v set verbose=true - if %%i==--verbose set verbose=true - if %%i==--gf set arg_gf_next=true - if %%i==--dest set arg_dest_next=true -) +:Loop + if '%1'=='' goto Continue + if %1==-v set verbose=true + if %1==--verbose set verbose=true + if %1==--gf set gf=%2 + if %1==--dest set dest=%2 + shift +goto Loop +:Continue REM Try to determine install location if "%dest%"=="" ( - set dest=%GF_LIB_PATH% + REM Separate paths with search path separator ; and pick first one + for %%p in ("%GF_LIB_PATH:;=";"%") do ( + set dest=%%~p + goto BreakLibPath + ) ) +:BreakLibPath + +set DATA_DIR=..\gf-core\DATA_DIR if "%dest%"=="" ( - REM TODO Look in ../gf-core/DATA=DIR + REM Look in already compiled GF folder + if exist %DATA_DIR% ( + for /f "delims=" %%x in (%DATA_DIR%) do ( + if not "%%x"=="" ( + set dest=%%x\lib + goto BreakDataDir + ) + ) + ) ) +:BreakDataDir + if "%dest%"=="" ( echo Unable to determine where to install the RGL. Please do one of the following: echo - Pass the --dest=... flag to this script @@ -52,17 +62,17 @@ if "%dest%"=="" ( REM A few more definitions before we get started set src=src set dist=dist -set gfc=gf --batch --gf-lib-path=%src% --quiet +set gfc=%gf% --batch --gf-lib-path=%src% -REM Redirect stderr if not verbose +REM Add quiet flag if not verbose if %verbose%==false ( - set gfc=2>NUL !gfc! + set gfc=%gfc% --quiet ) REM Make directories if not present -mkdir %dist%\prelude -mkdir %dist%\present -mkdir %dist%\alltenses +if not exist %dist%\prelude mkdir %dist%\prelude +if not exist %dist%\present mkdir %dist%\present +if not exist %dist%\alltenses mkdir %dist%\alltenses REM Build: prelude echo Building [prelude] @@ -74,13 +84,17 @@ REM Gather all language modules for building set modules= for %%l in (%langs%) do ( for %%m in (%modules_langs%) do ( - for /r %src% %%m in (*%%m%%l.gf) do ( - set modules=!modules! %%m + set patt=%%m%%l.gf + for /r %src% %%n in (!patt!) do ( + if exist %%n set modules=!modules! %%n ) ) +) +for %%l in (%langs%) do ( for %%m in (%modules_api%) do ( - for /r %src%\api %%m in (*%%m%%l.gf) do ( - set modules=!modules! %%m + set patt=%%m%%l.gf + for /r %src%\api %%n in (!patt!) do ( + if exist %%n set modules=!modules! %%n ) ) ) @@ -88,15 +102,25 @@ for %%l in (%langs%) do ( REM Build: present echo Building [present] for %%m in (%modules%) do ( + if %verbose%==true echo %%m %gfc% --no-pmcfg --gfo-dir=%dist%\present --preproc=mkPresent %%m ) REM Build: alltenses echo Building [alltenses] for %%m in (%modules%) do ( + if %verbose%==true echo %%m %gfc% --no-pmcfg --gfo-dir=%dist%\alltenses %%m ) +REM Make destination directories if not present +if not exist %dest% mkdir %dest% +if not exist %dest%\prelude mkdir %dest%\prelude +if not exist %dest%\present mkdir %dest%\present +if not exist %dest%\alltenses mkdir %dest%\alltenses + REM Copy echo Copying to %dest% -xcopy %dist% %dest% /d +copy %dist%\prelude\*.gfo %dest%\prelude\ +copy %dist%\present\*.gfo %dest%\present\ +copy %dist%\alltenses\*.gfo %dest%\alltenses\ diff --git a/Setup.hs b/Setup.hs index 2a40e398..76c21b35 100644 --- a/Setup.hs +++ b/Setup.hs @@ -2,19 +2,20 @@ -- | Main build script for RGL -import Data.List (find,isPrefixOf,isSuffixOf,(\\),unfoldr) +import Data.List (find,isPrefixOf,isSuffixOf,(\\)) import Data.Maybe (catMaybes) import System.IO (hPutStrLn,stderr) import System.IO.Error (catchIOError) import System.Exit (ExitCode(..),exitFailure) import System.Environment (getArgs,lookupEnv) import System.Process (rawSystem) -import System.FilePath (()) -- ,takeFileName,addExtension,dropExtension) +import System.FilePath ((),splitSearchPath) -- ,takeFileName,addExtension,dropExtension) import System.Directory (createDirectoryIfMissing,copyFile,getDirectoryContents,removeDirectoryRecursive,findFile) #if __GLASGOW_HASKELL__>=800 import System.Directory (getModificationTime,setModificationTime) #endif import Control.Monad (when,unless) +import Config main :: IO () main = do @@ -118,7 +119,7 @@ mkInfo = do -- Look for install location in a few different places let mflag = getFlag destination_flag args mbuilt <- catchIOError (readFile "../gf-core/DATA_DIR" >>= \d -> return (Just (d "lib"))) (\e -> return Nothing) - menvar <- lookupEnv "GF_LIB_PATH" + menvar <- lookupEnv "GF_LIB_PATH" >>= return . fmap (head . splitSearchPath) let inst_dir = case catMaybes [mflag,menvar,mbuilt] of @@ -347,57 +348,6 @@ verbose_switch_short = "-v" getFlag :: String -> [String] -> Maybe String getFlag flag args = fmap (drop (length flag)) $ find (isPrefixOf flag) args -------------------------------------------------------------------------------- --- Languages of the RGL - --- | Path to language config file -configFile :: FilePath -configFile = "languages.csv" - --- | Information about a language -data LangInfo = LangInfo - { langCode :: String -- ^ 3-letter ISO 639-2/B code - , langDir :: String -- ^ directory name - , langFunctor :: Maybe String -- ^ functor (not used) - , langUnlexer :: Maybe String -- ^ decoding for postprocessing linearizations - , langPresent :: Bool - , langAll :: Bool - , langTry :: Bool - , langSymbolic :: Bool - , langCompatibility :: Bool - } deriving (Show,Eq) - --- | Load language information from config file -loadLangs :: IO [LangInfo] -loadLangs = do - lns <- readFile configFile >>= return . lines - mapM mkLangInfo (tail lns) - where - maybeBit bits n = if length bits >= (n+1) && length (bits !! n) > 0 then Just (bits !! n) else Nothing - boolBit bits n def = if length bits >= (n+1) && length (bits !! n) > 0 then (if def then bits !! n /= "n" else bits !! n == "y") else def - mkLangInfo s = - let bits = separateBy ',' s in - if length bits < 2 - then die $ "Invalid entry in " ++ configFile ++ ": " ++ s - else return $ LangInfo - { langCode = bits !! 0 - , langDir = bits !! 1 - , langFunctor = maybeBit bits 2 - , langUnlexer = maybeBit bits 3 - , langPresent = boolBit bits 4 False - , langAll = boolBit bits 5 True - , langTry = boolBit bits 6 True - , langSymbolic = boolBit bits 7 True - , langCompatibility = boolBit bits 8 False - } - --- | Separate a string on a character --- Source: https://stackoverflow.com/a/4978733/98600 -separateBy :: Eq a => a -> [a] -> [[a]] -separateBy chr = unfoldr sep where - sep [] = Nothing - sep l = Just . fmap (drop 1) . break (== chr) $ l - ------------------------------------------------------------------------------- -- Executing GF @@ -444,7 +394,7 @@ execute command args = do -- | For parallel RGL module compilation -- Unfortunately, this has no effect unless compiled with -threaded -parallel_ :: (Foldable t, Monad m) => t (m a) -> m () +--parallel_ :: (Foldable t, Monad m) => t (m a) -> m () parallel_ ms = sequence_ ms -- do c <- newChan -- ts <- sequence [ forkIO (m >> writeChan c ()) | m <- ms] diff --git a/Setup.sh b/Setup.sh index ce73171c..6e8c5e1f 100755 --- a/Setup.sh +++ b/Setup.sh @@ -6,11 +6,11 @@ set -e # Get languages from config -langs=$(tail -n +2 languages.csv | awk -F ',' '{ if ($6 != "n") { print $1 } }') -langs_present=$(tail -n +2 languages.csv | awk -F ',' '{ if ($5 == "y") { print $1 } }') -langs_try=$(tail -n +2 languages.csv | awk -F ',' '{ if ($7 != "n") { print $1 } }') -langs_symbolic=$(tail -n +2 languages.csv | awk -F ',' '{ if ($8 != "n") { print $1 } }') -langs_compat=$(tail -n +2 languages.csv | awk -F ',' '{ if ($9 == "y") { print $1 } }') +langs=$(tail -n +2 languages.csv | awk -F ',' '{ if ($7 != "n") { print $1 } }') +langs_present=$(tail -n +2 languages.csv | awk -F ',' '{ if ($6 == "y") { print $1 } }') +langs_try=$(tail -n +2 languages.csv | awk -F ',' '{ if ($8 != "n") { print $1 } }') +langs_symbolic=$(tail -n +2 languages.csv | awk -F ',' '{ if ($9 != "n") { print $1 } }') +langs_compat=$(tail -n +2 languages.csv | awk -F ',' '{ if ($10 == "y") { print $1 } }') # Modules to compile for each language modules_langs="All Symbol Compatibility" @@ -36,7 +36,7 @@ done # Try to determine install location if [ -z "$dest" ]; then - dest="$GF_LIB_PATH" + dest=$(echo "$GF_LIB_PATH" | sed 's/:.*$//') fi if [ -z "$dest" ] && [ -f "../gf-core/DATA_DIR" ]; then dest=$(cat ../gf-core/DATA_DIR) diff --git a/doc/Makefile b/doc/Makefile index 556973dd..f766adca 100644 --- a/doc/Makefile +++ b/doc/Makefile @@ -1,4 +1,4 @@ -.PHONY: abstract synopsis index status +.PHONY: all index status synopsis abstract all: synopsis @@ -15,17 +15,17 @@ synopsis: synopsis.html S=../src -# List of languages extracted from MkSynopsis.hs -LANGS=Afr Ara Bul Cat Chi Dan Dut Eng Est Eus Fin Fre Ger Gre Hin Ice Ita Jpn Lav Mlt Mon Nep Nor Nno Pes Pnb Pol Por Ron Rus Snd Spa Swe Tha Urd +# List of languages extracted from languages.csv, with 'Synopsis' column == y +LANGS=$(shell cat ../languages.csv | cut -d',' -f1,11 | grep ',y' | cut -d',' -f1) # This list was constructed by observing what files MkSynopsis.hs reads -SRC_FILES=$S/abstract/Common.gf $S/abstract/Cat.gf $S/api/Constructors.gf $S/abstract/Structural.gf $(patsubst %,$S/*/Paradigms%.gf,$(LANGS)) +SRC_FILES=$(S)/abstract/Common.gf $(S)/abstract/Cat.gf $(S)/api/Constructors.gf $(S)/abstract/Structural.gf $(patsubst %,$S/*/Paradigms%.gf,$(LANGS)) EXAMPLES_OUT=$(patsubst %,api-examples-%.txt,$(LANGS)) INCLUDES=synopsis-intro.txt categories-intro.txt categories-imagemap.html synopsis-additional.txt synopsis-browse.txt synopsis-example.txt synopsis.html: MkSynopsis.hs MkExxTable.hs $(INCLUDES) $(EXAMPLES_OUT) $(SRC_FILES) - runghc MkSynopsis.hs + runghc -i.. MkSynopsis.hs categories.png: categories.dot dot -Tpng $^ > $@ @@ -37,11 +37,9 @@ abstract: $(GFDOC) -txthtml $S/abstract/*.gf mv $S/abstract/*.html abstract - api-examples.gfs: api-examples.txt MkExx.hs runghc MkExx.hs < $< > $@ - # Since .gfo files aren't self-contained, the dependencies given here are # incomplete. But I am thinking that the Try%.gfo file will always be newer # than any other files it depends on, so the rule will trigger when diff --git a/doc/MkExx.hs b/doc/MkExx.hs index 16eb38bf..0db35218 100644 --- a/doc/MkExx.hs +++ b/doc/MkExx.hs @@ -6,7 +6,7 @@ main = interact (unlines . concatMap mkScript . takeWhile (/="--.") . lines) mkScript l = case l of - ' ':_ -> + ' ':_ -> let ident = mkIdent $ unwords $ takeWhile (/="--") $ words l in [add $ psq ident] '-':_ -> [] @@ -30,45 +30,3 @@ mkIdent = concatMap unspec where ')' -> "" ':' -> "-" _ -> [c] - - - -langsCoding = [ - (("amharic", "Amh"),""), - (("arabic", "Ara"),""), - (("basque", "Eus"),""), - (("bulgarian","Bul"),""), - (("catalan", "Cat"),"Romance"), - (("danish", "Dan"),"Scand"), - (("dutch", "Dut"),""), - (("english", "Eng"),""), - (("finnish", "Fin"),""), - (("french", "Fre"),"Romance"), - (("hindi", "Hin"),"Hindustani"), - (("german", "Ger"),""), - (("interlingua","Ina"),""), - (("italian", "Ita"),"Romance"), - (("latin", "Lat"),""), - (("norwegian","Nor"),"Scand"), - (("polish", "Pol"),""), - (("punjabi", "Pnb"),""), - (("portuguese", "Por"), "Romance"), - (("romanian", "Ron"),""), - (("russian", "Rus"),""), - (("spanish", "Spa"),"Romance"), - (("swedish", "Swe"),"Scand"), - (("thai", "Tha"),""), - (("turkish", "Tur"),""), - (("urdu", "Urd"),"Hindustani") - ] - - -langs = map fst langsCoding - --- languagues for which Try is normally compiled -langsLang = langs `except` langsIncomplete - --- languages for which Lang can be compiled but which are incomplete -langsIncomplete = ["Amh","Ara","Hin","Lat","Pnb","Rus","Tha","Tur","Urd"] - -except ls es = filter (flip notElem es . snd) ls diff --git a/doc/MkExxTable.hs b/doc/MkExxTable.hs index a45effab..4cdab36c 100644 --- a/doc/MkExxTable.hs +++ b/doc/MkExxTable.hs @@ -4,10 +4,9 @@ module MkExxTable (getApiExx, ApiExx, prApiEx, mkEx) where import System.Environment(getArgs) import Control.Monad(when) import qualified Data.Map as M -import Data.Char main = do - xx <- getArgs + xx <- getArgs aexx <- getApiExx' True xx return () -- putStrLn $ prApiExx aexx @@ -16,7 +15,7 @@ getApiExx = getApiExx' False getApiExx' verbose xx = do s <- readFile (head xx) - let aet = getApiExxTrees $ filter validOutput $ mergeOutput $ lines s + let aet = getApiExxTrees $ filter validOutput $ mergeOutput $ lines s aeos <- mapM (readApiExxOne verbose) xx let aexx = mkApiExx $ ("API",aet) : aeos -- putStrLn $ prApiExx aexx @@ -57,7 +56,7 @@ cleanUp = dropWhile (flip elem " >") --- this makes txt2tags loop... mergeOutput ls = ls mergeOutputt ls = case ls of - l@('>':_):ll -> let (ll1,ll2) = span ((/=">") . take 1) ll in unwords (l : map (unwords . words) ll1) : mergeOutput ll2 + l@('>':_):ll -> let (ll1,ll2) = span ((/=">") . take 1) ll in unwords (l : map (unwords . words) ll1) : mergeOutput ll2 _:ll -> mergeOutput ll _ -> [] @@ -65,15 +64,15 @@ mergeOutputt ls = case ls of validOutput = (==">") . take 1 mkApiExx :: [(String,ApiExxOne)] -> ApiExx -mkApiExx ltes = - M.fromList [(t, - M.fromList [(l,maybe "NONE" id (M.lookup t te)) | (l,te) <- ltes]) +mkApiExx ltes = + M.fromList [(t, + M.fromList [(l,maybe "NONE" id (M.lookup t te)) | (l,te) <- ltes]) | t <- M.keys firstL] where firstL = snd (head ltes) prApiExx :: ApiExx -> String -prApiExx aexx = unlines +prApiExx aexx = unlines [unlines (t:prApiEx lexx) | (t,lexx) <- M.toList aexx] prApiEx :: M.Map String String -> [String] @@ -81,7 +80,7 @@ prApiEx apexx = case M.toList apexx of (a,e):lexx -> (a ++ ": ``" ++ unwords (words e) ++ "``"): [l ++ ": //" ++ mkEx l e ++ "//" | (l,e) <- lexx] -mkEx l = unws . bind . mkE . words where +mkEx l = unws . bind . mkE . words where unws = if elem l ["Chi","Jpn","Tha"] then concat else unwords -- remove spaces mkE e = case e of "atomic":"term":_ -> ["*"] @@ -101,6 +100,6 @@ bind ws = case ws of "&+":ws2 -> bind ws2 "Predef.BIND":ws2 -> bind ws2 "Predef.SOFT_BIND":ws2 -> bind ws2 - w : ws2 -> w : bind ws2 w : "++" : ws2 -> w : bind ws2 + w : ws2 -> w : bind ws2 _ -> ws diff --git a/doc/MkSynopsis.hs b/doc/MkSynopsis.hs index e2040fe4..777c3aad 100644 --- a/doc/MkSynopsis.hs +++ b/doc/MkSynopsis.hs @@ -1,27 +1,35 @@ import MkExxTable import System.Process(system) import System.Environment(getArgs) +import System.FilePath((),(<.>)) import Data.Char import Data.List -import qualified Data.ByteString.Char8 as BS import qualified Data.Map as M ----import Debug.Trace ---- +import Text.Printf +import Config type Cats = [(String,String,String)] type Rules = [(String,String,String)] -- the file generated +synopsis :: FilePath synopsis = "synopsis.txt" -- the language in which revealed examples are shown +revealedLang :: String revealedLang = "Eng" -- all languages shown (a copy of this list appears in Makefile) -apiExxFiles = ["api-examples-" ++ lang ++ ".txt" | lang <- words --- "Eng Chi" - "Afr Ara Bul Cat Chi Dan Dut Eng Est Eus Fin Fre Ger Gre Hin Ice Ita Jpn Lav Mlt Mon Nep Nor Nno Pes Pnb Pol Por Ron Rus Snd Spa Swe Tha Urd" - ] +apiExxFiles :: IO [FilePath] +apiExxFiles = do + langs <- loadLangsFrom (".." configFile) + return $ + [ "api-examples-" ++ (langCode lang) ++ ".txt" + | lang <- langs + , langSynopsis lang + ] +main :: IO () main = do xx <- getArgs let isLatex = case xx of @@ -31,7 +39,7 @@ main = do cs2 <- getCats catAPI let cs = sortCats (cs1 ++ cs2) writeFile synopsis "GF Resource Grammar Library: Synopsis" - append "B. Bringert, T. Hallgren, and A. Ranta" + -- append "B. Bringert, T. Hallgren, and A. Ranta" space append "%!Encoding:utf-8" append "%!style(html): ./revealpopup.css" @@ -66,7 +74,7 @@ main = do space link "Source 2:" structuralAPI space - apiExx <- getApiExx apiExxFiles + apiExx <- apiExxFiles >>= getApiExx rs <- getRules apiExx syntaxAPI --- putStrLn $ unlines ["p -cat=" ++ last (words t) ++ --- " \"" ++ e ++ "\"" | (_,t,e) <- rs, not (null e)] ---- @@ -83,7 +91,7 @@ main = do -- delimit rs space title "Lexical Paradigms" - mapM_ (putParadigms isLatex cs) paradigmFiles + paradigmFiles >>= mapM_ (putParadigms isLatex cs) space include "synopsis-additional.txt" space @@ -227,7 +235,6 @@ mkIdent = concatMap unspec where ':' -> "-" _ -> [c] - mkCatTable :: Bool -> Cats -> [String] mkCatTable isLatex cs = inChunks chsize (\rs -> header ++ map mk1 rs) cs where @@ -236,49 +243,36 @@ mkCatTable isLatex cs = inChunks chsize (\rs -> header ++ map mk1 rs) cs mk1 (name,expl,ex) = unwords ["|", showCat cs name, "|", expl, "|", typo ex, "|"] typo ex = if take 1 ex == "\"" then itf (init (tail ex)) else ex -srcPath = ("../src" ++) +srcPath = (() "../src") -commonAPI = srcPath "/abstract/Common.gf" -catAPI = srcPath "/abstract/Cat.gf" -syntaxAPI = srcPath "/api/Constructors.gf" -structuralAPI = srcPath "/abstract/Structural.gf" -paradigmFiles = [ - ("Afrikaans", srcPath "/afrikaans/ParadigmsAfr.gf"), - ("Arabic", srcPath "/arabic/ParadigmsAra.gf"), - ("Basque", srcPath "/basque/ParadigmsEus.gf"), - ("Bulgarian", srcPath "/bulgarian/ParadigmsBul.gf"), - ("Catalan", srcPath "/catalan/ParadigmsCat.gf"), - ("Chinese", srcPath "/chinese/ParadigmsChi.gf"), - ("Danish", srcPath "/danish/ParadigmsDan.gf"), - ("Dutch", srcPath "/dutch/ParadigmsDut.gf"), - ("English", srcPath "/english/ParadigmsEng.gf"), - ("Estonian", srcPath "/estonian/ParadigmsEst.gf"), - ("Finnish", srcPath "/finnish/ParadigmsFin.gf"), - ("French", srcPath "/french/ParadigmsFre.gf"), - ("German", srcPath "/german/ParadigmsGer.gf"), - ("Greek", srcPath "/greek/ParadigmsGre.gf"), - ("Hindi", srcPath "/hindi/ParadigmsHin.gf"), - ("Icelandic", srcPath "/icelandic/ParadigmsIce.gf"), --- ("Interlingua", srcPath "/interlingua/ParadigmsIna.gf"), - ("Italian", srcPath "/italian/ParadigmsIta.gf"), - ("Japanese", srcPath "/japanese/ParadigmsJpn.gf"), - ("Latvian", srcPath "/latvian/ParadigmsLav.gf"), - ("Maltese", srcPath "/maltese/ParadigmsMlt.gf"), - ("Mongolian", srcPath "/mongolian/ParadigmsMon.gf"), - ("Nepali", srcPath "/nepali/ParadigmsNep.gf"), - ("Norwegian", srcPath "/norwegian/ParadigmsNor.gf"), - ("Nynorsk", srcPath "/nynorsk/ParadigmsNno.gf"), - ("Polish", srcPath "/polish/ParadigmsPol.gf"), - ("Punjabi", srcPath "/punjabi/ParadigmsPnb.gf"), - ("Portuguese", srcPath "/portuguese/ParadigmsPor.gf"), - ("Romanian", srcPath "/romanian/ParadigmsRon.gf"), - ("Russian", srcPath "/russian/ParadigmsRus.gf"), - ("Sindhi", srcPath "/sindhi/ParadigmsSnd.gf"), - ("Spanish", srcPath "/spanish/ParadigmsSpa.gf"), - ("Swedish", srcPath "/swedish/ParadigmsSwe.gf"), - ("Thai", srcPath "/thai/ParadigmsTha.gf"), - ("Urdu", srcPath "/urdu/ParadigmsUrd.gf") - ] +commonAPI = srcPath "abstract/Common.gf" +catAPI = srcPath "abstract/Cat.gf" +syntaxAPI = srcPath "api/Constructors.gf" +structuralAPI = srcPath "abstract/Structural.gf" + +paradigmFiles :: IO [(String,FilePath)] +paradigmFiles = do + langs <- loadLangsFrom (".." configFile) + return $ + [ (name, srcPath $ printf "%s/Paradigms%s.gf" (langDir lang) (langCode lang)) + | lang <- langs + , langSynopsis lang + , let name = formatName (langDir lang) + ] + +-- | Format language name from directory name +-- "ancient_greek -> Ancient Greek" +formatName :: String -> String +formatName = unwords . map (\(s:ss) -> toUpper s : ss) . splitOn (=='_') + +-- | Split a string at given character, similar to words +splitOn :: (Char -> Bool) -> String -> [String] +splitOn _ "" = [] +splitOn f s = takeWhile (not.f) s : splitOn f rest + where + rest = case dropWhile (not.f) s of + "" -> [] + _:xs -> xs append s = appendFile synopsis ('\n':s) title s = append $ "=" ++ s ++ "=" @@ -339,7 +333,7 @@ showTyp cs = unwords . map f . words -- to work around GHC 6.12 file input readFileC cod file = do - let tmp = file ++ ".tmp" + let tmp = file <.> "tmp" case cod of "utf8" -> readFile file _ -> do diff --git a/doc/categories-imagemap.html b/doc/categories-imagemap.html deleted file mode 100644 index 31f01eef..00000000 --- a/doc/categories-imagemap.html +++ /dev/null @@ -1,53 +0,0 @@ - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - diff --git a/doc/language-list.txt b/doc/language-list.txt deleted file mode 100644 index 486353b6..00000000 --- a/doc/language-list.txt +++ /dev/null @@ -1,35 +0,0 @@ -Afrikaans -Amharic -Arabic -Bulgarian -Catalan -Chinese -Danish -Dutch -English -Finnish -French -German -Greek -Hebrew -Hindi -Interlingua -Japanese -Italian -Latin -Latvian -
