Handle embed elements properly.

3 files changed, 36 insertions, 8 deletions

diff --git a/ooo-help-parser.py b/ooo-help-parser.py
index bba303e..31c0a88 100755
--- a/ooo-help-parser.py
+++ b/ooo-help-parser.py
@@ -56,6 +56,7 @@ def walkDirs (fpaths):
 def parseAllXHPFiles (filepaths):
     filesParsed = 0
     rootNodes = {}
+    embedNodes = {}
     for fpath in filepaths:
         file = open(fpath, 'r')
         strm = file.read()
@@ -69,8 +70,9 @@ def parseAllXHPFiles (filepaths):
                 # We need to make this consistent.
                 p.filename = p.filename[1:]
             rootNodes[p.filename] = p.root
+            embedNodes[p.filename] = p.ids
         filesParsed += 1
-    return rootNodes, filesParsed
+    return rootNodes, embedNodes, filesParsed
 
 def main ():
     parser = optparse.OptionParser()
@@ -109,10 +111,10 @@ def main ():
     filepaths = walkDirs(args)
 
 
-    xhproots, filesParsed = parseAllXHPFiles(filepaths)
+    xhproots, embeds, filesParsed = parseAllXHPFiles(filepaths)
 
     if options.convert:
-        converter = docbook.DocBookConverter(treeroot, xhproots)
+        converter = docbook.DocBookConverter(treeroot, xhproots, embeds)
         converter.convert()
         converter.prettyPrint(fd)
     else:
diff --git a/source/docbook.py b/source/docbook.py
index 2d61390..e19767b 100644
--- a/source/docbook.py
+++ b/source/docbook.py
@@ -109,9 +109,10 @@ class DocBookConverter:
         else:
             return None
 
-    def __init__ (self, treeroot, xhproots):
+    def __init__ (self, treeroot, xhproots, embeds):
         self.treeroot = treeroot
         self.xhproots = xhproots
+        self.embeds = embeds
         self.root = node.Root()
         self.headingParent = None
 
@@ -152,11 +153,20 @@ class DocBookConverter:
         for xhpelem in xhpbody.getChildNodes():
             dest = self.__walkXHP(xhpelem, dest)
 
+    def __getEmbedNode (self, href):
+        filename, id = href.split('#')
+        if filename.startswith('/'):
+            filename = filename[1:]
+        if self.embeds.has_key(filename) and self.embeds[filename].has_key(id):
+            return self.embeds[filename][id]
+        return None
+
     def __walkXHP (self, src, dest):
 
         if src.nodeType == node.NodeType.Content:
             # content node.
-            if len(src.content.strip()) > 0:
+            content = src.content.strip()
+            if len(content) > 0:
                 dest.appendElement('warning').appendContent("unhandled content : '%s'"%src.content.strip())
 
         elif src.name == 'comment':
@@ -165,7 +175,13 @@ class DocBookConverter:
         elif src.name == 'embed':
             # embed takes its content from another xhp file.
             href = src.getAttr('href')
-            dest.appendElement('warning').appendContent('take embedded content from ' + href)
+            embed = self.__getEmbedNode(href)
+            if embed == None:
+                dest.appendElement('warning').appendContent('take embedded content from ' + href)
+            else:
+                dest = self.__walkXHP(embed, dest)
+        elif src.name == 'link':
+            dest.appendElement('para').appendContent(src.getContent())
 
         elif src.name == 'paragraph':
             # normal paragraph content.
@@ -182,9 +198,10 @@ class DocBookConverter:
                 para = dest.appendElement('para')
                 para.appendContent(src.getContent())
 
-        elif src.name == 'section':
+        elif src.name == 'section' or src.name == 'variable':
             # we can probably just ignore this element for now.  Section elements
-            # are used only to group multiple elements together.
+            # are used only to group multiple elements together.  Same with 
+            # the variable elements.
             for child in src.getChildNodes():
                 dest = self.__walkXHP(child, dest)
 
diff --git a/source/expatimpl.py b/source/expatimpl.py
index f9865d0..06d5375 100644
--- a/source/expatimpl.py
+++ b/source/expatimpl.py
@@ -55,6 +55,14 @@ class XHPParser(ParserBase):
     def __init__ (self, strm):
         ParserBase.__init__(self, strm)
         self.filename = None
+        self.ids = {}
+
+    def startElement(self, name, attrs):
+        ParserBase.startElement(self, name, attrs)
+        if attrs.has_key('id'):
+            # associate this node with this ID.
+            val = attrs['id']
+            self.ids[val] = self.nodestack[-1]
 
     def character(self, data):
         ParserBase.character(self, data)
@@ -62,6 +70,7 @@ class XHPParser(ParserBase):
             # For now, I just assume that the filename element is always at the correct position.
             self.filename = self.char
 
+
 class TreeParser(ParserBase):
 
     def __init__ (self, strm):