summaryrefslogtreecommitdiff
path: root/panfry/document.py
diff options
context:
space:
mode:
Diffstat (limited to 'panfry/document.py')
-rwxr-xr-xpanfry/document.py85
1 files changed, 69 insertions, 16 deletions
diff --git a/panfry/document.py b/panfry/document.py
index 2b1cf01..e03aca1 100755
--- a/panfry/document.py
+++ b/panfry/document.py
@@ -1,20 +1,24 @@
#!/usr/bin/env python
# -*- coding: utf-8 -*-
+from __future__ import print_function
import os
import re
import pandoc
from tempfile import NamedTemporaryFile
from panfry.util import *
from panfry.page import TextFile, Page
+from panfry.toc import TOCItem
class Document:
- def __init__(self, path):
- self.src_path = os.path.join(path, 'src')
+ def __init__(self, path, subdoc=''):
+ self.src_path = os.path.join(path, 'src', subdoc)
self.meta = self.get_meta(self.src_path)
self.pages = self.get_pages(self.src_path)
self.full_toc = True
+ self.standalone = True
+ self.json_toc = False
self.css_file = 'css/style.css'
self.assets_dir = 'assets'
@@ -24,6 +28,11 @@ class Document:
def set_simple_toc(self):
self.full_toc = False
+ @property
+ def set_json_toc(self):
+ self.json_toc = True
+ self.standalone = False
+
def set_pandoc_options(self, options):
self.pandoc_options = options
@@ -62,18 +71,18 @@ class Document:
for page in self.pages:
for tocitem in page.toc:
if tocitem.heading == heading.replace('\n', ' '):
- return '[%s](%s#%s)' % (heading,
- page.htmlfile,
- tocitem.header_id,
- )
+ return '[%s](%s%s)' % (heading,
+ page.htmlfile,
+ tocitem.header_id,
+ )
return '[%s]' % ref_txt
@property
def title(self):
- if self.header.firstline.startswith('%'):
- title = self.header.firstline.split('%')[1].strip()
- for line in self.header.lines[1:]:
+ if self.title_block.firstline.startswith('%'):
+ title = self.title_block.firstline.split('%')[1].strip()
+ for line in self.title_block.lines[1:]:
if line.startswith('%'):
break
title = '%s %s' % (title, line.strip())
@@ -98,7 +107,7 @@ class Document:
return pages
def get_meta(self, path):
- metafiles = ['TOC', 'HEADER']
+ metafiles = ['TOC', 'TITLE_BLOCK']
meta = {}
for filename in metafiles:
source = read_file(os.path.join(path, filename))
@@ -143,7 +152,7 @@ class Document:
toc = [page.toc[0]]
for tocitem in toc:
- link = '%s#%s' % (page.htmlfile, tocitem.header_id)
+ link = '%s%s' % (page.htmlfile, tocitem.header_id)
links.append(dict(link=unicode(link, "utf8"),
text=unicode(tocitem.heading, "utf8"),
level=tocitem.level,
@@ -152,6 +161,37 @@ class Document:
return links
@property
+ def toc_json(self):
+ lastchild = top = toc = TOCItem(self.title, 1, 'title_block.html')
+
+ curlevel = 0
+ for page in self.pages:
+ heading = ''
+ for line in page.markdown.split('\n'):
+ line = line.strip()
+ if heading and re.match('[=]{2}', line):
+ level = 1
+ elif heading and re.match('[-]{2}', line):
+ level = 2
+ elif re.match('#+.+[A-z|0-9]', line):
+ level = len(re.match('#+', line).group())
+ heading = line.split(' ', 1)[1].strip()
+ else:
+ heading = line.strip()
+ continue
+
+ if curlevel < level:
+ toc = lastchild
+ curlevel = toc.level
+ while toc.parent and curlevel >= level:
+ toc = toc.parent
+ curlevel = toc.level
+
+ lastchild = toc.add_child(heading, level, page.htmlfile)
+
+ return top.json
+
+ @property
def filename_base(self):
filename = self.title
filename = re.sub(r'\s+', ' ', filename)
@@ -173,7 +213,7 @@ class Document:
def publish_pdf(self, pub_path):
pdf_path = os.path.join(pub_path, self.pdf_filename)
- src = self.header.source
+ src = self.title_block.source
for page in self.pages:
src += '\n%s' % page.source
@@ -191,7 +231,7 @@ class Document:
def publish_epub(self, pub_path):
epub_path = os.path.join(pub_path, self.epub_filename)
- src = self.header.source
+ src = self.title_block.source
for page in self.pages:
src += '\n%s' % page.source
@@ -231,24 +271,37 @@ class Document:
doc.add_argument('toc')
doc.add_argument('template=%s' % template_file.name)
doc.add_argument('css=%s' % self.css_file)
+
for option in self.pandoc_options:
if option == 'number-sections':
continue
doc.add_argument(option)
- doc.markdown = '%s\n%s' % (self.header.source,
- self.expand_int_links(page.markdown))
+
+ if self.standalone:
+ markdown = '%s\n' % self.title_block.source
+ else:
+ markdown = ''
+
+ markdown += self.expand_int_links(page.markdown)
+ doc.markdown = markdown
content = doc.html5
write_file(os.path.join(pub_path, page.htmlfile),
unicode(content, 'utf-8'))
# If there is not explicit index.html, then link 'index.html'
# to the toplevel page.
- if not 'index.md' in self.pages:
+ htmlfiles = [page.htmlfile for page in self.pages]
+ if self.standalone and not 'index.html' in htmlfiles:
src = os.path.join(self.pages[0].htmlfile)
ref = os.path.join(pub_path, 'index.html')
+ print("Linking %s to %s" % (ref, src))
if os.path.exists(ref):
os.remove(ref)
os.symlink(src, ref)
+ if self.json_toc:
+ write_file(os.path.join(pub_path, 'toc.json'),
+ unicode(self.toc_json, 'utf-8'))
+
###### Copy any assets to publish directory
self.publish_assets(pub_path)