X-Git-Url: https://git.phdru.name/?a=blobdiff_plain;f=reindex_blog.py;h=197c1c6956db3b3972f1a41a01791b607a3fc1b8;hb=f1774f010b40c9f147e069c363bd0e88dcccdfe0;hp=4451baa740e25f20b60be92b3699fc69e4ca2413;hpb=b4c1e4de6a50d04c81519eaf3dc440cca1e38da4;p=phdru.name%2Fphdru.name.git
diff --git a/reindex_blog.py b/reindex_blog.py
index 4451baa..197c1c6 100755
--- a/reindex_blog.py
+++ b/reindex_blog.py
@@ -1,15 +1,15 @@
-#! /usr/local/bin/python -O
+#! /usr/bin/env python
# -*- coding: koi8-r -*-
-__version__ = "$Revision$"[11:-2]
-__revision__ = "$Id$"[5:-2]
-__date__ = "$Date$"[7:-2]
-__author__ = "Oleg BroytMann "
-__copyright__ = "Copyright (C) 2006 PhiloSoft Design"
-
+__author__ = "Oleg Broytman "
+__copyright__ = "Copyright (C) 2006-2013 PhiloSoft Design"
import sys, os
+blog_data_root = sys.argv[1]
+blog_root = sys.argv[2]
+blog_filename = os.path.join(blog_data_root, "blog_dict.pickle")
+
try:
import cPickle as pickle
except ImportError:
@@ -20,7 +20,6 @@ from Cheetah.Template import Template
# Load old blog
-blog_filename = sys.argv[1]
try:
blog_file = open(blog_filename, "rb")
except IOError:
@@ -36,27 +35,31 @@ else:
blog = {}
years = {}
+# bodies is a dictionary mapping file => body
+
+bodies = {}
+
# Walk the directory recursively
-for dirpath, dirs, files in os.walk(os.curdir):
+for dirpath, dirs, files in os.walk(blog_root):
d = os.path.basename(dirpath)
if not d.startswith("20") and not d.isdigit():
continue
for file in files:
- # Ignore index.tmpl and *.html files; supose all other files are *.tmpl
- if file == "index.tmpl" or file.endswith(".html"):
+ if not file.endswith(".tmpl"):
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 = getattr(template, "Tag", None)
+ 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.split(os.sep)[1:])
+ key = year, month, day = tuple(dirpath[len(blog_root):].split(os.sep)[1:])
if key in blog:
days = blog[key]
else:
@@ -75,6 +78,12 @@ for dirpath, dirs, files in os.walk(os.curdir):
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:
@@ -82,42 +91,32 @@ if blog <> old_blog:
pickle.dump(blog, blog_file, pickle.HIGHEST_PROTOCOL)
blog_file.close()
-
# Localized month names
import locale
-locale.setlocale(locale.LC_ALL, '')
+locale.setlocale(locale.LC_ALL, "ru_RU.KOI8-R")
from calendar import _localized_day, _localized_month
locale.setlocale(locale.LC_TIME, 'C')
months_names_en = list(_localized_month('%B'))
months_abbrs_en = list(_localized_month('%b'))
-locale.setlocale(locale.LC_TIME, '')
-months_names_ru = [month.lower() for month in _localized_month('%B')]
+locale.setlocale(locale.LC_TIME, "ru_RU.KOI8-R")
+#months_names_ru = list(_localized_month('%B'))
+
+months_names_ru = ['', "ÑÎ×ÁÒÑ", "ÆÅ×ÒÁÌÑ", "ÍÁÒÔÁ", "ÁÐÒÅÌÑ", "ÍÁÑ", "ÉÀÎÑ",
+ "ÉÀÌÑ", "Á×ÇÕÓÔÁ", "ÓÅÎÔÑÂÒÑ", "ÏËÔÑÂÒÑ", "ÎÏÑÂÒÑ", "ÄÅËÁÂÒÑ"
+]
months_names_ru0 = ['', "ÑÎ×ÁÒØ", "ÆÅ×ÒÁÌØ", "ÍÁÒÔ", "ÁÐÒÅÌØ", "ÍÁÊ", "ÉÀÎØ",
"ÉÀÌØ", "Á×ÇÕÓÔ", "ÓÅÎÔÑÂÒØ", "ÏËÔÑÂÒØ", "ÎÏÑÂÒØ", "ÄÅËÁÂÒØ"
]
+from news import write_if_changed
-def write_if_changed(filename, new_text):
- try:
- infile = open(filename, 'r')
- old_text = infile.read()
- infile.close()
- except IOError:
- old_text = None
-
- if old_text <> new_text:
- print "Writing", filename
- outfile = open(filename, 'w')
- outfile.write(new_text)
- outfile.close()
-
-def write_template(level, year, month, day, titles):
- path = []
+def write_template(level, year, month, day, titles, tags=None):
+ path = [blog_root]
if level >= 1:
path.append(year)
if level >= 2:
@@ -129,39 +128,47 @@ def write_template(level, year, month, day, titles):
