Implement paragraphs

3 files changed, 258 insertions, 189 deletions

diff --git a/wiki2html.py b/wiki2html.py
index 907e3b1..7fa97b7 100644
--- a/wiki2html.py
+++ b/wiki2html.py
@@ -28,3 +28,35 @@ class HtmlWikiMarkup (WikiMarkup):
     """
+
+    # FIXME: Awful kludge
+    image_kw = [ 'Image',
+                 'Grafika',
+                 'Bild',
+                 'Εικόνα',
+                 'Dosiero',
+                 'Slika',
+                 'Resim'
+                 ]
+    
+    ST_INIT = 0
+    ST_PARA = 1
+    ST_OPEN = 2
+    
+    state = []
+
+    def opara(self):
+        if self.state[-1] == self.ST_PARA:
+            self.state[-1] = self.ST_OPEN
+            return "<p>"
+        else:
+            return ""
         
+    def cpara(self):
+        state = self.state.pop();
+        self.state.append(self.ST_INIT)
+        if state == self.ST_OPEN:
+            return "</p>"
+        else:
+            return ""
+    
+            
     def target(self, t):
@@ -33,3 +65,3 @@ class HtmlWikiMarkup (WikiMarkup):
         if tgt != '':
-            if qual in ('Image', 'Grafika'):
+            if qual in self.image_kw:
                 t = self.image_base + urllib.quote(tgt) + '/250px-' + urllib.quote(tgt)
@@ -98,7 +130,26 @@ class HtmlWikiMarkup (WikiMarkup):
 
-    def str_seq(self, tok, env):
+    def str_para(self, tok, env):
+        s = self.cpara()
+        self.state.append(self.ST_PARA)
+        return s
+
+    def fmtok(self, tok, env):
+        if type(tok) != TupleType:
+            return ""
+        if tok[0] in [ self.ENV, self.HDR ]:
+            s = self.cpara()
+        elif tok[0] == self.BAR:
+            s = self.str_para(tok, env)
+        elif tok[0] in [ self.NIL, self.SEQ ]:
             s = ""
-            for t in tok[1:]:
-                s += self.fmtok(t, env)
-            return s
+        else:
+            s = self.opara()
+        s1 = WikiMarkup.fmtok(self, tok, env)
+        if s1:
+            s += s1
+        return s
+    
+    def __str__(self):
+        self.state = [ self.ST_PARA ]
+        return WikiMarkup.__str__(self) + self.cpara()
 
@@ -111,3 +162,3 @@ class HtmlWiktionaryMarkup (HtmlWikiMarkup):
  format templates. But "tries" does not mean "does". The heuristics
- used here is clearly not enogh to cope with it.
+ used here is clearly not enough to cope with it.
 
@@ -136,8 +187,10 @@ class HtmlWiktionaryMarkup (HtmlWikiMarkup):
     def str_seq(self, tok, env):
-            s = ""
+            str = ""
             self.seq_pos=0
             for t in tok[1:]:
-                s += self.fmtok(t, env)
-                self.seq_pos += 1
-            return s
+                s = self.fmtok(t, env)
+                if s:
+                    str += s
+                    self.seq_pos += 1
+            return str
     
diff --git a/wiki2text.py b/wiki2text.py
index e943f32..3669bd7 100644
--- a/wiki2text.py
+++ b/wiki2text.py
@@ -138,2 +138,5 @@ class TextWikiMarkup (WikiMarkup):
 
+    def str_para(self, tok, env):
+        return "\n"
+    
     def __str__(self):
diff --git a/wikimarkup.py b/wikimarkup.py
index d9ae7cc..e2a1cab 100644
--- a/wikimarkup.py
+++ b/wikimarkup.py
@@ -37,6 +37,7 @@ A base class for handling Wiki markups.
 It handles:
-   1. basic block markup (headers, numbered and unnumbered lists,
+   1. paragraphs;
+   2. basic block markup (headers, numbered and unnumbered lists,
       indentations);
-   2. basic inline markup (bold, italic);
-   3. basic reference markup (links, templates, external links).
+   3. basic inline markup (bold, italic);
+   4. basic reference markup (links, templates, external links).
  It does NOT handle:
@@ -92,2 +93,4 @@ It handles:
     SEQ  = 11
+    # Paragraph
+    PARA = 12
     
@@ -132,2 +135,6 @@ It handles:
 
+            if line == '\n':
+                yield(self.PARA,)
+                continue
+            
             m = eltbeg.match(line)
@@ -249,4 +256,2 @@ It handles:
             return toktype, tok[1], self.expandtok(tok[2])
-        elif toktype == self.BAR:
-            return tok
         elif toktype == self.ENV:
@@ -266,2 +271,4 @@ It handles:
                        subtree[1] if len(subtree) == 2 else None
+        else:
+            return tok
 
