X-Git-Url: https://git.phdru.name/?a=blobdiff_plain;f=reindex_blog.py;h=33e9077aefd0e6472959c3300d1f105ea59ef6b0;hb=f95205f4ff896df01ca7c9c3d672147408e0b119;hp=92a02272ea79c60a9f79b0ae21fe8733d6fa47f2;hpb=6d775fb85dad6e10928e0e418645124a804a4275;p=phdru.name%2Fphdru.name.git
diff --git a/reindex_blog.py b/reindex_blog.py
index 92a0227..33e9077 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 PhiloSoft Design"
-
+__author__ = "Oleg Broytman "
+__copyright__ = "Copyright (C) 2006-2014 PhiloSoft Design"
import sys, os
@@ -39,6 +35,10 @@ else:
blog = {}
years = {}
+# bodies is a dictionary mapping file => body
+
+bodies = {}
+
# Walk the directory recursively
for dirpath, dirs, files in os.walk(blog_root):
d = os.path.basename(dirpath)
@@ -49,13 +49,14 @@ for dirpath, dirs, files in os.walk(blog_root):
continue
fullpath = os.path.join(dirpath, file)
template = Template(file=fullpath)
- title_parts = template.Title.split()
+ title_parts = template.Title.decode('utf-8').encode('koi8-r').split()
title = ' '.join(title_parts[6:])
- lead = getattr(template, "Lead", None)
+ lead = template.Lead.decode('utf-8').encode('koi8-r')
tags = template.Tag
if isinstance(tags, basestring):
tags = (tags,)
+ tags = [tag.decode('utf-8').encode('koi8-r') for tag in tags]
if title:
key = year, month, day = tuple(dirpath[len(blog_root):].split(os.sep)[1:])
@@ -77,6 +78,12 @@ for dirpath, dirs, files in os.walk(blog_root):
if day not in days: days.append(day)
+ file = file[:-len("tmpl")] + "html"
+ key = (year, month, day, file)
+ body = template.body()
+ if isinstance(body, unicode):
+ body = body.encode('koi8-r')
+ bodies[key] = body
# Need to save the blog?
if blog <> old_blog:
@@ -121,7 +128,8 @@ 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
"""]
@@ -130,8 +138,12 @@ def write_template(level, year, month, day, titles, tags=None):
#attr $Title = "Oleg Broytman's blog"
#attr $Description = "Broytman Russian Blog Index Document"
#attr $Copyright = %(cyear)s
-#attr $alternates = (("application/atom+xml", "News [Atom 1.0]", "atom_10.xml"),
- ("application/rss+xml", "News [RSS 2.0]", "rss_20.xml")
+#attr $alternates = (("îÏ×ÏÓÔÉ [Atom 1.0] ÔÏÌØËÏ ÚÁÇÏÌÏ×ËÉ", "application/atom+xml", "atom_10_titles.xml"),
+ ("îÏ×ÏÓÔÉ [Atom 1.0]", "application/atom+xml", "atom_10.xml"),
+ ("îÏ×ÏÓÔÉ [Atom 1.0] ÐÏÌÎÙÅ ÔÅËÓÔÙ", "application/atom+xml", "atom_10_full.xml"),
+ ("îÏ×ÏÓÔÉ [RSS 2.0] ÔÏÌØËÏ ÚÁÇÏÌÏ×ËÉ", "application/rss+xml", "rss_20_titles.xml"),
+ ("îÏ×ÏÓÔÉ [RSS 2.0]", "application/rss+xml", "rss_20.xml"),
+ ("îÏ×ÏÓÔÉ [RSS 2.0] ÐÏÌÎÙÅ ÔÅËÓÔÙ", "application/rss+xml", "rss_20_full.xml"),
)
##
#def body_html
@@ -167,14 +179,9 @@ def write_template(level, year, month, day, titles, tags=None):
iday = int(day)
imonth = int(month)
- new_text.append("""\
-#attr $Next = "%s"
-""" % titles[0][3])
-
-
if len(titles) == 1:
new_text.append("""\
-#attr $refresh = "0; URL=%s"
+#attr $Refresh = "0; URL=%s"
""" % titles[0][3])
new_text.append("""\
@@ -212,23 +219,25 @@ def write_template(level, year, month, day, titles, tags=None):
else:
new_text.append('\n%s %s
' % (day, months_names_ru[int(month)]))
save_date = year, month, day
- if lead:
- lead = lead + ' '
- else:
- lead = ''
new_text.append('''
%s%s.