  • Maltese -Nepali -Norwegian -Persian -Polish -Punjabi -Romanian -Russian -Sindhi -Spanish -Swahili -Swedish -Thai -Turkish -Urdu diff --git a/languages.csv b/languages.csv index 5af6e0a1..0bc0de3b 100644 --- a/languages.csv +++ b/languages.csv @@ -1,45 +1,45 @@ -Code,Directory,Functor,Unlexer,Present,All,Try,Symbolic,Compatibility -Afr,afrikaans,,,,,,n, -Amh,amharic,,,,,n,n, -Ara,arabic,,,,,,y, -Eus,basque,,,,,,, -Bul,bulgarian,,,y,,,, -Cat,catalan,Romance,,y,,,,y -Chi,chinese,,,,,,, -Dan,danish,Scand,,y,,,, -Dut,dutch,,,y,,,, -Eng,english,,,y,,,,y -Est,estonian,,,,,,, -Fin,finnish,,,y,,,,y -Fre,french,Romance,,y,,,,y -Grc,ancient_greek,,,y,,n,n, -Gre,greek,,,,,,, -Heb,hebrew,,,,,n,n, -Hin,hindi,Hindustani,to_devanagari,y,,,, -Hun,hungarian,,,y,n,n,n, -Ger,german,,,,,,, -Ice,icelandic,,,,,,n, -Ina,interlingua,,,y,,n,n, -Ita,italian,Romance,,y,,,,y -Jpn,japanese,,,,,,, -Lat,latin,,,y,,n,n, -Lav,latvian,,,,,,,y -Mlt,maltese,,,,,,, -Mon,mongolian,,,,,,n, -Nep,nepali,,,,,,n, -Nor,norwegian,Scand,,y,,,, -Nno,nynorsk,,,y,,,, -Pes,persian,,,,,,, -Pol,polish,,,,,,, -Por,portuguese,Romance,,y,,,,y -Pnb,punjabi,,,y,,,, -Ron,romanian,,,y,,,, -Rus,russian,,,y,,,, -Snd,sindhi,,,,,,, -Spa,spanish,Romance,,y,,,,y -Swa,swahili,,,,n,n,n,y -Swe,swedish,Scand,,y,,,,y -Tel,telugu,,,y,n,n,n, -Tha,thai,,to_thai,,,,, -Tur,turkish,,,,,n,n, -Urd,urdu,Hindustani,,,,,, +Code,Name,Directory,Functor,Unlexer,Present,All,Try,Symbolic,Compatibility,Synopsis +Afr,Afrikaans,afrikaans,,,,,,n,,y +Amh,Amharic,amharic,,,,,n,n,,n +Ara,Arabic,arabic,,,,,,y,,y +Bul,Bulgarian,bulgarian,,,y,,,,,y +Cat,Catalan,catalan,Romance,,y,,,,y,y +Chi,Chinese (simplified),chinese,,,,,,,,y +Dan,Danish,danish,Scand,,y,,,,,y +Dut,Dutch,dutch,,,y,,,,,y +Eng,English,english,,,y,,,,y,y +Est,Estonian,estonian,,,,,,,,y +Eus,Basque,basque,,,,,,,,y +Fin,Finnish,finnish,,,y,,,,y,y +Fre,French,french,Romance,,y,,,,y,y +Ger,German,german,,,,,,,,y +Grc,Ancient Greek,ancient_greek,,,y,,n,n,,n +Gre,Greek,greek,,,,,,,,y +Heb,Hebrew,hebrew,,,,,n,n,,n +Hin,Hindi,hindi,Hindustani,to_devanagari,y,,,,,y +Hun,Hungarian,hungarian,,,y,n,n,n,,n +Ice,Icelandic,icelandic,,,,,,n,,y +Ina,Interlingua,interlingua,,,y,,n,n,,n +Ita,Italian,italian,Romance,,y,,,,y,y +Jpn,Japanese,japanese,,,,,,,,y +Lat,Latin,latin,,,y,,n,n,,n +Lav,Latvian,latvian,,,,,,,y,y +Mlt,Maltese,maltese,,,,,,,,y +Mon,Mongolian,mongolian,,,,,,n,,y +Nep,Nepali,nepali,,,,,,n,,y +Nno,Norwegian (nynorsk),nynorsk,,,y,,,,,y +Nor,Norwegian (bokmål),norwegian,Scand,,y,,,,,y +Pes,Persian,persian,,,,,,,,y +Pnb,Punjabi,punjabi,,,y,,,,,y +Pol,Polish,polish,,,,,,,,y +Por,Portuguese,portuguese,Romance,,y,,,,y,y +Ron,Pomanian,romanian,,,y,,,,,y +Rus,Russian,russian,,,y,,,,,y +Snd,Sindhi,sindhi,,,,,,,,y +Spa,Spanish,spanish,Romance,,y,,,,y,y +Swa,Swahili,swahili,,,,n,n,n,y,n +Swe,Swedish,swedish,Scand,,y,,,,y,y +Tel,Telugu,telugu,,,y,n,n,n,,n +Tha,Thai,thai,,to_thai,,,,,,y +Tur,Turkish,turkish,,,,,n,n,,n +Urd,Urdu,urdu,Hindustani,,,,,,,y diff --git a/src/abstract/Common.gf b/src/abstract/Common.gf index 0e7a5763..3e852978 100644 --- a/src/abstract/Common.gf +++ b/src/abstract/Common.gf @@ -1,7 +1,7 @@ --1 Common: Structures with Common Implementations. --- This module defines the categories that uniformly have the --- linearization type ${s:Str}$ in all languages. +-- This module defines the categories that uniformly have the same +-- linearization type (usually ${s:Str}$) in all languages. abstract Common = { diff --git a/src/abstract/Construction.gf b/src/abstract/Construction.gf index ca2b39d7..69f5d1b4 100644 --- a/src/abstract/Construction.gf +++ b/src/abstract/Construction.gf @@ -78,7 +78,8 @@ cat Year ; fun - timeunitAdv : Card -> Timeunit -> Adv ; -- (for) three hours + timeunitAdv : Card -> Timeunit -> Adv ; -- (for) three hours + timeunitRange : Card -> Card -> Timeunit -> Adv ; -- (cats live) ten to twenty years oneHour : Hour ; twoHour : Hour ; @@ -105,8 +106,8 @@ fun twentyThreeHour : Hour ; twentyFourHour : Hour ; - timeHour : Hour -> Adv ; -- at three (o'clock / am / pm) - timeHourMinute : Hour -> Card -> Adv ; -- at forty past six + timeHour : Hour -> Adv ; -- at three a.m./p.m. + timeHourMinute : Hour -> Card -> Adv ; -- at six forty a.m./p.m. weekdayPunctualAdv : Weekday -> Adv ; -- on Monday weekdayHabitualAdv : Weekday -> Adv ; -- on Mondays diff --git a/src/abstract/Sentence.gf b/src/abstract/Sentence.gf index 6068223c..fc3c760b 100644 --- a/src/abstract/Sentence.gf +++ b/src/abstract/Sentence.gf @@ -70,7 +70,7 @@ abstract Sentence = Cat ** { -- This covers subjunctive clauses, but they can also be added to the end. - SSubjS : S -> Subj -> S -> S ; -- I go home if she comes + SSubjS : S -> Subj -> S -> S ; -- I go home, if she comes -- A sentence can be modified by a relative clause referring to its contents. diff --git a/src/arabic/AdjectiveAra.gf b/src/arabic/AdjectiveAra.gf index baae0935..93c64a42 100644 --- a/src/arabic/AdjectiveAra.gf +++ b/src/arabic/AdjectiveAra.gf @@ -21,7 +21,7 @@ concrete AdjectiveAra of Adjective = CatAra ** open ResAra, Prelude in { -- $SuperlA$ belongs to determiner syntax in $Noun$. -- ComplA2 a np = { - s = \\sp,g,n,st,c => a.s ! APosit g n st c ++ a.c2 ++ np.s ! Gen ; + s = \\sp,g,n,st,c => a.s ! APosit g n st c ++ a.c2.s ++ np.s ! a.c2.c ; } ; -- -- ReflA2 a = { @@ -37,7 +37,13 @@ concrete AdjectiveAra of Adjective = CatAra ** open ResAra, Prelude in { AdAP ada ap = { s = \\sp,g,n,st,c => ada.s ++ ap.s ! sp ! g ! n ! st ! c } ; --- --- UseA2 a = a ; --- + + UseA2 = PositA ; + + UseComparA a = { + s = \\h,g,n,d,c => a.s ! AComp d c + }; + + -- : Ord -> AP ; -- warmest + AdjOrd ord = {s = \\h,g,n,s,c => ord.s ! g ! s ! c} ; } diff --git a/src/arabic/AdverbAra.gf b/src/arabic/AdverbAra.gf index d7a90387..c02aa593 100644 --- a/src/arabic/AdverbAra.gf +++ b/src/arabic/AdverbAra.gf @@ -2,21 +2,24 @@ concrete AdverbAra of Adverb = CatAra ** open ResAra, Prelude in { flags coding=utf8; lin + PositAdvAdj a = {s = a.s ! APosit Masc Sg Indef Acc} ; --- ComparAdvAdj cadv a np = { --- s = cadv.s ++ a.s ! AAdv ++ "مِنْ" ++ np.s ! Gen --- } ; --- ComparAdvAdjS cadv a s = { --- s = cadv.s ++ a.s ! AAdv ++ "تهَن" ++ s.s --- } ; + -- ComparAdvAdj cadv a np = { + -- s = cadv.s ++ a.s ! AAdv ++ "مِنْ" ++ np.s ! Gen + -- } ; + -- ComparAdvAdjS cadv a s = { + -- s = cadv.s ++ a.s ! AAdv ++ "مِنْ" ++ s.s + -- } ; - PrepNP prep np = {s = prep.s ++ np.s ! Gen} ; + PrepNP prep np = {s = prep.s ++ np.s ! prep.c} ; + + AdAdv ad av = cc2 av ad ; + + -- : Subj -> S -> Adv ; -- when she sleeps + SubjS subj s = {s = subj.s ++ s.s ! subj.o} ; --- AdAdv = cc2 ; --- --- SubjS = cc2 ; -- AdvSC s = s ; --- this rule give stack overflow in ordinary parsing --- --- AdnCAdv cadv = {s = cadv.s ++ "تهَن"} ; --- + + AdnCAdv cadv = {s = cadv.s ++ "مِنْ"} ; + } diff --git a/src/arabic/AllAra.gf b/src/arabic/AllAra.gf index c331a6af..dc20a645 100644 --- a/src/arabic/AllAra.gf +++ b/src/arabic/AllAra.gf @@ -1,3 +1,3 @@ ---# -path=.:../abstract:../common:../prelude +--# -path=.:../abstract:../common:../api:../prelude concrete AllAra of AllAraAbs = LangAra ; diff --git a/src/arabic/CatAra.gf b/src/arabic/CatAra.gf index 3a1ee258..6cb7ccdb 100644 --- a/src/arabic/CatAra.gf +++ b/src/arabic/CatAra.gf @@ -10,35 +10,35 @@ concrete CatAra of Cat = CommonX - [Utt] ** open ResAra, Prelude, ParamX in { -- Tensed/Untensed - S = {s : Str} ; + SSlash, + S = {s : Order => Str} ; -- subordinate clause has nominal word order and subject in acc QS = {s : QForm => Str} ; --- RS = {s : Agr => Str} ; + RS = {s : Agr => Case => Str} ; -- case because the relative pronoun inflects in case -- Sentence - Cl = ResAra.Cl ; -- {s : ResAra.Tense => Polarity => Order => Str} ; + Cl = ResAra.Cl ; -- {s : Tense => Polarity => Order => Str} ; ClSlash = ResAra.ClSlash ; Imp = {s : Polarity => Gender => ResAra.Number => Str} ; -- Question - QCl = ResAra.QCl ; -- {s : ResAra.Tense => Polarity => QForm => Str} ; - IP, - IDet, - IComp = ResAra.IP ; -- {s : Gender => State => Case => Str ; n : ResAra.Number} ; - -- IAdv = {s : Str} ; + QCl = ResAra.QCl ; -- {s : Tense => Polarity => QForm => Str} ; + IDet = ResAra.IDet ; -- {s : Gender => State => Case => Str ; n : Number} ; + IP = ResAra.IP ; -- {s : (isPred : Bool) => State => Case => Str ; n : Number} ; + IComp = ResAra.IComp ; -- IQuant = {s : State => Case => Str} ; --- ----- Relative --- --- RCl = {s : Tense => Anteriority => Polarity => Agr => Str} ; --- RP = {s : Case => Str ; a : RAgr} ; --- + +-- Relative + + RCl = ResAra.RCl ; + RP = ResAra.RP ; + -- Verb VP = ResAra.VP ; - VPSlash = ResAra.VPSlash ; -- VP ** {c2:Str} - Comp = ResAra.Comp ; --{s : AAgr => Case => Str} ; + VPSlash = ResAra.VPSlash ; -- VP ** {c2:Preposition} + Comp = ResAra.Comp ** {obj : Obj ; isNP : Bool} ; -- SC = {s : Str} ; -- -- Adjective @@ -72,26 +72,30 @@ concrete CatAra of Cat = CommonX - [Utt] ** open ResAra, Prelude, ParamX in { Conj = {s : Str ; n : ResAra.Number} ; -- DConj = {s1,s2 : Str ; n : ResAra.Number} ; --- Subj = {s : Str} ; - Prep = {s : Str} ; + Subj = {s : Str ; o : Order} ; + Prep = ResAra.Preposition ; -- Open lexical classes, e.g. Lexicon V, VS, VQ, VA = ResAra.Verb ; -- = {s : VForm => Str} ; - V2, V2A = ResAra.Verb ** {c2 : Str} ; - VV, V2V, V2S, V2Q = ResAra.Verb ** {c2 : Str} ; --- AR - V3 = ResAra.Verb ** {c2, c3 : Str} ; + V2, V2A = ResAra.Verb2 ; + VV = ResAra.Verb2 ** {sc : Preposition} ; -- c2 is for verb + V2S, V2Q = ResAra.Verb2 ; + V3 = ResAra.Verb3 ; + V2V = ResAra.Verb3 ** {sc : Preposition} ; -- c3 is for verb, c2 is for dir.obj A = ResAra.Adj ; - A2 = ResAra.Adj ** {c2 : Str} ; + A2 = ResAra.Adj2 ; N = ResAra.Noun ; - N2 = ResAra.Noun ** {c2 : Str} ; - N3 = ResAra.Noun ** {c2, c3 : Str} ; + N2 = ResAra.Noun2 ; + N3 = ResAra.Noun3 ; PN = {s : Case => Str; g : Gender; h : Species} ; linref CN = \cn -> uttCN cn ! Masc ; + N = \n -> uttCN (useN n) ! Masc ; + VP = \vp -> uttVP vp ! Masc ; } diff --git a/src/arabic/ConjunctionAra.gf b/src/arabic/ConjunctionAra.gf index f7683c6c..0db6abd1 100644 --- a/src/arabic/ConjunctionAra.gf +++ b/src/arabic/ConjunctionAra.gf @@ -1,45 +1,82 @@ concrete ConjunctionAra of Conjunction = CatAra ** open ResAra, Coordination, Prelude in { --- --- flags optimize=all_subs ; --- --- lin --- --- ConjS = conjunctSS ; --- DConjS = conjunctDistrSS ; --- --- ConjAdv = conjunctSS ; --- DConjAdv = conjunctDistrSS ; --- --- ConjNP conj ss = conjunctTable Case conj ss ** { --- a = {n = conjNumber conj.n ss.a.n ; p = ss.a.p} --- } ; --- DConjNP conj ss = conjunctDistrTable Case conj ss ** { --- a = {n = conjNumber conj.n ss.a.n ; p = ss.a.p} --- } ; --- --- ConjAP conj ss = conjunctTable Agr conj ss ** { --- isPre = ss.isPre --- } ; --- DConjAP conj ss = conjunctDistrTable Agr conj ss ** { --- isPre = ss.isPre --- } ; --- ----- These fun's are generated from the list cat's. --- --- BaseS = twoSS ; --- ConsS = consrSS comma ; --- BaseAdv = twoSS ; --- ConsAdv = consrSS comma ; --- BaseNP x y = twoTable Case x y ** {a = conjAgr x.a y.a} ; --- ConsNP xs x = consrTable Case comma xs x ** {a = conjAgr xs.a x.a} ; --- BaseAP x y = twoTable Agr x y ** {isPre = andB x.isPre y.isPre} ; --- ConsAP xs x = consrTable Agr comma xs x ** {isPre = andB xs.isPre x.isPre} ; --- --- lincat --- [S] = {s1,s2 : Str} ; --- [Adv] = {s1,s2 : Str} ; --- [NP] = {s1,s2 : Case => Str ; a : Agr} ; --- [AP] = {s1,s2 : Agr => Str ; isPre : Bool} ; --- + +lincat + + [S] = {s1,s2 : Order => Str} ; + [Adv] = {s1,s2 : Str} ; + [NP] = {s1,s2 : Case => Str ; a : Agr ; empty : Str} ; + [AP] = {s1,s2 : Species => Gender => Number => State => Case => Str} ; + +lin + + + BaseAdv = twoSS ; + ConsAdv = consrSS comma ; + ConjAdv = conjunctSS ; + + BaseS = twoTable Order ; + ConsS = consrTable Order comma ; + ConjS = conjunctTable Order ; + + BaseNP x y = twoTable Case x y ** { + a = conjAgr x.a y.a ; + empty = [] + } ; + ConsNP xs x = consrTable Case comma xs x ** { + a = conjAgr xs.a x.a ; + empty = [] + } ; + ConjNP conj ss = conjunctTable Case conj ss ** { + a = let gn = pgn2gn ss.a.pgn in + {pgn = Per3 gn.g (conjNumber conj.n gn.n) ; isPron = False} ; + empty = [] + } ; + + BaseAP = twoTable5 Species Gender Number State Case ; + ConsAP = consrTable5 Species Gender Number State Case comma ; + ConjAP = conjunctTable5 Species Gender Number State Case ; + + +oper + conjAgr : Agr -> Agr -> Agr = \a,b -> { + isPron = False ; + pgn = let gnA = pgn2gn a.pgn ; gnB = pgn2gn b.pgn in + Per3 (conjGender gnA.g gnB.g) (conjNumber gnA.n gnB.n) + } ; + + conjGender : Gender -> Gender -> Gender = \g,h -> + case g of {Fem => h ; _ => Masc} ; + + conjNumber : Number -> Number -> Number = \m,n -> + case m of {Sg => n ; _ => Pl} ; + + -- move to predef? + + ListTable5 : PType -> PType -> PType -> PType -> PType -> Type = \P,Q,R,T,S -> + {s1,s2 : P => Q => R => T => S => Str} ; + + twoTable5 : (P,Q,R,T,S : PType) -> (_,_ : {s : P => Q => R => T => S => Str}) -> + ListTable5 P Q R T S = + \_,_,_,_,_,x,y -> + {s1 = x.s ; s2 = y.s} ; + + consrTable5 : + (P,Q,R,T,S : PType) -> Str -> {s : P => Q => R => T => S => Str} -> + ListTable5 P Q R T S -> ListTable5 P Q R T S = + \P,Q,R,T,S,c,x,xs -> + {s1 = \\p,q,r,t,s => xs.s1 ! p ! q ! r ! t ! s ++ c ++ xs.s2 ! p ! q ! r ! t ! s ; + s2 = x.s + } ; + + conjunctTable5 : + (P,Q,R,T,S : PType) -> Conjunction -> ListTable5 P Q R T S -> {s : P => Q => R => T => S => Str} = + \P,Q,R,T,S,or,xs -> + {s = \\p,q,r,t,s => xs.s1 ! p ! q ! r ! t ! s ++ or.s ++ xs.s2 ! p ! q ! r ! t ! s} ; + + -- conjunctDistrTable5 : + -- (P,Q,R,T,S : PType) -> ConjunctionDistr -> ListTable5 P Q R T S -> + -- {s : P => Q => R => T => S => Str} = + -- \P,Q,R,T,S,or,xs -> + -- {s = \\p,q,r,t,s => or.s1++ xs.s1 ! p ! q ! r ! t ! s ++ or.s2 ++ xs.s2 ! p ! q ! r ! t ! s} ; } diff --git a/src/arabic/ConstructionAra.gf b/src/arabic/ConstructionAra.gf new file mode 100644 index 00000000..95a37061 --- /dev/null +++ b/src/arabic/ConstructionAra.gf @@ -0,0 +1,156 @@ +concrete ConstructionAra of Construction = CatAra ** open + Prelude, + ParadigmsAra, + SyntaxAra, + SymbolicAra, + StructuralAra, + (E=ExtendAra), + (R=ResAra), + (L=LexiconAra) in { + +lincat + Timeunit = N ; + Weekday = N ; + Monthday = NP ; + Month = N ; + Year = NP ; + +lin + + timeunitAdv n time = + let n_card : Card = n ; + n_hours_NP : NP = mkNP n_card time ; + in SyntaxAra.mkAdv during_Prep n_hours_NP | ParadigmsAra.mkAdv (n_hours_NP.s ! R.Nom) ; + + -- random guesses + weekdayPunctualAdv w = SyntaxAra.mkAdv on_Prep (mkNP w) ; -- on Sunday + weekdayHabitualAdv w = SyntaxAra.mkAdv on_Prep (mkNP w) ; -- on Sundays + weekdayNextAdv w = SyntaxAra.mkAdv on_Prep (mkNP w) ; -- next Sunday + weekdayLastAdv w = SyntaxAra.mkAdv on_Prep (mkNP w) ; -- last Sunday + + monthAdv m = SyntaxAra.mkAdv in_Prep (mkNP m) ; + yearAdv y = SyntaxAra.mkAdv in_Prep y ; + + -- dummy + dayMonthAdv d m = SyntaxAra.mkAdv on_Prep (mkNP d) ; -- on 17 May + monthYearAdv m y = SyntaxAra.mkAdv on_Prep (mkNP m) ; -- in May 2012 + dayMonthYearAdv d m y = SyntaxAra.mkAdv on_Prep y ; -- on 17 May 2013 + + intYear = symb ; + intMonthday = symb ; + +-- n_units_AP + + +oper + -- hack used in the name constructions + toNP : Bool -> NP -> NP = \b -> if_then_else NP b R.emptyNP ; + +lin + -- : NP -> NP -> Cl + have_name_Cl np nm = + let subjPron : Pron = R.np2pron np ; + me : NP = toNP np.a.isPron np ; + myName : NP = E.ApposNP me (mkNP (mkDet subjPron) L.name_N) ; + in mkCl myName nm ; + + -- : NP -> QCl + what_name_QCl np = + let subjPron : Pron = R.np2pron np ; + me : R.NP = toNP np.a.isPron np ; + myName : NP = E.ApposNP me (mkNP (mkDet subjPron) L.name_N) ; + what_IP : R.IP = R.mkIP "مَا هُوَ" R.Sg ; + in mkQCl what_IP myName ; + +-- how_old_QCl + +-- hungry_VP = +-- thirsty_VP = + +lincat Language = N ; + +lin InLanguage l = mkAdv in_Prep (mkNP l) ; + +lin + weekdayN w = w ; + monthN m = m ; + + weekdayPN w = mkPN w ; + monthPN m = mkPN m ; + + languageCN l = mkCN l ; + languageNP l = mkNP l ; + + +oper mkLanguage : Str -> N = mkN ; + +---------------------------------------------- +---- lexicon of snpcial names + +-- TODO in arabic +lin second_Timeunit = mkN "second" ; +lin minute_Timeunit = mkN "minute" ; +lin hour_Timeunit = mkN "hour" ; +lin day_Timeunit = mkN "day" ; +lin week_Timeunit = mkN "week" ; +lin month_Timeunit = mkN "month" ; +lin year_Timeunit = mkN "year" ; + +lin monday_Weekday = mkN "Monday" ; +lin tuesday_Weekday = mkN "Tuesday" ; +lin wednesday_Weekday = mkN "Wednesday" ; +lin thursday_Weekday = mkN "Thursday" ; +lin friday_Weekday = mkN "Friday" ; +lin saturday_Weekday = mkN "Saturday" ; +lin sunday_Weekday = mkN "Sunday" ; + +lin january_Month = mkN "January" ; +lin february_Month = mkN "February" ; +lin march_Month = mkN "March" ; +lin april_Month = mkN "April" ; +lin may_Month = mkN "May" ; +lin june_Month = mkN "June" ; +lin july_Month = mkN "July" ; +lin august_Month = mkN "August" ; +lin september_Month = mkN "September" ; +lin october_Month = mkN "October" ; +lin november_Month = mkN "November" ; +lin december_Month = mkN "December" ; + +-- lin afrikaans_Language = mkLanguage "Afrikaans" ; +-- lin amharic_Language = mkLanguage "Amharic" ; +lin arabic_Language = mkLanguage "عَرَبِيَّة" ; +-- lin bulgarian_Language = mkLanguage "Bulgarian" ; +-- lin catalan_Language = mkLanguage "Catalan" ; +-- lin chinese_Language = mkLanguage "Chinese" ; +-- lin danish_Language = mkLanguage "Danish" ; +-- lin dutch_Language = mkLanguage "Dutch" ; +lin english_Language = mkLanguage "إنْجلِيزيْة" ; +-- lin estonian_Language = mkLanguage "Estonian" ; +lin finnish_Language = mkLanguage "فِنْلَنْدِيّة" ; +-- lin french_Language = mkLanguage "French" ; +-- lin german_Language = mkLanguage "German" ; +-- lin greek_Language = mkLanguage "Greek" ; +-- lin hebrew_Language = mkLanguage "Hebrew" ; +-- lin hindi_Language = mkLanguage "Hindi" ; +-- lin japanese_Language = mkLanguage "Japanese" ; +-- lin italian_Language = mkLanguage "Italian" ; +-- lin latin_Language = mkLanguage "Latin" ; +-- lin latvian_Language = mkLanguage "Latvian" ; +-- lin maltese_Language = mkLanguage "Maltese" ; +-- lin nepali_Language = mkLanguage "Nepali" ; +-- lin norwegian_Language = mkLanguage "Norwegian" ; +lin nprsian_Language = mkLanguage "فَارِسيّة" ; +-- lin polish_Language = mkLanguage "Polish" ; +-- lin punjabi_Language = mkLanguage "Punjabi" ; +-- lin romanian_Language = mkLanguage "Romanian" ; +-- lin russian_Language = mkLanguage "Russian" ; +-- lin sindhi_Language = mkLanguage "Sindhi" ; +-- lin spanish_Language = mkLanguage "Spanish" ; +-- lin swahili_Language = mkLanguage "Swahili" ; +lin swedish_Language = mkLanguage "سُويدِيّة" ; +-- lin thai_Language = mkLanguage "Thai" ; +-- lin turkish_Language = mkLanguage "Turkish" ; +-- lin urdu_Language = mkLanguage "Urdu" ; + +} diff --git a/src/arabic/ExtendAra.gf b/src/arabic/ExtendAra.gf index 43c35f59..a371dd23 100644 --- a/src/arabic/ExtendAra.gf +++ b/src/arabic/ExtendAra.gf @@ -4,16 +4,56 @@ concrete ExtendAra of Extend = CatAra ** ExtendFunctor - [ GenNP, SlashBareV2S, PredAPVP, GenModNP, ExistsNP, StrandRelSlash, ExistPluralCN, ExistMassCN, ExistCN, EmptyRelSlash, DetNPMasc, DetNPFem, - ComplBareVS, ComplDirectVS, ComplDirectVQ + ComplBareVS, ComplDirectVS, ComplDirectVQ, + ICompAP, + VPS, MkVPS, PredVPS, BaseVPS, ConsVPS, ConjVPS, + ApposNP ] with (Grammar=GrammarAra) ** open + ParamX, + ResAra, Prelude, - ResAra + Coordination in { lin - GenNP np = { s = \\_,_,_,_ => np.s ! Gen ; d = Const ; isNum, isPron = False } ; -} ; + GenNP np = {s = \\_,_,_,_ => np.s ! Gen ; d = Const ; isNum,isPron,is1sg = False} ; + + -- : NP -> NP -> NP + ApposNP np1 np2 = np2 ** {s = \\c => np1.s ! c ++ np2.s ! c} ; + + -- : AP -> IComp ; -- "how old" + ICompAP ap = {s = \\gn => "كَمْ" ++ ap.s ! NoHum ! gn.g ! gn.n ! Indef ! Acc} ; + + lincat + + VPS = {s : PerGenNum => Str} ; -- finite VP's with tense and polarity + [VPS] = {s1,s2 : PerGenNum => Str} ; + lin + -- : Temp -> Pol -> VP -> VPS ; -- hasn't slept + MkVPS t p vp = { + s = \\pgn => let vps = + wordOrderNoSubj + Nominal -- Nominal (=SVO) generalises best for ConjVPS. + vp.obj.a.isPron + (vStr vp pgn t.t p.p) + (case of { + => BIND ++ vp.obj.s ; + _ => vp.obj.s }) + (pred vp pgn t.t p.p) + vp.s2 + in vps.before ++ vps.after -- word order is SVO, so this is safe for just this case. + } ; + + BaseVPS = twoTable PerGenNum ; + ConsVPS = consrTable PerGenNum comma ; + ConjVPS = conjunctTable PerGenNum ; + + PredVPS np vps = { + s = \\_ => np.s ! Nom ++ vps.s ! np.a.pgn -- first quick version with order always Nominal. + } ; -- if necessary, change VPS into {s : PerGenNum => Order => {before,after : Str}} + +} diff --git a/src/arabic/IdiomAra.gf b/src/arabic/IdiomAra.gf index fbfc5f9b..7a18b133 100644 --- a/src/arabic/IdiomAra.gf +++ b/src/arabic/IdiomAra.gf @@ -1,15 +1,63 @@ -concrete IdiomAra of Idiom = CatAra ** open Prelude, ResAra in { - flags coding=utf8; --- --- flags optimize=all_subs ; --- --- lin --- ExistNP np = --- mkClause "تهري" (agrP3 np.a.n) (insertObj (\\_ => np.s ! Acc) (predAux auxBe)) ; --- ImpersCl vp = mkClause "ِت" (agrP3 Sg) vp ; --- GenericCl vp = mkClause "ْني" (agrP3 Sg) vp ; --- --- ProgrVP vp = insertObj (\\a => vp.ad ++ vp.prp ++ vp.s2 ! a) (predAux auxBe) ; --- -} +concrete IdiomAra of Idiom = CatAra ** open + Prelude, + ResAra, + VerbAra, + ParadigmsAra + in { + + lin + + -- : VP -> Cl ; -- it is hot + ImpersCl vp = + let it : ResAra.NP = case vp.isPred of { + True => pron2np (pgn2pron vp.obj.a.pgn) ; + False => pgn2pron vp.obj.a.pgn } ; -- if no obj, Per3 Masc Sg chosen by default + in predVP it vp ; + + -- : VP -> Cl ; -- one sleeps + GenericCl = predVP (regNP "المَرْء" Sg) ; + + -- : NP -> RS -> Cl ; -- it is I who did it + --CleftNP np rs = + + -- : Adv -> S -> Cl ; -- it is here she slept + CleftAdv adv s = + let comp : Comp = CompAdv (lin Adv {s = adv.s ++ s.s ! Verbal}) ; -- no idea about word order /IL + pass_V = mkV "مضي" va vi ; -- switch to copula or some other verb if better /IL + in predVP emptyNP (UseV pass_V ** {isPred=True ; pred=comp}) ; -- very hacky /IL + + -- : NP -> Cl ; -- there is a house + ExistNP np = + predVP (emptyNP ** {s=\\c=>"هُنَاكَ"}) (UseComp (CompNP np)) ; -- IL + + -- ExistIP : IP -> QCl ; -- which houses are there + +-- 7/12/2012 generalizations of these + + -- : NP -> Adv -> Cl ; -- there is a house in Paris + ExistNPAdv np adv = + predVP (emptyNP ** {s=\\c=>"هُنَاكَ"}) (AdvVP (UseComp (CompNP np)) adv) ; -- IL + + -- ExistIPAdv : IP -> Adv -> QCl ; -- which houses are there in Paris + + -- ProgrVP : VP -> VP ; -- be sleeping + + -- ImpPl1 : VP -> Utt ; -- let's go + + -- ImpP3 : NP -> VP -> Utt ; -- let John walk + +-- 3/12/2013 non-reflexive uses of "self" + + -- : VP -> VP ; -- is at home himself; is himself at home + SelfAdvVP, + SelfAdVVP = \vp -> vp ** { + s = \\pgn,vf => vp.s ! pgn ! vf ++ reflPron Nom pgn + } ; + + -- : NP -> NP ; -- the president himself (is at home) + SelfNP np = np ** { + s = \\c => np.s ! c ++ reflPron c (np.a.pgn) + } ; + +} diff --git a/src/arabic/LangAra.gf b/src/arabic/LangAra.gf index 55c9008b..bf4be4c7 100644 --- a/src/arabic/LangAra.gf +++ b/src/arabic/LangAra.gf @@ -1,8 +1,9 @@ ---# -path=.:../abstract:../common:../prelude +--# -path=.:../abstract:../common:../api:../prelude concrete LangAra of Lang = GrammarAra, - LexiconAra + LexiconAra, + ConstructionAra ** { flags startcat = Phr ; unlexer = text ; lexer = text ; coding = utf8 ; diff --git a/src/arabic/LexiconAra.gf b/src/arabic/LexiconAra.gf index 7d4e9507..ca5571cf 100644 --- a/src/arabic/LexiconAra.gf +++ b/src/arabic/LexiconAra.gf @@ -3,7 +3,6 @@ concrete LexiconAra of Lexicon = CatAra ** open ParadigmsAra, ResAra, - MorphoAra, --shouldn't open it here, only needed reg &sndf Prelude in { flags @@ -19,13 +18,13 @@ flags ask_V2Q = dirV2 (regV "يَسءَل") ; -- ask_V2Q = dirV2 (v1 "سءل" a a) ; baby_N = brkN "طفل" "فِعل" "أَفعَال" Masc Hum; --- bad_A = sndA "سوء" "فَيِّع" ; - bad_A = degrA "سَيِّئ" "سَيِّئَة" "سَيِّئِين" ; +-- bad_A = sndA "سوء" "فَيِّع" ; + bad_A = degrA "سَيِّئ" "سَيِّئَة" "سَيِّئِين" ; bank_N = brkN "بنك" "فَعل" "فُعُول" Masc NoHum ; beautiful_A = sndA "جمل" "فَعِيل" ; become_VA = mkVA (v4 "صبح") ; beer_N = sdfN "بير" "فِعلة" Fem NoHum ; - beg_V2V = dirV2 (v5 "وسل") ; + beg_V2V = mkV2V (mkVV (v5 "وسل")) noPrep ; big_A = sndA "كبر" "فَعِيل" ; bike_N = sdfN "درج" "فَعّالة" Fem NoHum ; bird_N = brkN "طير" "فَعل" "فُعُول" Masc NoHum; @@ -42,11 +41,11 @@ flags -- break_V2 = dirV2 (v1 "كسر" a u) ; broad_A = sndA "وسع" "فاعِل" ; brother_N2 = mkN2 (brkN "ءخو" "فَع" "فِعلة" Masc Hum) ; --FIXME dual - brown_A = sndA "بني" "فُعِّل"; + brown_A = sndA "بني" "فُعِّل"; butter_N = sdfN "سبد" "فُعلة" Fem NoHum ; buy_V2 = dirV2 (v8 "شري") ; camera_N = sdfN "كمر" "فاعِيلا" Fem NoHum ; -- |Alö taSwIr - cap_N = sdfN "قبع" "فُعَّلة" Fem NoHum ; --qalnUsö + cap_N = sdfN "قبع" "فُعَّلة" Fem NoHum ; --qalnUsö car_N = sdfN "سير" "فَعّالة" Fem NoHum ; carpet_N = sdfN "سجد" "فَعّالة" Fem NoHum ; cat_N = brkN "هرّ" "فِعّ" "فِعَلَة" Fem NoHum ; @@ -74,7 +73,7 @@ flags door_N = brkN "بوب" "فاع" "أَفعَال" Masc NoHum ; drink_V2 = dirV2 (regV "شَرِب") ; -- drink_V2 = dirV2 (v1 "شرب" i a) ; - easy_A2V = mkA2 (sndA "سهل" "فَعل") "لِ" ; + easy_A2V = mkA2 (sndA "سهل" "فَعل") liPrep ; eat_V2 = dirV2 (mkV "ءكل" FormI) ; empty_A = sndA "فرغ" "فاعِل" ; enemy_N = brkN "عدو" "فَعُلّ" "أَفعَاء" Masc Hum ; @@ -93,13 +92,13 @@ flags girl_N = brkN "بنت" "فِعل" "فَعَال" Fem Hum ; glove_N = sdfN "قفز" "فُعّال" Masc NoHum ; gold_N = sdfN "ذهب" "فَعَل" Masc NoHum ; - good_A = sndA "جود" "فَيِّع" ; -- Hasan, HisAn + good_A = sndA "جود" "فَيِّع" ; -- Hasan, HisAn go_V = regV "يَذهَب" ; -- go_V = v1 "ذهب" a a ; green_A = clrA "خضر" ; harbour_N = brkN "رفء" "مَفعَل" "مَفاعِل" Masc NoHum ; --mInA', marsaY hate_V2 = dirV2 (regV "كَرِه") ; - hat_N = sdfN "قبع" "فُعَّلة" Fem NoHum ; + hat_N = sdfN "قبع" "فُعَّلة" Fem NoHum ; -- have_V2 = dirV2 (v1 "ملك" a i) ; hear_V2 = dirV2 (regV "سَمِع") ; -- hear_V2 = dirV2 (v1 "سمع" i a) ; @@ -130,14 +129,14 @@ flags love_N = brkN "حبّ" "فُعّ" "فُعّ" Masc NoHum ; -- no plur love_V2 = dirV2 (v1 "حبّ" a i) ; man_N = brkN "رجل" "فَعُل" "فِعَال" Masc Hum ; - married_A2 = mkA2 (sndA "زوج" "مُتَفَعِّل") "مِن" ; + married_A2 = mkA2 (sndA "زوج" "مُتَفَعِّل") "مِن" ; meat_N = brkN "لحم" "فَعلة" "فُعُول" Masc NoHum ; milk_N = brkN "حلب" "فَعِيل" "فَعِيل" Masc NoHum ; --no plur moon_N = brkN "قمر" "فَعَل" "أَفعَال" Masc NoHum ; - mother_N2 = mkN2 (sdfN "ءم" "فُعَّ" Fem Hum) ; + mother_N2 = mkN2 (sdfN "ءم" "فُعَّ" Fem Hum) ; mountain_N = brkN "جبل" "فَعَل" "فِعَال" Masc NoHum ; music_N = mkN (reg "مُوسِيقَى" "مُوسِيقَى") Fem NoHum ; --no plur - narrow_A = sndA "ضيق" "فَعِّل" ; + narrow_A = sndA "ضيق" "فَعِّل" ; new_A = mkA "جدد" "فَعِيل" "فُعُل" ; newspaper_N = brkN "صحف" "فَعِيلة" "فُعُل" Fem NoHum ; oil_N = brkN "زيت" "فَعل" "فُعُول" Masc NoHum ; @@ -155,7 +154,7 @@ flags play_V2 = dirV2 (regV "لَعِب") ; -- play_V2 = dirV2 (v1 "لعب" i a) ; policeman_N = sdmN "شرط" "فِعلِي" Masc Hum ; - priest_N = brkN "قسّ" "فِعِّيل" "أَفِعّة" Masc Hum ; + priest_N = brkN "قسّ" "فِعِّيل" "أَفِعّة" Masc Hum ; probable_AS = mkAS (sndA "مكن" "مُفعِل") ; queen_N = sdfN "ملك" "فَعِلة" Fem Hum ; radio_N = mkN (sndf "راديُو") Masc NoHum ; @@ -169,7 +168,7 @@ flags river_N = brkN "نهر" "فَعل" "أَفعَال" Masc NoHum ; rock_N = brkN "صخر" "فَعلة" "فُعُول" Fem NoHum ; roof_N = brkN "سطح" "فَعل" "أَفعُل" Masc NoHum ; - rubber_N = brkN "مطط" "فَعَّال" "فَعَّال" Masc NoHum ; -- no hum + rubber_N = brkN "مطط" "فَعَّال" "فَعَّال" Masc NoHum ; -- no hum run_V = regV "يَركُض" ; -- run_V = v1 "ركض" a u ; say_VS = mkVS (v1 "قول" a u) ; --check @@ -187,11 +186,11 @@ flags shoe_N = brkN "حذو" "فِعَاء" "أَفعِية" Masc NoHum ; shop_N = brkN "تجر" "مَفعَل" "مَفاعِل" Masc NoHum ; short_A = sndA "قصر" "فَعِيل" ; - silver_N = brkN "فضض" "فِعَّة" "فِعَل" Fem NoHum ; + silver_N = brkN "فضض" "فِعَّة" "فِعَل" Fem NoHum ; sister_N = brkN "ءخو" "فُعت" "فَعَوَات" Fem Hum ; --FIXME sleep_V = v1 "نوم" i a ; --check small_A = sndA "صغر" "فَعِيل" ; - snake_N = sdfN "حيّ" "فَعَّة" Fem NoHum ; + snake_N = sdfN "حيّ" "فَعَّة" Fem NoHum ; sock_N = brkN "جرب" "فَوعَل" "فَواعِل" Masc NoHum ; speak_V2 = dirV2 (v5 "كلم") ; star_N = brkN "نجم" "فَعل" "فُعُول" Masc NoHum ; --najmö @@ -204,8 +203,8 @@ flags switch8off_V2 = dirV2 (v4 "طفء") ; switch8on_V2 = dirV2 (v4 "شعل") ; table_N = sdfN "طول" "فاعِلة" Fem NoHum ; - talk_V3 = mkV3 (v5 "حدث") "لِ" "عَن" ; - teacher_N = sdmN "علم" "مُفَعِّل" Masc Hum ; --mucal~imö + talk_V3 = mkV3 (v5 "حدث") liPrep (mkPrep "عَن") ; + teacher_N = sdmN "علم" "مُفَعِّل" Masc Hum ; --mucal~imö teach_V2 = dirV2 (v2 "علم") ; television_N = mkN (sndf "تِلِفِزيُون") Masc NoHum ; thick_A = sndA "سمك" "فَعِيل" ; @@ -261,7 +260,7 @@ flags heavy_A = sndA "ثقل" "فَعِيل" ; near_A = sndA "قرب" "فَعِيل" ; rotten_A = sndA "فسد" "فاعِل" ; - round_A = sndA "دور" "مُفَعَّل" ; + round_A = sndA "دور" "مُفَعَّل" ; sharp_A = sndA "حدّ" "فاعّ" ; smooth_A = sndA "نعم" "فاعِل" ; straight_A = sndA "قوم" "مُستَفِيع" ; @@ -277,7 +276,7 @@ flags bone_N = brkN "عظم" "فَعلة" "فِعَال" Fem NoHum; breast_N = brkN "صدر" "فَعل" "فُعُول" Masc NoHum; cloud_N = brkN "غيم" "فَعلة" "فُعُول" Fem NoHum; - day_N = brkN "يوم" "فَعل" "أَفَّاع" Masc NoHum; + day_N = brkN "يوم" "فَعل" "أَفَّاع" Masc NoHum; dust_N = brkN "غبر" "فُعَال" "أَفعِلة" Masc NoHum; ear_N = brkN "ءذن" "فُعل" "أَفعَال" Fem NoHum; earth_N = brkN "ترب" "فُعلة" "فُعَل" Fem NoHum; @@ -296,7 +295,7 @@ flags hair_N = sdfN "شعر" "فَعلة" Fem NoHum ; hand_N = brkN "يد" "فَع" "أَفَاعِي" Fem NoHum ; head_N = brkN "رءس" "فَعل" "فُعُول" Masc NoHum; - heart_N = brkN "قلب" "فَعل" "فُعُول" Masc NoHum; + heart_N = brkN "قلب" "فَعْل" "فُعُول" Masc NoHum; horn_N = brkN "قرن" "فَعل" "فُعُول" Masc NoHum; husband_N = brkN "زوج" "فَعل" "أَفعَال" Masc NoHum; ice_N = brkN "ثلج" "فَعل" "فُعُول" Masc NoHum; diff --git a/src/arabic/MissingAra.gf b/src/arabic/MissingAra.gf index 7d82c4eb..e714eb0e 100644 --- a/src/arabic/MissingAra.gf +++ b/src/arabic/MissingAra.gf @@ -3,43 +3,21 @@ resource MissingAra = open GrammarAra, Prelude in { -- temporary definitions to enable the compilation of RGL API oper AdAdv : AdA -> Adv -> Adv = notYet "AdAdv" ; oper AdVVP : AdV -> VP -> VP = notYet "AdVVP" ; -oper AdjOrd : Ord -> AP = notYet "AdjOrd" ; oper AdnCAdv : CAdv -> AdN = notYet "AdnCAdv" ; -oper AdvCN : CN -> Adv -> CN = notYet "AdvCN" ; oper AdvIAdv : IAdv -> Adv -> IAdv = notYet "AdvIAdv" ; -oper AdvS : Adv -> S -> S = notYet "AdvS" ; -oper BaseAP : AP -> AP -> ListAP = notYet "BaseAP" ; -oper BaseAdv : Adv -> Adv -> ListAdv = notYet "BaseAdv" ; -oper BaseNP : NP -> NP -> ListNP = notYet "BaseNP" ; oper BaseRS : RS -> RS -> ListRS = notYet "BaseRS" ; -oper BaseS : S -> S -> ListS = notYet "BaseS" ; oper CAdvAP : CAdv -> AP -> NP -> AP = notYet "CAdvAP" ; -oper CleftAdv : Adv -> S -> Cl = notYet "CleftAdv" ; oper CleftNP : NP -> RS -> Cl = notYet "CleftNP" ; oper ComparAdvAdj : CAdv -> A -> NP -> Adv = notYet "ComparAdvAdj" ; oper ComparAdvAdjS : CAdv -> A -> S -> Adv = notYet "ComparAdvAdjS" ; -oper ComplVA : VA -> AP -> VP = notYet "ComplVA" ; -oper ComplVQ : VQ -> QS -> VP = notYet "ComplVQ" ; -oper ComplVS : VS -> S -> VP = notYet "ComplVS" ; -oper ConjAP : Conj -> ListAP -> AP = notYet "ConjAP" ; -oper ConjAdv : Conj -> ListAdv -> Adv = notYet "ConjAdv" ; -oper ConjNP : Conj -> ListNP -> NP = notYet "ConjNP" ; oper ConjRS : Conj -> ListRS -> RS = notYet "ConjRS" ; -oper ConjS : Conj -> ListS -> S = notYet "ConjS" ; -oper ConsAP : AP -> ListAP -> ListAP = notYet "ConsAP" ; -oper ConsAdv : Adv -> ListAdv -> ListAdv = notYet "ConsAdv" ; -oper ConsNP : NP -> ListNP -> ListNP = notYet "ConsNP" ; oper ConsRS : RS -> ListRS -> ListRS = notYet "ConsRS" ; -oper ConsS : S -> ListS -> ListS = notYet "ConsS" ; oper DetNP : Det -> NP = notYet "DetNP" ; oper EmbedQS : QS -> SC = notYet "EmbedQS" ; oper EmbedS : S -> SC = notYet "EmbedS" ; oper EmbedVP : VP -> SC = notYet "EmbedVP" ; oper ExistIP : IP -> QCl = notYet "ExistIP" ; -oper ExistNP : NP -> Cl = notYet "ExistNP" ; oper FunRP : Prep -> NP -> RP -> RP = notYet "FunRP" ; -oper GenericCl : VP -> Cl = notYet "GenericCl" ; -oper IdRP : RP = notYet "IdRP" ; oper ImpPl1 : VP -> Utt = notYet "ImpPl1" ; oper ImpersCl : VP -> Cl = notYet "ImpersCl" ; oper PConjConj : Conj -> PConj = notYet "PConjConj" ; @@ -48,25 +26,15 @@ oper PredSCVP : SC -> VP -> Cl = notYet "PredSCVP" ; oper ProgrVP : VP -> VP = notYet "ProgrVP" ; oper ReflA2 : A2 -> AP = notYet "ReflA2" ; oper ReflVP : VPSlash -> VP = notYet "ReflVP" ; -oper RelCN : CN -> RS -> CN = notYet "RelCN" ; -oper RelCl : Cl -> RCl = notYet "RelCl" ; -oper RelNP : NP -> RS -> NP = notYet "RelNP" ; -oper RelSlash : RP -> ClSlash -> RCl = notYet "RelSlash" ; -oper RelVP : RP -> VP -> RCl = notYet "RelVP" ; oper SentAP : AP -> SC -> AP = notYet "SentAP" ; oper SentCN : CN -> SC -> CN = notYet "SentCN" ; +oper SlashPrep : Cl -> Prep -> ClSlash = notYet "SlashPrep" ; oper Slash2V3 : V3 -> NP -> VPSlash = notYet "Slash2V3" ; oper SlashV2A : V2A -> AP -> VPSlash = notYet "SlashV2A" ; oper SlashV2Q : V2Q -> QS -> VPSlash = notYet "SlashV2Q" ; oper SlashV2S : V2S -> S -> VPSlash = notYet "SlashV2S" ; -oper SlashV2V : V2V -> VP -> VPSlash = notYet "SlashV2V" ; -oper SlashV2VNP : V2V -> NP -> VPSlash -> VPSlash = notYet "SlashV2VNP" ; oper SlashVS : NP -> VS -> SSlash -> ClSlash = notYet "SlashVS" ; oper SubjS : Subj -> S -> Adv = notYet "SubjS" ; -oper UseA2 : A2 -> AP = notYet "UseA2" ; -oper UseComparA : A -> AP = notYet "UseComparA" ; -oper UseRCl : Temp -> Pol -> RCl -> RS = notYet "UseRCl" ; -oper UseSlash : Temp -> Pol -> ClSlash -> SSlash = notYet "UseSlash" ; oper VocNP : NP -> Voc = notYet "VocNP" ; oper pot3plus : Sub1000 -> Sub1000 -> Sub1000000 = notYet "pot3plus" ; diff --git a/src/arabic/MorphoAra.gf b/src/arabic/MorphoAra.gf index b55549b8..5d0a0657 100644 --- a/src/arabic/MorphoAra.gf +++ b/src/arabic/MorphoAra.gf @@ -34,7 +34,7 @@ flags optimize = all ;--noexpand; case g of { Masc => waHid; Fem => waHida - } in defArt state waHid + word + dec1sg ! state ! c; + } in defArt state c waHid + word + dec1sg ! state ! c; n = num; d = state; isPron = False; diff --git a/src/arabic/NounAra.gf b/src/arabic/NounAra.gf index d20b3e78..174a1d81 100644 --- a/src/arabic/NounAra.gf +++ b/src/arabic/NounAra.gf @@ -6,7 +6,12 @@ lin DetCN det cn = let { cas : Case -> Case = if_then_else Case det.is1sg Bare ; - number = sizeToNumber det.n ; + number = case cn.isDual of { + True => + case sizeToNumber det.n of { + Sg => Sg ; + _ => Dl } ; + False => sizeToNumber det.n } ; determiner : Case -> Str = \c -> det.s ! cn.h ! (detGender cn.g det.n) ! c ; noun : Case -> Str = \c -> @@ -14,9 +19,9 @@ lin ! nounState det.d number ! nounCase c det.n det.d ; adj : Case -> Str = \c -> - cn.adj ! number - ! (definite ! det.d) -- Indef remains Indef, rest become Def - ! c + cn.s2 ! number + ! (definite ! det.d) -- Indef remains Indef, rest become Def + ! c } in { s = \\c => case cnB4det det.isPron det.isNum det.n det.d of { @@ -25,27 +30,31 @@ lin ++ adj c ++ cn.np ! c ; True => noun (cas c) -- deal with possessive suffix - ++ determiner c - ++ adj c + ++ determiner c -- (nounCase c det.n det.d) --?? + ++ adj c ++ cn.np ! c }; a = { pgn = agrP3 cn.h cn.g number; - isPron = False } + isPron = False } ; + empty = [] }; UsePN pn = { s = pn.s; - a = {pgn = (Per3 pn.g Sg); isPron = False } + a = {pgn = Per3 pn.g Sg ; isPron = False} ; + empty = [] }; UsePron p = p ; - PredetNP pred np = { + DetNP det = emptyNP ** {s = det.s ! NoHum ! Masc} ; ---- + + PredetNP pred np = np ** { s = \\c => case pred.isDecl of { True => pred.s!c ++ np.s ! Gen ; -- akvaru l-awlAdi False => pred.s!c ++ np.s ! c - }; - a = np.a + } ; + a = np.a ** {isPron=False} } ; {- @@ -62,26 +71,15 @@ lin AdvNP np adv = np ** { s = \\c => np.s ! c ++ adv.s }; -{- - DetSg quant ord = { - s = \\h,g,c => - quant.s ! Sg ! h ! g ! c ++ ord.s ! g ! quant.d ! c ; - n = One; - d = quant.d; - isPron = quant.isPron; - isNum = - case ord.n of { - None => False; - _ => True - } - } ; --} DetQuantOrd quant num ord = quant ** { - s = \\h,g,c => quant.s ! Pl ! h ! g ! c - ++ num.s ! g ! (toDef quant.d num.n) ! c + s = \\h,g,c => let d = toDef quant.d num.n in + quant.s ! Pl ! h ! g ! c + ++ num.s ! g ! d ! c --FIXME check this: - ++ ord.s ! g ! (toDef quant.d num.n) ! c ; + ++ ord.s ! g + ! case d of {Poss => Def ; _ => d} + ! c ; n = num.n; isNum = orB num.isNum ord.isNum ; -- ord may come from OrdDigits or OrdNumeral @@ -90,7 +88,7 @@ lin } ; DetQuant quant num = quant ** { - s = \\h,g,c => quant.s ! Pl ! h ! g ! c + s = \\h,g,c => quant.s ! sizeToNumber num.n ! h ! g ! c ++ num.s ! g ! (toDef quant.d num.n) ! c ; n = num.n; isNum = -- Num may come from NumCard : Card -> Num @@ -101,7 +99,7 @@ lin } ; PossPron p = { - s = \\_,_,_,_ => p.s ! Gen; + s = \\_,_,_,_ => BIND ++ p.s ! Gen; d = Poss; is1sg = case p.a.pgn of { Per1 Sing => True ; _ => False } ; isPron = True; @@ -165,42 +163,42 @@ lin isNum,isPron,is1sg = False } ; - MassNP cn = ---- AR - {s = \\c => cn.s ! Sg ! Indef ! c ++ cn.np ! c ++ cn.adj ! Sg ! Indef ! c ; - a = {pgn = Per3 cn.g Sg ; isPron = False}} ; + MassNP cn = + {s = \\c => cn2str cn Sg Indef c ; + a = {pgn = Per3 cn.g Sg ; isPron = False} ; + empty = []} ; --- MassDet = {s = \\_,_,_,_ => [] ; d = Indef; --- isNum = False; isPron = False} ; UseN, - UseN2 = \n -> n ** { - adj = \\_,_,_ => []; - np = \\_ => []}; + UseN2 = useN ; Use2N3 n3 = n3 ; Use3N3 n3 = n3 ** {c2 = n3.c3} ; - ComplN2 n2 np = UseN n2 ** --- IL - {s = \\n,s,c => n2.s ! n ! s ! c ++ n2.c2 ++ np.s ! Gen} ; - + ComplN2 n2 np = UseN n2 ** {np = \\c => n2.c2.s ++ np.s ! n2.c2.c} ; ComplN3 n3 np = ComplN2 n3 np ** {c2 = n3.c3} ; AdjCN ap cn = cn ** { - adj = \\n,d,c => ap.s ! cn.h ! cn.g ! n ! (definite ! d) ! c + s2 = \\n,d,c => cn.s2 ! n ! d ! c ++ ap.s ! cn.h ! cn.g ! n ! (definite ! d) ! c }; - -- RelCN cn rs = {s = \\n,c => cn.s ! n ! c ++ rs.s ! {n = n ; p = P3}} ; - -- AdvCN cn ad = {s = \\n,c => cn.s ! n ! c ++ ad.s} ; - -- - -- SentCN cn sc = {s = \\n,c => cn.s ! n ! c ++ sc.s} ; + + RelCN cn rs = cn ** { + s2 = \\n,s,c => cn.s2 ! n ! s ! c ++ rs.s ! {pgn=Per3 cn.g n ; isPron=False} ! c}; + + RelNP np rs = np ** {s = \\c => np.s ! c ++ rs.s ! np.a ! c} ; + + AdvCN, + SentCN = \cn,ss -> cn ** {s2 = \\n,d,c => cn.s2 ! n ! d ! c ++ ss.s} ; + ApposCN cn np = cn ** { np = \\c => cn.np ! c ++ np.s ! c } ; -- : CN -> NP -> CN ; -- house of Paris, house of mine PossNP cn np = cn ** { - s = \\n,_d,c => cn.s ! n ! Const ! c ; + s = \\n,_d,c => cn.s ! n ! Const ! c ; + s2 = \\n,_d,c => cn.s2 ! n ! Const ! Gen ; -- unsure about this /IL np = \\c => cn.np ! c ++ np.s ! Gen }; - -- : CN -> NP -> CN ; -- glass of wine --PartNP } diff --git a/src/arabic/OrthoAra.gf b/src/arabic/OrthoAra.gf index a7dc5d66..8bd8f4f3 100644 --- a/src/arabic/OrthoAra.gf +++ b/src/arabic/OrthoAra.gf @@ -2,58 +2,76 @@ resource OrthoAra = open Prelude, Predef in { flags coding=utf8 ; - oper +oper - vow : pattern Str = #("َ" | "ِ" | "ُ" | "ً" | "ٍ" | "ٌ") ; + vow : pattern Str = #("َ" | "ِ" | "ُ" | "ً" | "ٍ" | "ٌ") ; - -- "Sun letters": assimilate with def. article - sun : pattern Str = #("ت"|"ث"|"د"|"ذ"|"ر"|"ز"|"س"|"ش"|"ص"|"ض"|"ط"|"ظ"|"ل"|"ن") ; + weak : pattern Str = #("و"|"ي") ; - -- Shadda: https://www.unicode.org/L2/L2017/17253-arabic-ordering.pdf - fixShd : Str -> Str -> Str = \word,suffix -> - case of { - => x + v + "ّ" + y ; - _ => word + suffix - } ; + -- "Sun letters": assimilate with def. article + sun : pattern Str = #("ت"|"ث"|"د"|"ذ"|"ر"|"ز"|"س"|"ش"|"ص"|"ض"|"ط"|"ظ"|"ل"|"ن") ; - -- Hamza - hamza : pattern Str = #("ء"|"؟") ; +-- Shadda: https://www.unicode.org/L2/L2017/17253-arabic-ordering.pdf + fixShd : Str -> Str -> Str = \word,suffix -> + case of { + -- => x + v + "ّ" + y ; + => x + "ّ" + v + y ; + _ => word + suffix + } ; - rectifyHmz: Str -> Str = \word -> - case word of { - l@(""|"ال") + ("أ"|"أَ") + #hamza + "ْ" + tail => l + "آ" + tail; - l@(""|"ال") + ("أ"|"أَ") + #hamza + tail => l + "آ" + tail; - l@(""|"ال") + #hamza + v@("َ"|"ُ") + tail => l + "أ" + v + tail; - l@(""|"ال") + #hamza + v@("ِ") + tail => l + "إ" + v + tail; +-- IL: using this to reuse patterns for weak verbs, might be strange/wrong + rmSukun : Str -> Str = \s -> case s of { + x + "ْ" + y => x + y ; + _ => s + } ; - head + v1@("ِ"|"ُ"|"َ"|"ْ"|"ا"|"ي"|"و") + #hamza + v2@(""|"ُ"|"َ"|"ْ"|"ِ") => head + v1 + (tHmz v1) + v2; - head + #hamza + tail => head + (bHmz (dp 2 head) (take 2 tail)) + tail; --last head , take 1 tail - _ => word - }; +-- Hamza + hamza : pattern Str = #("ء"|"؟") ; - --hamza at beginning of word (head) - hHmz : Str -> Str = \d -> - case d of { - "ِ" => "إ"; - _ => "أ" - }; + rectifyHmz : Str -> Str = \word -> + case word of { + l@(""|"ال") + ("أ"|"أَ") + #hamza + "ْ" + tail => l + "آ" + tail; + l@(""|"ال") + ("أ"|"أَ") + #hamza + tail => l + "آ" + tail; + l@(""|"ال") + #hamza + v@("َ"|"ُ") + tail => l + "أ" + v + tail; + l@(""|"ال") + #hamza + v@("ِ") + tail => l + "إ" + v + tail; + head + v1@(#vow|"ْ"|"ا"|"ي"|"و") + + #hamza + v2@(#vow|"ْ") + tail => + case v2 of { "ْ" => head + v1 + bHmz v1 v2 + tail ; -- unsure about this /IL + _ => head + v1 + bHmz v1 v2 + v2 + tail } ; + head + v1@(#vow|"ْ"|"ا"|"ي"|"و") -- the same but it ends in vowel + + #hamza + v2@(#vow|"ْ") => + case v2 of { "ْ" => head + v1 + tHmz v1 ; + _ => head + v1 + tHmz v1 + v2 } ; + head + v1@(#vow|"ْ"|"ا"|"ي"|"و") -- the same but it ends without vowel + + #hamza => head + v1 + tHmz v1 ; - --hamza in middle of word (body) - bHmz : Str -> Str -> Str = \d1,d2 -> - case of { - <"ِ",_> | <_,"ِ"> => "ئ"; - <"ُ",_> | <_,"ُ"> => "ؤ"; - <"َ",_> | <_,"َ"> => "أ"; - _ => "ء" - }; + head + #hamza + tail => head + (bHmz (dp 2 head) (take 2 tail)) + tail; --last head , take 1 tail + _ => word + }; - --hamza carrier sequence - tHmz : Str -> Str = \d -> - case d of { - "ِ" => "ئ"; - "ُ" => "ؤ"; - "َ" => "أ"; - "ْ"|"ا"|"و"|"ي" => "ء" - }; + --hamza at beginning of word (head) + hHmz : Str -> Str = \d -> + case d of { + "ِ" => "إ"; + _ => "أ" + }; + + --hamza in middle of word (body) + bHmz : Str -> Str -> Str = \d1,d2 -> + case of { + <"ِ",_> | <_,"ِ"> => "ئ"; + <"ُ",_> | <_,"ُ"> => "ؤ"; + <"َ",_> | <_,"َ"> => "أ"; + _ => "ء" + }; + + --hamza carrier sequence + tHmz : Str -> Str = \d -> + case d of { + "ِ" => "ئ"; + "ُ" => "ؤ"; + "َ" => "أ"; + "ْ"|"ا"|"و"|"ي" => "ء" + }; } diff --git a/src/arabic/ParadigmsAra.gf b/src/arabic/ParadigmsAra.gf index 9dea1146..b00c1844 100644 --- a/src/arabic/ParadigmsAra.gf +++ b/src/arabic/ParadigmsAra.gf @@ -35,31 +35,54 @@ resource ParadigmsAra = open oper --- Prepositions are used in many-argument functions for rection. + Case : Type ; + nom : Case ; + acc : Case ; + gen : Case ; +-- Prepositions are used in many-argument functions for rection. Preposition : Type ; + noPrep : Preposition ; + casePrep : Case -> Preposition ; + --- TODO: continue, add all over the grammar + + Gender : Type ; + masc : Gender ; + fem : Gender ; + + Number : Type ; + sg : Number ; + pl : Number ; + + Species : Type ; + hum : Species ; + nohum : Species ; + + Vowel : Type ; + va : Vowel ; + vi : Vowel ; + vu : Vowel ; --2 Nouns -- Overloaded operator for main cases - mkN = overload { - mkN : (sg : Str) -> N -- non-human regular nouns - = smartN ; - mkN : Species -> N -> N - = \p,n -> n ** {h = p} ; - mkN : (sg,pl : Str) -> Gender -> Species -> N - = \sg,pl -> mkFullN (reg sg pl) ; - mkN : NTable -> Gender -> Species -> N -- loan words, irregular - = mkFullN ; - mkN : (root,sgPatt,brokenPlPatt : Str) -> Gender -> Species -> N -- broken plural - = brkN ; - mkN : N -> (attr : Str) -> N -- Compound nouns - = \n,attr -> n ** { s = \\num,s,c => n.s ! num ! s ! c ++ attr } ; --- IL (TODO: all kinds of compounds) + + mkN : overload { + mkN : (sg : Str) -> N ; -- non-human regular nouns + mkN : Species -> N -> N ; + mkN : (sg,pl : Str) -> Gender -> Species -> N ; + mkN : NTable -> Gender -> Species -> N ; -- loan words, irregular + mkN : (root,sgPatt,brokenPlPatt : Str) -> Gender -> Species -> N ; -- broken plural + mkN : N -> (attr : Str) -> N ; -- Compound noun with invariant attribute + mkN : N -> N -> N ; -- Compound noun where attribute inflects in state and case. Attribute in singular. + mkN : Number -> N -> N -> N ; -- Compound noun where attribute inflects in state and case. Attribute's number specified by 1st arg. --- mkN : (root,sgPatt : Str) -> Gender -> Species -> N -- sound feminine plural --- = sdfN ; } ; + dualN : N -> N ; -- Force the plural of the N into dual (e.g. "twins") + --This is used for loan words or anything that has untreated irregularities --in the interdigitization process of its words mkFullN : NTable -> Gender -> Species -> N ; @@ -82,6 +105,8 @@ resource ParadigmsAra = open mkPN = overload { mkPN : Str -> PN -- Fem Hum if ends with ة, otherwise Masc Hum = smartPN ; + mkPN : N -> PN + = \n -> lin PN (n ** {s = \\c => n.s ! Sg ! Const ! c ++ n.s2 ! Sg ! Const ! c }) ; -- no idea /IL mkPN : Str -> Gender -> Species -> PN = mkFullPN ; } ; @@ -92,13 +117,17 @@ resource ParadigmsAra = open --3 Relational nouns - mkN2 = overload { - mkN2 : N -> Preposition -> N2 = prepN2 ; - mkN2 : N -> N2 = \n -> lin N2 (n ** {c2 = []}) ; - mkN2 : Str -> N2 = \str -> lin N2 (smartN str ** {c2 = []}) + mkN2 : overload { + mkN2 : N -> Preposition -> N2 ; -- ready-made preposition + mkN2 : N -> Str -> N2 ; -- preposition given as a string + mkN2 : N -> N2 ; -- no preposition + mkN2 : Str -> N2 ; -- no preposition, predictable inflection } ; - mkN3 : N -> Preposition -> Preposition -> N3 ; + mkN3 : overload { + mkN3 : N -> Preposition -> Preposition -> N3 ; -- ready-made prepositions + mkN3 : N -> Str -> Str -> N3 ; -- prepositions given as strings + } ; --2 Adjectives @@ -106,16 +135,15 @@ resource ParadigmsAra = open -- Overloaded operator for main cases mkA = overload { - mkA : (root,patt : Str) -> A + mkA : (root,sg : Str) -> A -- adjective with sound plural; takes root string and sg. pattern string = \r,p -> lin A (sndA r p); - mkA : (root : Str) -> A -- forms adjectives with positive form aFCal + mkA : (root : Str) -> A -- adjective with positive form aFCal = \r -> lin A (clrA r); - mkA : (root,sg,pl : Str) -> A + mkA : (root,sg,pl : Str) -> A -- adjective with broken plural = \r,s,p -> lin A (brkA r s p) ; - -- mkA : (posit,compar,plur : Str) -> A - -- = degrA ; } ; + degrA : (posit,compar,plur : Str) -> A ; --Takes a root string and a pattern string sndA : (root,patt : Str) -> Adj ; @@ -123,11 +151,16 @@ resource ParadigmsAra = open --Takes a root string only clrA : (root : Str) -> Adj ; -- forms adjectives of type aFCal + nisbaA : Str -> Adj ; -- forms relative adjectives by adding the suffix ِيّ + --3 Two-place adjectives -- -- Two-place adjectives need a preposition for their second argument. - mkA2 : A -> Preposition -> A2 ; + mkA2 : overload { + mkA2 : A -> Preposition -> A2 ; + mkA2 : A -> Str -> A2 + } ; --2 