unlexer bug ; "vilka" ; lib path warning ; p -lines result

2006-03-21 13:57:38 +00:00
parent ed047e9429
commit 836026a3ce
8 changed files with 196 additions and 23 deletions
--- a/lib/resource-1.0/Makefile
+++ b/lib/resource-1.0/Makefile
@@ -29,7 +29,7 @@ multimodal:
 	mv multimodal/*.gfc multimodal/*.gfr ../multimodal

 compiled:
-	cd .. ; tar cvfz compiled.tgz alltenses/ mathematical/ multimodal/ present/ 
+	cd .. ; tar cvfz compiled.tgz alltenses/ mathematical/ multimodal/ present/ prelude/*.gf?

 treebank:
 	gf <mkTreebank.gfs
--- a/lib/resource-1.0/doc/index.html
+++ b/lib/resource-1.0/doc/index.html
@@ -7,7 +7,7 @@
 <P ALIGN="center"><CENTER><H1>GF Resource Grammar Library v. 1.0</H1>
 <FONT SIZE="4">
 <I>Author: Aarne Ranta &lt;aarne (at) cs.chalmers.se&gt;</I><BR>
-Last update: Fri Mar 10 13:27:21 2006
+Last update: Tue Mar 21 14:49:24 2006
 </FONT></CENTER>

 <P></P>
@@ -35,7 +35,8 @@ Last update: Fri Mar 10 13:27:21 2006
      <LI><A HREF="#toc14">Tram</A>
      <LI><A HREF="#toc15">Animals</A>
      </UL>
-    <LI><A HREF="#toc16">More reading</A>
+    <LI><A HREF="#toc16">Known bugs and missing components</A>
+    <LI><A HREF="#toc17">More reading</A>
    </UL>

 <P></P>
@@ -60,9 +61,12 @@ we refer to <A HREF="../../resource/">v. 0.9</A>.
 <A NAME="toc1"></A>
 <H2>Authors</H2>
 <P>
+Inger Andersson and Therese Soderberg (Spanish morphology),
+Nicolas Barth and Sylvain Pogodalla (French verb list),
 Janna Khegai (Russian modules, forthcoming),
 Bjorn Bringert (many Swadesh lexica),
 Carlos Gonzalia (Spanish cardinals), 
+Harald Hammarstrom (German morphology),
 Partik Jansson (Swedish cardinals),
 Aarne Ranta.
 </P>
@@ -70,12 +74,16 @@ Aarne Ranta.
 We are grateful for contributions and 
 comments to several other people who have used this and 
 the previous versions of the resource library, including
+Ludmilla Bogavac,
 Ana Bove,
 David Burke,
 Lauri Carlson,
 Gloria Casanellas,
 Karin Cavallin,
+Robin Cooper,
 Hans-Joachim Daniels,
+Elisabet Engdahl,
+Markus Forsberg,
 Kristofer Johannisson,
 Anni Laine,
 Wanjiku Ng'ang'a,
@@ -130,15 +138,26 @@ Go to the main directory, compile the grammars, and run a test.
 </PRE>
 <P>
 This will take quite some time. An alternative is to use the
-<A HREF="../../compiled.tgz">precompiled grammar package</A>. Just do
+precompiled grammar package from GF download page. This package 
+has the necessary <CODE>gfc</CODE> and <CODE>gfr</CODE> files directly under <CODE>GF/lib</CODE>.
 </P>
 <PRE>
-    cd GF/lib/resource-1.0
-    make pretest
+    GF/lib/alltenses
+    GF/lib/mathematical
+    GF/lib/multimodal
+    GF/lib/present
 </PRE>
 <P>
-For more examples, see the 
-<A HREF="clt2006.html">Overview slides</A>. 
+Do for instance
+</P>
+<PRE>
+    cd GF/lib/
+    gf
+      &gt; i -path=present:prelude present/LangEng.gfc
+      &gt; gr -cat=S -number=3 -cf | tb
+</PRE>
+<P>
+For more examples, see the <A HREF="clt2006.html">Overview slides</A>. 
 </P>
 <A NAME="toc5"></A>
 <H3>The language independent ground API</H3>
@@ -328,6 +347,84 @@ Its purpose is to serve as a prototype for example-based
 grammar writing.
 </P>
 <A NAME="toc16"></A>
+<H2>Known bugs and missing components</H2>
+<P>
+This bugs should be fixed before the final release of v. 1.0.
+</P>
+<P>
+Danish
+</P>
+<UL>
+<LI>API 1.0 not implemented
+</UL>
+
+<P>
+English 
+</P>
+<UL>
+<LI>only contracted negation forms
+</UL>
+
+<P>
+Finnish
+</P>
+<UL>
+<LI>no possessive suffixes
+<LI>wrong cases in passive constructions
+</UL>
+
+<P>
+French
+</P>
+<UL>
+<LI>only direct word order in questions
+</UL>
+
+<P>
+German
+</P>
+<UL>
+<LI>the Swadesh part of lexicon missing
+<LI>no list of irregular verbs
+</UL>
+
+<P>
+Italian
+</P>
+<UL>
+<LI>no contraction of infinitives before clitics
+<LI>no list of irregular verbs
+</UL>
+
+<P>
+Norwegian
+</P>
+<UL>
+<LI>word order in compound verbs sometimes wrong
+</UL>
+
+<P>
+Russian
+</P>
+<UL>
+<LI>API 1.0 not implemented
+</UL>
+
+<P>
+Spanish
+</P>
+<UL>
+<LI>no ordinal numbers
+</UL>
+
+<P>
+Swedish
+</P>
+<UL>
+<LI>word order in compound verbs sometimes wrong
+</UL>
+
+<A NAME="toc17"></A>
 <H2>More reading</H2>
 <P>
 <A HREF="gslt-sem-2006.html">Grammars as Software Libraries</A>. Slides
--- a/lib/resource-1.0/doc/index.txt
+++ b/lib/resource-1.0/doc/index.txt
@@ -25,21 +25,28 @@ we refer to [v. 0.9 ../../resource/].