-''' % (lead, href, title))
+''' % (lead+' ' if lead else '', href, title))
if level == 0:
new_text.append("""
îÏ×ÏÓÔÅ×ÁÑ ÌÅÎÔÁ × ÆÏÒÍÁÔÁÈ
-Atom 1.0
-É RSS 2.0 .
+
+Atom 1.0 ÔÏÌØËÏ ÚÁÇÏÌÏ×ËÉ /
+Atom 1.0 /
+Atom 1.0 ÐÏÌÎÙÅ ÔÅËÓÔÙ
+É
+RSS 2.0 ÔÏÌØËÏ ÚÁÇÏÌÏ×ËÉ /
+RSS 2.0 /
+RSS 2.0 ÐÏÌÎÙÅ ÔÅËÓÔÙ.
""")
@@ -250,18 +259,25 @@ def write_template(level, year, month, day, titles, tags=None):
''')
max_year = int(sorted(years.keys())[-1])
- years = range(2005, max_year+1)
+ years = range(max_year, 2005, -1)
new_text.append('''
ðÏ ÇÏÄÁÍ:
''')
+
+ year_counts = {}
+ for year, month, day, file, title, lead in all_titles:
+ year_counts[year] = 0
+ for year, month, day, file, title, lead in all_titles:
+ year_counts[year] += 1
+
first_year = True
for year in years:
if first_year:
first_year = False
else:
new_text.append(' - ')
- new_text.append('%s' % (year, year))
+ new_text.append('%s (%d)' % (year, year, year_counts[str(year)]))
new_text.append('''
''')
@@ -273,7 +289,7 @@ def write_template(level, year, month, day, titles, tags=None):
new_text.append("""\
#end def
-$phd_pp_ru.respond(self)
+$phd_site.respond(self)
""")
write_if_changed(index_name, ''.join(new_text))
@@ -326,7 +342,8 @@ 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"
@@ -335,38 +352,88 @@ new_text = ["""\
#def body_html
ôÅÇÉ
-
+
+æÏÒÍÁ ÐÏÉÓËÁ ÐÏÚ×ÏÌÑÅÔ ÉÓËÁÔØ ÓÏÏÂÝÅÎÉÑ × ÂÌÏÇÅ, ÓÏÏÔ×ÅÔÓÔ×ÕÀÝÉÅ ×ÙÒÁÖÅÎÉÀ.
+óÉÎÔÁËÓÉÓ ×ÙÒÁÖÅÎÉÑ:
+
+ - ôÅÇ - ÅÓÌÉ ÔÁËÏÊ ÔÅÇ ÓÕÝÅÓÔ×ÕÅÔ, ÐÒÏÉÚÏÊÄ£Ô ÐÅÒÅÎÁÐÒÁ×ÌÅÎÉÅ ÎÁ ÓÔÒÁÎÉÃÕ ÔÅÇÁ.
+ - ïÐÅÒÁÔÏÒ '!' (NOT, not) - ÉÝÅÔ ÚÁÐÉÓÉ, × ËÏÔÏÒÙÈ ÎÅÔ ÜÔÏÇÏ ÔÅÇÁ.
+ - ïÐÅÒÁÔÏÒ '&' (AND, and) - ÉÝÅÔ ÚÁÐÉÓÉ, × ËÏÔÏÒÙÈ ÅÓÔØ ÏÂÁ ÔÅÇÁ.
+ - ïÐÅÒÁÔÏÒ '|' (OR, or) - ÉÝÅÔ ÚÁÐÉÓÉ, × ËÏÔÏÒÙÈ ÅÓÔØ ÌÀÂÙÅ ÉÚ ÔÅÇÏ×.
+ - óËÏÂËÉ '()' ÐÏÚ×ÏÌÑÀÔ ÏÂßÅÄÉÎÑÔØ ×ÙÒÁÖÅÎÉÑ.