Adverbs @@ -141,75 +174,65 @@ resource ParadigmsAra = open mkAdA : Str -> AdA ; + mkInterj : Str -> Interj ; + + mkSubj : overload { + mkSubj : Str -> Subj ; -- Default order Subord (=noun first and in accusative) + mkSubj : Str -> Order -> Subj -- Specify word order + } ; + --2 Prepositions -- -- A preposition as used for rection in the lexicon, as well as to --- build $PP$s in the resource API, just requires a string. - - mkPrep : Str -> Prep - = \s -> lin Prep {s = mkPreposition s} ; -- preposition in the sense of RGL abstract syntax - - mkPreposition : Str -> Preposition ; -- just a string, for internal use - +-- build $PP$s in the resource API. Requires a string and a case. + mkPrep : overload { + mkPrep : Str -> Prep ; + mkPrep : Str -> Case -> Prep + } ; -- preposition in the sense of RGL abstract syntax --2 Verbs -- Overloaded operations - mkV = overload { - mkV : (imperfect : Str) -> V - = regV ; - mkV : (root : Str) -> (perf,impf : Vowel) -> V -- verb form I ; vowel = a|i|u - = v1 ; - mkV : (root : Str) -> VerbForm -> V -- FormI .. FormX (no VII, IX) ; default vowels a u for I - = formV ; + mkV : overload { + mkV : (imperfect : Str) -> V ; -- The verb in Per3 Sg Masc imperfect tense gives the most information + mkV : (root : Str) -> (perf,impf : Vowel) -> V ; -- verb form I ; vowel = a|i|u + mkV : (root : Str) -> VerbForm -> V ; -- FormI .. FormX (no VII, IX) ; default vowels a u for I + mkV : V -> (particle : Str) -> V -- V with a non-inflecting particle/phrasal verb } ; --- The verb in the imperfect tense gives the most information + -- regV : Str -> V ; - regV : Str -> V ; + reflV : V -> V ; -- نَفْس in the proper case and with possessive suffix, e.g. نَفْسَكِ ---Verb Form I : fa`ala, fa`ila, fa`ula + v1 : Str -> Vowel -> Vowel -> V ; -- Verb Form I : fa`ala, fa`ila, fa`ula - v1 : Str -> Vowel -> Vowel -> V ; + v2 : Str -> V ; -- Verb Form II : fa``ala ---Verb Form II : fa``ala + v3 : Str -> V ; -- Verb Form III : faa`ala - v2 : Str -> V ; + v4 : Str -> V ; -- Verb Form IV : 'af`ala ---Verb Form III : faa`ala + v5 : Str -> V ; -- Verb Form V : tafa``ala - v3 : Str -> V ; + v6 : Str -> V ; -- Verb Form VI : tafaa`ala ---Verb Form IV : 'af`ala + v7 : Str -> V ; -- Verb Form VII : infa`ala - v4 : Str -> V ; + v8 : Str -> V ; -- Verb Form VIII ifta`ala ---Verb Form V : tafa``ala - - v5 : Str -> V ; - ---Verb Form VI : tafaa`ala - - v6 : Str -> V ; - ---Verb Form VIII 'ifta`ala - - v8 : Str -> V ; - --- Verb Form X 'istaf`ala - - v10 : Str -> V ; + v10 : Str -> V ; -- Verb Form X 'istaf`ala --3 Two-place verbs -- Two-place verbs need a preposition, except the special case with direct object. -- (transitive verbs). Notice that a particle comes from the $V$. - mkV2 = overload { - mkV2 : V -> V2 = dirV2 ; - mkV2 : V -> Preposition -> V2 = prepV2 ; - mkV2 : Str -> V2 = strV2; + mkV2 : overload { + mkV2 : V -> V2 ; -- No preposition + mkV2 : V -> Str -> V2 ; -- Preposition as string, default case genitive + mkV2 : V -> Preposition -> V2 ; -- Ready-made preposition + mkV2 : Str -> V2 ; -- Predictable verb conjugation, no preposition } ; dirV2 : V -> V2 ; @@ -219,9 +242,15 @@ resource ParadigmsAra = open -- Three-place (ditransitive) verbs need two prepositions, of which -- the first one or both can be absent. - mkV3 : V -> Preposition -> Preposition -> V3 ; -- speak, with, about - dirV3 : V -> Preposition -> V3 ; -- give,_,to - dirdirV3 : V -> V3 ; -- give,_,_ + mkV3 : overload { + mkV3 : V -> Preposition -> Preposition -> V3 ; -- speak, with, about + mkV3 : V -> (to : Str) -> (about:Str) -> V3 -- like above, but with strings as arguments (default complement case genitive) + } ; + dirV3 : overload { + dirV3 : V -> Preposition -> V3 ; -- give,_,to + dirV3 : V -> (to : Str) -> V3 -- like above, but with string as argument (default complement case genitive) + } ; + dirdirV3 : V -> V3 ; -- give,_,_ --3 Other complement patterns -- @@ -233,9 +262,15 @@ resource ParadigmsAra = open mkV2S : V -> Str -> V2S ; mkVV = overload { mkVV : V -> VV = regVV ; - mkVV : V -> Str -> VV = c2VV + mkVV : V -> Str -> VV = c2VV ; + mkVV : V -> Preposition -> VV = prepVV ; + mkVV : V -> Preposition -> Preposition -> VV = prep2VV } ; - mkV2V : V -> Str -> Str -> V2V ; + mkV2V : overload { + mkV2V : V -> Str -> Str -> V2V ; + mkV2V : V -> Preposition -> Preposition -> V2V ; + mkV2V : VV -> Preposition -> V2V + } ; mkVA : V -> VA ; mkV2A : V -> Str -> V2A ; mkVQ : V -> VQ ; @@ -260,6 +295,94 @@ resource ParadigmsAra = open -- The definitions should not bother the user of the API. So they are -- hidden from the document. + Case = ResAra.Case ; + nom = ResAra.Nom ; + acc = ResAra.Acc ; + gen = ResAra.Gen ; + +-- Prepositions are used in many-argument functions for rection. + + Preposition = ResAra.Preposition ; + noPrep = {s=[]; c=nom} ; + casePrep c = {s=[]; c=c} ; + + Gender = ResAra.Gender ; + masc = ResAra.Masc ; + fem = ResAra.Fem ; + + Number = ResAra.Number ; + sg = ResAra.Sg ; + pl = ResAra.Pl ; + + Species = ResAra.Species ; + hum = ResAra.Hum ; + nohum = ResAra.NoHum ; + + Vowel = ResAra.Vowel ; + va = ResAra.a ; + vu = ResAra.u ; + vi = ResAra.i ; + + mkPrep = overload { + mkPrep : Str -> Prep = \s -> + lin Prep (mkPreposition s) ; + mkPrep : Str -> Case -> Prep = \s,c -> + lin Prep (mkPreposition s c) + } ; + + + mkV2 = overload { + mkV2 : V -> V2 = dirV2 ; + mkV2 : V -> Str -> V2 = \v,p -> prepV2 v (mkPreposition p); + mkV2 : V -> Preposition -> V2 = prepV2 ; + mkV2 : Str -> V2 = strV2; + } ; + + prepV2 : V -> Preposition -> V2 = \v,p -> v ** {s = v.s ; c2 = p ; lock_V2 = <>} ; + strV2 : Str -> V2 = \str -> dirV2 (mkV str) ; + + mkN = overload { + mkN : (sg : Str) -> N -- non-human regular nouns + = smartN ; + mkN : Species -> N -> N + = \p,n -> n ** {h = p} ; + mkN : (sg,pl : Str) -> Gender -> Species -> N + = \sg,pl -> mkFullN (reg sg pl) ; + mkN : NTable -> Gender -> Species -> N -- loan words, irregular + = mkFullN ; + mkN : (root,sgPatt,brokenPlPatt : Str) -> Gender -> Species -> N -- broken plural + = brkN ; + mkN : N -> (attr : Str) -> N -- Compound nouns with noninflecting attribute + = \n,attr -> n ** {s2 = \\n,s,c => attr} ; + mkN : N -> N -> N -- Compound nouns where attribute inflects in state and case but not number + = attrN Sg ; + mkN : Number -> N -> N -> N -- Compound nouns where attribute inflects in state, case and number + = attrN ; + } ; + + attrN : Number -> N -> N -> N = \num,n1,n2 -> n1 ** { + s = \\n,_,c => n1.s ! n ! Const ! c ; + s2 = \\n,s,c => let c' = case c of {Dat => Gen; _ => c} in -- the Dat with liPrep hack only applies to the first word + n1.s2 ! num ! s ! c' -- attribute doesn't change + ++ n2.s ! num ! s ! c' + ++ n2.s2 ! num ! s ! c'} ; + + dualN : N -> N = \n -> n ** {isDual=True} ; + + proDrop : NP -> NP ; -- Force a NP to lose its string, only contributing with its agreement. + + mkPron : (_,_,_ : Str) -> PerGenNum -> Pron ; + + mkV = overload { + mkV : (imperfect : Str) -> V + = regV ; + mkV : (root : Str) -> (perf,impf : Vowel) -> V -- verb form I ; vowel = a|i|u + = v1 ; + mkV : (root : Str) -> VerbForm -> V -- FormI .. FormX (no VII, IX) ; default vowels a u for I + = formV ; + mkV : V -> (particle : Str) -> V = \v,p -> + v ** { s = \\vf => v.s ! vf ++ p } ; + } ; regV : Str -> V = \wo -> let rau : Str * Vowel * Vowel = @@ -274,27 +397,19 @@ resource ParadigmsAra = open v1 = \rootStr,vPerf,vImpf -> let { raw = v1' rootStr vPerf vImpf } in - { s = \\vf => - case rootStr of { - _ + #hamza + _ => rectifyHmz(raw.s ! vf); - _ => raw.s ! vf - }; - lock_V = <> - } ; - - va : Vowel = ResAra.a ; + lin V { s = \\vf =>rectifyHmz (raw.s ! vf) } ; v1' : Str -> Vowel -> Vowel -> Verb = \rootStr,vPerf,vImpf -> - let { root = mkRoot3 rootStr } in - case of { - <"ّ", _> => v1geminate rootStr vPerf vImpf ; - <"و"|"ي",_> => case vPerf of { - i => v1defective_i root vImpf ; - _ => v1defective_a root vImpf } ; - <_,"و"|"ي"> => v1hollow root vImpf ; - _ => v1sound root vPerf vImpf - }; + let root = mkRoot3 rootStr + in case rootStr of { + _ + "ّ" => v1geminate rootStr vPerf vImpf ; + ? + #hamza + #weak => v1doubleweak root ; + ? + ? + #weak => case vPerf of { + i => v1defective_i root vImpf ; + _ => v1defective_a root vImpf } ; + ? + #weak + ? => v1hollow root vImpf ; + _ => v1sound root vPerf vImpf } ; v2 = \rootStr -> @@ -303,7 +418,7 @@ resource ParadigmsAra = open } in { s = case root.l of { - "و"|"ي" => (v2defective root).s; + #weak => (v2defective root).s; _ => (v2sound root).s }; lock_V = <> @@ -320,16 +435,13 @@ resource ParadigmsAra = open v4 = \rootStr -> - let { - root = mkRoot3 rootStr - } in { - s = - case root.l of { - "و"|"ي" => (v4defective root).s; - _ => (v4sound root).s - }; - lock_V = <> - }; + let root : Root3 = mkRoot3 rootStr ; + verb : Verb = case rootStr of { + ? + #hamza + #weak => v4doubleweak root ; + ? + #weak + ? => v4hollow root ; + _ + #weak => v4defective root ; + _ => v4sound root } ; + in lin V verb ; v5 = @@ -360,31 +472,43 @@ resource ParadigmsAra = open lock_V = <> }; + v7 = + \rootStr -> + let { + fcl = mkRoot3 rootStr ; + v7fun = v7geminate ; -- TODO add rest + } in lin V (v7fun fcl) ; + v8 = \rootStr -> let { rbT = mkRoot3 rootStr ; v8fun = case rbT.f of { ("و"|"ي"|"ّ") => v8assimilated ; - _ => v8sound } + _ => + case rbT.c of { + #weak => v8hollow ; + _ => v8sound }} } in lin V (v8fun rbT) ; v10 = \rootStr -> let { rbT = mkRoot3 rootStr ; - v10fun = case rbT.c of { - ("و"|"ي") => v10hollow ; - _ => v10sound } + v10fun : Root3 -> Verb = case rootStr of { + ? + #weak + ? => v10hollow ; + ? + ? + #weak => v10defective ; + _ => v10sound } } in lin V (v10fun rbT) ; - Preposition = Str ; + reflV v = lin V (ResAra.reflV v) ; - mkFullN nsc gen spec = + mkFullN nsc gen spec = lin N { s = nsc; --NTable + s2 = emptyNTable; g = gen; h = spec; - lock_N = <> + isDual = False }; brkN' : Str -> Str -> Str -> Gender -> Species -> N = @@ -394,14 +518,12 @@ resource ParadigmsAra = open } in mkFullN (reg kitAb kutub) gen spec; brkN root sg pl gen spec = - let { raw = brkN' root sg pl gen spec} in + let { raw = brkN' root sg pl gen spec} in raw ** { s = \\n,d,c => case root of { _ + #hamza + _ => rectifyHmz(raw.s ! n ! d ! c); _ => raw.s ! n ! d ! c - }; - g = gen; - h = spec ; lock_N = <> + } }; sdfN = @@ -409,7 +531,7 @@ resource ParadigmsAra = open let { kalimaStr = mkWord sg root; kalimaRaw = sndf kalimaStr; kalima : NTable = \\n,d,c => case root of { - _ + #hamza + _ + _ + #hamza + _ => rectifyHmz (kalimaRaw ! n ! d ! c); _ => kalimaRaw ! n ! d ! c }; @@ -427,35 +549,36 @@ resource ParadigmsAra = open lock_PN = <> }; + mkN2 = overload { + mkN2 : N -> Preposition -> N2 = prepN2 ; + mkN2 : N -> Str -> N2 = \n,s -> prepN2 n (mkPreposition s); + mkN2 : N -> N2 = \n -> lin N2 (n ** {c2 = noPrep}) ; + mkN2 : Str -> N2 = \str -> lin N2 (smartN str ** {c2 = noPrep}) + } ; - prepN2 : N -> Str -> N2 = \n,p -> lin N2 (n ** {c2 = p}) ; + prepN2 : N -> Preposition -> N2 = \n,p -> lin N2 (n ** {c2 = p}) ; - mkN3 = \n,p,q -> n ** {lock_N3 = <> ; c2 = p ; c3 = q} ; + mkN3 = overload { + mkN3 : N -> Preposition -> Preposition -> N3 = \n,p,q -> + lin N3 (n ** {c2 = p ; c3 = q}) ; + mkN3 : N -> Str -> Str -> N3 = \n,p,q -> + lin N3 (n ** {c2 = mkPreposition p ; c3 = mkPreposition q}) ; + } ; - mkPron : (_,_,_ : Str) -> PerGenNum -> NP = \ana,nI,I,pgn -> - { s = - table { - Acc => BIND ++ nI; -- object suffix - Gen => BIND ++ I; -- possessive suffix - _ => ana - }; - a = {pgn = pgn; isPron = True }; - lock_NP = <> - }; + mkPron : (_,_,_ : Str) -> PerGenNum -> Pron = \ana,nI,I,pgn -> + lin Pron (ResAra.mkPron ana nI I pgn) ; + + proDrop : NP -> NP = \np -> lin NP (ResAra.proDrop np) ; -- e.g. al-jamii3, 2a7ad - regNP : Str -> Number -> NP = \word,n -> - { s = \\c => fixShd word (dec1sg ! Def ! c) ; - a = {pgn = Per3 Masc n; isPron = False }; - lock_NP = <> - }; + regNP : Str -> Number -> NP = \word,n -> lin NP (emptyNP ** { + s = \\c => fixShd word (dec1sg ! Def ! c) + }); -- e.g. hadha, dhaalika - indeclNP : Str -> Number -> NP = \word,n -> - { s = \\c => word ; - a = {pgn = Per3 Masc n; isPron = False }; - lock_NP = <> - }; + indeclNP : Str -> Number -> NP = \word,n -> lin NP (emptyNP ** { + s = \\c => word + }); mkQuant7 : (_,_,_,_,_,_,_ : Str) -> State -> Quant = \hava,havihi,havAn,havayn,hAtAn,hAtayn,hA'ulA,det -> lin Quant (baseQuant ** @@ -493,9 +616,9 @@ resource ParadigmsAra = open mascTbl = reg jadId judud ; femTbl = reg jadIda judud ; in { s = table { - APosit Masc n d c => mascTbl ! n ! d ! c ; - APosit Fem n d c => femTbl ! n ! d ! c ; - AComp d c => indeclN akbar ! d ! c } + APosit Masc n d c => rectifyHmz (mascTbl ! n ! d ! c) ; + APosit Fem n d c => rectifyHmz (femTbl ! n ! d ! c) ; + AComp d c => rectifyHmz (indeclN akbar ! d ! c) } } ; degrA : (posit,compar,plur : Str) -> A @@ -521,6 +644,14 @@ resource ParadigmsAra = open } }; + nisbaA : Str -> Adj = \Haal -> + let Haaliyy = Haal + "ِيّ" in { + s = table { + APosit g n d c => positAdj Haaliyy ! g ! n ! d ! c ; + AComp d c => "أَكْثَر" ++ indeclN Haaliyy ! d ! c + } + } ; + clrA root = let { eaHmar = mkWord "أَفعَل" root; HamrA' = mkWord "فَعلاء" root; @@ -529,45 +660,73 @@ resource ParadigmsAra = open s = clr eaHmar HamrA' Humr; }; - mkA2 a p = a ** {c2 = p ; lock_A2 = <>} ; + mkA2 = overload { + mkA2 : A -> Preposition -> A2 = prepA2 ; + mkA2 : A -> Str -> A2 = \a,p -> prepA2 a (mkPreposition p) + } ; - mkAdv x = ss x ** {lock_Adv = <>} ; - mkAdV x = ss x ** {lock_AdV = <>} ; - mkAdA x = ss x ** {lock_AdA = <>} ; + prepA2 : A -> Preposition -> A2 = \a,p -> lin A2 (a ** {c2 = p}) ; - mkPreposition p = p ; + mkAdv x = lin Adv (ss x) ; + mkAdV x = lin AdV (ss x) ; + mkAdA x = lin AdA (ss x) ; + mkInterj x = lin Interj (ss x) ; - prepV2 : V -> Preposition -> V2 = \v,p -> v ** {s = v.s ; c2 = p ; lock_V2 = <>} ; - strV2 : Str -> V2 = \str -> dirV2 (mkV str) ; + mkSubj = overload { + mkSubj : Str -> Subj = \s -> lin Subj {s = s ; o = Subord} ; + mkSubj : Str -> Order -> Subj = \s,o -> lin Subj {s = s ; o = o} ; + } ; - dirV2 v = prepV2 v [] ; + dirV2 v = prepV2 v (casePrep acc) ; - mkV3 v p q = v ** {s = v.s ; c2 = p ; c3 = q ; lock_V3 = <>} ; - dirV3 v p = mkV3 v [] p ; - dirdirV3 v = dirV3 v [] ; + mkV3 = overload { + mkV3 : V -> Preposition -> Preposition -> V3 = \v,p,q -> + lin V3 (prepV3 v p q) ; + mkV3 : V -> Str -> Str -> V3 = \v,p,q -> + lin V3 (v ** {s = v.s ; c2 = mkPreposition p ; c3 = mkPreposition q}) + } ; + + prepV3 : V -> Preposition -> Preposition -> Verb3 = \v,p,q -> + v ** {s = v.s ; c2 = p ; c3 = q} ; + + dirV3 = overload { + dirV3 : V -> Preposition -> V3 = \v,p -> mkV3 v (casePrep acc) p ; + dirV3 : V -> Str -> V3 = \v,s -> mkV3 v (casePrep acc) (mkPreposition s) + } ; + + dirdirV3 v = dirV3 v (casePrep acc) ; mkVS v = v ** {lock_VS = <>} ; mkVQ v = v ** {lock_VQ = <>} ; - regVV : V -> VV = \v -> lin VV v ** {c2 = "أَنْ"} ; - c2VV : V -> Str -> VV = \v,prep -> regVV v ** {c2 = prep} ; - + regVV : V -> VV = \v -> lin VV v ** {c2 = mkPreposition "أَنْ" ; sc = noPrep} ; + c2VV : V -> Str -> VV = \v,prep -> regVV v ** {c2 = mkPreposition prep ; sc = noPrep} ; + prepVV : V -> Preposition -> VV = \v,prep -> regVV v ** {c2=prep; sc=noPrep} ; + prep2VV : V -> (_,_ : Preposition) -> VV = \v,p1,p2 -> regVV v ** {c2=p1; sc=p2} ; V0 : Type = V ; ---- V2S, V2V, V2Q, V2A : Type = V2 ; AS, A2S, AV : Type = A ; A2V : Type = A2 ; - mkV0 v = v ** {lock_V = <>} ; - mkV2S v p = mkV2 v p ** {lock_V2S = <>} ; - mkV2V v p t = mkV2 v p ** {s4 = t ; lock_V2V = <>} ; - mkVA v = v ** {lock_VA = <>} ; - mkV2A v p = mkV2 v p ** {lock_V2A = <>} ; - mkV2Q v p = mkV2 v p ** {lock_V2Q = <>} ; + mkV0 v = v ; + mkV2S v p = lin V2S (prepV2 v (mkPreposition p)) ; + mkV2V = overload { + mkV2V : V -> Str -> Str -> V2V = \v,p,q -> + lin V2V (prepV3 v (mkPreposition p) (mkPreposition q) ** {sc = noPrep}) ; + mkV2V : V -> Preposition -> Preposition -> V2V = \v,p,q -> + lin V2V (prepV3 v p q ** {sc = noPrep}) ; + mkV2V : VV -> Preposition -> V2V = \vv,p -> + lin V2V (vv ** {c2 = p ; c3 = vv.c2}) ; + } ; + mkVA v = v ** {lock_VA = <>} ; + mkV2A v p = lin V2A (prepV2 v (mkPreposition p)); + mkV2Q v p = lin V2Q (prepV2 v (mkPreposition p)); + + mkAS, + mkAV = \a -> a ; + mkA2S, + mkA2V = \a,p -> prepA2 a (mkPreposition p) ; - mkAS v = v ** {lock_A = <>} ; - mkA2S v p = mkA2 v p ** {lock_A = <>} ; - mkAV v = v ** {lock_A = <>} ; - mkA2V v p = mkA2 v p ** {lock_A2 = <>} ; smartN : Str -> N = \s -> case s of { @@ -588,12 +747,12 @@ formV : (root : Str) -> VerbForm -> V = \s,f -> case f of { FormIV => v4 s ; FormV => v5 s ; FormVI => v6 s ; ---- FormVII => v7 s ; + FormVII => v7 s ; FormVIII => v8 s ; FormX => v10 s } ; param VerbForm = - FormI | FormII | FormIII | FormIV | FormV | FormVI | FormVIII | FormX ; + FormI | FormII | FormIII | FormIV | FormV | FormVI | FormVII | FormVIII | FormX ; } ; diff --git a/src/arabic/PatternsAra.gf b/src/arabic/PatternsAra.gf index 26c5e8c5..8fa311af 100644 --- a/src/arabic/PatternsAra.gf +++ b/src/arabic/PatternsAra.gf @@ -35,9 +35,13 @@ flags coding=utf8 ; fuci = { h = "" ; m1 = "ُ" ; m2 = ""; t = "ِ" } ; fucu = { h = "" ; m1 = "ُ" ; m2 = ""; t = "ُ" } ; fUc = { h = "" ; m1 = "ُو"; m2 = ""; t = "" } ; - ufAc = { h = "ُ" ; m1 = "َا"; m2 = ""; t = "" } ; ufca = { h = "ُ" ; m1 = "ْ" ; m2 = ""; t = "َ" } ; + eafAc = fAc ** { h = "أَ" } ; + eafac = fac ** { h = "أَ" } ; + eafIc = fIc ** { h = "أَ" } ; + eafic = fic ** { h = "أَ" } ; + eafAcil = { h = "أَ"; m1 = "َا" ; m2 = "ِ" ; t = "" } ; eafAcIl = { h = "أَ"; m1 = "َا" ; m2 = "ِي" ; t = "" } ; eafcilp = { h = "أَ"; m1 = "ْ" ; m2 = "ِ" ; t = "َة" } ; @@ -47,8 +51,14 @@ flags coding=utf8 ; eafcul = { h = "أَ"; m1 = "ْ" ; m2 = "ُ" ; t = "" } ; eiftacal = { h = "إِ"; m1 = "ْتَ" ; m2 = "َ" ; t = "" } ; eufcil = { h = "أُ"; m1 = "ْ" ; m2 = "ِ" ; t = "" } ; + eufic = fic ** { h = "أُ" } ; + eufIc = fIc ** { h = "أُ" } ; + ufic = fic ** { h = "ُ" } ; + ufIc = fIc ** { h = "ُ" } ; + ufac = fac ** { h = "ُ" } ; + ufAc = fAc ** { h = "ُ" } ; euftucil = { h = "أُ"; m1 = "ْتُ" ; m2 = "ِ" ; t = "" } ; - euttucil = euftucil ** { h = "اُتُّ" ; m1 = "ِ" } ; ---- IL assimilated VIII + euttucil = euftucil ** { h = "اُتُّ" ; m1 = "ِ" } ; ---- IL assimilated VIII afcul = { h = "َ" ; m1 = "ْ" ; m2 = "ُ" ; t = "" } ; faccalo = { h = "" ; m1 = "َ" ; m2 = "َّ" ; t = "ْ" } ; facal = { h = "" ; m1 = "َ" ; m2 = "َ" ; t = "" } ; @@ -77,9 +87,12 @@ flags coding=utf8 ; ficAl = { h = "" ; m1 = "ِ" ; m2 = "َا" ; t = "" } ; ficlp = { h = "" ; m1 = "ِ" ; m2 = "ْ" ; t = "َة" } ; ftacal = { h = "" ; m1 = "ْتَ" ; m2 = "َ" ; t = "" } ; + ftical = ftacal ** { m1 = "ْتِ" } ; -- IL hollow VIII + ftAcal = ftacal ** { m1 = "ْتَا" } ; -- IL hollow VIII + ftIcal = ftacal ** { m1 = "ْتِي" } ; -- IL hollow VIII ftacil = { h = "" ; m1 = "ْتَ" ; m2 = "ِ" ; t = "" } ; - ttacal = ftacal ** { m1 = "" ; h = "تَّ" } ; ---- IL assimilated VIII - ttacil = ftacil ** { m1 = "" ; h = "تَّ" } ; ---- IL assimilated VIII + ttacal = ftacal ** { m1 = "" ; h = "تَّ" } ; ---- IL assimilated VIII + ttacil = ftacil ** { m1 = "" ; h = "تَّ" } ; ---- IL assimilated VIII fuccAl = { h = "" ; m1 = "ُ" ; m2 = "َّا" ; t = "" } ; fuccil = { h = "" ; m1 = "ُ" ; m2 = "ِّ" ; t = "" } ; fuccilo = { h = "" ; m1 = "ُ" ; m2 = "ِّ" ; t = "ْ" } ; diff --git a/src/arabic/PhraseAra.gf b/src/arabic/PhraseAra.gf index 9d24e689..40482e3a 100644 --- a/src/arabic/PhraseAra.gf +++ b/src/arabic/PhraseAra.gf @@ -10,15 +10,16 @@ concrete PhraseAra of Phrase = CatAra ** open UttQS qs = {s = \\g => qs.s ! QDir} ; UttImpSg pol imp = {s = \\g => imp.s ! pol.p ! g ! ResAra.Sg ++ pol.s} ; UttImpPl,UttImpPol = \pol,imp -> {s = \\g => imp.s ! pol.p ! g ! ResAra.Pl ++ pol.s} ; + UttInterj i = {s = \\g => i.s} ; - UttIP ip = {s = \\g => ip.s ! g ! Def ! Nom} ; --IL + UttIP ip = {s = \\g => ip.s ! False ! g ! Def ! Nom} ; --IL UttAP ap = {s = ResAra.uttAP ap} ; --IL UttCard c = {s = ResAra.uttNum c} ; --IL UttCN cn = {s = ResAra.uttCN cn } ; --IL UttNP np = {s = \\_ => np.s ! Nom} ; - UttVP vp = {s = \\g => (compVP vp).s ! {g=g ; n=Sg} ! Nom} ; --IL - UttS, + UttVP vp = {s = uttVP vp} ; --IL + UttS s = {s = \\_ => s.s ! Verbal} ; UttAdv, UttIAdv = \s -> {s = \\_ => s.s} ; ---- OK? AR -- diff --git a/src/arabic/QuestionAra.gf b/src/arabic/QuestionAra.gf index 38b2296c..e6bcf6a6 100644 --- a/src/arabic/QuestionAra.gf +++ b/src/arabic/QuestionAra.gf @@ -1,4 +1,4 @@ -concrete QuestionAra of Question = CatAra ** open ResAra, ParamX, Prelude, VerbAra in { +concrete QuestionAra of Question = CatAra ** open ResAra, ParamX, Prelude, VerbAra, SentenceAra in { flags optimize=all_subs ; coding = utf8 ; @@ -7,97 +7,72 @@ concrete QuestionAra of Question = CatAra ** open ResAra, ParamX, Prelude, VerbA QuestCl cl = { s = \\t,p => table { - QIndir => "إِذا" ++ cl.s ! t ! p ! Verbal ; - QDir => "هَل" ++ cl.s ! t ! p ! Verbal + QIndir => "إِذا" ++ cl.s ! t ! p ! toOrder QIndir ; + QDir => "هَلْ" ++ cl.s ! t ! p ! toOrder QDir } }; -- ComplSlashIP vps ip = {} ; --- AR copied from PredVP + --IL guessed QuestVP qp vp = - { s =\\t,p,_ => - let { - ---- o = Verbal ; ---- AR - objgn = pgn2gn vp.obj.a.pgn ; - np = {s = qp.s ! objgn.g ! Def ; ----IL just guessing state - a ={pgn = Per3 Masc qp.n ; isPron = False}} ; - pgn = np.a.pgn ; - gn = pgn2gn pgn; - kataba = vp.s ! pgn ! VPPerf ; - yaktubu = vp.s ! pgn ! VPImpf Ind ; - yaktuba = vp.s ! pgn ! VPImpf Cnj ; - yaktub = vp.s ! pgn ! VPImpf Jus ; - vStr : ResAra.Tense -> Polarity -> Str = - \tn,pl -> case of { - => yaktubu ; - => "لَا" ++ yaktubu ; - => "" ; --no verb "to be" in present - => "لَيسَ" ;--same here, just add negation particle - <_, ResAra.Past, Pos> => kataba ; - <_, ResAra.Past, Neg> => "لَمْ" ++ yaktub ; - <_, ResAra.Fut, Pos> => "سَ" ++ yaktubu ; - <_, ResAra.Fut, Neg> => "لَنْ" ++ yaktuba - }; - pred : ResAra.Tense -> Polarity -> Str = - \tn,pl -> case of { - => vp.pred.s ! gn ! Nom; --xabar marfooc - _ => vp.pred.s ! gn ! Acc --xabar kaana wa laysa manSoob - } ; + let np = ip2np qp vp.isPred ; + cl = PredVP np vp ; + in { s = \\t,p,qf => cl.s ! t ! p ! toOrder qf } ; - } in ---- case o of { ----- _ => - case of { ----- AR workaround 18/12/2008 case of { - -- ya2kuluhu - => (vStr t p) ++ vp.obj.s ++ vp.s2 ++ (pred t p); - -- ya2kuluhu al-waladu, yakuluhu al-2awlaadu - => (vStr t p) ++ np.s ! Nom ++ vp.obj.s ++ vp.s2 ++ (pred t p); - => (vStr t p) ++ vp.obj.s ++ np.s ! Nom ++ vp.s2 ++ (pred t