From a8cd24f0f5cbefccdefd2a4a5166b89c6c8f7a54 Mon Sep 17 00:00:00 2001
From: Sergey Poznyakoff <gray@gnu.org.ua>
Date: Thu, 5 Mar 2009 20:25:52 +0200
Subject: Avoid losing newlines while parsing the input stream. Provide some
 rudimentary parsing for wiktionary templates

---
 wiki2html.py  | 53 +++++++++++++++++++++++++++++++++++++++++++++++------
 wikicvt.py    |  0
 wikimarkup.py |  6 +++++-
 3 files changed, 52 insertions(+), 7 deletions(-)
 mode change 100644 => 100755 wikicvt.py

diff --git a/wiki2html.py b/wiki2html.py
index 5a8fdcc..faab18b 100644
--- a/wiki2html.py
+++ b/wiki2html.py
@@ -18,6 +18,7 @@
 from wikimarkup import *
 from types import TupleType
 from wikins import wiki_ns_re, wiki_ns
+import re
 import urllib
 
 class HtmlWikiMarkup (WikiMarkup):
@@ -45,13 +46,53 @@ class HtmlWikiMarkup (WikiMarkup):
         if not lang:
             lang = self.lang
         return self.html_base % { 'lang' : lang } + urllib.quote(tgt)
+
+    def tmpl_term(self, s):
+        if len(s) == 2:
+            return s[1]
+        text = None
+        trans = None
+        for x in s[1:]:
+            m = re.match('(\w+)=', x)
+            if m:
+                if m.group(1) == "tr":
+                    trans = x[m.end(1)+1:]
+            elif not text:
+                text = x
+        if text:
+            if trans:
+                text += ' <span class="trans">[' + trans + ']</span>'
+        return text
+
+    def tmpl_proto(self, s):
+        text = '<span class="proto-lang">Proto-' + s[1] + '</span>'
+        if len(s) >= 4:
+            n = 0
+            for x in s[2:-2]:
+                if n > 0:
+                    text += ','
+                n += 1
+                text += ' <span class="proto">' + x + '</span>'
+                text += ' <span class="meaning">(' + s[-2] + ')</span>'
+        return text
+                
     
     def fmtlink(self, elt, istmpl):
         arg = self.format(elt[1][0])
+        text = None
         if len(elt[1]) > 1:
-            text = self.format(elt[1][1])
-        else:
-            text = None
+            text = '<span class="template">' + self.format(elt[1][1]) + '</span>'
+            if istmpl:
+                s = map(self.format, elt[1])
+                if re.match("t[+-]$", s[0]):
+                    if len(s) > 2:
+                        text = s[2]
+                elif s[0] == "term":
+                    text = self.tmpl_term(s)
+                elif s[0] == "proto":
+                    text = self.tmpl_proto(s)
+                return text
+            
         (qual,sep,tgt) = arg.partition(':')
         if tgt != '':
             ns = self.wiki_ns_name(qual)
@@ -89,10 +130,10 @@ class HtmlWikiMarkup (WikiMarkup):
                                                else arg)
                 
     def str_link(self, elt):
-        return self.fmtlink(elt, False) + " "
+        return self.fmtlink(elt, False)
 
     def str_tmpl(self, elt):
-        return self.fmtlink(elt, True) + " "
+        return self.fmtlink(elt, True)
 
     def str_ref(self, elt):
         target = elt[1]
@@ -104,7 +145,7 @@ class HtmlWikiMarkup (WikiMarkup):
     def concat(self, eltlist):
         string = ""
         for x in eltlist:
-            string += " " + self.format(x)
+            string += self.format(x)
         return string
     
     def str_it(self, elt):
diff --git a/wikicvt.py b/wikicvt.py
old mode 100644
new mode 100755
diff --git a/wikimarkup.py b/wikimarkup.py
index 716bc4a..a340628 100644
--- a/wikimarkup.py
+++ b/wikimarkup.py
@@ -183,6 +183,7 @@ class BaseWikiMarkup:
                 if self.peektkn()[0] == NL:
                     self.dprint(80, "LEAVE parse_fontmod=None")
                     return None
+                seq.append((TEXT, '\n'))
             else:
                 self.dprint(80, "LEAVE parse_fontmod=None")
                 return None
@@ -243,6 +244,7 @@ class BaseWikiMarkup:
             elif tok[0] == TEXT:
                 list.append(tok)
             elif tok[0] == NL:
+                list.append((TEXT, '\n'))
                 continue
             else:
                 self.dprint(80, "LEAVE parse_ref=%s", "None")
@@ -299,6 +301,7 @@ class BaseWikiMarkup:
                     self.ungetkn()
                     if self.is_block_delim(tok):
                         break
+                textlist.append('\n')
             elif tok[0] == NIL:
                 break
             elif tok[0] == DELIM:
@@ -431,7 +434,8 @@ class BaseWikiMarkup:
                 self.ungetkn()
                 return self.parse_para()
         elif toktype == NL:
-            return self.parse0()
+            return (TEXT, '\n')
+#            return self.parse0()
     
     def parse(self):
         if not self.toklist:
-- 
cgit v1.2.1