new_text = ["""\
## THIS FILE IS AUTOMATICALLY GENERATED. DO NOT EDIT.
-#extends phd_pp_ru
+#encoding koi8-r
+#extends phd_site
#implements respond
"""]
if level == 0:
new_text.append("""\
-#attr $Title = "Oleg BroytMann's blog"
-#attr $Description = "BroytMann Russian Blog Index Document"
+#attr $Title = "Oleg Broytman's blog"
+#attr $Description = "Broytman Russian Blog Index Document"
#attr $Copyright = %(cyear)s
+#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
-öÕÒÎÁÌ
+öÕÒÎÁÌ
""" % {"cyear": year or 2005})
elif level == 1:
new_text.append("""\
-#attr $Title = "Oleg BroytMann's blog: %(year)s"
-#attr $Description = "BroytMann Russian Blog %(year)s Index Document"
+#attr $Title = "Oleg Broytman's blog: %(year)s"
+#attr $Description = "Broytman Russian Blog %(year)s Index Document"
#attr $Copyright = %(cyear)s
##
#def body_html
-öÕÒÎÁÌ: %(year)s
+öÕÒÎÁÌ: %(year)s
""" % {"year": year, "cyear": year or 2005})
elif level == 2:
imonth = int(month)
new_text.append("""\
-#attr $Title = "Oleg BroytMann's blog: %(month_abbr_en)s %(year)s"
-#attr $Description = "BroytMann Russian Blog %(month_name_en)s %(year)s Index Document"
+#attr $Title = "Oleg Broytman's blog: %(month_abbr_en)s %(year)s"
+#attr $Description = "Broytman Russian Blog %(month_name_en)s %(year)s Index Document"
#attr $Copyright = %(cyear)s
##
#def body_html
-öÕÒÎÁÌ: %(month_name_ru0)s %(year)s
+öÕÒÎÁÌ: %(month_name_ru0)s %(year)s
""" % {
"year": year, "cyear": year or 2005,
"month_abbr_en": months_abbrs_en[imonth], "month_name_en": months_names_en[imonth],
@@ -172,34 +179,29 @@ def write_template(level, year, month, day, titles):
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("""\
-#attr $Title = "Oleg BroytMann's blog: %(day)d %(month_abbr_en)s %(year)s"
-#attr $Description = "BroytMann Russian Blog %(day)d %(month_name_en)s %(year)s Index Document"
+#attr $Title = "Oleg Broytman's blog: %(day)d %(month_abbr_en)s %(year)s"
+#attr $Description = "Broytman Russian Blog %(day)d %(month_name_en)s %(year)s Index Document"
#attr $Copyright = %(cyear)s
##
#def body_html
-öÕÒÎÁÌ: %(day)d %(month_name_ru0)s %(year)s
+öÕÒÎÁÌ: %(day)d %(month_name_ru)s %(year)s
""" % {
"year": year, "cyear": year or 2005,
"month_abbr_en": months_abbrs_en[imonth], "month_name_en": months_names_en[imonth],
- "month_name_ru0": months_names_ru0[imonth],
+ "month_name_ru": months_names_ru[imonth],
"day": iday
})
save_titles = titles[:]
titles.reverse()
- save_day = None
+ save_date = None
for year, month, day, file, title, lead in titles:
href = []
if level == 0:
@@ -211,33 +213,59 @@ def write_template(level, year, month, day, titles):
href.append(file)
href = '/'.join(href)
if day[0] == '0': day = day[1:]
- if save_day <> day:
+ if save_date <> (year, month, day):
if level == 0:
new_text.append('\n%s %s %s
' % (day, months_names_ru[int(month)], year))
else:
new_text.append('\n%s %s
' % (day, months_names_ru[int(month)]))
- save_day = day
- if lead:
- lead = lead + ' '
- else:
- lead = ''
+ save_date = year, month, day
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 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.
