Cleanup and prepare for full level TOC generation

2 files changed, 63 insertions, 29 deletions

diff --git a/panfry/document.py b/panfry/document.py
index 7bb06b4..0b450da 100755
--- a/panfry/document.py
+++ b/panfry/document.py
@@ -5,6 +5,7 @@ import re
 import pandoc
 from tempfile import NamedTemporaryFile
 from panfry.util import *
+from panfry.page import TextFile, Page
 
 
 class Document:
@@ -19,16 +20,20 @@ class Document:
 
     @property
     def title(self):
-        header = self.meta['HEADER'].strip()
-        if header.startswith('%'):
-            return header.split('%')[1].strip()
+        if self.header.firstline.startswith('%'):
+            return self.header.firstline.split('%')[1].strip()
         else:
             return pages[0].title
 
     def get_pages(self, path):
         pages = []
-        filelist = get_lines(self.meta['TOC'])
+        filelist = self.toc.lines
+        print filelist
         for filename in filelist:
+            print("Reading in %s..." % filename)
+            if not filename:  # blank line in toc
+                print("WARNING: %s contains blank lines" % self.toc.filename)
+                continue
             source = read_file(os.path.join(path, filename))
             if source:
                 pages.append(Page(filename, source))
@@ -43,7 +48,8 @@ class Document:
         for filename in metafiles:
             source = read_file(os.path.join(path, filename))
             if source:
-                meta[filename] = source
+                metafile = TextFile(filename, source)
+                setattr(self, filename.lower(), metafile)
             else:
                 print("!E: %s not found or is empty. Aborting..." % filename)
                 exit(1)
@@ -90,7 +96,6 @@ class Document:
     @property
     def pdf_filename(self):
         filename = self.title
-        filename = self.meta['HEADER'].split('%')[1].strip()
         filename = re.sub(r'\s+', ' ', filename)
         filename = re.sub(r'[\n ]', '_', filename)
         filename = re.sub(r'[:,]', '-', filename)
@@ -102,7 +107,7 @@ class Document:
 
     def publish_pdf(self, pub_path):
         pdf_path = os.path.join(pub_path, self.pdf_filename)
-        src = self.meta['HEADER']
+        src = self.header.source
         for page in self.pages:
             src += '\n%s' % page.source
 
@@ -146,7 +151,7 @@ class Document:
             doc.add_argument('toc')
             doc.add_argument('template=%s' % template_file.name)
             doc.add_argument('css=%s' % self.css_file)
-            doc.markdown = '%s\n%s' % (self.meta['HEADER'], page.source)
+            doc.markdown = '%s\n%s' % (self.header.source, page.source)
             content = doc.html
             write_file(os.path.join(pub_path, page.htmlfile),
                        unicode(content, 'utf-8'))
diff --git a/panfry/page.py b/panfry/page.py
index 27b4cb1..2c82e04 100755
--- a/panfry/page.py
+++ b/panfry/page.py
@@ -1,18 +1,35 @@
 #!/usr/bin/env python
 # -*- coding: utf-8 -*-
 import re
-from panfry.utils import get_lines
 
 
-class Page:
+class TextFile:
     '''
-    Represents single source file.
+    Basic text file class
     '''
     def __init__(self, filename, source):
         self.filename = filename
         self.source = source
 
     @property
+    def lines(self):
+        '''
+        Returns list of lines from source.
+        Leading and trailing blank lines are removed.
+        '''
+        return self.source.strip().split('\n')
+
+    @property
+    def firstline(self):
+        return self.lines[0]
+
+
+class Page(TextFile):
+    '''
+    Represents single page source file.
+    '''
+
+    @property
     def title(self):
         '''
         Returns a tuple containing a string representing the page title
@@ -32,27 +49,39 @@ class Page:
         the filename is returned as the title with underscores changed
         to spaces.
         '''
-        title = ''
-        for line in get_lines(self.source):
-            if line.startswith('% '):
-                title = line.split(' ', 1)[1].strip()
-                if '(' and ')' in line:
-                    num = line.split('(')[1][0]
-                    return 'man(%s) %s' % (num, title.split('(')[0].strip())
-                else:
-                    return (title, 1)
-            if re.match('[=]{2}', line):
-                return (title, 1)
-            if re.match('[-]{2}', line):
-                return (title, 2)
+        if self.lines[0].startswith('%'):
+            title = (self.lines[0].split(' ', 1)[1].strip(), 1)
+        elif self.toc:
+            title = self.toc[0]
+        else:
+            title = (self.filename.replace('_', ' '), 1)
+
+        return title
+
+    @property
+    def toc(self):
+        '''
+        Returns a list of tuples representing the headings/sub-heading
+        structure of the page.
+        The first element of each tuple is the heading text.
+        The second element of each tuple is the heading level.
+        '''
+        toc = []
+        heading = ''
+        for line in self.lines:
+            if heading and re.match('[=]{2}', line):
+                toc.append((heading, 1))
+                continue
+            if heading and re.match('[-]{2}', line):
+                toc.append((heading, 2))
+                continue
             if re.match('#+.+[A-z|0-9]', line):
                 level = len(re.match('#+', line).group())
-                return (line.split(' ', 1)[1].strip(), level)
-            title = line.strip()
-
-        title = self.filename.replace('_', ' ')
+                toc.append((line.split(' ', 1)[1].strip(), level))
+                continue
+            heading = line.strip()
 
-        return (title, 1)
+        return toc
 
     @property
     def htmlfile(self):