X-Git-Url: https://git.phdru.name/?a=blobdiff_plain;ds=inline;f=reindex_blog.py;h=ec6adc05a227b99c1e4bda9fc1a048686dfeff64;hb=98cddf6d41746dc555d879d664824c62b8761c4f;hp=fd82c1839111ae59ddcd7a7e86eace33d7b9a715;hpb=d68f2dd8fdab1ec22049d14f274ec34a2ed8431f;p=phdru.name%2Fphdru.name.git
diff --git a/reindex_blog.py b/reindex_blog.py
index fd82c18..ec6adc0 100755
--- a/reindex_blog.py
+++ b/reindex_blog.py
@@ -1,12 +1,8 @@
#! /usr/bin/env python
# -*- coding: koi8-r -*-
-__version__ = "$Revision$"[11:-2]
-__revision__ = "$Id$"[5:-2]
-__date__ = "$Date$"[7:-2]
-__author__ = "Oleg Broytman "
-__copyright__ = "Copyright (C) 2006-2010 PhiloSoft Design"
-
+__author__ = "Oleg Broytman "
+__copyright__ = "Copyright (C) 2006-2012 PhiloSoft Design"
import sys, os
@@ -39,9 +35,8 @@ else:
blog = {}
years = {}
-# excerpts nd bodies are dictionaries mapping file => excerpt/body
+# bodies is a dictionary mapping file => body
-excerpts = {}
bodies = {}
# Walk the directory recursively
@@ -56,7 +51,7 @@ for dirpath, dirs, files in os.walk(blog_root):
template = Template(file=fullpath)
title_parts = template.Title.split()
title = ' '.join(title_parts[6:])
- lead = getattr(template, "Lead", None)
+ lead = template.Lead
tags = template.Tag
if isinstance(tags, basestring):
@@ -84,8 +79,7 @@ for dirpath, dirs, files in os.walk(blog_root):
file = file[:-len("tmpl")] + "html"
key = (year, month, day, file)
- excerpts[key] = template.get_first_p()
- bodies[key] = template.Body
+ bodies[key] = template.body()
# Need to save the blog?
if blog <> old_blog:
@@ -130,7 +124,8 @@ def write_template(level, year, month, day, titles, tags=None):
new_text = ["""\
## THIS FILE IS AUTOMATICALLY GENERATED. DO NOT EDIT.
-#extends phd_pp_ru
+#encoding koi8-r
+#extends phd_site
#implements respond
"""]
@@ -182,7 +177,7 @@ def write_template(level, year, month, day, titles, tags=None):
if len(titles) == 1:
new_text.append("""\
-#attr $refresh = "0; URL=%s"
+#attr $Refresh = "0; URL=%s"
""" % titles[0][3])
new_text.append("""\
@@ -220,15 +215,11 @@ def write_template(level, year, month, day, titles, tags=None):
else:
new_text.append('\n%s %s
' % (day, months_names_ru[int(month)]))
save_date = year, month, day
- if lead:
- lead = lead + ' '
- else:
- lead = ''
new_text.append('''
%s%s.
-''' % (lead, href, title))
+''' % (lead+' ' if lead else '', href, title))
if level == 0:
new_text.append("""
@@ -264,7 +255,7 @@ def write_template(level, year, month, day, titles, tags=None):
''')
max_year = int(sorted(years.keys())[-1])
- years = range(2005, max_year+1)
+ years = range(max_year, 2005, -1)
new_text.append('''
ðÏ ÇÏÄÁÍ:
@@ -287,7 +278,7 @@ def write_template(level, year, month, day, titles, tags=None):
new_text.append("""\
#end def
-$phd_pp_ru.respond(self)
+$phd_site.respond(self)
""")
write_if_changed(index_name, ''.join(new_text))
@@ -340,7 +331,8 @@ write_template(0, year, month, day, all_titles[-20:], all_tags)
new_text = ["""\
## THIS FILE IS AUTOMATICALLY GENERATED. DO NOT EDIT.
-#extends phd_pp_ru
+#encoding koi8-r
+#extends phd_site
#implements respond
#attr $Title = "Oleg Broytman's blog: tags"
#attr $Description = "Broytman Russian Blog Tags Index Document"
@@ -371,7 +363,8 @@ for i, (count, tag, links) in enumerate(all_tags):