+
+""")
+
years = {}
for year, month, day, file, title, lead in save_titles:
years[year] = True
- first_year = True
new_text.append('''
-
+ôÅÇÉ:
+''')
+ first_tag = True
+ for count, tag, links in all_tags:
+ if first_tag:
+ first_tag = False
+ else:
+ new_text.append(' - ')
+ new_text.append("""%s (%d)""" % (tag, tag, count))
+ new_text.append('''
+
+''')
+
+ max_year = int(sorted(years.keys())[-1])
+ years = range(max_year, 2005, -1)
-
+ new_text.append('''
+
ðÏ ÇÏÄÁÍ:
''')
- for year in sorted(years.keys()):
+ first_year = True
+ for year in years:
if first_year:
first_year = False
else:
@@ -247,16 +275,22 @@ def write_template(level, year, month, day, titles):
''')
+ new_text.append("""
+
+öö
+""")
+
new_text.append("""\
#end def
-$phd_pp_ru.respond(self)
+$phd_site.respond(self)
""")
write_if_changed(index_name, ''.join(new_text))
-all_titles = []
all_tags = {}
+all_titles = []
+all_titles_tags = []
for year in sorted(years.keys()):
year_titles = []
@@ -270,6 +304,7 @@ for year in sorted(years.keys()):
for file, title, lead, tags in blog[key]:
if file.endswith(".tmpl"): file = file[:-len("tmpl")] + "html"
value = (year, month, day, file, title, lead)
+ all_titles_tags.append((year, month, day, file, title, lead, tags))
all_titles.append(value)
year_titles.append(value)
month_titles.append(value)
@@ -283,54 +318,94 @@ for year in sorted(years.keys()):
write_template(3, year, month, day, day_titles)
write_template(2, year, month, day, month_titles)
write_template(1, year, month, day, year_titles)
-write_template(0, year, month, day, all_titles[-20:])
+
+def by_count_rev_tag_link(t1, t2):
+ """Sort all_tags by count in descending order,
+ and by tags and links in ascending order
+ """
+ r = cmp(t1[0], t2[0])
+ if r:
+ return -r
+ return cmp((t1[1], t1[2]), (t2[1], t2[2]))
all_tags = [(len(links), tag, links) for (tag, links) in all_tags.items()]
-all_tags.sort()
+all_tags.sort(by_count_rev_tag_link)
+
+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 BroytMann's blog: tags"
-#attr $Description = "BroytMann Russian Blog Tags Index Document"
+#attr $Title = "Oleg Broytman's blog: tags"
+#attr $Description = "Broytman Russian Blog Tags Index Document"
#attr $Copyright = 2006
##
#def body_html
-
ôÅÇÉ
+ôÅÇÉ
"""]
-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 BroytMann's blog: tag %s"
-#attr $Description = "BroytMann Russian Blog Tag %s Index Document"
+#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 $Copyright = 2006
##
#def body_html
-%s
+%s
-""" % (tag, tag, tag)]
+""" % tag)
count = 0
for year, month, day, filename, title, lead in reversed(links):
- if lead:
- lead = lead + ' '
- else:
- lead = ''
link = "../%s/%s/%s/%s" % (year, month, day, filename)
- item_text = """- %s/%s/%s: %s%s
""" % (link, year, month, day, lead, title)
+ item_text = """- %s/%s/%s: %s%s
""" % (link, year, month, day, lead+' ' if lead else '', title)
count += 1
if count <= 5:
@@ -342,14 +417,162 @@ for count, tag, links in all_tags:
#end def
-$phd_pp_ru.respond(self)
+$phd_site.respond(self)
""")
- write_if_changed(os.path.join("tags", tag+".tmpl"), ''.join(tag_text))
+ write_if_changed(os.path.join(blog_root, "tags", tag+".tmpl"), ''.join(tag_text))
new_text.append("""\
#end def
-$phd_pp_ru.respond(self)
+$phd_site.respond(self)
""")
-write_if_changed(os.path.join("tags", "index.tmpl"), ''.join(new_text))
+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://phdru.name/%s/" % blog_root
+else:
+ baseURL = "http://phdru.name/"
+
+items = []
+for item in tuple(reversed(all_titles_tags))[:10]:
+ year, month, day, file, title, lead, tags = item
+ 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+' ' 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)
+ body = body.decode('koi8-r').encode('utf-8')
+ item.body = body
+ item.excerpt = get_first_p(body)
+
+namespace = {
+ "title": "Oleg Broytman's blog",
+ "baseURL": baseURL,
+ "indexFile": "",
+ "description": "",
+ "lang": "ru",
+ "author": "Oleg Broytman",
+ "email": "phd@phdru.name",
+ "generator": os.path.basename(sys.argv[0]),
+ "posts": items,
+}
+
+# For english dates
+locale.setlocale(locale.LC_TIME, 'C')
+
+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 = 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)