+
+
+ðÒÉÍÅÒÙ ×ÙÒÁÖÅÎÉÊ: linux - ÐÒÏÉÚÏÊÄ£Ô ÐÅÒÅÎÁÐÒÁ×ÌÅÎÉÅ
+ÎÁ ÓÔÒÁÎÉÃÕ linux.html; linux&!audio - ÉÓËÁÔØ ÚÁÐÉÓÉ × ËÏÔÏÒÙÈ ÅÓÔØ ÔÅÇ
+linux É ÎÅÔ ÔÅÇÁ audio; linux and not audio - ÔÏ ÖÅ ÓÁÍÏÅ.
+
+
+
+
+
+
"""]
-for count, tag, links in all_tags:
+for i, (count, tag, links) in enumerate(all_tags):
new_text.append("""\
- %s (%d)
""" % (tag, tag, count))
+ first = all_tags[0][1]
+ if i == 0:
+ prev = None
+ else:
+ prev = all_tags[i-1][1]
+ if i >= len(all_tags)-1:
+ next = None
+ else:
+ next = all_tags[i+1][1]
+ last = all_tags[-1][1]
+
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"
+""" % (tag, tag)]
+
+ tag_text.append("""\
+#attr $First = "%s"
+""" % first)
+
+ if prev:
+ tag_text.append("""\
+#attr $Prev = "%s"
+""" % prev)
+
+ if next:
+ tag_text.append("""\
+#attr $Next = "%s"
+""" % next)
+
+ tag_text.append("""\
+#attr $Last = "%s"
+""" % last)
+
+ tag_text.append("""\
#attr $Copyright = 2006
##
#def body_html
%s
-
-""" % (tag, tag, tag)]
+""" % tag)
count = 0
for year, month, day, filename, title, lead in reversed(links):
- if lead:
- lead = lead + ' '
- else:
- lead = ''
link = "../%s/%s/%s/%s" % (year, month, day, filename)
- item_text = """- %s/%s/%s: %s%s
""" % (link, year, month, day, lead, title)
+ item_text = """- %s/%s/%s: %s%s
""" % (link, year, month, day, lead+' ' if lead else '', title)
count += 1
if count <= 5:
@@ -376,45 +443,128 @@ for count, tag, links in all_tags:
tag_text.append("""\
-
#end def
-$phd_pp_ru.respond(self)
+$phd_site.respond(self)
""")
write_if_changed(os.path.join(blog_root, "tags", tag+".tmpl"), ''.join(tag_text))
new_text.append("""\
-
#end def
-$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 = ''
+ lead = lead.decode('koi8-r').encode('utf-8')
+ title = title.decode('koi8-r').encode('utf-8')
+ 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
+ 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",
@@ -423,7 +573,7 @@ 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,
}
@@ -431,7 +581,23 @@ namespace = {
# For english dates
locale.setlocale(locale.LC_TIME, 'C')
-atom_tmpl = str(atom_10(searchList=[namespace]))
+atom_tmpl = unicode(atom_10(searchList=[namespace])).encode('koi8-r')
write_if_changed(os.path.join(blog_root, "atom_10.xml"), atom_tmpl)
-rss_tmpl = str(rss_20(searchList=[namespace]))
+rss_tmpl = unicode(rss_20(searchList=[namespace])).encode('koi8-r')
write_if_changed(os.path.join(blog_root, "rss_20.xml"), rss_tmpl)
+
+for item in items:
+ item.excerpt = None
+
+atom_tmpl = unicode(atom_10(searchList=[namespace])).encode('koi8-r')
+write_if_changed(os.path.join(blog_root, "atom_10_titles.xml"), atom_tmpl)
+rss_tmpl = unicode(rss_20(searchList=[namespace])).encode('koi8-r')
+write_if_changed(os.path.join(blog_root, "rss_20_titles.xml"), rss_tmpl)
+
+for item in items:
+ item.content = item.body
+
+atom_tmpl = unicode(atom_10(searchList=[namespace])).encode('koi8-r')
+write_if_changed(os.path.join(blog_root, "atom_10_full.xml"), atom_tmpl)
+rss_tmpl = unicode(rss_20(searchList=[namespace])).encode('koi8-r')
+write_if_changed(os.path.join(blog_root, "rss_20_full.xml"), rss_tmpl)