tag_text = ["""\
## THIS FILE IS AUTOMATICALLY GENERATED. DO NOT EDIT.
-#extends phd_pp_ru
+#encoding koi8-r
+#extends phd_site
#implements respond
#attr $Title = "Oleg Broytman's blog: tag %s"
#attr $Description = "Broytman Russian Blog Tag %s Index Document"
@@ -407,12 +400,8 @@ for i, (count, tag, links) in enumerate(all_tags):
count = 0
for year, month, day, filename, title, lead in reversed(links):
- if lead:
- lead = lead + ' '
- else:
- lead = ''
link = "../%s/%s/%s/%s" % (year, month, day, filename)
- item_text = """
%s/%s/%s: %s%s""" % (link, year, month, day, lead, title)
+ item_text = """%s/%s/%s: %s%s""" % (link, year, month, day, lead+' ' if lead else '', title)
count += 1
if count <= 5:
@@ -424,7 +413,7 @@ for i, (count, tag, links) in enumerate(all_tags):
#end def
-$phd_pp_ru.respond(self)
+$phd_site.respond(self)
""")
write_if_changed(os.path.join(blog_root, "tags", tag+".tmpl"), ''.join(tag_text))
@@ -432,37 +421,118 @@ new_text.append("""\
#end def
-$phd_pp_ru.respond(self)
+$phd_site.respond(self)
""")
write_if_changed(os.path.join(blog_root, "tags", "index.tmpl"), ''.join(new_text))
+from HTMLParser import HTMLParseError
+import cgi
+from urlparse import urljoin
+from m_lib.net.www.html import HTMLParser as _HTMLParser
+
+class HTMLDone(Exception): pass
+
+
+class FirstPHTMLParser(_HTMLParser):
+ def __init__(self):
+ _HTMLParser.__init__(self)
+ self.first_p = None
+
+ def start_p(self, attrs):
+ self.accumulator = ''
+
+ def end_p(self):
+ self.first_p = self.accumulator + '
'
+ raise HTMLDone()
+
+def get_first_p(body):
+ parser = FirstPHTMLParser()
+
+ try:
+ parser.feed(body)
+ except (HTMLParseError, HTMLDone):
+ pass
+
+ try:
+ parser.close()
+ except (HTMLParseError, HTMLDone):
+ pass
+
+ return parser.first_p
+
+
+class AbsURLHTMLParser(_HTMLParser):
+ def __init__(self, base):
+ _HTMLParser.__init__(self)
+ self.base = base
+
+ def start_a(self, attrs):
+ self.accumulator += ''
+
+ def end_a(self):
+ self.accumulator += ''
+
+ def start_img(self, attrs):
+ self.accumulator += '
'
+
+ def end_img(self):
+ pass
+
+def absolute_urls(body, base):
+ parser = AbsURLHTMLParser(base)
+
+ try:
+ parser.feed(body)
+ except HTMLParseError:
+ pass
+
+ try:
+ parser.close()
+ except HTMLParseError:
+ pass
+
+ return parser.accumulator
+
+
from atom_10 import atom_10
from rss_20 import rss_20
from news import NewsItem
if blog_root:
- baseURL = "http://phd.pp.ru/%s/" % blog_root
+ baseURL = "http://phdru.name/%s/" % blog_root
else:
- baseURL = "http://phd.pp.ru/"
+ baseURL = "http://phdru.name/"
items = []
for item in tuple(reversed(all_titles_tags))[:10]:
year, month, day, file, title, lead, tags = item
- if lead:
- lead = lead + ' '
- else:
- lead = ''
+ url_path = "%s/%s/%s/%s" % (year, month, day, file)
item = NewsItem(
"%s-%s-%s" % (year, month, day),
- "%s%s" % (lead, title),
- "%s/%s/%s/%s" % (year, month, day, file)
- )
+ "%s%s" % (lead+' ' if lead else '', title),
+ url_path)
items.append(item)
item.baseURL = baseURL
item.categoryList = tags
- item.excerpt = excerpts[(year, month, day, file)]
- item.body = bodies[(year, month, day, file)]
+ body = bodies[(year, month, day, file)]
+ body = absolute_urls(body, baseURL + url_path)
+ item.body = body
+ item.excerpt = get_first_p(body)
namespace = {
"title": "Oleg Broytman's blog",
@@ -471,7 +541,7 @@ namespace = {
"description": "",
"lang": "ru",
"author": "Oleg Broytman",
- "email": "phd@phd.pp.ru",
+ "email": "phd@phdru.name",
"generator": os.path.basename(sys.argv[0]),
"posts": items,
}