More on converting xhp to docbook.

2 files changed, 59 insertions, 23 deletions

diff --git a/source/docbook.py b/source/docbook.py
index 4668726..2d61390 100644
--- a/source/docbook.py
+++ b/source/docbook.py
@@ -91,33 +91,35 @@ class FilePathSorter:
                 self.__walkToContent(child)
 
 
-def appendDocBookElement (src):
-    if src.name == 'book':
-        return src.appendElement('chapter')
-    elif src.name == 'chapter':
-        return src.appendElement('sect1')
-    elif src.name == 'sect1':
-        return src.appendElement('sect2')
-    elif src.name == 'sect2':
-        return src.appendElement('sect3')
-    elif src.name == 'sect3':
-        return src.appendElement('sect4')
-    else:
-        return None
-
-
 class DocBookConverter:
 
+    elemRelTable = {
+        'book': 'chapter',
+        'chapter': 'sect1',
+        'sect1': 'sect2',
+        'sect2': 'sect3',
+        'sect3': 'sect4',
+        'sect4': 'sect5'
+    }
+
+    @staticmethod
+    def startSubSection (src):
+        if DocBookConverter.elemRelTable.has_key(src.name):
+            return src.appendElement(DocBookConverter.elemRelTable[src.name])
+        else:
+            return None
+
     def __init__ (self, treeroot, xhproots):
         self.treeroot = treeroot
         self.xhproots = xhproots
         self.root = node.Root()
+        self.headingParent = None
 
     def convert (self):
         book = self.root.appendElement('book')
         bookinfo = book.appendElement('bookinfo')
         title = bookinfo.appendElement('title')
-        title.appendContent("OpenOffice.org Help")
+        title.appendContent(globals.productName + " Help")
         
         for section in self.treeroot.getChildNodes():
             self.__walk(section, book)
@@ -126,14 +128,14 @@ class DocBookConverter:
         # src element is either 'section' or 'content'.
         if src.name == 'section':
             title = src.getAttr('title')
-            elem = appendDocBookElement(dest)
+            elem = DocBookConverter.startSubSection(dest)
             elem.appendElement('title').appendContent(title)
             for child in src.getChildNodes():
                 self.__walk(child, elem)
         elif src.name == 'content':
             # this element has 'title' and 'path' attributes, and has no more child nodes.
             title = src.getAttr('title')
-            sect = appendDocBookElement(dest)
+            sect = DocBookConverter.startSubSection(dest)
             sect.appendElement('title').appendContent(title)
             xhppath = src.getAttr('path')
             if self.xhproots.has_key(xhppath):
@@ -144,21 +146,54 @@ class DocBookConverter:
     def __getContentFromXHP (self, xhppath, dest):
         xhproot = self.xhproots[xhppath]
         xhpbody = xhproot.firstChild().firstChildByName('body')
+        # parent element for the xhp 'heading' paragraphs which should start a 
+        # new sub-section when converting to docbook.
+        self.headingParent = None 
         for xhpelem in xhpbody.getChildNodes():
-            self.__walkXHP(xhpelem, dest)
+            dest = self.__walkXHP(xhpelem, dest)
 
     def __walkXHP (self, src, dest):
+
         if src.nodeType == node.NodeType.Content:
+            # content node.
             if len(src.content.strip()) > 0:
                 dest.appendElement('warning').appendContent("unhandled content : '%s'"%src.content.strip())
-            return
 
-        if src.name == 'paragraph':
-            para = dest.appendElement('para')
-            para.appendContent(src.getContent())
+        elif src.name == 'comment':
+            # we can ignore these elements.
+            pass
+        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)
+
+        elif src.name == 'paragraph':
+            # normal paragraph content.
+            role = src.getAttr('role')
+            if role != None and role == 'heading':
+                # finish the current sub-section, adn start a new one.
+                title = src.getContent()
+                if self.headingParent == None:
+                    self.headingParent = dest
+                elem = DocBookConverter.startSubSection(self.headingParent)
+                elem.appendElement('title').appendContent(title)
+                return elem
+            else:
+                para = dest.appendElement('para')
+                para.appendContent(src.getContent())
+
+        elif src.name == 'section':
+            # we can probably just ignore this element for now.  Section elements
+            # are used only to group multiple elements together.
+            for child in src.getChildNodes():
+                dest = self.__walkXHP(child, dest)
+
         else:
+            # unhandled element.  My work is done when these elements go away.
             dest.appendElement('warning').appendContent("unhandled element '%s'"%src.name)
 
+        return dest
+
 
     def prettyPrint (self, fd):
         node.prettyPrint(fd, self.root)
diff --git a/source/globals.py b/source/globals.py
index 4afb82c..0322bd4 100644
--- a/source/globals.py
+++ b/source/globals.py
@@ -13,3 +13,4 @@ class Exception:
 
 # global parameters
 productName = 'OpenOffice.org'
+#productName = 'Go-OO'