 ==Authors==

+Inger Andersson and Therese Soderberg (Spanish morphology),
+Nicolas Barth and Sylvain Pogodalla (French verb list),
 Janna Khegai (Russian modules, forthcoming),
 Bjorn Bringert (many Swadesh lexica),
 Carlos Gonzalia (Spanish cardinals), 
+Harald Hammarstrom (German morphology),
 Partik Jansson (Swedish cardinals),
 Aarne Ranta.

 We are grateful for contributions and 
 comments to several other people who have used this and 
 the previous versions of the resource library, including
+Ludmilla Bogavac,
 Ana Bove,
 David Burke,
 Lauri Carlson,
 Gloria Casanellas,
 Karin Cavallin,
+Robin Cooper,
 Hans-Joachim Daniels,
+Elisabet Engdahl,
+Markus Forsberg,
 Kristofer Johannisson,
 Anni Laine,
 Wanjiku Ng'ang'a,
@@ -83,13 +90,22 @@ Go to the main directory, compile the grammars, and run a test.
  make test
 ```
 This will take quite some time. An alternative is to use the
-[precompiled grammar package ../../compiled.tgz]. Just do
+precompiled grammar package from GF download page. This package 
+has the necessary ``gfc`` and ``gfr`` files directly under ``GF/lib``.
 ```
-  cd GF/lib/resource-1.0
-  make pretest
+  GF/lib/alltenses
+  GF/lib/mathematical
+  GF/lib/multimodal
+  GF/lib/present
 ```
-For more examples, see the 
-[Overview slides clt2006.html]. 
+Do for instance
+```
+  cd GF/lib/
+  gf
+    > i -path=present:prelude present/LangEng.gfc
+    > gr -cat=S -number=3 -cf | tb
+```
+For more examples, see the [Overview slides clt2006.html]. 



@@ -265,6 +281,54 @@ Its purpose is to serve as a prototype for example-based
 grammar writing.


+==Known bugs and missing components==
+
+This bugs should be fixed before the final release of v. 1.0.
+
+Danish
+- API 1.0 not implemented
+
+
+English 
+- only contracted negation forms
+
+
+Finnish
+- no possessive suffixes
+- wrong cases in passive constructions
+
+
+French
+- only direct word order in questions
+
+
+German
+- the Swadesh part of lexicon missing
+- no list of irregular verbs
+
+
+Italian
+- no contraction of infinitives before clitics
+- no list of irregular verbs
+
+
+Norwegian
+- word order in compound verbs sometimes wrong
+
+
+Russian
+- API 1.0 not implemented
+
+
+Spanish
+- no ordinal numbers
+
+
+Swedish
+- word order in compound verbs sometimes wrong
+
+
+
 ==More reading==

 [Grammars as Software Libraries gslt-sem-2006.html]. Slides
--- a/lib/resource-1.0/norwegian/StructuralNor.gf
+++ b/lib/resource-1.0/norwegian/StructuralNor.gf
@@ -107,7 +107,7 @@ concrete StructuralNor of Structural = CatNor **
  whichPl_IDet = {s = \\_ => "hvilke" ; n = Pl ; det = DIndef} ;
  whichSg_IDet = {s = genderForms "hvilken" "hvilket" ; n = Sg ; det = DIndef} ;
  whoSg_IP = {s = vem.s ; gn = SgUtr} ;
-  whoPl_IP = {s = vem.s ; gn = Plg} ;
+  whoPl_IP = {s = \\_ => "hvilke" ; gn = Plg} ;
  why_IAdv = ss "hvorfor" ;
  without_Prep = ss "uten" ;
  with_Prep = ss "med" ;
--- a/lib/resource-1.0/swedish/StructuralSwe.gf
+++ b/lib/resource-1.0/swedish/StructuralSwe.gf
@@ -106,7 +106,7 @@ concrete StructuralSwe of Structural = CatSwe **
  whichPl_IDet = {s = \\_ => "vilka" ; n = Pl ; det = DIndef} ;
  whichSg_IDet = {s = genderForms "vilken" "vilket" ; n = Sg ; det = DIndef} ;
  whoSg_IP = {s = vem.s ; gn = SgUtr} ;
-  whoPl_IP = {s = vem.s ; gn = Plg} ;