@@ -316,2 +323,4 @@ It handles:
                 self.prtok(tok[1], indent+1)
+            elif toktype == self.PARA:
+                print "PARA"
     
@@ -379,20 +388,20 @@ class WikiMarkup (BaseWikiMarkup):
     langtab = {
-        "aa": "Afar", # Afar
-        "ab": "Аҧсуа", # Abkhazian
-        "ae": None, # Avestan
-        "af": "Afrikaans", # Afrikaans
-        "ak": "Akana", # Akan # or ak_CI
+        "aa": "Afar",            # Afar
+        "ab": "Аҧсуа",           # Abkhazian
+        "ae": None,              # Avestan
+        "af": "Afrikaans",       # Afrikaans
+        "ak": "Akana",           # Akan
         "als": "Alemannisch",
-        "am": "አማርኛ", # Amharic
-        "an": "Aragonés", # Aragonese
+        "am": "አማርኛ",            # Amharic
+        "an": "Aragonés",        # Aragonese
         "ang": "Englisc",
-        "ar": "العربية" , # Arabic
+        "ar": "العربية" ,        # Arabic
         "arc": "ܐܪܡܝܐ",
-        "as": "অসমীয়া", # Assamese
+        "as": "অসমীয়া",         # Assamese
         "ast": "Asturian", 
-        "av": "Авар", # Avaric # Spoken mainly in Dagestan
-        "ay": "Aymar", # Aymara
-        "az": "Azərbaycan" , # Azerbaijani
+        "av": "Авар",            # Avaric
+        "ay": "Aymar",           # Aymara
+        "az": "Azərbaycan" ,     # Azerbaijani
 
-        "ba": "Башҡорт", # Bashkir
+        "ba": "Башҡорт",         # Bashkir
         "bar": "Boarisch",      
@@ -400,13 +409,13 @@ class WikiMarkup (BaseWikiMarkup):
         "bcl": "Bikol",
-        "be": "Беларуская", # Byelorussian; Belarusian
+        "be": "Беларуская",      # Byelorussian; Belarusian
         "be-x-old": "Беларуская (тарашкевіца)",
-        "bg": "Български", # Bulgarian
-        "bh": "भोजपुरी", # Bihari
-        "bi": "Bislama", # Bislama
-        "bm": "Bamanankan", # Bambara
-        "bn": "বাংলা" , # Bengali; Bangla
-        "bo": "བོད་སྐད", # Tibetan
+        "bg": "Български",       # Bulgarian
+        "bh": "भोजपुरी",  # Bihari
+        "bi": "Bislama",         # Bislama
+        "bm": "Bamanankan",      # Bambara
+        "bn": "বাংলা" ,          # Bengali; Bangla
+        "bo": "བོད་སྐད",         # Tibetan
         "bpy": "ইমার ঠার/বিষ্ণুপ্রিয়া মণিপুরী" ,
-        "br": "Brezhoneg" , # Breton
-        "bs": "Bosanski" , # Bosnian
+        "br": "Brezhoneg" ,      # Breton
+        "bs": "Bosanski" ,       # Bosnian
         "bug": "Basa Ugi",
@@ -414,3 +423,3 @@ class WikiMarkup (BaseWikiMarkup):
 
-        "ca": "Català" , # Catalan
+        "ca": "Català" ,         # Catalan
         "cbk-zam": "Chavacano de Zamboanga",
@@ -418,67 +427,67 @@ class WikiMarkup (BaseWikiMarkup):
         "cho": "Choctaw",
-        "ce": "Нохчийн", # Chechen
+        "ce": "Нохчийн",         # Chechen
         "ceb": "Sinugboanong Binisaya" , # Cebuano
-        "ch": "Chamor", # Chamorro
+        "ch": "Chamor",          # Chamorro
         "chr": "ᏣᎳᎩ",
         "chy": "Tsetsêhestâhese",
-        "co": "Cors", # Corsican
-        "cr": "Nehiyaw", # Cree
+        "co": "Cors",            # Corsican
+        "cr": "Nehiyaw",         # Cree
         "crh": "Qırımtatarca",
-        "cs": "Česky" , # Czech
+        "cs": "Česky" ,          # Czech
         "csb": "Kaszëbsczi",
-        "c": "Словѣньскъ", # Church Slavic
-        "cv": "Чăваш", # Chuvash
-        "cy": "Cymraeg" , # Welsh
+        "c": "Словѣньскъ",       # Church Slavic
+        "cv": "Чăваш",           # Chuvash
+        "cy": "Cymraeg" ,        # Welsh
 
-        "da": "Dansk" , # Danish
-        "de": "Deutsch" , # German
-        "diq": "Zazaki", # Dimli (Southern Zazaki)
+        "da": "Dansk" ,          # Danish
+        "de": "Deutsch" ,        # German
+        "diq": "Zazaki",         # Dimli (Southern Zazaki)
         "dsb": "Dolnoserbski",