p); - => (vStr t p) ++ vp.obj.s ++ vp.s2 ++ (pred t p) - }; - ---- Nominal => - ---- np.s ! Nom ++ (vStr t p) ++ vp.obj.s ++ vp.s2 ++ (pred t p) - } - ; ---- }; ---- AR guessed - QuestIAdv iadv cl = {s = \\t,p,_ => iadv.s ++ cl.s ! t ! p ! Verbal} ; + QuestIAdv iadv cl = {s = \\t,p,qf => iadv.s ++ cl.s ! t ! p ! toOrder qf} ; ---- IL guessed - QuestIComp icomp np = - let vp = kaan (CompNP np) in - QuestVP icomp vp ; + -- : IComp -> NP -> QCl + QuestIComp ic np = + let vp = UseComp (CompNP np) ; -- puts NP in nominative + ip : ResAra.IP = np ** { -- NP's s is already present in VP, we only want its agr + s = \\_,_,_,_ => ic.s ! pgn2gn np.a.pgn } ; + in QuestVP ip vp ; - CompIP ip = ip ; - -- old, when IComp = Comp { s = \\{g=g ; n=_},c => ip.s ! g ! Def ! c } ; ---- - - CompIAdv iadv = mkIP iadv.s ResAra.Sg ; - - -- QCl = {s : R.Tense => Polarity => QForm => Str} ; - QuestSlash ip cl = { ----IL just guessing - s = \\t,p,qf => case qf of { - QDir => cl.s ! t ! p ! Verbal ++ cl.c2 ++ ip.s ! Masc ! Def ! Nom ; --VSO (purely guessing) - QIndir => cl.s ! t ! p ! Nominal ++ cl.c2 ++ ip.s ! Masc ! Def ! Nom } --SVO (purely guessing) + -- : IP -> IComp ; + CompIP ip = ip ** { + s = \\gn => ip.s ! True -- True=IP will be a subject of predicative sentence + ! gn.g -- IComp agrees in gender with eventual head + ! Def ! Nom ; -- IP will be a subject } ; - PrepIP p ip = {s = p.s ++ ip.s ! Masc ! Def ! Acc} ; ----IL + CompIAdv iadv = { s = \\_ => iadv.s ; a = ResAra.Sg } ; - AdvIP ip adv = ip ** { - s = \\g,s,c => ip.s ! g ! s ! c ++ adv.s ; - n = ip.n + -- QCl = {s : Tense => Polarity => QForm => Str} ; + QuestSlash ip cls = { ----IL just guessing + s = \\t,p,qf => + let cl : ResAra.Cl = complClSlash cls ; -- dummy conversion to Cl + o = toOrder qf + in cls.c2.s ++ ip.s ! False ! Masc ! Def ! Nom ++ cl.s ! t ! p ! o + } ; + + --IL guessed + PrepIP p ip = { + s = p.s ++ ip.s ! False -- not used as a subject of predicative sentence + ! Masc ---- + ! Def ! Gen + } ; + + AdvIP ip adv = ip ** { + s = \\isPred,g,s,c => ip.s ! isPred ! g ! s ! c ++ adv.s ; + } ; + + -- : IDet -> IP + IdetIP idet = idet ** { + s = \\isPred => idet.s ; + a = { pgn = agrP3 NoHum Masc idet.n ; isPron = False } } ; -----IL guessed with help of L and Google translate - -- : IDet -> IP - IdetIP idet = idet ; -- Gender still matters if turned into IComp - -- : IDet -> CN -> IP - IdetCN idet cn = idet ** { - s = \\g,s,c => idet.s ! cn.g ! s ! c ++ -- gender is determined by the CN - cn.s ! idet.n ! Indef ! Gen ; --idaafa + IdetCN idet cn = { + s = \\isPred,g,s,c + => idet.s ! cn.g ! s ! c ++ + cn2str cn idet.n idet.d Gen ; + a = { pgn = agrP3 NoHum cn.g idet.n ; isPron = False } } ; -- : IQuant -> Num -> IDet @@ -105,6 +80,7 @@ concrete QuestionAra of Question = CatAra ** open ResAra, ParamX, Prelude, VerbA s = \\g,s,c => let gend = detGender g num.n -- gender flips with some numbers in iquant.s ! s ! c ++ num.s ! gend ! s ! c ; - n = sizeToNumber num.n + n = sizeToNumber num.n ; + d = Indef ---- TODO check } ; } diff --git a/src/arabic/RelativeAra.gf b/src/arabic/RelativeAra.gf index 3cc0a9cd..25f1fe64 100644 --- a/src/arabic/RelativeAra.gf +++ b/src/arabic/RelativeAra.gf @@ -1,35 +1,50 @@ -concrete RelativeAra of Relative = CatAra ** open ResAra in { - flags coding=utf8; --- --- flags optimize=all_subs ; --- --- lin --- --- RelCl cl = { --- s = \\t,a,p,_ => "سُعه" ++ "تهَت" ++ cl.s ! t ! a ! p ! ODir --- } ; --- --- RelVP rp vp = { --- s = \\t,ant,b,ag => --- let --- agr = case rp.a of { --- RNoAg => ag ; --- RAg a => a --- } ; --- cl = mkClause (rp.s ! Nom) agr vp --- in --- cl.s ! t ! ant ! b ! ODir --- } ; --- --- RelSlash rp slash = { --- s = \\t,a,p,_ => slash.c2 ++ rp.s ! Acc ++ slash.s ! t ! a ! p ! ODir --- } ; +concrete RelativeAra of Relative = CatAra ** + open ResAra in { + flags coding=utf8; + + lin + + RelCl cl = { + s = \\t,p,agr,c => IdRP.s ! agr2ragr agr c ++ cl.s ! t ! p ! Nominal + } ; + + -- : RP -> VP -> RCl ; -- who loves John + RelVP rp vp = { + s = \\t,p,agr,c => + let + npS : Case => Str = \\_ => rp.s ! agr2ragr agr c ; + np : ResAra.NP = agrNP agr ** {s = npS} ; + cl = predVP np vp ; + in + cl.s ! t ! p ! Nominal + } ; + + -- : RP -> ClSlash -> RCl ; -- whom John loves + RelSlash rp cls = cls ** { + s = \\t,p,agr,c => + let --empty : Agr -> NP = emptyNP ; + obj : ResAra.NP = pgn2pron agr.pgn ; -- head is repeated as a clitic object pronoun + cl : ResAra.Cl = complClSlash obj cls ; + in rp.s ! agr2ragr agr c ++ cl.s ! t ! p ! VOS + } ; -- -- FunRP p np rp = { -- s = \\c => np.s ! c ++ p.s ++ rp.s ! Acc ; -- a = RAg np.a -- } ; --- --- IdRP = mkIP "وهِعه" "وهِعه" "وهْسي" Sg ** {a = RNoAg} ; --- + + IdRP = + { s = table { + RSg Masc => "اَلَّذِي" ; + RSg Fem => "اَلَّتِي" ; + RPl Masc => "اَلَّذِين" ; + RPl Fem => "اَللَّاتِي" ; + RDl Masc Bare => "اَللَّذَيْن" ; + RDl Masc Nom => "اَللَّذَانِ" ; + RDl Masc _ => "اَللَّذَيْنِ" ; + RDl Fem Bare => "اَللَّتَيْن" ; + RDl Fem Nom => "اَللَّتَانِ" ; + RDl Fem _ => "اَللَّتَيْنِ" + } + } ; } diff --git a/src/arabic/ResAra.gf b/src/arabic/ResAra.gf index d4f08f63..6156d435 100644 --- a/src/arabic/ResAra.gf +++ b/src/arabic/ResAra.gf @@ -18,8 +18,8 @@ resource ResAra = PatternsAra ** open Prelude, Predef, OrthoAra, ParamX in { Number = Sg | Dl | Pl; Gender = Masc | Fem ; Case = Nom | Acc | Gen - | Bare ; -- 1st person poss. suff. overrides case - Person = P1 | P2 | P3 ; + | Bare -- 1st person poss. suff. overrides case + | Dat ; -- Hack to make the preposition لِ contract Species = NoHum | Hum ; State = Def | Indef | Const | Poss ; -- ة turns into ت @@ -27,12 +27,14 @@ resource ResAra = PatternsAra ** open Prelude, Predef, OrthoAra, ParamX in { -- case vowel retained Mood = Ind | Cnj | Jus ; Voice = Act | Pas ; - Tense = Pres | Past | Fut ; - Order = Verbal | Nominal ; + Order = Verbal | Nominal + | VOS -- Relative clauses with resumptive pronouns + | Subord ; -- Nominal word order but subject in accusative oper - --roots, patterns, and making words: +----------------------------------------------------------------------------- +-- General morphology with roots, patterns, and making words: Pattern : Type = {h, m1, m2, t : Str}; Root : Type = {f : Str}; @@ -107,19 +109,57 @@ resource ResAra = PatternsAra ** open Prelude, Predef, OrthoAra, ParamX in { --types of open classes: NTable = Number => State => Case => Str; + emptyNTable : NTable = \\n,s,c => [] ; + + Preposition : Type = {s : Str ; c : Case} ; + Noun : Type = { + s,s2 : NTable ; + g : Gender ; + h : Species ; + isDual : Bool -- whether it takes dual instead of plural: eyes, twins, ... + } ; + Noun2 : Type = Noun ** {c2 : Preposition} ; + Noun3 : Type = Noun2 ** {c3 : Preposition} ; + + mkPreposition = overload { + mkPreposition : Str -> Case -> Preposition = \s,c -> {s=s;c=c} ; + mkPreposition : Str -> Preposition = \s -> {s=s;c=Gen} ; + } ; + + noPrep : Preposition = mkPreposition [] Nom ; + liPrep : Preposition = mkPreposition ( + pre { #pronSuffAndOther => "لِ" ; + #pronSuff => "لَ" ; + _ => "لِ" + } ++ BIND) Dat ; + biPrep : Preposition = mkPreposition ("بِ"++BIND) ; + + pronSuff : pattern Str = #("كَ"|"كِ"|"كُمَا"|"كُمْ"|"كُنَّ"|"هُ"|"ها"|"هُمَا"|"هُمْ"|"هُنَّ") ; + pronSuffAndOther : pattern Str = #( "كَم" ) ; -- TODO list words that begin like pron.suff. but aren't - Noun : Type = {s : NTable ; g : Gender; h : Species} ; --- Adj : Type = {s : Gender => NTable} ; Adj : Type = {s : AForm => Str} ; + Adj2 : Type = Adj ** {c2 : Preposition} ; + Verb : Type = {s : VForm => Str} ; + Verb2 : Type = Verb ** {c2 : Preposition} ; + Verb3 : Type = Verb2 ** {c3 : Preposition} ; AP : Type = {s : Species => Gender => NTable } ; uttAP : AP -> (Gender => Str) ; uttAP ap = \\g => ap.s ! NoHum ! g ! Sg ! Def ! Nom ; ----IL - CN : Type = Noun ** {adj : NTable ; np : Case => Str}; + CN : Type = Noun ** {np : Case => Str}; + + -- All fields of NP + cn2str : CN -> Number -> State -> Case -> Str = \cn,n,s,c -> + cn.s ! n ! s ! c ++ + cn.s2 ! n ! s ! c ++ + cn.np ! c ; + + useN : Noun -> CN = \n -> n ** {np = \\_ => []} ; + uttCN : CN -> (Gender => Str) ; - uttCN cn = \\_ => cn.s ! Sg ! Indef ! Bare ; + uttCN cn = \\_ => cn2str cn Sg Indef Bare ; NumOrdCard : Type = { s : Gender => State => Case => Str ; @@ -127,7 +167,6 @@ resource ResAra = PatternsAra ** open Prelude, Predef, OrthoAra, ParamX in { isNum : Bool } ; - uttNum : NumOrdCard -> (Gender => Str) ; uttNum n = \\g => n.s ! g ! Def ! Nom ; ----IL @@ -263,8 +302,11 @@ oper } ; --macro for defective verbs: - verbDef : DefForms -> Vowel -> Verb = - \vforms,vowImpf -> + verbDef : DefForms -> Vowel -> Verb = verbDefBool False ; + verbDoubleDef : DefForms -> Vowel -> Verb = verbDefBool True ; + + verbDefBool : Bool -> DefForms -> Vowel -> Verb = + \isDoubleDef,vforms,vowImpf -> let { rama = vforms ! 0 ; -- VPerf Act (Per3 Masc Sg) ramay = vforms ! 1 ; -- VPerf Act (Per3 Fem Pl) @@ -280,13 +322,14 @@ oper patPerf = patDefPerf rama ramay rumi rumu rumiy ; patImpfAct = patDefImpfAct armi armu ; - patImp = patDefImp Irmi Irmu + patImp = patDefImp Irmi Irmu ; + suffixImpf = case isDoubleDef of {True => suffixImpfDoubleDef ; _ => suffixImpfDef} } in { s = table { - VPerf v pgn => patPerf ! v ! pgn + suffixPerfDef v ! pgn ; - VImpf m Act pgn => prefixImpf ! pgn + patImpfAct ! pgn + suffixImpfDef Act vowImpf ! m ! pgn ; - VImpf m Pas pgn => prefixImpf ! pgn + urma + suffixImpfDef Pas vowImpf ! m ! pgn ; - VImp g n => patImp ! g ! n + suffixImpfDef Act vowImpf ! Jus ! Per2 g n ; + VPerf v pgn => patPerf ! v ! pgn + suffixPerfDef v ! pgn ; + VImpf m Act pgn => prefixImpf ! pgn + patImpfAct ! pgn + suffixImpf Act vowImpf ! m ! pgn ; + VImpf m Pas pgn => prefixImpf ! pgn + urma + suffixImpf Pas vowImpf ! m ! pgn ; + VImp g n => patImp ! g ! n + suffixImpf Act vowImpf ! Jus ! Per2 g n ; VPPart => ppart } } ; @@ -352,7 +395,6 @@ oper Per1 Plur => "نَا" } ; - suffixImpfDef : Voice -> Vowel -> Mood => PerGenNum => Str = \vc,vw -> let { default : Mood -> Str = \m -> @@ -389,6 +431,10 @@ oper } } ; + -- does this even happen other than with رءي? /IL + suffixImpfDoubleDef : Voice -> Vowel -> Mood => PerGenNum => Str = \vc,vw -> + \\m,p => rmSukun (suffixImpfDef vc vw ! m ! p) ; + --now is used for the sound, assimilated (weak C1), and when C1 = hamza: v1sound : Root3 -> Vowel -> Vowel -> Verb = @@ -406,8 +452,8 @@ v1sound : Root3 -> Vowel -> Vowel -> Verb = }; uktab = mkStrong ufcal fcl ; euktub = case fcl.f of { - "؟"|"و"|"ي" => qif ; - _ => prefixImp ! vowImpf + ktub + "ء"|"و"|"ي" => qif ; + _ => prefixImp ! vowImpf + ktub }; maktUb = mkStrong mafcUl fcl } in @@ -506,7 +552,7 @@ toDefForms : (x1,_,_,_,_,_,_,_,_,_,x11 : Str) -> DefForms = 7 => h ; 8 => i ; 9 => j ; 10 => k } ; -def1Forms_perfA : Root3 -> Vowel -> DefForms = \rmy,vowImpf -> +v1DefForms_perfA : Root3 -> Vowel -> DefForms = \rmy,vowImpf -> let { _rmi = mkDefective (patDef1 ! vowImpf) rmy ; _rmu = mkDefective (patDef2 ! vowImpf) rmy ; @@ -524,11 +570,11 @@ def1Forms_perfA : Root3 -> Vowel -> DefForms = \rmy,vowImpf -> } in toDefForms rama ramay rumi rumu rumiy armi armu urma eirmi eirmu marmiy ; v1defective_a : Root3 -> Vowel -> Verb = \rmy,vowImpf -> - let vforms = def1Forms_perfA rmy vowImpf + let vforms = v1DefForms_perfA rmy vowImpf in verbDef vforms vowImpf ; v1defective_i : Root3 -> Vowel -> Verb = \bqy,vowImpf -> -- IL (conjugation 1d4) - let vforms_a = def1Forms_perfA bqy vowImpf ; + let vforms_a = v1DefForms_perfA bqy vowImpf ; baqI = mkDefective facIl bqy ; baqiy = mkDefective facil bqy ; vforms_i = table { 0 => baqI ; @@ -536,6 +582,19 @@ v1defective_i : Root3 -> Vowel -> Verb = \bqy,vowImpf -> -- IL (conjugation 1d4) x => vforms_a ! x } ; in verbDef vforms_i vowImpf ; +v1doubleweak : Root3 -> Verb = \r'y -> + let ry = r'y ** {c = ""} ; + vforms_doubleweak : DefForms = \\x => rmSukun (v1DefForms_perfA ry a ! x) ; -- only remove the first sukun + vforms_weak : DefForms = v1DefForms_perfA r'y a ; + vforms = table { 0 => vforms_weak ! 0 ; -- all perfect forms + 1 => vforms_weak ! 1 ; + 2 => vforms_weak ! 2 ; + 3 => vforms_weak ! 3 ; + 4 => vforms_weak ! 4 ; + x => vforms_doubleweak ! x } ; + in verbDoubleDef vforms a ; -- sukun in suffixes is removed in verbDoubleDef + + patDef1 : Vowel => Pattern = table { u => fcu ; @@ -598,24 +657,54 @@ v4sound : Root3 -> Verb = } in verb eaqnac euqnic uqnic uqnac eaqnic muqnac; +v4hollow : Root3 -> Verb = + \rwd -> + let { + earad = mkHollow eafac rwd ; -- VPerf Act (Per3 Fem Pl) etc. + earAd = mkHollow eafAc rwd ; -- VPerf Act + eurid = mkHollow eufic rwd ; -- VPerf Pas (Per3 Fem Pl) etc. + eurId = mkHollow eufIc rwd ; -- VPerf Pas + + urid = mkHollow ufic rwd ; -- VImpf Act (Per2/Per3 Fem Pl) + urId = mkHollow ufIc rwd ; -- VImpf Act + urad = mkHollow ufac rwd ; -- VImpf Pas (Per2/Per3 Fem Pl) + urAd = mkHollow ufAc rwd ; -- VImpf Pas + + earid = mkHollow eafic rwd ; -- VImp (Sg Masc / Pl Fem) + earId = mkHollow eafIc rwd ; -- VImp (Pl Masc / Sg Fem) + + ppart = "م" + urAd ; + + } in verbHollow (toDefForms + earAd earad eurId eurid + urId urid urAd urad + earId earid ppart) ; + + v4DefForms : Root3 -> DefForms = \cTy -> + let { + _cTa = mkDefective fca cTy; + _cTu = mkDefective fcu cTy; + _cTi = mkDefective fci cTy; + eacTa = "أَ" + _cTa; -- VPerf Act (Per3 Masc Sg) + eacTay = mkStrong eafcal cTy ; -- VPerf Act (Per3 Fem Pl) + eucTi = "أُ" + _cTi; -- VPerf Pas (Per3 _ Sg) + eucTu = "أُ" + _cTu; -- VPerf Pas (Per3 Masc Pl) + eucTiy = mkStrong eufcil cTy ; -- VPerf Pas (Per3 Fem Pl) + ucTi = "ُ" + _cTi; -- VImpf Act + ucTu = "ُ" + _cTu; -- VImpf Act (Per2/3 Masc Pl) + ucTa = "ُ" + _cTa; -- VImpf Pas + eacTi = "أَ" + _cTi; -- VImp (Masc Sg / Fem _) + eacTu = "أَ" + _cTu; -- VImp Masc Pl + mucTaY = "م" + ucTa +"ى" + } in toDefForms eacTa eacTay eucTi eucTu eucTiy ucTi ucTu ucTa eacTi eacTu mucTaY ; v4defective : Root3 -> Verb = \cTy -> - let { - cTa = mkDefective fca cTy; - cTu = mkDefective fcu cTy; - cTi = mkDefective fci cTy; - eacTa = "أَ" + cTa; - eacTay = mkStrong eafcal cTy ; - ucTi = "ُ" + cTi; - eucTi = "أُ" + cTi; - ucTu = "ُ" + cTu; - eucTu = "أُ" + cTu; - eucTiy = mkStrong eufcil cTy ; - ucTa = "ُ" + cTa; - eacTi = "أَ" + cTi; - eacTu = "أَ" + cTu; - mucTaY = "م" + ucTa +"ى" - } in verbDef (toDefForms eacTa eacTay eucTi eucTu eucTiy ucTi ucTu ucTa eacTi eacTu mucTaY) i; + verbDef (v4DefForms cTy) i ; + +v4doubleweak : Root3 -> Verb = \r'y -> + let ry = r'y ** {c = ""} ; + vforms : DefForms = \\x => rmSukun (v4DefForms ry ! x) ; -- only remove the first sukun + in verbDoubleDef vforms i ; -- sukun in suffixes is removed in verbDoubleDef v5sound : Root3 -> Verb = \nfs -> @@ -637,6 +726,33 @@ v6sound : Root3 -> Verb = mutafAqam = "م" + utafAqam } in verb tafAqam tufUqim atafAqam utafAqam tafAqam mutafAqam; +-- v7sound : Root3 -> Verb = -- TODO 7s +-- \fcl -> +-- let { +-- _facal = mkStrong facal fcl ; +-- _facil = mkStrong facil fcl; +-- infacal = "اِنْ" + _facal ; -- VPerf Act +-- ; -- VPerf Pas +-- anfacil = "َنْ" + _facil ; -- VImpf _ Act +-- ; -- VImpf _ Pas +-- ; -- VImp +-- -- VPPart +-- } in +-- verb ; + +v7geminate : Root3 -> Verb = -- IL 7g -- very likely wrong + \fcl -> + let { + _nfacc = "نْ" + mkHollow facc fcl ; + infacal = "اِنْ" + mkStrong facal fcl ; -- VPerf Act -- TODO use another constructor, this is wrong for 3rd person + unfucc = "اُنْ" + mkHollow fucc fcl ; -- VPerf Pas + anfacc = "َ" + _nfacc ; -- VImpf _ Act + unfacc = "ُ" + _nfacc ; -- VImpf _ Pas + infacc = "اِ" + _nfacc ; -- VImp + munfacc = "مُ" +_nfacc -- VPPart + } in + verb infacal unfucc anfacc unfacc infacc munfacc ; + v8sound : Root3 -> Verb = \rbT -> let { @@ -663,11 +779,33 @@ v8assimilated : Root3 -> Verb = --- IL 8a1 muttafaq = "م" + uttafaq } in verb eittafaq euttufiq attafiq uttafaq eittafiq muttafaq; -v10sound : Root3 -> Verb = ---- IL 10s -- to be checked +v8hollow : Root3 -> Verb = -- IL + \Hwj -> + let { + _Htaj = mkHollow ftacal Hwj ; + _HtAj = mkHollow ftAcal Hwj ; + _Htij = mkHollow ftical Hwj ; + _HtIj = mkHollow ftIcal Hwj ; + iHtaj = "اِ" + _Htaj ; -- VPerf Act (Per3 Fem Pl) + iHtAj = "اِ" + _HtAj ; -- VPerf Act _ + uHtij = "اُ" + _Htij ; -- VPerf Pas (Per3 Fem Pl) + uHtIj = "اُ" + _HtIj ; -- VPerf Pas _ + aHtaj = "َ" + _Htaj ; -- VImpf Act (Per2/Per3 Fem Pl) + aHtAj = "َ" + _HtAj ; -- VImpf Act _ + uHtaj = "ُ" + _Htaj ; -- VImpf Pas (Per2/Per3 Fem Pl) + uHtAj = "ُ" + _Htaj ; -- VImpf Pas _ + -- iHtaj again -- VImp Sg Masc / Pl Fem + -- iHtAj again -- VImp Pl Masc / Sg Fem + ppart = "مُ" + _HtAj -- PPart + + } in verbHollow (toDefForms + iHtAj iHtaj uHtIj uHtij aHtAj aHtaj + uHtAj uHtaj iHtAj iHtaj ppart) ; +v10sound : Root3 -> Verb = -- IL 10s -- to be checked \qtl -> let { - _staqtal = "َستَ" + mkStrong fcal qtl ; - _staqtil = "َستَ" + mkStrong fcil qtl; + _staqtal = "ستَ" + mkStrong fcal qtl ; + _staqtil = "ستَ" + mkStrong fcil qtl; istaqtal = "اِ" + _staqtal ; -- VPerf Act ustuqtil = "اُسْتُ" + mkStrong fcil qtl; -- VPerf Pas astaqtil = "َ" + _staqtil ; -- VImpf _ Act @@ -677,7 +815,7 @@ v10sound : Root3 -> Verb = ---- IL 10s -- to be checked } in verb istaqtal ustuqtil astaqtil astaqtal istaqtil mustaqtal ; -v10hollow : Root3 -> Verb = ---- IL 10h -- to be checked +v10hollow : Root3 -> Verb = -- IL 10h -- to be checked \xwf -> let { _staxaf = "سْتَ" + mkHollow fac xwf ; @@ -696,10 +834,33 @@ v10hollow : Root3 -> Verb = ---- IL 10h -- to be checked ustaxAf = "ُ" + _staxAf ; -- VImpf Pas _ ppart = "مُ" + _staxIf -- PPart ("weird anomalies" here too?) - } in verbHollow (toDefForms + } in verbHollow (toDefForms istaxAf istaxaf ustuxIf ustuxif astaxIf astaxif ustaxAf ustaxaf istaxif istaxIf ppart) ; +v10defective : Root3 -> Verb = -- IL + \lqy -> + let { + _stalqa = "سْتَ" + mkDefective fca lqy ; + _stalqu = "سْتَ" + mkDefective fcu lqy ; + _stalqi = "سْتَ" + mkDefective fci lqy ; + _stulqi = "سْتُ" + mkDefective fci lqy ; + + istalqa = "اِ" + _stalqa ; -- VPerf Act (Per3 Masc Sg) + istalqay = "اِسْتَ" + mkStrong fcal lqy ; -- VPerf Act (Per3 Fem Pl) + ustulqi = "اُ" + _stulqi; -- VPerf Pas (Per3 _ _) + + astalqu = "َ" + _stalqu ; -- VImpf Act (Per2/3 Masc Pl) + astalqi = "َ" + _stalqi ; -- VImpf Act _ + ustalqa = "ُ" + _stalqa ; -- VImpf Pas _ + istalqi = "اِ" + _stalqi; -- VImp (Masc Sg / Fem _) + istalqu = "اِ" + _stalqu; -- VImp Masc Pl + mustalqin = "مُ" + _stalqi + "ت" ; + + } in verbDef (toDefForms + istalqa istalqay ustulqi ustulqi ustulqi + astalqi astalqu ustalqa istalqi istalqu mustalqin) i ; + patV1Perf : Vowel => Pattern = table { a => facal ; --katab @@ -723,8 +884,8 @@ endVowel : Mood => Str = prefixImp : Vowel => Str = table { - u => "أُ" ; - _ => "إِ" + u => "اُ" ; + _ => "اِ" } ; patHollowPerf : (_,_,_,_ :Str) -> Voice => PerGenNum => Str = \xAf,xif,xIf,xuf -> @@ -795,7 +956,7 @@ patHollowImp : (_,_ :Str) -> Gender => Number => Str =\xaf,xAf -> Bare => [] ; Nom => "ُ"; Acc => "َ"; - Gen => "ِ" + _Gen => "ِ" -- dat is the same as gen, except in definite before لِ }; --takes the adjective lemma and gives the Posit table @@ -825,7 +986,7 @@ patHollowImp : (_,_ :Str) -> Gender => Number => Str =\xaf,xAf -> -- indeclinable nominal word (mamnuu3 mina S-Sarf) indeclN : Str -> State => Case => Str = - \aHmar -> \\s,c => defArt s aHmar + indecl!c; + \aHmar -> \\s,c => defArt s c aHmar + indecl!c; -- takes 2 words, singular and broken plural, and gives the -- complete noun inflection table @@ -860,12 +1021,14 @@ patHollowImp : (_,_ :Str) -> Gender => Number => Str =\xaf,xAf -> -- takes a singular or broken plural word and tests the ending to -- determine the declension and gives the corresponding inf table sing : Str -> State => Case => Str = \word -> - \\s,c => defArt s (case word of { - lemma + "ِي" => fixShd lemma (dec2sg ! s ! c) ; - _ + ("ا"|"ى") => fixShd word (dec3sg ! s ! c) ; - lemma + "ة" => case s of { + \\s,c => defArt s c (case word of { + lemma + "ِيّ" => fixShd word (decNisba ! s ! c) ; + lemma + "ِي" => fixShd lemma (dec2sg ! s ! c) ; + _ + ("ا"|"ى") => fixShd word (dec3sg ! s ! c) ; + lemma + ("ء"|"أ"|"ئ"|"ؤ") => word + dec1sgNoDoubleAlif ! s ! c ; + lemma + "ة" => case s of { Poss => lemma + "ت" + dec1sg ! s ! c ; - _ => word + dec1sg ! s ! c + _ => word + dec1sgNoDoubleAlif ! s ! c } ; _ => fixShd word (dec1sg ! s ! c) }) ; @@ -874,7 +1037,7 @@ patHollowImp : (_,_ :Str) -> Gender => Number => Str =\xaf,xAf -> -- takes a singular word and tests the ending to -- determine the declension and gives the corresponding dual inf table dual : Str -> State => Case => Str = \caSaA -> - \\s,c => defArt s (case caSaA of { + \\s,c => defArt s c (case caSaA of { lemma + ("ا"|"ى") => lemma + "ي" + dl ! s ! c ; lemma + "ة" => lemma + "ت" + dl ! s ! c ; _ => fixShd caSaA (dl ! s ! c) @@ -884,13 +1047,13 @@ patHollowImp : (_,_ :Str) -> Gender => Number => Str =\xaf,xAf -> --plural feminine table plurF : Str -> State => Case => Str = \kalima -> - \\s,c => defArt s (mkAt kalima) + f_pl ! s ! c ; + \\s,c => defArt s c (mkAt kalima) + f_pl ! s ! c ; -- takes a singular word and gives the corresponding sound --plural masculine table. FIXME: consider declension 3 plurM : Str -> State => Case => Str = \mucallim -> - \\s,c => defArt s mucallim + m_pl ! s ! c ; + \\s,c => defArt s c mucallim + m_pl ! s ! c ; -- to add the Al prefix for Definite words Al : State => Str = @@ -899,14 +1062,14 @@ patHollowImp : (_,_ :Str) -> Gender => Number => Str =\xaf,xAf -> _ => "" }; - defArt : State -> Str -> Str = \st,stem -> -- IL -- to be checked + defArt : State -> Case -> Str -> Str = \st,c,stem -> -- IL -- to be checked let al = "ال" in - case st of { - Def => + case of { + => "ل" + stem ; -- only happens before the preposition لِ + => case stem of { - s@#sun + v@#vow + x => al + s + v + "ّ" + x ; -- vowel before shadda - s@#sun + x => al + s + "ّ" + x; - x => al + x } ; + s@#sun + x => fixShd (al + s) ("ّ" + x) ; + x => al + x } ; _ => stem }; @@ -917,18 +1080,25 @@ patHollowImp : (_,_ :Str) -> Gender => Number => Str =\xaf,xAf -> table { Bare => []; Nom => "ٌ"; - Acc => "ً"; - Gen => "ٍ" + Acc => "اً"; + _Gen => "ٍ" }; _ => caseTbl --think of ?axU, ?axA, (the five nouns) }; + -- if a word ends in ء or ة, don't add alif for indef acc. + dec1sgNoDoubleAlif : State => Case => Str = \\s,c => + case of { + => "ً" ; + _ => dec1sg ! s ! c + }; + --indeclinables (mamnuu3 mina S-Sarf) indecl : Case => Str = table { - Gen => "َ" ; - x => caseTbl ! x + (Gen|Dat) => "َ" ; + x => caseTbl ! x }; @@ -942,6 +1112,7 @@ patHollowImp : (_,_ :Str) -> Gender => Number => Str =\xaf,xAf -> _ => "ِي" }; + --declension 3 (ending in alif) dec3sg : State => Case => Str = \\s,c => case of { @@ -950,6 +1121,15 @@ patHollowImp : (_,_ :Str) -> Gender => Number => Str =\xaf,xAf -> _ => [] }; + --declension 2 (ends in yaa') + decNisba : State => Case => Str = \\s,c => + case of { + <_, Bare> => [] ; + => "اً" ; + => "ٍ" ; + <_, Acc> => "َ" ; + _ => [] + }; --dual suffixes dl : State => Case => Str = @@ -1051,7 +1231,7 @@ patHollowImp : (_,_ :Str) -> Gender => Number => Str =\xaf,xAf -> => Acc; => Gen; => Gen; - <_,Const> => Gen; + <_,Const> => Gen; -- not sure if this is an actual rule /IL _ => c }; @@ -1102,9 +1282,14 @@ patHollowImp : (_,_ :Str) -> Gender => Number => Str =\xaf,xAf -> } }; + gn2pgn : {g : Gender; n : Number} -> PerGenNum = \gn -> + case gn of { {g = gn; n = nm} => Per3 gn nm } ; + + -- these are chosen in many places, trying to be consistent + toOrder : QForm -> Order = \qf -> + case qf of { QIndir => Nominal ; + QDir => Verbal } ; - mkIP : Str -> Number -> IP = - \s,n -> {s = \\_g,_s,_c => s ; n = n} ; mkOrd : (_,_ : Str) -> Size -> NumOrdCard = \aysar,yusra,sz -> @@ -1118,7 +1303,9 @@ patHollowImp : (_,_ :Str) -> Gender => Number => Str =\xaf,xAf -> }; - oper + +----------------------------------------------------------------------------- +-- Det, Quant BaseQuant : Type = { d : State; @@ -1148,53 +1335,172 @@ patHollowImp : (_,_ :Str) -> Gender => Number => Str =\xaf,xAf -> Agr = { pgn : PerGenNum; isPron : Bool} ; AAgr = { g : Gender ; n : Number} ; - Comp : Type = { - s : AAgr => Case => Str - } ; - Obj : Type = { - s : Str ; - a : Agr - }; +----------------------------------------------------------------------------- +-- NP, Pron NP : Type = { s : Case => Str ; - a : Agr + a : Agr ; + empty : Str -- to prevent ambiguities with prodrop } ; + mkPron : (_,_,_ : Str) -> PerGenNum -> NP = \ana,nI,I,pgn -> + { s = + table { + (Nom|Bare) => ana; + Acc => nI ; -- object suffix + Gen => I ; -- possessive suffix + Dat => I -- will only be used with preposition لِ + }; + a = {pgn = pgn; isPron = True }; + empty = [] + }; + + proDrop : NP -> NP = \np -> + case np.a.isPron of { + True => np ** {s = table {Nom => [] ; x => np.s ! x}}; + _ => np + } ; + + emptyNP : NP = { + s = \\_ => [] ; + a = {pgn = Per3 Masc Sg ; isPron = False} ; + empty = [] } ; + + agrNP : Agr -> NP = \agr -> emptyNP ** {a = agr} ; + + i_Pron : NP = mkPron "أَنَا" "نِي" "ي" (Per1 Sing) ; + we_Pron : NP = mkPron "نَحنُ" "نا" "نا" (Per1 Plur) ; + + youSgMasc_Pron : NP = mkPron "أَنتَ" "كَ" "كَ" (Per2 Masc Sg) ; + youSgFem_Pron : NP = mkPron "أَنتِ" "كِ" "كِ" (Per2 Fem Sg) ; + youDlMasc_Pron : NP = mkPron "أَنتُمَا" "كُمَا" "كُمَا" (Per2 Masc Dl) ; + youDlFem_Pron : NP = mkPron "أَنتُمَا" "كُمَا" "كُمَا" (Per2 Fem Dl) ; + youPlMasc_Pron : NP = mkPron "أَنتُمْ" "كُمْ" "كُمْ" (Per2 Masc Pl) ; + youPlFem_Pron : NP = mkPron "أَنتُنَّ" "كُنَّ" "كُنَّ" (Per2 Fem Pl) ; + + he_Pron : NP = mkPron "هُوَ" "هُ" "هُ" (Per3 Masc Sg) ; + she_Pron : NP = mkPron "هِيَ" "ها" "ها" (Per3 Fem Sg) ; + theyDlMasc_Pron : NP = mkPron "هُمَا" "هُمَا" "هُمَا" (Per3 Masc Dl) ; + theyDlFem_Pron : NP = mkPron "هُمَا" "هُمَا" "هُمَا" (Per3 Fem Dl) ; + theyMasc_Pron : NP = mkPron "هُمْ" "هُمْ" "هُمْ" (Per3 Masc Pl) ; + theyFem_Pron : NP = mkPron "هُنَّ" "هُنَّ" "هُنَّ" (Per3 Fem Pl) ; + + + -- Used e.g. to encode the subject as an object clitic + -- or to find a possessive suffix corresponding to the NP. + -- If the NP is a pronoun, just use itself. + np2pron : NP -> NP = \np -> case np.a.isPron of { + True => np ; + False => pgn2pron np.a.pgn + } ; + + pgn2pron : PerGenNum -> NP = \pgn -> + case pgn of { + Per1 Sing => i_Pron ; + Per1 Plur => we_Pron ; + Per2 Fem Sg => youSgFem_Pron ; + Per2 Masc Sg => youSgMasc_Pron ; + Per2 Fem Dl => youDlFem_Pron ; + Per2 Masc Dl => youDlMasc_Pron ; + Per2 Fem Pl => youPlFem_Pron ; + Per2 Masc Pl => youPlMasc_Pron ; + Per3 Fem Sg => she_Pron ; + Per3 Masc Sg => he_Pron ; + Per3 Fem Dl => theyDlFem_Pron ; + Per3 Masc Dl => theyDlMasc_Pron ; + Per3 Fem Pl => theyFem_Pron ; + Per3 Masc Pl => theyMasc_Pron + } ; + + pron2np : NP -> NP = \np -> np ** { + a = np.a ** {isPron=False} -- hack, sometimes we *don't* want prodrop + } ; + + reflPron : Case -> PerGenNum -> Str = \c,pgn -> + let pron : NP = pgn2pron pgn + in "نَفْس" + caseTbl ! c ++ pron.s ! Gen ; + + reflV : Verb -> Verb = \v -> v ** { + s = \\vf => case vf of { + VPerf _ pgn => v.s ! vf ++ reflPron Acc pgn ; + VImpf _ _ pgn => v.s ! vf ++ reflPron Acc pgn ; + VImp g n => v.s ! vf ++ reflPron Acc (Per2 g n) ; + VPPart => v.s ! vf ++ reflPron Acc (Per3 Masc Sg) ---- + } + } ; +----------------------------------------------------------------------------- +-- IP, questions + IP : Type = { - s : Gender -- because of CompIP + s : Bool -- different forms for "what is this" and "what do you do" + => Gender -- because an IP can be made into an IComp => State => Case -- because of PrepIP: e.g. "in which" chooses definite accusative => Str ; - n : Number + a : Agr -- can be both subject and object of a QCl, needs full agr. info (stupid given that s depends on gender but meh) } ; - param VPForm = + mkIP = overload { + mkIP : Str -> Number -> IP = \maa,n -> { + s = \\_p,_g,_s,_c => maa ; + a = { pgn = agrP3 NoHum Masc n ; isPron = False } + } ; + mkIP : (_,_ : Str) -> Number -> IP = \maa,maadhaa,n -> { + s = table { True => \\_g,_s,_c => maa ; + False => \\_g,_s,_c => maadhaa } ; + a = { pgn = agrP3 NoHum Masc n ; isPron = False } + } + } ; + + ip2np : IP -> Bool -> NP = \ip,isPred -> ip ** { s = ip.s ! isPred ! Masc ! Def ; empty = [] } ; + np2ip : NP -> IP = \np -> np ** {s = \\_,_,_ => np.s} ; + + IDet : Type = { + s : Gender -- IdetCN needs to choose the gender of the CN + => State -- Needs to be retained variable for IP; PrepIP chooses the state of IP + => Case => Str ; + n : Number ; + d : State -- in IdetCN, chooses the state of the CN + } ; + + IQuant : Type = { + s : State => Case => Str + } ; + + IComp : Type = { + s : AAgr -- "how old": masc or fem for adjective + -- no need for Case, IComp is only used by QuestIComp, as grammatical subject + => Str ; + } ; + +----------------------------------------------------------------------------- +-- VP + + param VPForm = VPPerf | VPImpf Mood | VPImp ; oper - VP : Type = { - s : PerGenNum => VPForm => Str; + BaseVP : Type = { -- to minimise duplication of code for VPS + sc : Preposition ; -- subject case: e.g. يُمْكِنُ *لِ*Xِ obj : Obj; pred : Comp; isPred : Bool; --indicates if there is a predicate (xabar) s2 : Str - }; - - -- For complements of VV. - -- TODO: does verbal complement agree with the noun - compVP : VP -> Comp = \vp -> ---- IL - { s = table { - aagr@{g=g ; n=n} => \\c => - vp.s ! Per3 g n ! VPImpf Ind ---- IL guesswork + https://arabic.desert-sky.net/g_modals.html - ++ vp.s2 - ++ vp.pred.s ! aagr ! Acc - ++ vp.obj.s } } ; + VP : Type = BaseVP ** { + s : PerGenNum => VPForm => Str ; + } ; + + uttVP : VP -> (Gender=>Str) = \vp -> + \\g => vp.s ! Per3 g Sg ! VPPerf + ++ vp.obj.s ++ vp.pred.s ! {n = Sg ; g = g} ! Nom + ++ vp.s2 ; + predV : Verb -> VP = \v -> { s = \\pgn,vf => let gn = pgn2gn pgn in @@ -1203,17 +1509,94 @@ patHollowImp : (_,_ :Str) -> Gender => Number => Str =\xaf,xAf -> VPImpf m => v.s ! (VImpf m Act pgn); VPImp => v.s ! (VImp gn.g gn.n) }; - obj = { - s = [] ; - a = {pgn = Per3 Masc Sg ; isPron = False} - }; --or anything! + sc = noPrep ; + obj = emptyObj ; s2 = []; - pred = { s = \\_,_ => []}; + pred = {s = \\_,_ => []} ; isPred = False }; - predVSlash : Verb ** {c2 : Str} -> VPSlash = \v -> - predV v ** {c2 = v.c2} ; + passPredV : Verb -> VP = \v -> + let actVP = predV v in actVP ** { + s = \\pgn,vf => + case vf of { + VPPerf => v.s ! (VPerf Pas pgn) ; + VPImpf m => v.s ! (VImpf m Pas pgn) ; + _ => actVP.s ! pgn ! vf + } + }; + + predVP : NP -> VP -> Cl = \np,vp -> + { s =\\t,p,o => + let { + pgn = + case of { + => verbalAgr np.a.pgn; + _ => np.a.pgn + }; + sc : Preposition = case o of { -- very unsure of this /IL + Subord => {s=[]; c=Acc} ; -- to prevent weird stuff with VVs + _ => case np.a.isPron of {True => noPrep; _ => vp.sc} + } ; + subj = np.empty ++ sc.s + ++ case vp.isPred of { + False => (proDrop np).s ! sc.c ; -- prodrop if it's not predicative + True => np.s ! sc.c + } ; + } in wordOrder o + vp.obj.a.isPron np.a.isPron + (vStr vp pgn t p) + vp.obj.s + (pred vp pgn t p) + vp.s2 + subj + } ; + + -- seems complicated, but this is to share code with VPS and other similar structures + wordOrder : Order -> (objIsPron,subjIsPron : Bool) -> (verb,obj,pred,adv,subj : Str) -> Str = + \o,objIsPron,subjIsPron,verb,obj,pred,adv,subj -> + let cl = wordOrderNoSubj o objIsPron verb obj pred adv in + case o of { + Subord => + let bind = if_then_Str subjIsPron BIND [] -- in subord. clause, subj. pronoun binds to the main verb + in cl.before ++ bind ++ subj ++ cl.after ; + _ => cl.before ++ subj ++ cl.after + } ; + + wordOrderNoSubj : Order -> (objIsPron : Bool) -> (verb,obj,pred,adv : Str) -> {before,after : Str} = + \o,objIsPron,verb,obj,pred,adv -> + case o of { + VOS => {before = verb ++ obj ++ pred ++ adv; after = []} ; + Verbal => case objIsPron of { + True => {before = verb ++ obj ; after = adv ++ pred} ; -- obj. clitic attaches directly to the verb + False => {before = verb ; after = obj ++ adv ++ pred} + } ; + (Nominal|Subord) => {before = [] ; after = verb ++ obj ++ adv ++ pred} + } ; + + pred : VP -> PerGenNum -> ParamX.Tense -> Polarity -> Str = \vp,pgn,tn,pl -> + let gn = pgn2gn pgn + in case of { + => vp.pred.s ! gn ! Nom; --xabar marfooc + _ => vp.pred.s ! gn ! Acc --xabar kaana wa laysa manSoob + } ; + + vStr : VP -> PerGenNum -> ParamX.Tense -> Polarity -> Str = \vp,pgn,tn,pl -> + let kataba = vp.s ! pgn ! VPPerf ; + yaktubu = vp.s ! pgn ! VPImpf Ind ; + yaktuba = vp.s ! pgn ! VPImpf Cnj ; + yaktub = vp.s ! pgn ! VPImpf Jus ; + in case of { + => yaktubu ; + => "لَا" ++ yaktubu ; + => "" ; --no verb "to be" in present + => "لَيسَ" ;--same here, just add negation particle + <_, Past, Pos> => kataba ; + <_, Past, Neg> => "لَمْ" ++ yaktub ; + <_, Cond, _ > => yaktuba ; + <_, Fut, Pos> => glue "سَ" yaktubu ; + <_, Fut, Neg> => "لَنْ" ++ yaktuba + } ; -- in verbal sentences, the verb agrees with the subject -- in Gender but not in number @@ -1223,10 +1606,38 @@ patHollowImp : (_,_ :Str) -> Gender => Number => Str =\xaf,xAf -> _ => pgn }; - insertObj : NP -> VPSlash -> VP = \np,vp -> vp ** - { obj = {s = vp.obj.s ++ vp.c2 ++ np.s ! Acc ; a = np.a} }; +----------------------------------------------------------------------------- +-- Comp, arguments for VP - insertPred : {s : AAgr => Case => Str} -> VP -> VP = \p,vp -> vp ** + Comp : Type = { + s : AAgr => Case => Str ; + } ; + + Obj : Type = { + s : Str ; + a : Agr -- default Agr in a VP without real Obj is Per3 Masc Sg. + }; -- need isPron for word order in predVP, and pgn for ImpersCl + + Subj : Type = {s : Case => Str ; isPron : Bool} ; + + np2subj : NP -> Subj = \np -> np ** {isPron = np.a.isPron} ; + subj2np : Subj -> NP = \su -> su ** {a = {pgn = emptyNP.a.pgn ; isPron = su.isPron} ; empty=[]} ; + emptyObj : Obj = emptyNP ** {s=[]} ; + + insertObj : NP -> VPSlash -> VP = \np,vp -> vp ** { + obj = {s = vp.obj.s -- old object, if there was one + ++ bindIfPron np vp -- new object, bind if pronoun and not pred + ++ vp.agrObj ! np.a.pgn ; -- only used for SlashV2V + a = np.a} + } ; + + bindIfPron : NP -> {c2:Preposition; isPred:Bool} -> Str = \np,vp -> + let bind = case of { + => BIND ; + _ => [] } + in vp.c2.s ++ bind ++ np.s ! vp.c2.c ; + + insertPred : Comp -> VP -> VP = \p,vp -> vp ** { pred = p; isPred = True }; @@ -1235,16 +1646,69 @@ patHollowImp : (_,_ :Str) -> Gender => Number => Str =\xaf,xAf -> { s2 = vp.s2 ++ str }; kaan : {s : AAgr => Case => Str} -> VP = \xabar -> - insertPred xabar (predV (v1hollow {f = "ك"; c = "و" ; l = "ن"} u) ); + insertPred xabar (predV copula); - -- Slash categories - VPSlash : Type = VP ** {c2 : Str} ; - ClSlash : Type = Cl ** {c2 : Str} ; + copula : Verb = v1hollow {f = "ك"; c = "و" ; l = "ن"} u ; + +----------------------------------------------------------------------------- +-- Slash categories + + VPSlash : Type = VP ** {c2 : Preposition ; agrObj : PerGenNum => Str} ; + ClSlash : Type = VPSlash ** {subj : Subj} ; + + emptyVPslash : VP -> VPSlash = \vp -> vp ** { + c2 = noPrep ; agrObj = \\_ => [] + } ; + + slashV2 : Verb2 -> VPSlash = \v -> + predV v ** {c2 = v.c2 ; agrObj = \\_ => []} ; + + -- Add subject string, fix agreement to the subject, + -- but keep the structure as VP, because later on + -- we might need different word orders for the ClSlash. + predVPSlash : NP -> VPSlash -> ClSlash = \np,v -> v ** { + subj = np2subj np ; + s = \\_pgn,vf => v.s ! np.a.pgn ! vf -- so we can throw away subject's pgn + } ; + + complClSlash = overload { + complClSlash : NP -> ClSlash -> Cl = \obj,cls -> + predVP (subj2np cls.subj) (insertObj obj cls) ; + complClSlash : ClSlash -> Cl = \cls -> + predVP (subj2np cls.subj) (insertObj emptyNP cls) -- Empty subject and object + } ; Cl : Type = {s : Tense => Polarity => Order => Str} ; QCl : Type = {s : Tense => Polarity => QForm => Str} ; ---TODO: slashRCl : ClSlash -> RP -> RCl ; + forceOrder : Order -> Cl -> Cl = \o,cl -> + {s = \\t,p,_ => cl.s ! t ! p ! o} ; + +----------------------------------------------------------------------------- +-- Relative + + param + RAgr = RSg Gender | RPl Gender | RDl Gender Case ; + + oper + agr2ragr = overload { + agr2ragr : Agr -> Case -> RAgr = \a,c -> + let gn = pgn2gn a.pgn in case of { + => RSg x ; + => RDl x c ; + => RPl x } ; + agr2ragr : Number -> Case -> Gender -> RAgr = \n,c,g -> + case n of { + Sg => RSg g ; + Dl => RDl g c ; + Pl => RPl g } + } ; + + RCl : Type = {s : Tense => Polarity => Agr => Case => Str} ; + RP : Type = {s : RAgr => Str } ; + +----------------------------------------------------------------------------- +-- Num param @@ -1268,23 +1732,23 @@ patHollowImp : (_,_ :Str) -> Gender => Number => Str =\xaf,xAf -> \wAhid,awwal,Ula -> let wAhida : Str = case wAhid of { x + "ة" => mkAt wAhid ; - _ => wAhid + "َة" } + _ => wAhid + "َة" } in { s= table { unit => table { NCard => table { Masc => \\s,c => (sing wAhid) ! s ! c ; --all fem are first declension: - Fem => \\s,c => defArt s wAhida + dec1sg ! s ! c + Fem => \\s,c => defArt s c wAhida + dec1sgNoDoubleAlif ! s ! c }; NOrd => table { - Masc => \\s,c => defArt s awwal + dec1sg ! s ! c; + Masc => \\s,c => defArt s c awwal + dec1sg ! s ! c; Fem => \\s,c => (sing Ula) ! s ! c } }; ten => table { - NCard => \\_,s,c => defArt s wAhid + m_pl ! Indef ! c; - NOrd => \\_,s,c => defArt s awwal + m_pl ! Indef ! c + NCard => \\_,s,c => defArt s c wAhid + m_pl ! Indef ! c; + NOrd => \\_,s,c => defArt s c awwal + m_pl ! Indef ! c } } }; @@ -1333,6 +1797,4 @@ patHollowImp : (_,_ :Str) -> Gender => Number => Str =\xaf,xAf -> Masc => Fem; Fem => Masc }; - - } diff --git a/src/arabic/SentenceAra.gf b/src/arabic/SentenceAra.gf index 1a26da8c..854daaf4 100644 --- a/src/arabic/SentenceAra.gf +++ b/src/arabic/SentenceAra.gf @@ -26,67 +26,16 @@ concrete SentenceAra of Sentence = CatAra ** open } }; -} - PredVP np vp = - { s =\\t,p,o => - let { - pgn = - case of { - => verbalAgr np.a.pgn; - _ => np.a.pgn - }; - gn = pgn2gn pgn; - kataba = vp.s ! pgn ! VPPerf ; - yaktubu = vp.s ! pgn ! VPImpf Ind ; - yaktuba = vp.s ! pgn ! VPImpf Cnj ; - yaktub = vp.s ! pgn ! VPImpf Jus ; - vStr : ResAra.Tense -> Polarity -> Str = - \tn,pl -> case of { - => yaktubu ; - => "لَا" ++ yaktubu ; - => "" ; --no verb "to be" in present - => "لَيسَ" ;--same here, just add negation particle - <_, ResAra.Past, Pos> => kataba ; - <_, ResAra.Past, Neg> => "لَمْ" ++ yaktub ; - <_, ResAra.Fut, Pos> => "سَ" ++ yaktubu ; - <_, ResAra.Fut, Neg> => "لَنْ" ++ yaktuba - }; - pred : ResAra.Tense -> Polarity -> Str = - \tn,pl -> case of { - => vp.pred.s ! gn ! Nom; --xabar marfooc - _ => vp.pred.s ! gn ! Acc --xabar kaana wa laysa manSoob - }; - - } in - case o of { - Verbal => - --case of { ---- AR workaround 18/12/2008 - case of { - {- IL: I don't think we should do prodrop here. vStr drops the copula in present tense, - so there's hardly anything left for a predicative clause: e.g. - PredVP (UsePron i_Pron) (UseComp (CompCN (UseN car_N))) "I am a car" - would be linearised just as "car", if we have both prodrop and copula drop. - Leaving it up to someone who knows Arabic to decide what is better. - Original here: - => (vStr t p) ++ vp.obj.s ++ vp.s2 ++ (pred t p) ; - -- ya2kuluhu - => (vStr t p) ++ vp.obj.s ++ vp.s2 ++ (pred t p); -} - -- ya2kuluhu al-waladu, yakuluhu al-2awlaadu - => (vStr t p) ++ np.s ! Nom ++ vp.obj.s ++ vp.s2 ++ (pred t p); - => (vStr t p) ++ vp.obj.s ++ np.s ! Nom ++ vp.s2 ++ (pred t p) - }; - Nominal => - np.s ! Nom ++ (vStr t p) ++ vp.obj.s ++ vp.s2 ++ (pred t p) - } - }; + PredVP = predVP ; -- PredSCVP sc vp = mkClause sc.s (agrP3 Sg) vp ; ImpVP vp = { s = \\p,g,n => case p of { - Pos => vp.s ! (Per2 g n) ! VPImp ++ vp.obj.s ++ vp.s2 ; - Neg => "لا" ++ vp.s ! (Per2 g n) ! (VPImpf Jus) ++ vp.obj.s ++ vp.s2 - } + Pos => vp.s ! Per2 g n ! VPImp ; + Neg => "لَا" ++ vp.s ! Per2 g n ! VPImpf Jus + } ++ vp.obj.s ++ vp.pred.s ! {g=g;n=n} ! Acc ++ vp.s2 }; -- @@ -100,9 +49,11 @@ concrete SentenceAra of Sentence = CatAra ** open -- ClSlash - SlashVP np vps = PredVP np vps ** { c2 = vps.c2 } ; + SlashVP = predVPSlash ; AdvSlash slash adv = slash ** { s2 = slash.s2 ++ adv.s } ; - SlashPrep cl prep = cl ** {c2 = prep.s} ; + +-- : Cl -> Prep -> ClSlash +-- SlashPrep cl prep = TODO -- SlashVS np vs sslash = TODO @@ -113,22 +64,26 @@ concrete SentenceAra of Sentence = CatAra ** open -- UseCl t p cl = - {s = case of { --- IL guessed tenses - <(Pres|Cond),Simul> => cl.s ! ResAra.Pres ! p.p ! Verbal ; - => cl.s ! ResAra.Fut ! p.p ! Verbal ; - <_ ,_ > => cl.s ! ResAra.Past ! p.p ! Verbal + {s = \\o => t.s ++ p.s ++ + case of { --- IL guessed tenses + => cl.s ! Pres ! p.p ! o ; + => cl.s ! Past ! p.p ! o ; + => cl.s ! x ! p.p ! o } }; UseQCl t p qcl = - {s = \\q => - case of { --- IL guessed tenses - <(Pres|Cond),Simul> => qcl.s ! ResAra.Pres ! p.p ! q ; - => qcl.s ! ResAra.Fut ! p.p ! q ; - <_ ,_ > => qcl.s ! ResAra.Past ! p.p ! q + {s = \\q => t.s ++ p.s ++ + case of { --- IL guessed tenses + => qcl.s ! Pres ! p.p ! q ; + => qcl.s ! Past ! p.p ! q ; + => qcl.s ! x ! p.p ! q } }; --- UseRCl t a p cl = {s = \\r => t.s ++ a.s ++ p.s ++ cl.s ! t.t ! a.a ! p.p ! r} ; + UseRCl t p cl = {s = \\agr,c => t.s ++ p.s ++ cl.s ! t.t ! p.p ! agr ! c} ; + UseSlash t p cl = UseCl t p (complClSlash cl) ; + + AdvS adv s = s ** {s = \\o => adv.s ++ s.s ! o} ; } diff --git a/src/arabic/StructuralAra.gf b/src/arabic/StructuralAra.gf index 7931b813..c6c8aaaf 100644 --- a/src/arabic/StructuralAra.gf +++ b/src/arabic/StructuralAra.gf @@ -4,8 +4,8 @@ concrete StructuralAra of Structural = CatAra ** flags optimize=all ; coding=utf8 ; lin - above_Prep = ss "فَوْقَ" ; - after_Prep = ss "بَعْدَ" ; + above_Prep = mkPrep "فَوْقَ" ; + after_Prep = mkPrep "بَعْدَ" ; all_Predet = mkPredet "كُلّ" True ; almost_AdA = ss "تَقْرِيباً"; almost_AdN = ss "حَوَالي" ; -- or "تَقرِيبا" @@ -13,20 +13,20 @@ concrete StructuralAra of Structural = CatAra ** -- always_AdV = ss "َلوَيس" ; and_Conj = ss "وَ" ** {n = Pl} ; -- because_Subj = ss "بعَُسي" ; - before_Prep = ss "قَبْلَ" ; - behind_Prep = ss "خَلْفَ" ; - between_Prep = ss "بَيْنَ" ; + before_Prep = mkPrep "قَبْلَ" ; + behind_Prep = mkPrep "خَلْفَ" ; + between_Prep = mkPrep "بَيْنَ" ; -- both7and_DConj = sd2 "بْته" "َند" ** {n = Pl} ; -- but_PConj = ss "بُت" ; - by8agent_Prep = ss "بِ" ; - by8means_Prep = ss "بِ" ; + by8agent_Prep, + by8means_Prep = biPrep ; can_VV = mkVV (mkV "طوع" FormX) ; -- can8know_VV = { -- s = table VVForm [["بي َبلي تْ"] ; "عَن" ; "عُْلد" ; -- ["بّن َبلي تْ"] ; ["بِنغ َبلي تْ"] ; "عَنءت" ; "عُْلدنءت"] ; -- isAux = True -- } ; - during_Prep = ss "خِلَالَ" ; + during_Prep = mkPrep "خِلَالَ" ; -- either7or_DConj = sd2 "ِتهر" "ْر" ** {n = Sg} ; everybody_NP = regNP "الجَمِيع" Pl ; every_Det = mkDet "كُلّ" Sg Const ; @@ -34,18 +34,23 @@ concrete StructuralAra of Structural = CatAra ** -- everywhere_Adv = ss "ثريوهري" ; few_Det = mkDet "بَعض" Pl Const ; -- first_Ord = ss "فِرست" ; - from_Prep = ss "مِنَ" ; - he_Pron = mkPron "هُوَ" "هُ" "هُ" (Per3 Masc Sg) ; + for_Prep = liPrep ; + from_Prep = mkPrep "مِنَ" ; + he_Pron = ResAra.he_Pron ; here_Adv = ss "هُنا" ; -- here7to_Adv = ss ["تْ هري"] ; -- here7from_Adv = ss ["فرْم هري"] ; how_IAdv = ss "كَيفَ" ; --- how8many_IDet = mkDet "كَمْ" Pl Const ; -- IL: check (was ["هْو مَني"]) ; --- if_Subj = ss "ِف" ; - in8front_Prep = ss "مُقَابِلَ" ; - i_Pron = mkPron "أَنَا" "نِي" "ي" (Per1 Sing); - in_Prep = ss "فِي" ; - it_Pron = he_Pron ; -- was: it_Pron = mkPron "ِت" "ِت" "ِتس" (Per3 Masc Sg); + how8many_IDet = { + s = \\g,s,c => "كَمْ عَدَد" + caseTbl ! c ; + n = Pl ; d = Def + } ; -- IL + + if_Subj = mkSubj "إِذَا" Verbal ; + in8front_Prep = mkPrep "مُقَابِلَ" ; + i_Pron = ResAra.i_Pron ; + in_Prep = mkPrep "فِي" ; + it_Pron = emptyNP ** {s = \\_ => "هَذَا"} ; -- was: it_Pron = mkPron "ِت" "ِت" "ِتس" (Per3 Masc Sg); -- less_CAdv = ss "لسّ" ; many_Det = mkDet "جَمِيع" Pl Const ; -- more_CAdv = ss "مْري" ; @@ -57,16 +62,15 @@ concrete StructuralAra of Structural = CatAra ** -- isAux = True -- } ; no_Utt = {s = \\_ => "لا"} ; - on_Prep = ss "عَلى" ; ---- DEPREC one_Quant = mkQuantNum "واحِد" Sg Indef ; + on_Prep = mkPrep "عَلَى" ; only_Predet = mkPredet "فَقَط" False; -- or_Conj = ss "ْر" ** {n = Sg} ; -- otherwise_PConj = ss "ْتهروِسي" ; - part_Prep = ss "مِنَ" ; + part_Prep = mkPrep "مِنَ" ; -- please_Voc = ss "ةلَسي" ; - possess_Prep = ss "ل" ; + possess_Prep = liPrep ; -- quite_Adv = ss "قُِتي" ; - she_Pron = mkPron "هِيَ" "ها" "ها" (Per3 Fem Sg) ; + she_Pron = ResAra.she_Pron ; -- so_AdA = ss "سْ" ; somebody_NP = regNP "أَحَد" Sg ; someSg_Det = mkDet "أَحَد" Sg Const ; @@ -74,48 +78,49 @@ concrete StructuralAra of Structural = CatAra ** something_NP = regNP "شَيْء" Sg ; -- somewhere_Adv = ss "سْموهري" ; that_Quant = mkQuant3 "ذَلِكَ" "تِلكَ" "أُلٱِكَ" Def; + that_Subj = mkSubj "أنَّ" ; ----b that_NP = indeclNP "ذَلِكَ" Sg ; there_Adv = ss "هُناك" ; -- there7to_Adv = ss "تهري" ; -- there7from_Adv = ss ["فرْم تهري"] ; -- therefore_PConj = ss "تهرفْري" ; ----b these_NP = indeclNP "هَؤُلَاء" Pl ; - they_Pron = mkPron "هُمْ" "هُمْ" "هُمْ" (Per3 Masc Pl) ; + they_Pron = theyMasc_Pron ; this_Quant = mkQuant7 "هَذا" "هَذِهِ" "هَذَان" "هَذَيْن" "هَاتَان" "هَاتَيْن" "هَؤُلَاء" Def; ----b this_NP = indeclNP "هَذا" Sg ; ----b those_NP = indeclNP "هَؤُلَاءكَ" Pl ; - through_Prep = ss "عَبْرَ" ; + through_Prep = mkPrep "عَبْرَ" ; -- too_AdA = ss "تّْ" ; - to_Prep = ss "إِلى" ; - under_Prep = ss "تَحْتَ" ; + to_Prep = mkPrep "إِلى" ; + under_Prep = mkPrep "تَحْتَ" ; -- very_AdA = ss "ثري" ; --- want_VV = P.mkVV (P.regV "وَنت") ; - we_Pron = mkPron "نَحنُ" "نا" "نا" (Per1 Plur) ; - whatPl_IP = mkIP "ماذا" Pl ; - whatSg_IP = mkIP "ماذا" Sg ; + want_VV = mkVV (mkV "رود" FormIV) ; + we_Pron = ResAra.we_Pron ; + whatPl_IP = mkIP "ما" "ماذا" Pl ; + whatSg_IP = mkIP "ما" "ماذا" Sg ; when_IAdv = ss "مَتَى" ; --- when_Subj = ss "وهن" ; + when_Subj = mkSubj "عِنْدَمَا" Verbal ; where_IAdv = ss "أَينَ" ; which_IQuant = { s = \\s,c => case of { => "أيّ" ; - => "أيٌّ" ; - => "أيُّ" ; + => "أيٌّ" ; + => "أيُّ" ; => "أيّاً" ; - => "أيَّ" ; - => "أيٍّ" ; - => "أيِّ" + => "أيَّ" ; + <_Gen,Indef> => "أيٍّ" ; + <_Gen,_> => "أيِّ" } } ; - whoSg_IP = mkIP "مَنْ" Sg ; - whoPl_IP = mkIP "مَنْ" Pl ; + whoSg_IP = mkIP "مَنْ" "مَنْ" Sg ; + whoPl_IP = mkIP "مَنْ" "مَنْ" Pl ; -- why_IAdv = ss "وهي" ; - without_Prep = ss "بِدُونِ" ; - with_Prep = ss "مَع" ; + without_Prep = mkPrep "بِدُونِ" ; + with_Prep = mkPrep "مَع" ; yes_Utt = {s = \\_ => "نَعَم"} ; - youSg_Pron = mkPron "أَنتَ" "كَ" "كَ" (Per2 Masc Sg) ; - youPl_Pron = mkPron "أَنتُمْ" "كُمْ" "كُمْ" (Per2 Masc Sg) ; - youPol_Pron = mkPron "أَنتِ" "كِ" "كِ" (Per2 Fem Sg) ; + youSg_Pron = youSgMasc_Pron ; + youPl_Pron = youPlMasc_Pron ; + youPol_Pron = youPlFem_Pron ; -- arbitrary? have_V2 = dirV2 (regV "يَملِك") ; diff --git a/src/arabic/SymbolAra.gf b/src/arabic/SymbolAra.gf index c239b087..10e2b7bb 100644 --- a/src/arabic/SymbolAra.gf +++ b/src/arabic/SymbolAra.gf @@ -9,21 +9,23 @@ lin FloatPN i = {s = \\c => i.s ; g = Masc ; h = NoHum } ; --IL NumPN i = {s = \\c => uttNum i ! Masc ; g = Masc ; h = NoHum } ; --IL -- CNIntNP cn i = { - -- s = \\c => cn.s ! Sg ! Def ! c ++ uttNum i ! Masc ; + -- s = \\c => cn2str cn Sg Def c ++ uttNum i ! cn.g ; -- a = dummyAgrP3 Sg ; -- } ; --IL TODO: check out some opers regarding state in ResAra. These are just dummy values. CNSymbNP det cn xs = let g = cn.g ; n = sizeToNumber det.n in { - s = \\c => det.s ! NoHum ! g ! c ++ cn.s ! Sg ! Def ! c ++ cn.adj ! n ! Def ! c ++ xs.s; ----IL word order?? Seems to be nontrivial according to ResAra comments. + s = \\c => det.s ! NoHum ! g ! c ++ cn2str cn n Def c ++ xs.s; ----IL word order?? Seems to be nontrivial according to ResAra comments. a = dummyAgrP3 n ; + empty = [] } ; CNNumNP cn i = { - s = \\c => cn.s ! Sg ! Def ! c ++ uttNum i ! Masc ; + s = \\c => cn2str cn Sg Def c ++ uttNum i ! cn.g ; a = dummyAgrP3 Sg ; + empty = [] } ; - SymbS sy = sy ; + SymbS sy = {s = \\_ => sy.s} ; SymbOrd n = {s = \\_,_,_ => n.s ; n = One ; isNum = False } ; diff --git a/src/arabic/VerbAra.gf b/src/arabic/VerbAra.gf index 23878f9b..abfdc218 100644 --- a/src/arabic/VerbAra.gf +++ b/src/arabic/VerbAra.gf @@ -1,4 +1,4 @@ -concrete VerbAra of Verb = CatAra ** open Prelude, ResAra in { +concrete VerbAra of Verb = CatAra ** open Prelude, ResAra, ParamX in { flags optimize=all_subs ; @@ -6,47 +6,96 @@ concrete VerbAra of Verb = CatAra ** open Prelude, ResAra in { UseV = predV ; SlashVV vv vps = vps ** predV vv ; ----IL - SlashV2a v = predVSlash v ; - Slash3V3 v np = insertObj np (predVSlash v) ** {c2 = v.c3}; + + -- : V2V -> VP -> VPSlash ; -- beg (her) to go + SlashV2V v2v vp = let v2vVP = predV v2v in -- IL + vp ** { + s = v2vVP.s ; + agrObj = \\pgn => v2v.c3.s -- أَنْ + ++ vp.s ! pgn ! VPImpf Cnj ; + isPred = False ; + c2 = v2v.c2 ; -- preposition for the direct object + sc = v2v.sc + } ; + + -- : V2V -> NP -> VPSlash -> VPSlash ; -- beg me to buy + SlashV2VNP v2v np vps = let v2vVP = slashV2 v2v in -- IL + vps ** { + s = \\pgn,vpf => v2vVP.s ! pgn ! vpf -- main verb agrees with subject + ++ bindIfPron np v2vVP + ++ v2v.c3.s -- أَنْ + ++ vps.s ! np.a.pgn ! VPImpf Cnj -- verb from old VP agrees with object + ++ vps.obj.s ; -- otherwise obj appears in a weird place /IL + obj = emptyObj ; + isPred = False ; + -- preposition for the direct object comes from VP + sc = v2v.sc + } ; + + SlashV2a = slashV2 ; + Slash3V3 v np = insertObj np (slashV2 v) ** {c2 = v.c3 ; agrObj = \\_ => []}; ComplSlash vp np = insertObj np vp ; --- Complv3 v np np2 = insertObj np2 (insertObj np (predV v)) ; - -{-{s = \\_ => v.c2 ++ np.s ! Acc ++ v.c3 ++ np2.s ! Acc ; - a = {pgn = Per3 Masc Sg ; isPron = False} } --FIXME - (predV v) ;-} - - ComplVV vv vp = let vvVP = predV vv in --- IL + -- : VV -> VP -> VP ; -- want to run + ComplVV vv vp = let vvVP = predV vv in -- IL vp ** { s = \\pgn,vpf => vvVP.s ! pgn ! vpf - ++ vv.c2 -- أَنْ - ++ vp.s ! pgn ! VPImpf Cnj + ++ vv.c2.s -- أَنْ + ++ vp.s ! pgn ! VPImpf Cnj ; + isPred = False ; + sc = vv.sc } ; --- ComplVS v s = insertObj (\\_ => conjThat ++ s.s) (predV v) ; --- ComplVQ v q = insertObj (\\_ => q.s ! QIndir) (predV v) ; --- --- ComplVA v ap = insertObj (ap.s) (predV v) ; + -- : VS -> S -> VP ; -- say that she runs + ComplVS vs s = predV vs ** { -- IL + obj = emptyObj ** {s = s.s ! Subord} + } ; + + -- : VQ -> QS -> VP ; -- wonder who runs + ComplVQ vq qs = predV vq ** { -- IL + obj = emptyObj ** {s = qs.s ! QIndir} + } ; + + -- : VA -> AP -> VP ; -- they become red + ComplVA v ap = predV v ** {pred = CompAP ap} ; + -- ComplV2A v np ap = -- insertObj (\\_ => v.c2 ++ np.s ! Acc ++ ap.s ! np.a) (predV v) ; -- - UseComp xabar = kaan xabar ; + UseComp xabar = + case xabar.isNP of { + False => kaan xabar ; + True => predV copula ** {obj = xabar.obj ; isPred=True} + } ; + + UseCopula = predV copula ; + + -- : VP -> Prep -> VPSlash ; -- live in (it) + VPSlashPrep vp prep = vp ** { + c2 = prep ; + agrObj = \\_ => [] + } ; AdvVP vp adv = insertStr adv.s vp ; --- AdVVP adv vp = insertAdV adv.s vp ; + AdVVP adv = insertStr adv.s ; + AdVVPSlash adv vps = vps ** insertStr adv.s vps ; -- -- ReflV2 v = insertObj (\\a => v.c2 ++ reflPron ! a) (predV v) ; -- - PassV2 v = kaan {s = \\_,_ => v.s ! VPPart} ; ---- IL guessed + PassV2 = passPredV ; -- -- UseVS, UseVQ = \vv -> {s = vv.s ; c2 = [] ; isRefl = vv.isRefl} ; -- no - CompCN cn = {s = \\agr,c => cn.s ! agr.n ! Indef ! c ++ cn.np ! c ++ cn.adj ! agr.n ! Indef ! c} ; ----IL - CompAP ap = {s = \\agr,c => ap.s ! Hum ! agr.g ! agr.n ! Indef ! c} ; --FIXME - CompNP np = {s = \\_,c => np.s ! c}; - CompAdv a = {s = \\_,_ => a.s} ; + CompAP ap = {s = \\agr,c => ap.s ! Hum ! agr.g ! agr.n ! Indef ! c ; --FIXME + obj = emptyObj ; isNP = False} ; + CompAdv a = {s = \\_,_ => a.s ; + obj = emptyObj ; isNP = False} ; + + CompCN cn = {s = \\agr,c => cn2str cn agr.n Indef Nom ; + obj = emptyObj ; isNP = False} ; + CompNP np = {s = \\_,_ => [] ; obj = np ** {s = np.s ! Nom} ; isNP = True} ; -- -- } diff --git a/src/english/ConstructionEng.gf b/src/english/ConstructionEng.gf index d0e6b591..a12cc46c 100644 --- a/src/english/ConstructionEng.gf +++ b/src/english/ConstructionEng.gf @@ -74,6 +74,8 @@ lincat n_hours_NP : NP = mkNP n_card time ; in Sy.mkAdv for_Prep n_hours_NP | mkAdv (n_hours_NP.s ! R.npNom) ; + timeunitRange l u time = {s = l.s ! True ! R.Nom ++ to_Prep.s ++ u.s ! True ! R.Nom ++ time.s ! R.Pl ! R.Nom} ; + oneHour = mkHour "1" True ; twoHour = mkHour "2" True ; threeHour = mkHour "3" True ; diff --git a/src/portuguese/BeschPor.gf b/src/portuguese/BeschPor.gf index cf3c5ea8..e8607591 100644 --- a/src/portuguese/BeschPor.gf +++ b/src/portuguese/BeschPor.gf @@ -71,7 +71,11 @@ oper -- or in =Ryan, M. A. Conjugação dos Verbos em Português. ática, -- 1991.= + -- the numbers in the comments below the oper declaration are the + -- verb's paradigm numbers in bescherelle and in Ryan + oper ter_Besch : Str -> Verbum = \ter -> +-- 1 | r3 let x_ = Predef.tk 3 ter in {s = table { VI Infn => x_ + "ter" ; @@ -141,6 +145,7 @@ oper ter_Besch : Str -> Verbum = \ter -> } ; oper haver_Besch : Str -> Verbum = \haver -> +-- 2 | r4 let x_ = Predef.tk 5 haver in {s = table { VI Infn => x_ + "haver" ; @@ -210,6 +215,7 @@ oper haver_Besch : Str -> Verbum = \haver -> } ; oper ser_Besch : Str -> Verbum = \ser -> +-- 3 | r1 let x_ = Predef.tk 3 ser in {s = table { VI Infn => x_ + "ser" ; @@ -279,6 +285,7 @@ oper ser_Besch : Str -> Verbum = \ser -> } ; oper comprar_Besch : Str -> Verbum = \comprar -> +-- 4 | r5 let compr_ = Predef.tk 2 comprar in {s = table { VI Infn => compr_ + "ar" ; @@ -348,6 +355,7 @@ oper comprar_Besch : Str -> Verbum = \comprar -> } ; oper vender_Besch : Str -> Verbum = \vender -> +-- 5 | r6 let vend_ = Predef.tk 2 vender in {s = table { VI Infn => vend_ + "er" ; @@ -417,6 +425,7 @@ oper vender_Besch : Str -> Verbum = \vender -> } ; oper garantir_Besch : Str -> Verbum = \garantir -> +-- 6 | r7 let garant_ = Predef.tk 2 garantir in {s = table { VI Infn => garant_ + "ir" ; @@ -486,6 +495,7 @@ oper garantir_Besch : Str -> Verbum = \garantir -> } ; oper estar_Besch : Str -> Verbum = \estar -> +-- 10 | r2 let est_ = Predef.tk 2 estar in {s = table { VI Infn => est_ + "ar" ; @@ -555,6 +565,7 @@ oper estar_Besch : Str -> Verbum = \estar -> } ; oper dar_Besch : Str -> Verbum = \dar -> +-- 11 | r59 let x_ = Predef.tk 3 dar in {s = table { VI Infn => x_ + "dar" ; @@ -624,6 +635,7 @@ oper dar_Besch : Str -> Verbum = \dar -> } ; oper ficar_Besch : Str -> Verbum = \ficar -> +-- 12 | r23 let x_ = Predef.tk 5 ficar in {s = table { VI Infn => x_ + "ficar" ; @@ -693,75 +705,77 @@ oper ficar_Besch : Str -> Verbum = \ficar -> } ; oper começar_Besch : Str -> Verbum = \começar -> - let x_ = Predef.tk 7 começar in +-- 13 | r24 + let come_ = Predef.tk 3 começar in {s = table { - VI Infn => x_ + "começar" ; - VI Ger => x_ + "começando" ; - VI Part => x_ + "começado" ; - VPB (Pres Ind Sg P1) => x_ + "começo" ; - VPB (Pres Ind Sg P2) => x_ + "começas" ; - VPB (Pres Ind Sg P3) => x_ + "começa" ; - VPB (Pres Ind Pl P1) => x_ + "começamos" ; - VPB (Pres Ind Pl P2) => x_ + "começais" ; - VPB (Pres Ind Pl P3) => x_ + "começam" ; - VPB (Pres Sub Sg P1) => x_ + "comece" ; - VPB (Pres Sub Sg P2) => x_ + "comeces" ; - VPB (Pres Sub Sg P3) => x_ + "comece" ; - VPB (Pres Sub Pl P1) => x_ + "comecemos" ; - VPB (Pres Sub Pl P2) => x_ + "comeceis" ; - VPB (Pres Sub Pl P3) => x_ + "comecem" ; - VPB (PretI Ind Sg P1) => x_ + "começava" ; --# notpresent - VPB (PretI Ind Sg P2) => x_ + "começavas" ; --# notpresent - VPB (PretI Ind Sg P3) => x_ + "começava" ; --# notpresent - VPB (PretI Ind Pl P1) => x_ + "começávamos" ; --# notpresent - VPB (PretI Ind Pl P2) => x_ + "começáveis" ; --# notpresent - VPB (PretI Ind Pl P3) => x_ + "começavam" ; --# notpresent - VPB (PretI Sub Sg P1) => x_ + "começasse" ; --# notpresent - VPB (PretI Sub Sg P2) => x_ + "começasses" ; --# notpresent - VPB (PretI Sub Sg P3) => x_ + "começasse" ; --# notpresent - VPB (PretI Sub Pl P1) => x_ + "começássemos" ; --# notpresent - VPB (PretI Sub Pl P2) => x_ + "começasseis" ; --# notpresent - VPB (PretI Sub Pl P3) => x_ + "começassem" ; --# notpresent - VPB (MQPerf Sg P1) => x_ + "começara" ; --# notpresent - VPB (MQPerf Sg P2) => x_ + "começaras" ; --# notpresent - VPB (MQPerf Sg P3) => x_ + "começara" ; --# notpresent - VPB (MQPerf Pl P1) => x_ + "começáramos" ; --# notpresent - VPB (MQPerf Pl P2) => x_ + "começáreis" ; --# notpresent - VPB (MQPerf Pl P3) => x_ + "começaram" ; --# notpresent - VPB (PretP Sg P1) => x_ + "comecei" ; --# notpresent - VPB (PretP Sg P2) => x_ + "começaste" ; --# notpresent - VPB (PretP Sg P3) => x_ + "começou" ; --# notpresent - VPB (PretP Pl P1) => x_ + vars "começamos" "começámos" ; --# notpresent - VPB (PretP Pl P2) => x_ + "começastes" ; --# notpresent - VPB (PretP Pl P3) => x_ + "começaram" ; --# notpresent - VPB (Fut Ind Sg P1) => x_ + "começarei" ; --# notpresent - VPB (Fut Ind Sg P2) => x_ + "começarás" ; --# notpresent - VPB (Fut Ind Sg P3) => x_ + "começará" ; --# notpresent - VPB (Fut Ind Pl P1) => x_ + "começaremos" ; --# notpresent - VPB (Fut Ind Pl P2) => x_ + "começareis" ; --# notpresent - VPB (Fut Ind Pl P3) => x_ + "começarão" ; --# notpresent - VPB (Fut Sub Sg P1) => x_ + "começar" ; --# notpresent - VPB (Fut Sub Sg P2) => x_ + "começares" ; --# notpresent - VPB (Fut Sub Sg P3) => x_ + "começar" ; --# notpresent - VPB (Fut Sub Pl P1) => x_ + "começarmos" ; --# notpresent - VPB (Fut Sub Pl P2) => x_ + "começardes" ; --# notpresent - VPB (Fut Sub Pl P3) => x_ + "começarem" ; --# notpresent - VPB (Cond Sg P1) => x_ + "começaria" ; --# notpresent - VPB (Cond Sg P2) => x_ + "começarias" ; --# notpresent - VPB (Cond Sg P3) => x_ + "começaria" ; --# notpresent - VPB (Cond Pl P1) => x_ + "começaríamos" ; --# notpresent - VPB (Cond Pl P2) => x_ + "começarieis" ; --# notpresent - VPB (Cond Pl P3) => x_ + "começariam" ; --# notpresent - VPB (Imper Sg P2) => x_ + "começa" ; - VPB (Imper Sg P3) => x_ + "comece" ; - VPB (Imper Pl P1) => x_ + "comecemos" ; - VPB (Imper Pl P2) => x_ + "começai" ; - VPB (Imper Pl P3) => x_ + "comecem" ; + VI Infn => come_ + "çar" ; + VI Ger => come_ + "çando" ; + VI Part => come_ + "çado" ; + VPB (Pres Ind Sg P1) => come_ + "ço" ; + VPB (Pres Ind Sg P2) => come_ + "ças" ; + VPB (Pres Ind Sg P3) => come_ + "ça" ; + VPB (Pres Ind Pl P1) => come_ + "çamos" ; + VPB (Pres Ind Pl P2) => come_ + "çais" ; + VPB (Pres Ind Pl P3) => come_ + "çam" ; + VPB (Pres Sub Sg P1) => come_ + "ce" ; + VPB (Pres Sub Sg P2) => come_ + "ces" ; + VPB (Pres Sub Sg P3) => come_ + "ce" ; + VPB (Pres Sub Pl P1) => come_ + "cemos" ; + VPB (Pres Sub Pl P2) => come_ + "ceis" ; + VPB (Pres Sub Pl P3) => come_ + "cem" ; + VPB (PretI Ind Sg P1) => come_ + "çava" ; --# notpresent + VPB (PretI Ind Sg P2) => come_ + "çavas" ; --# notpresent + VPB (PretI Ind Sg P3) => come_ + "çava" ; --# notpresent + VPB (PretI Ind Pl P1) => come_ + "çávamos" ; --# notpresent + VPB (PretI Ind Pl P2) => come_ + "çáveis" ; --# notpresent + VPB (PretI Ind Pl P3) => come_ + "çavam" ; --# notpresent + VPB (PretI Sub Sg P1) => come_ + "çasse" ; --# notpresent + VPB (PretI Sub Sg P2) => come_ + "çasses" ; --# notpresent + VPB (PretI Sub Sg P3) => come_ + "çasse" ; --# notpresent + VPB (PretI Sub Pl P1) => come_ + "çássemos" ; --# notpresent + VPB (PretI Sub Pl P2) => come_ + "çasseis" ; --# notpresent + VPB (PretI Sub Pl P3) => come_ + "çassem" ; --# notpresent + VPB (MQPerf Sg P1) => come_ + "çara" ; --# notpresent + VPB (MQPerf Sg P2) => come_ + "çaras" ; --# notpresent + VPB (MQPerf Sg P3) => come_ + "çara" ; --# notpresent + VPB (MQPerf Pl P1) => come_ + "çáramos" ; --# notpresent + VPB (MQPerf Pl P2) => come_ + "çáreis" ; --# notpresent + VPB (MQPerf Pl P3) => come_ + "çaram" ; --# notpresent + VPB (PretP Sg P1) => come_ + "cei" ; --# notpresent + VPB (PretP Sg P2) => come_ + "çaste" ; --# notpresent + VPB (PretP Sg P3) => come_ + "çou" ; --# notpresent + VPB (PretP Pl P1) => come_ + vars "çamos" "çámos" ; --# notpresent + VPB (PretP Pl P2) => come_ + "çastes" ; --# notpresent + VPB (PretP Pl P3) => come_ + "çaram" ; --# notpresent + VPB (Fut Ind Sg P1) => come_ + "çarei" ; --# notpresent + VPB (Fut Ind Sg P2) => come_ + "çarás" ; --# notpresent + VPB (Fut Ind Sg P3) => come_ + "çará" ; --# notpresent + VPB (Fut Ind Pl P1) => come_ + "çaremos" ; --# notpresent + VPB (Fut Ind Pl P2) => come_ + "çareis" ; --# notpresent + VPB (Fut Ind Pl P3) => come_ + "çarão" ; --# notpresent + VPB (Fut Sub Sg P1) => come_ + "çar" ; --# notpresent + VPB (Fut Sub Sg P2) => come_ + "çares" ; --# notpresent + VPB (Fut Sub Sg P3) => come_ + "çar" ; --# notpresent + VPB (Fut Sub Pl P1) => come_ + "çarmos" ; --# notpresent + VPB (Fut Sub Pl P2) => come_ + "çardes" ; --# notpresent + VPB (Fut Sub Pl P3) => come_ + "çarem" ; --# notpresent + VPB (Cond Sg P1) => come_ + "çaria" ; --# notpresent + VPB (Cond Sg P2) => come_ + "çarias" ; --# notpresent + VPB (Cond Sg P3) => come_ + "çaria" ; --# notpresent + VPB (Cond Pl P1) => come_ + "çaríamos" ; --# notpresent + VPB (Cond Pl P2) => come_ + "çarieis" ; --# notpresent + VPB (Cond Pl P3) => come_ + "çariam" ; --# notpresent + VPB (Imper Sg P2) => come_ + "ça" ; + VPB (Imper Sg P3) => come_ + "ce" ; + VPB (Imper Pl P1) => come_ + "cemos" ; + VPB (Imper Pl P2) => come_ + "çai" ; + VPB (Imper Pl P3) => come_ + "cem" ; VPB (Imper Sg P1) => nonExist } } ; oper chegar_Besch : Str -> Verbum = \chegar -> +-- 14 | r26 let cheg_ = Predef.tk 2 chegar in {s = table { VI Infn => cheg_ + "ar" ; @@ -831,6 +845,7 @@ oper chegar_Besch : Str -> Verbum = \chegar -> } ; oper recear_Besch : Str -> Verbum = \recear -> +-- 15 | r46 let rec_ = Predef.tk 3 recear in {s = table { VI Infn => rec_ + "ear" ; @@ -900,6 +915,7 @@ oper recear_Besch : Str -> Verbum = \recear -> } ; oper anunciar_Besch : Str -> Verbum = \anunciar -> +-- 16 | r46 let anunci_ = Predef.tk 2 anunciar in {s = table { VI Infn => anunci_ + "ar" ; @@ -969,6 +985,7 @@ oper anunciar_Besch : Str -> Verbum = \anunciar -> } ; oper odiar_Besch : Str -> Verbum = \odiar -> +-- 17 | r46 let od_ = Predef.tk 3 odiar in {s = table { VI Infn => od_ + "iar" ; @@ -1038,6 +1055,7 @@ oper odiar_Besch : Str -> Verbum = \odiar -> } ; oper comerciar_Besch : Str -> Verbum = \comerciar -> +-- 18 | r36 let comerc_ = Predef.tk 3 comerciar in {s = table { VI Infn => comerc_ + "iar" ; @@ -1107,6 +1125,7 @@ oper comerciar_Besch : Str -> Verbum = \comerciar -> } ; oper saudar_Besch : Str -> Verbum = \saudar -> +-- 19 | r16 let sa_ = Predef.tk 4 saudar in {s = table { VI Infn => sa_ + "udar" ; @@ -1176,6 +1195,7 @@ oper saudar_Besch : Str -> Verbum = \saudar -> } ; oper perdoar_Besch : Str -> Verbum = \perdoar -> +-- 20 | r38 let perd_ = Predef.tk 3 perdoar in {s = table { VI Infn => perd_ + "oar" ; @@ -1245,6 +1265,7 @@ oper perdoar_Besch : Str -> Verbum = \perdoar -> } ; oper averiguar_Besch : Str -> Verbum = \averiguar -> +-- 21 | r30 let averigu_ = Predef.tk 2 averiguar in {s = table { VI Infn => averigu_ + "ar" ; @@ -1314,6 +1335,7 @@ oper averiguar_Besch : Str -> Verbum = \averiguar -> } ; oper neviscar_Besch : Str -> Verbum = \neviscar -> +-- 22 | r23 let n_ = Predef.tk 7 neviscar in {s = table { VI Infn => n_ + "eviscar" ; @@ -1383,6 +1405,7 @@ oper neviscar_Besch : Str -> Verbum = \neviscar -> } ; oper adequar_Besch : Str -> Verbum = \adequar -> +-- 23 | r82 let adequ_ = Predef.tk 2 adequar in {s = table { VI Infn => adequ_ + "ar" ; @@ -1452,6 +1475,7 @@ oper adequar_Besch : Str -> Verbum = \adequar -> } ; oper relampaguear_Besch : Str -> Verbum = \relampaguear -> +-- 24 let relamp_ = Predef.tk 6 relampaguear in {s = table { VI Infn => relamp_ + "aguear" ; @@ -1521,6 +1545,7 @@ oper relampaguear_Besch : Str -> Verbum = \relampaguear -> } ; oper aquecer_Besch : Str -> Verbum = \aquecer -> +-- 25 | r25 let aque_ = Predef.tk 3 aquecer in {s = table { VI Infn => aque_ + "cer" ; @@ -1590,6 +1615,7 @@ oper aquecer_Besch : Str -> Verbum = \aquecer -> } ; oper proteger_Besch : Str -> Verbum = \proteger -> +-- 26 | r27 let prote_ = Predef.tk 3 proteger in {s = table { VI Infn => prote_ + "ger" ; @@ -1659,6 +1685,7 @@ oper proteger_Besch : Str -> Verbum = \proteger -> } ; oper erguer_Besch : Str -> Verbum = \erguer -> +-- 27 | r32 let erg_ = Predef.tk 3 erguer in {s = table { VI Infn => erg_ + "uer" ; @@ -1728,6 +1755,7 @@ oper erguer_Besch : Str -> Verbum = \erguer -> } ; oper moer_Besch : Str -> Verbum = \moer -> +-- 28 | r39 let m_ = Predef.tk 3 moer in {s = table { VI Infn => m_ + "oer" ; @@ -1797,6 +1825,7 @@ oper moer_Besch : Str -> Verbum = \moer -> } ; oper dizer_Besch : Str -> Verbum = \dizer -> +-- 29 | r60 let di_ = Predef.tk 3 dizer in {s = table { VI Infn => di_ + "zer" ; @@ -1866,6 +1895,7 @@ oper dizer_Besch : Str -> Verbum = \dizer -> } ; oper trazer_Besch : Str -> Verbum = \trazer -> +-- 30 | r73 let tr_ = Predef.tk 4 trazer in {s = table { VI Infn => tr_ + "azer" ; @@ -1935,6 +1965,7 @@ oper trazer_Besch : Str -> Verbum = \trazer -> } ; oper fazer_Besch : Str -> Verbum = \fazer -> +-- 31 | r61 let f_ = Predef.tk 4 fazer in {s = table { VI Infn => f_ + "azer" ; @@ -2004,6 +2035,7 @@ oper fazer_Besch : Str -> Verbum = \fazer -> } ; oper aprazer_Besch : Str -> Verbum = \aprazer -> +-- 32 | r55 let apr_ = Predef.tk 4 aprazer in {s = table { VI Infn => apr_ + "azer" ; @@ -2073,6 +2105,7 @@ oper aprazer_Besch : Str -> Verbum = \aprazer -> } ; oper jazer_Besch : Str -> Verbum = \jazer -> +-- 33 | r43 let jaz_ = Predef.tk 2 jazer in {s = table { VI Infn => jaz_ + "er" ; @@ -2142,6 +2175,7 @@ oper jazer_Besch : Str -> Verbum = \jazer -> } ; oper caber_Besch : Str -> Verbum = \caber -> +-- 34 | r56 let c_ = Predef.tk 4 caber in {s = table { VI Infn => c_ + "aber" ; @@ -2211,6 +2245,7 @@ oper caber_Besch : Str -> Verbum = \caber -> } ; oper saber_Besch : Str -> Verbum = \saber -> +-- 35 | r72 let s_ = Predef.tk 4 saber in {s = table { VI Infn => s_ + "aber" ; @@ -2280,6 +2315,7 @@ oper saber_Besch : Str -> Verbum = \saber -> } ; oper poder_Besch : Str -> Verbum = \poder -> +-- 36 | r66 let p_ = Predef.tk 4 poder in {s = table { VI Infn => p_ + "oder" ; @@ -2349,6 +2385,7 @@ oper poder_Besch : Str -> Verbum = \poder -> } ; oper crer_Besch : Str -> Verbum = \crer -> +-- 37 | r58 let cr_ = Predef.tk 2 crer in {s = table { VI Infn => cr_ + "er" ; @@ -2418,6 +2455,7 @@ oper crer_Besch : Str -> Verbum = \crer -> } ; oper querer_Besch : Str -> Verbum = \querer -> +-- 38 | r69 let qu_ = Predef.tk 4 querer in {s = table { VI Infn => qu_ + "erer" ; @@ -2487,6 +2525,7 @@ oper querer_Besch : Str -> Verbum = \querer -> } ; oper requerer_Besch : Str -> Verbum = \requerer -> +-- 39 | r70 let reque_ = Predef.tk 3 requerer in {s = table { VI Infn => reque_ + "rer" ; @@ -2556,6 +2595,7 @@ oper requerer_Besch : Str -> Verbum = \requerer -> } ; oper ver_Besch : Str -> Verbum = \ver -> +-- 40 | r25 let v_ = Predef.tk 2 ver in {s = table { VI Infn => v_ + "er" ; @@ -2625,6 +2665,7 @@ oper ver_Besch : Str -> Verbum = \ver -> } ; oper prover_Besch : Str -> Verbum = \prover -> +-- 41 | r68 let prov_ = Predef.tk 2 prover in {s = table { VI Infn => prov_ + "er" ; @@ -2694,6 +2735,7 @@ oper prover_Besch : Str -> Verbum = \prover -> } ; oper ler_Besch : Str -> Verbum = \ler -> +-- 42 | r58 let l_ = Predef.tk 2 ler in {s = table { VI Infn => l_ + "er" ; @@ -2763,6 +2805,7 @@ oper ler_Besch : Str -> Verbum = \ler -> } ; oper valer_Besch : Str -> Verbum = \valer -> +-- 43 | r74 let val_ = Predef.tk 2 valer in {s = table { VI Infn => val_ + "er" ; @@ -2832,6 +2875,7 @@ oper valer_Besch : Str -> Verbum = \valer -> } ; oper perder_Besch : Str -> Verbum = \perder -> +-- 44 | r65 let per_ = Predef.tk 3 perder in {s = table { VI Infn => per_ + "der" ; @@ -2901,6 +2945,7 @@ oper perder_Besch : Str -> Verbum = \perder -> } ; oper pôr_Besch : Str -> Verbum = \pôr -> +-- 45 | r67 let p_ = Predef.tk 2 pôr in {s = table { VI Infn => p_ + "ôr" ; @@ -2970,6 +3015,7 @@ oper pôr_Besch : Str -> Verbum = \pôr -> } ; oper acontecer_Besch : Str -> Verbum = \acontecer -> +-- 46 | r25 let aconte_ = Predef.tk 3 acontecer in {s = table { VI Infn => aconte_ + "cer" ; @@ -3039,6 +3085,7 @@ oper acontecer_Besch : Str -> Verbum = \acontecer -> } ; oper chover_Besch : Str -> Verbum = \chover -> +-- 47 | r6 let chov_ = Predef.tk 2 chover in {s = table { VI Infn => chov_ + "er" ; @@ -3108,6 +3155,7 @@ oper chover_Besch : Str -> Verbum = \chover -> } ; oper doer_Besch : Str -> Verbum = \doer -> +-- 48 | r83 let d_ = Predef.tk 3 doer in {s = table { VI Infn => d_ + "oer" ; @@ -3177,6 +3225,7 @@ oper doer_Besch : Str -> Verbum = \doer -> } ; oper prazer_Besch : Str -> Verbum = \prazer -> +-- 49 | r55 let pr_ = Predef.tk 4 prazer in {s = table { VI Infn => pr_ + "azer" ; @@ -3246,6 +3295,7 @@ oper prazer_Besch : Str -> Verbum = \prazer -> } ; oper precaver_Besch : Str -> Verbum = \precaver -> +-- 50 | r85 let precav_ = Predef.tk 2 precaver in {s = table { VI Infn => precav_ + "er" ; @@ -3315,6 +3365,7 @@ oper precaver_Besch : Str -> Verbum = \precaver -> } ; oper reaver_Besch : Str -> Verbum = \reaver -> +-- 51 | r86 let re_ = Predef.tk 4 reaver in {s = table { VI Infn => re_ + "aver" ; @@ -3384,6 +3435,7 @@ oper reaver_Besch : Str -> Verbum = \reaver -> } ; oper redigir_Besch : Str -> Verbum = \redigir -> +-- 52 | r28 let red_ = Predef.tk 4 redigir in {s = table { VI Infn => red_ + "igir" ; @@ -3453,6 +3505,7 @@ oper redigir_Besch : Str -> Verbum = \redigir -> } ; oper extinguir_Besch : Str -> Verbum = \extinguir -> +-- 53 | r33 let extin_ = Predef.tk 4 extinguir in {s = table { VI Infn => extin_ + "guir" ; @@ -3522,6 +3575,7 @@ oper extinguir_Besch : Str -> Verbum = \extinguir -> } ; oper servir_Besch : Str -> Verbum = \servir -> +-- 54 | r47 let s_ = Predef.tk 5 servir in {s = table { VI Infn => s_ + "ervir" ; @@ -3591,6 +3645,7 @@ oper servir_Besch : Str -> Verbum = \servir -> } ; oper seguir_Besch : Str -> Verbum = \seguir -> +-- 55 | r50 let s_ = Predef.tk 5 seguir in {s = table { VI Infn => s_ + "eguir" ; @@ -3660,6 +3715,7 @@ oper seguir_Besch : Str -> Verbum = \seguir -> } ; oper sentir_Besch : Str -> Verbum = \sentir -> +-- 56 | r47 let s_ = Predef.tk 5 sentir in {s = table { VI Infn => s_ + "entir" ; @@ -3729,6 +3785,7 @@ oper sentir_Besch : Str -> Verbum = \sentir -> } ; oper preferir_Besch : Str -> Verbum = \preferir -> +-- 57 | r47 let pref_ = Predef.tk 4 preferir in {s = table { VI Infn => pref_ + "erir" ; @@ -3798,6 +3855,7 @@ oper preferir_Besch : Str -> Verbum = \preferir -> } ; oper agredir_Besch : Str -> Verbum = \agredir -> +-- 58 | r48 let agr_ = Predef.tk 4 agredir in {s = table { VI Infn => agr_ + "edir" ; @@ -3867,6 +3925,7 @@ oper agredir_Besch : Str -> Verbum = \agredir -> } ; oper dormir_Besch : Str -> Verbum = \dormir -> +-- 59 | r51 let d_ = Predef.tk 5 dormir in {s = table { VI Infn => d_ + "ormir" ; @@ -3936,6 +3995,7 @@ oper dormir_Besch : Str -> Verbum = \dormir -> } ; oper polir_Besch : Str -> Verbum = \polir -> +-- 60 | r81 let p_ = Predef.tk 4 polir in {s = table { VI Infn => p_ + "olir" ; @@ -4005,6 +4065,7 @@ oper polir_Besch : Str -> Verbum = \polir -> } ; oper acudir_Besch : Str -> Verbum = \acudir -> +-- 61 | r53 let ac_ = Predef.tk 4 acudir in {s = table { VI Infn => ac_ + "udir" ; @@ -4074,6 +4135,7 @@ oper acudir_Besch : Str -> Verbum = \acudir -> } ; oper fugir_Besch : Str -> Verbum = \fugir -> +-- 62 | r54 let f_ = Predef.tk 4 fugir in {s = table { VI Infn => f_ + "ugir" ; @@ -4143,6 +4205,7 @@ oper fugir_Besch : Str -> Verbum = \fugir -> } ; oper frigir_Besch : Str -> Verbum = \frigir -> +-- 63 | r28 let fr_ = Predef.tk 4 frigir in {s = table { VI Infn => fr_ + "igir" ; @@ -4212,6 +4275,7 @@ oper frigir_Besch : Str -> Verbum = \frigir -> } ; oper divergir_Besch : Str -> Verbum = \divergir -> +-- 64 | r49 let div_ = Predef.tk 5 divergir in {s = table { VI Infn => div_ + "ergir" ; @@ -4281,6 +4345,7 @@ oper divergir_Besch : Str -> Verbum = \divergir -> } ; oper refletir_Besch : Str -> Verbum = \refletir -> +-- 65 | r47 let refl_ = Predef.tk 4 refletir in {s = table { VI Infn => refl_ + vars "etir" "ectir" ; @@ -4350,6 +4415,7 @@ oper refletir_Besch : Str -> Verbum = \refletir -> } ; oper ir_Besch : Str -> Verbum = \ir -> +-- 66 | r53 let x_ = Predef.tk 2 ir in {s = table { VI Infn => x_ + "ir" ; @@ -4419,6 +4485,7 @@ oper ir_Besch : Str -> Verbum = \ir -> } ; oper vir_Besch : Str -> Verbum = \vir -> +-- 67 | r63 let v_ = Predef.tk 2 vir in {s = table { VI Infn => v_ + "ir" ; @@ -4488,6 +4555,7 @@ oper vir_Besch : Str -> Verbum = \vir -> } ; oper sair_Besch : Str -> Verbum = \sair -> +-- 68 | r42 let sa_ = Predef.tk 2 sair in {s = table { VI Infn => sa_ + "ir" ; @@ -4557,6 +4625,7 @@ oper sair_Besch : Str -> Verbum = \sair -> } ; oper rir_Besch : Str -> Verbum = \rir -> +-- 69 | r48 let r_ = Predef.tk 2 rir in {s = table { VI Infn => r_ + "ir" ; @@ -4626,6 +4695,7 @@ oper rir_Besch : Str -> Verbum = \rir -> } ; oper pedir_Besch : Str -> Verbum = \pedir -> +-- 70 | r63 let pe_ = Predef.tk 3 pedir in {s = table { VI Infn => pe_ + "dir" ; @@ -4695,6 +4765,7 @@ oper pedir_Besch : Str -> Verbum = \pedir -> } ; oper ouvir_Besch : Str -> Verbum = \ouvir -> +-- 71 | r63 let ou_ = Predef.tk 3 ouvir in {s = table { VI Infn => ou_ + "vir" ; @@ -4764,6 +4835,7 @@ oper ouvir_Besch : Str -> Verbum = \ouvir -> } ; oper traduzir_Besch : Str -> Verbum = \traduzir -> +-- 72 | r44 let traduz_ = Predef.tk 2 traduzir in {s = table { VI Infn => traduz_ + "ir" ; @@ -4833,6 +4905,7 @@ oper traduzir_Besch : Str -> Verbum = \traduzir -> } ; oper distribuir_Besch : Str -> Verbum = \distribuir -> +-- 73 | r40 let distribu_ = Predef.tk 2 distribuir in {s = table { VI Infn => distribu_ + "ir" ; @@ -4902,6 +4975,7 @@ oper distribuir_Besch : Str -> Verbum = \distribuir -> } ; oper destruir_Besch : Str -> Verbum = \destruir -> +-- 74 | r57 let destr_ = Predef.tk 3 destruir in {s = table { VI Infn => destr_ + "uir" ; @@ -4971,6 +5045,7 @@ oper destruir_Besch : Str -> Verbum = \destruir -> } ; oper arguir_Besch : Str -> Verbum = \arguir -> +-- 75 | r31 let arg_ = Predef.tk 3 arguir in {s = table { VI Infn => arg_ + "uir" ; @@ -5040,6 +5115,7 @@ oper arguir_Besch : Str -> Verbum = \arguir -> } ; oper reunir_Besch : Str -> Verbum = \reunir -> +-- 76 | r18 let re_ = Predef.tk 4 reunir in {s = table { VI Infn => re_ + "unir" ; @@ -5109,6 +5185,7 @@ oper reunir_Besch : Str -> Verbum = \reunir -> } ; oper proibir_Besch : Str -> Verbum = \proibir -> +-- 77 | r20 let pro_ = Predef.tk 4 proibir in {s = table { VI Infn => pro_ + "ibir" ; @@ -5178,6 +5255,7 @@ oper proibir_Besch : Str -> Verbum = \proibir -> } ; oper imergir_Besch : Str -> Verbum = \imergir -> +-- 78 | r49 let imerg_ = Predef.tk 2 imergir in {s = table { VI Infn => imerg_ + "ir" ; @@ -5247,6 +5325,7 @@ oper imergir_Besch : Str -> Verbum = \imergir -> } ; oper falir_Besch : Str -> Verbum = \falir -> +-- 79 | r81 let fal_ = Predef.tk 2 falir in {s = table { VI Infn => fal_ + "ir" ; @@ -5316,6 +5395,7 @@ oper falir_Besch : Str -> Verbum = \falir -> } ; oper remir_Besch : Str -> Verbum = \remir -> +-- 80 | r49 let rem_ = Predef.tk 2 remir in {s = table { VI Infn => rem_ + "ir" ; @@ -5385,6 +5465,7 @@ oper remir_Besch : Str -> Verbum = \remir -> } ; oper viajar_Besch : Str -> Verbum = \viajar -> +-- r22 let viaj_ = Predef.tk 2 viajar in {s = table { VI Infn => viaj_ + "ar" ; @@ -5454,6 +5535,7 @@ oper viajar_Besch : Str -> Verbum = \viajar -> } ; oper suar_Besch : Str -> Verbum = \suar -> +-- r37 let su_ = Predef.tk 2 suar in {s = table { VI Infn => su_ + "ar" ; @@ -5522,4 +5604,74 @@ oper suar_Besch : Str -> Verbum = \suar -> } } ; +oper peneirar_Besch : Str -> Verbum = \peneirar -> +-- r10 + let peneir_ = Predef.tk 2 peneirar in + {s = table { + VI Infn => peneir_ + "ar" ; + VI Ger => peneir_ + "ando" ; + VI Part => peneir_ + "ado" ; + VPB (Pres Ind Sg P1) => peneir_ + "o" ; + VPB (Pres Ind Sg P2) => peneir_ + "as" ; + VPB (Pres Ind Sg P3) => peneir_ + "a" ; + VPB (Pres Ind Pl P1) => peneir_ + "amos" ; + VPB (Pres Ind Pl P2) => peneir_ + "ais" ; + VPB (Pres Ind Pl P3) => peneir_ + "am" ; + VPB (Pres Sub Sg P1) => peneir_ + "e" ; + VPB (Pres Sub Sg P2) => peneir_ + "es" ; + VPB (Pres Sub Sg P3) => peneir_ + "e" ; + VPB (Pres Sub Pl P1) => peneir_ + "emos" ; + VPB (Pres Sub Pl P2) => peneir_ + "eis" ; + VPB (Pres Sub Pl P3) => peneir_ + "em" ; + VPB (PretI Ind Sg P1) => peneir_ + "ava" ; --# notpresent + VPB (PretI Ind Sg P2) => peneir_ + "avas" ; --# notpresent + VPB (PretI Ind Sg P3) => peneir_ + "ava" ; --# notpresent + VPB (PretI Ind Pl P1) => peneir_ + "ávamos" ; --# notpresent + VPB (PretI Ind Pl P2) => peneir_ + "áveis" ; --# notpresent + VPB (PretI Ind Pl P3) => peneir_ + "avam" ; --# notpresent + VPB (PretI Sub Sg P1) => peneir_ + "asse" ; --# notpresent + VPB (PretI Sub Sg P2) => peneir_ + "asses" ; --# notpresent + VPB (PretI Sub Sg P3) => peneir_ + "asse" ; --# notpresent + VPB (PretI Sub Pl P1) => peneir_ + "ássemos" ; --# notpresent + VPB (PretI Sub Pl P2) => peneir_ + "ásseis" ; --# notpresent + VPB (PretI Sub Pl P3) => peneir_ + "assem" ; --# notpresent + VPB (MQPerf Sg P1) => peneir_ + "ara" ; --# notpresent + VPB (MQPerf Sg P2) => peneir_ + "aras" ; --# notpresent + VPB (MQPerf Sg P3) => peneir_ + "ara" ; --# notpresent + VPB (MQPerf Pl P1) => peneir_ + "áramos" ; --# notpresent + VPB (MQPerf Pl P2) => peneir_ + "áreis" ; --# notpresent + VPB (MQPerf Pl P3) => peneir_ + "aram" ; --# notpresent + VPB (PretP Sg P1) => peneir_ + "ei" ; --# notpresent + VPB (PretP Sg P2) => peneir_ + "aste" ; --# notpresent + VPB (PretP Sg P3) => peneir_ + "ou" ; --# notpresent + VPB (PretP Pl P1) => peneir_ + "amos" ; --# notpresent + VPB (PretP Pl P2) => peneir_ + "astes" ; --# notpresent + VPB (PretP Pl P3) => peneir_ + "aram" ; --# notpresent + VPB (Fut Ind Sg P1) => peneir_ + "arei" ; --# notpresent + VPB (Fut Ind Sg P2) => peneir_ + "arás" ; --# notpresent + VPB (Fut Ind Sg P3) => peneir_ + "ará" ; --# notpresent + VPB (Fut Ind Pl P1) => peneir_ + "aremos" ; --# notpresent + VPB (Fut Ind Pl P2) => peneir_ + "areis" ; --# notpresent + VPB (Fut Ind Pl P3) => peneir_ + "arão" ; --# notpresent + VPB (Fut Sub Sg P1) => peneir_ + "ar" ; --# notpresent + VPB (Fut Sub Sg P2) => peneir_ + "ares" ; --# notpresent + VPB (Fut Sub Sg P3) => peneir_ + "ar" ; --# notpresent + VPB (Fut Sub Pl P1) => peneir_ + "armos" ; --# notpresent + VPB (Fut Sub Pl P2) => peneir_ + "ardes" ; --# notpresent + VPB (Fut Sub Pl P3) => peneir_ + "arem" ; --# notpresent + VPB (Cond Sg P1) => peneir_ + "aria" ; --# notpresent + VPB (Cond Sg P2) => peneir_ + "aries" ; --# notpresent + VPB (Cond Sg P3) => peneir_ + "aria" ; --# notpresent + VPB (Cond Pl P1) => peneir_ + "aríamos" ; --# notpresent + VPB (Cond Pl P2) => peneir_ + "aríeis" ; --# notpresent + VPB (Cond Pl P3) => peneir_ + "ariam" ; --# notpresent + VPB (Imper Sg P2) => peneir_ + "a" ; + VPB (Imper Sg P3) => peneir_ + "e" ; + VPB (Imper Pl P1) => peneir_ + "emos" ; + VPB (Imper Pl P2) => peneir_ + "ai" ; + VPB (Imper Pl P3) => peneir_ + "em" ; + VPB (Imper Sg P1) => nonExist + } + } ; + } ; diff --git a/src/portuguese/ConstructionPor.gf b/src/portuguese/ConstructionPor.gf index f3409985..5803e813 100644 --- a/src/portuguese/ConstructionPor.gf +++ b/src/portuguese/ConstructionPor.gf @@ -69,6 +69,8 @@ lin let n_card : Card = lin Card n; n_hours_NP : NP = mkNP n_card time ; in S.mkAdv for_Prep n_hours_NP | S.mkAdv to_Prep n_hours_NP ;--| S.mkAdv (n_hours_NP.s ! R.Nom) ; + timeunitRange l u time = {s = "de" ++ l.s ! time.g + ++ "a" ++ u.s ! time.g ++ time.s ! u.n } ; oneHour = mkHour "1" Manha Sg ; twoHour = mkHour "2" Manha Pl ; @@ -95,12 +97,12 @@ lin twentyThreeHour = mkHour "23" Noite Pl ; twentyFourHour = {s = "meia-noite" ; pe = None ; n = Sg} ; - timeHour h = mkAdv (a ! h.n ++ h.s ++ period ! h.pe) ; + timeHour h = mkAdv (R.a ! Fem ! h.n ++ h.s ++ period ! h.pe) ; timeHourMinute h m = let min = m.s ! Masc in - mkAdv (a ! h.n ++ h.s ++ "e" ++ min ++ period ! h.pe) ; + mkAdv (R.a ! Fem ! h.n ++ h.s ++ "e" ++ min ++ period ! h.pe) ; oper mkHour : Str -> Period -> Number -> {s : Str ; pe : Period ; n : Number} ; @@ -114,9 +116,6 @@ lin None => "" } ; - a : Number => Str ; - a = numForms "à" "às" ; - lin weekdayPunctualAdv w = lin Adv {s = w.s ! C.Sg} ; -- lundi weekdayHabitualAdv w = SyntaxPor.mkAdv noPrep (mkNP the_Det w) ; -- il lunedí ---- diff --git a/src/portuguese/DiffPor.gf b/src/portuguese/DiffPor.gf index 5cfcd7c6..16cd4d11 100644 --- a/src/portuguese/DiffPor.gf +++ b/src/portuguese/DiffPor.gf @@ -106,12 +106,18 @@ instance DiffPor of DiffRomance - [partAgr,vpAgrSubj,vpAgrClits] = open CommonRo partitive = \_,c -> prepCase c ; oper + a : Gender => Number => Str ; + a = genNumForms "a" "à" "aos" "às" ; + + de : Gender => Number => Str ; + de = genNumForms "do" "da" "dos" "das" ; + artDef : Bool -> Gender -> Number -> Case -> Str ; -- not sure if isNP is relevant artDef _isNP g n c = case c of { Nom | Acc => genNumForms "o" "a" "os" "as" ; - CPrep P_de => genNumForms "do" "da" "dos" "das" ; - CPrep P_a => genNumForms "ao" "à" "aos" "às" ; + CPrep P_de => de ; + CPrep P_a => a ; CPrep P_em => genNumForms "no" "na" "nos" "nas" ; CPrep P_por => genNumForms "pelo" "pela" "pelos" "pelas" } ! g ! n ; @@ -148,7 +154,7 @@ instance DiffPor of DiffRomance - [partAgr,vpAgrSubj,vpAgrClits] = open CommonRo subjIf = "se" ; oper - relPron : Bool => AAgr => Case => Str = \\b,a,c => + relPron : Bool => AAgr => Case => Str = \\_b,a,c => case c of { Nom | Acc => "que" ; CPrep P_a => "cujo" ; diff --git a/src/portuguese/ExtendPor.gf b/src/portuguese/ExtendPor.gf index 12ab235b..52cb8e90 100644 --- a/src/portuguese/ExtendPor.gf +++ b/src/portuguese/ExtendPor.gf @@ -4,6 +4,8 @@ concrete ExtendPor of Extend = CatPor ** ExtendFunctor - [ AdAdV, + AdjAsCN, + AdjAsNP, ApposNP, BaseVPS, ByVP, @@ -13,6 +15,7 @@ concrete ExtendPor of Extend = CompoundAP, CompoundN, CompVP, + ConjVPS, ConsVPS, --EmptyRelSlash, ExistsNP, @@ -27,11 +30,13 @@ concrete ExtendPor of Extend = ICompAP, InOrderToVP, ListVPS, + MkVPS, PassAgentVPSlash, PassVPSlash, PastPartAP, PastPartAgentAP, PositAdVAdj, + PredVPS, PresPartAP, ProDrop, PurposeVP, @@ -52,13 +57,13 @@ concrete ExtendPor of Extend = with (Grammar = GrammarPor), (Syntax = SyntaxPor) ** open - GrammarPor, - ResPor, - MorphoPor, - Coordination, - Prelude, - ParadigmsPor, - (S = StructuralPor) in { + GrammarPor, + ResPor, + MorphoPor, + Coordination, + Prelude, + ParadigmsPor, + (S = StructuralPor) in { lin GenNP np = @@ -85,17 +90,28 @@ concrete ExtendPor of Extend = c = Nom } ; + ---- these come from ExtraRomance: how to avoid the repetition? + ---- can't seem to be able to use two functors lincat - VPS = {s : Agr => Mood => Str} ; - [VPS] = {s1,s2 : Agr => Mood => Str} ; --- VPI = {s : VType => Agr => Str } ; + VPS = {s : Mood => Agr => Bool => Str} ; + [VPS] = {s1,s2 : Mood => Agr => Bool => Str} ; lin - BaseVPS = twoTable2 Agr Mood ; - ConsVPS = consrTable2 Agr Mood comma ; + BaseVPS x y = twoTable3 Mood Agr Bool x y ; + ConsVPS = consrTable3 Mood Agr Bool comma ; --- MkVPS t p vp = mkVPS (lin Temp t) (lin Pol p) (lin VP vp) ; ---TODO: write mkVPS oper + PredVPS np vpi = { + s = \\m => (np.s ! Nom).comp ++ vpi.s ! m ! np.a ! np.isNeg + } ; + + MkVPS tm p vp = { + s = \\m,agr,isNeg => + tm.s ++ p.s ++ + (mkClausePol (orB isNeg vp.isNeg) [] False False agr vp).s + ! DDir ! tm.t ! tm.a ! p.p ! m + } ; + + ConjVPS = conjunctDistrTable3 Mood Agr Bool ; lin ProDrop p = { @@ -148,13 +164,23 @@ concrete ExtendPor of Extend = ComplBareVS = ComplVS ; + AdjAsCN ap = { + s =\\n => ap.s ! AF Masc n ; + g = Masc + } ; + + AdjAsNP ap = heavyNP { + s = \\_c => ap.s ! AF Masc Sg ; + a = Ag Masc Sg P3 + } ; + oper pastPartAP : VPSlash -> Str -> AP ; pastPartAP vps agent = lin AP { s = \\af => vps.comp ! (aform2aagr af ** {p = P3}) ++ vps.s.s ! VPart (aform2gender af) (aform2number af) ++ agent ; isPre = False } ; - + passVPSlash : VPSlash -> Str -> VP ; passVPSlash vps agent = let auxvp = predV auxPassive diff --git a/src/portuguese/LexiconPor.gf b/src/portuguese/LexiconPor.gf index 96b9ee2a..699a8fa6 100644 --- a/src/portuguese/LexiconPor.gf +++ b/src/portuguese/LexiconPor.gf @@ -10,7 +10,7 @@ flags lin easy_A2V = mkA2V (mkA "fácil") dative genitive ; married_A2 = mkA2 (mkA "casado") dative ; - probable_AS = mkAS (mkA "provável" "provável" "prováveis" "prováveis" "provavelmente") ; + probable_AS = mkAS (prefA (mkA "provável" "provavelmente")) ; fun_AV = mkAV (mkA "divertido") genitive ; -- A bad_A = prefA (mkA (mkA "mau") (mkA "pior")) ; @@ -38,7 +38,7 @@ lin narrow_A = mkA "estreito" ; near_A = mkA "perto" ; new_A = prefA (mkA "novo") ; - old_A = prefA (mkA "velho") ; + old_A = prefA (mkA "velho") ; ready_A = mkA "pronto" ; red_A = mkA "vermelho" ; rotten_A = mkA "podre" ; @@ -58,7 +58,7 @@ lin white_A = compADeg (mkA "branco") ; wide_A = mkA "largo" ; -- extenso yellow_A = mkA "amarelo" ; - young_A = prefA (mkA "jovem" "jovem" "jovens" "jovens" "juvenilmente") ; + young_A = prefA (mkA "jovem" "juvenilmente") ; already_Adv = mkAdv "já" ; far_Adv = mkAdv "longe" ; ----? now_Adv = mkAdv "agora" ; diff --git a/src/portuguese/MorphoPor.gf b/src/portuguese/MorphoPor.gf index 54b31b10..e4c20604 100644 --- a/src/portuguese/MorphoPor.gf +++ b/src/portuguese/MorphoPor.gf @@ -124,16 +124,16 @@ oper } } ; - mkAdj2N : (_,_: N) -> Str -> Adj = \mascN, femN, burramente -> - {s = table { - AF Masc n => mascN.s ! n ; - AF Fem n => femN.s ! n ; - AA => burramente - } + mkAdj2 : (_,_: Str) -> Adj ; + mkAdj2 aj av = let + adj = mkAdjReg aj + in { + s = table { + AF g n => adj.s ! AF g n ; + AA => av + } } ; - mkAdjN : N -> Str -> Adj = \n, burramente -> mkAdj2N n n burramente ; - -- Then the regular and invariant patterns. adjPreto : Str -> Adj = \preto -> @@ -174,7 +174,7 @@ oper "ã" => "a" } ; alemvo : Str = alem + v + "o" ; - in mkAdj alemão alemã (alemã + "s") (alemã + "es") (alemã + "amente") ; + in mkAdj alemão alemã (alemã + "s") (alemã + "es") (alemã + "mente") ; adjEuropeu : Str -> Adj = \europeu -> let europe = init europeu in mkAdj europeu (europe + "ia") (europeu + "s") (europe + "ias") @@ -183,11 +183,13 @@ oper mkAdjReg : Str -> Adj = \a -> case a of { pret + "o" => adjPreto a ; - anarquist + v@("e" | "a") => adjUtil (anarquist + v) (anarquist + v + "s") ; + anarquist + v@("e" | "a") => adjUtil a (a + "s") ; ouvido + "r" => adjOuvidor a (ouvido + "ra") ; chin + "ês" => adjFrances a ; europ + "eu" => adjEuropeu a ; alem + "ão" => adjVo a ; + provav + v@("e" | "i") + "l" => adjUtil a (provav + "eis") ; + jove + "m" => adjUtil a (jove + "ns") ; _ => adjUtil a (a + "s") } ; diff --git a/src/portuguese/ParadigmsPor.gf b/src/portuguese/ParadigmsPor.gf index 38de771b..6066a16b 100644 --- a/src/portuguese/ParadigmsPor.gf +++ b/src/portuguese/ParadigmsPor.gf @@ -203,9 +203,9 @@ oper regA a = compADeg {s = \\_ => (mkAdjReg a).s ; isPre = False ; lock_A = <>} ; - mk2A : (espanhol,espanhola : Str) -> A ; - mk2A a b = compADeg {s = \\_ => (mkAdj2N (mkN a) (mkN b) (b + "mente")).s ; isPre = False ; - lock_A = <>} ; + mk2A : (único,unicamente : Str) -> A ; + mk2A adj adv = compADeg {s = \\_ => (mkAdj2 adj adv).s ; isPre = False ; + lock_A = <>} ; mk5A : (preto,preta,pretos,pretas,pretamente : Str) -> A ; mk5A a b c d e = compADeg {s = \\_ => (mkAdj a b c d e).s ; diff --git a/src/portuguese/ResPor.gf b/src/portuguese/ResPor.gf index f75f9df3..dcd9c98d 100644 --- a/src/portuguese/ResPor.gf +++ b/src/portuguese/ResPor.gf @@ -9,4 +9,5 @@ instance ResPor of ResRomance = DiffPor ** open CommonRomance, Prelude in { oper vowel : pattern Str = #("a" | "e" | "i" | "o" | "u") ; + } ; diff --git a/src/somali/SentenceSom.gf b/src/somali/SentenceSom.gf index 93d52274..5714f975 100644 --- a/src/somali/SentenceSom.gf +++ b/src/somali/SentenceSom.gf @@ -70,13 +70,13 @@ lin -} -- : Temp -> Pol -> Cl -> S ; - UseCl temp pol cl = { s = cl.s ! temp.t ! temp.a ! pol.p } ; + UseCl t p cl = { s = t.s ++ p.s ++ cl.s ! t.t ! t.a ! p.p } ; {- -- : Temp -> Pol -> RCl -> RS ; - UseRCl temp pol cl = { s = cl.s ! temp.t ! temp.a ! pol.p } ; + UseRCl t p cl = { s = t.s ++ p.s ++ cl.s ! t.t ! t.a ! p.p } ; -- : Temp -> Pol -> QCl -> QS ; - UseQCl temp pol qcl = { s = qcl.s ! temp.t ! temp.a ! pol.p } ; + UseQCl t p cl = { s = t.s ++ p.s ++ cl.s ! t.t ! t.a ! p.p } ; -- An adverb can be added to the beginning of a sentence, either with comma ("externally") -- or without: @@ -89,7 +89,7 @@ lin -- There's an SubjS already in AdverbSom -- should this be deprecated? -- : S -> Subj -> S -> S ; - SSubjS s1 subj s2 = AdvS (AE.SubjS subj s2) s1 ; + SSubjS s1 subj s2 = AdvS (AS.SubjS subj s2) s1 ; -- A sentence can be modified by a relative clause referring to its contents. diff --git a/src/swedish/ExtendSwe.gf b/src/swedish/ExtendSwe.gf index 5dfcf9a2..37acbeca 100644 --- a/src/swedish/ExtendSwe.gf +++ b/src/swedish/ExtendSwe.gf @@ -8,7 +8,7 @@ concrete ExtendSwe of Extend = CatSwe ** PassVPSlash, PassAgentVPSlash, UttVPShort, ByVP, InOrderToVP, MkVPI, BaseVPI, ConsVPI, ConjVPI, ComplVPIVV, MkVPS, BaseVPS, ConsVPS, ConjVPS, PredVPS, - ICompAP, + ICompAP,ProDrop, AdAdV, PositAdVAdj, GerundCN, GerundNP, GerundAdv, PresPartAP, PastPartAP, PastPartAgentAP, RNP, RNPList, ReflRNP, ReflPron, ReflPoss, PredetRNP, ConjRNP, Base_rr_RNP, Base_nr_RNP, Base_rn_RNP, Cons_rr_RNP, Cons_nr_RNP, @@ -122,7 +122,7 @@ concrete ExtendSwe of Extend = CatSwe ** ICompAP ap = {s = \\a => hur_IAdv.s ++ ap.s ! a} ; - + ProDrop pro = pro ** {s = \\_ => []} ; lincat RNP = {s : Agr => Str ; isPron : Bool} ; ---- inherent Agr needed: han färgar sitt hår vitt. But also depends on subject