+  whoPl_IP = {s = \\_ => "vilka" ; gn = Plg} ;
  why_IAdv = ss "varför" ;
  without_Prep = ss "utan" ;
  with_Prep = ss "med" ;
--- a/src/GF.hs
+++ b/src/GF.hs
@@ -31,11 +31,11 @@ import GF.Text.UTF8

 import GF.Today (today,version)
 import GF.System.Arch
-import System (getArgs,system)
+import System (getArgs,system,getEnv)
 import Control.Monad (foldM,liftM)
 import Data.List (nub)

-- AR 19/4/2000 -- 28/4/2005
+-- AR 19/4/2000 -- 21/3/2006

 main :: IO ()
 main = do
@@ -52,7 +52,8 @@ main = do
      putStrLnFlush $ encodeUTF8 helpMsg

    _ | opt forJava -> do
-      putStrLnFlush $ encodeUTF8 welcomeMsg 
+      welcome <- welcomeMsgLib
+      putStrLnFlush $ encodeUTF8 welcome 
      st <- useIOE st0 $ 
              foldM (shellStateFromFiles os) st0 fs
      sessionLineJ True st
@@ -77,8 +78,8 @@ main = do
      if opt beSilent then return () else putStrLnFlush "</gfbatch>"
      return ()
    _ -> do
-
-      ifNotSil $ putStrLnFlush $ welcomeMsg
+      welcome <- welcomeMsgLib
+      ifNotSil $ putStrLnFlush $ welcome
      st <- useIOE st0 $ 
              foldM (shellStateFromFiles os) st0 fs
      if null fs then return () else (ifNotSil putCPU) 
@@ -110,9 +111,16 @@ helpMsg = unlines [
  "which suppresses all messages except the output and fatal errors."
  ]

-welcomeMsg = 
+welcomeMsgLib = do
+  lib <- catch 
+    (getEnv "GF_LIB_PATH" >>= return . ("GF_LIB_PATH is set to" +++)) 
+    (const (return "Warning: GF_LIB_PATH is not defined."))
+  return $ welcomeMsg lib
+
+welcomeMsg lib = 
  "Welcome to " ++ authorMsg ++++ 
  "If ä and ö (umlaut letters) look strange, see 'h -coding'." ++
+  "\n" ++ lib ++
  "\n\nType 'h' for help, and 'h [Command] for more detailed help.\n"

 authorMsg = unlines [
--- a/src/GF/Shell.hs
+++ b/src/GF/Shell.hs
@@ -24,6 +24,7 @@ import qualified GF.Grammar.Lookup as L
 import qualified GF.Canon.GFC as GFC
 import qualified GF.Canon.Look as Look
 import qualified GF.Canon.CMacros as CMacros
+import qualified GF.Grammar.MMacros as MMacros
 import qualified GF.Compile.GrammarToCanon as GrammarToCanon
 import GF.Grammar.Values
 import GF.UseGrammar.GetTree
@@ -236,7 +237,10 @@ execC co@(comm, opts0) sa@(sh@(st,(h,_,_,_)),a) = checkOptions st co >> case com
        let ss = (if oElem showAll opts then id else filter (not . null)) $ 
                     lines $ prCommandArg a
        mts <- mapM parse ss
-        let a' = ATrms [t | (_,ATrms ts) <- mts, t <- ts]
+        let mark s ts = case ts of
+              [] -> [MMacros.uTree] -- to leave a trace of unparsed line
+              _  -> ts
+        let a' = ATrms [t | (s,(_,ATrms ts)) <- zip ss mts, t <- mark s ts]
        changeArg (const a') sa
    | otherwise -> parse $ prCommandArg a
   where 
--- a/src/GF/Text/Text.hs
+++ b/src/GF/Text/Text.hs
@@ -60,7 +60,7 @@ formatAsCodeLit :: String -> String
 formatAsCodeLit = formatAsCode . unwords . map unStringLit . words 

 formatAsText,formatAsHTML,formatAsLatex :: String -> String
-formatAsText  = formatAsTextGen (=="&-") (=="&-") 
+formatAsText  = formatAsTextGen (const False) (=="&-") 
 formatAsHTML  = formatAsTextGen (\s -> take 1 s == "<" || last s == '>') (const False) 
 formatAsLatex = formatAsTextGen ((=="\\") . take 1)  (const False)