-#! /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 <phd@phd.pp.ru>"
-__copyright__ = "Copyright (C) 2006 PhiloSoft Design"
-
+__author__ = "Oleg Broytman <phd@phdru.name>"
+__copyright__ = "Copyright (C) 2006-2024 PhiloSoft Design"
import sys, os
-
try:
- import cPickle as pickle
+ from urllib.parse import quote
except ImportError:
- import pickle
+ from urllib import quote
from Cheetah.Template import Template
+from Cheetah.compat import string_type
+from blog_db import blog_root, load_blog, save_blog
-# Load old blog
+old_blog = load_blog()
-blog_filename = sys.argv[1]
-try:
- blog_file = open(blog_filename, "rb")
-except IOError:
- old_blog = {}
-else:
- old_blog = pickle.load(blog_file)
- blog_file.close()
+# blog is a dictionary mapping
+# (year, month, day) => [list of (file, title, lead, tags)]
-
-# blog is a dictionary mapping (year, month, day) => (filename, title, lead)
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 = template.Tag
+ if isinstance(tags, string_type):
+ 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:
days = blog[key] = []
- days.append((file, title, lead))
+ days.append((file, title, lead, tags))
if year in years:
months = years[year]
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:
- blog_file = open(blog_filename, "wb")
- pickle.dump(blog, blog_file, pickle.HIGHEST_PROTOCOL)
- blog_file.close()
-
+if blog != old_blog:
+ save_blog(blog)
# Localized month names
import locale
-locale.setlocale(locale.LC_ALL, '')
-from calendar import _localized_day, _localized_month
+locale.setlocale(locale.LC_ALL, "ru_RU.KOI8-R")
+from calendar import _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 = ['', "января", "февраля", "марта", "апреля", "мая", "июня",
+ "июля", "августа", "сентября", "октября", "ноября", "декабря"
+]
months_names_ru0 = ['', "январь", "февраль", "март", "апрель", "май", "июнь",
"июль", "август", "сентябрь", "октябрь", "ноябрь", "декабрь"
]
+from news import write_if_changed
+
+def encode_tag(tag):
+ return quote(tag.replace(' ', '_'))
-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:
path.append(day)
path.append("index.tmpl")
index_name = os.path.join(*path)
- try:
- index_tmpl = open(index_name, 'r')
- old_text = index_tmpl.read()
- index_tmpl.close()
- except IOError:
- old_text = None
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 = (("Новости [Atom 1.0] только заголовки", "application/atom+xml", "atom_10_titles.xml"),
+ ("Новости [Atom 1.0]", "application/atom+xml", "atom_10.xml"),
+ ("Новости [Atom 1.0] полные тексты", "application/atom+xml", "atom_10_full.xml"),
+ ("Новости [RSS 2.0] только заголовки", "application/rss+xml", "rss_20_titles.xml"),
+ ("Новости [RSS 2.0]", "application/rss+xml", "rss_20.xml"),
+ ("Новости [RSS 2.0] полные тексты", "application/rss+xml", "rss_20_full.xml"),
+)
##
#def body_html
-<H1>Журнал</H1>
+<h1>Журнал</h1>
""" % {"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
-<H1>Журнал: %(year)s</H1>
+<h1>Журнал: %(year)s</h1>
""" % {"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
-<H1>Журнал: %(month_name_ru0)s %(year)s</H1>
+<h1>Журнал: %(month_name_ru0)s %(year)s</h1>
""" % {
"year": year, "cyear": year or 2005,
"month_abbr_en": months_abbrs_en[imonth], "month_name_en": months_names_en[imonth],
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
-<H1>Журнал: %(day)d %(month_name_ru0)s %(year)s</H1>
+<h1>Журнал: %(day)d %(month_name_ru)s %(year)s</h1>
""" % {
"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
- for year, month, day, filename, title, lead in titles:
+ save_date = None
+ for year, month, day, file, title, lead in titles:
href = []
if level == 0:
href.append(year)
href.append(month)
if level <= 2:
href.append(day)
- href.append(filename)
+ 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<h2>%s %s %s</h2>' % (day, months_names_ru[int(month)], year))
else:
new_text.append('\n<h2>%s %s</h2>' % (day, months_names_ru[int(month)]))
- save_day = day
- if lead:
- lead = lead + ' '
- else:
- lead = ''
+ save_date = year, month, day
new_text.append('''
<p class="head">
%s<a href="%s">%s</a>.
</p>
-''' % (lead, href, title))
+''' % (lead+' ' if lead else '', href, title))
if level == 0:
+ new_text.append("""
+<hr>
+
+<p class="head">Новостевая лента в форматах
+<img src="../../Graphics/atom_10.jpg" border=0>
+<A HREF="atom_10_titles.xml">Atom 1.0 только заголовки</A> /
+<A HREF="atom_10.xml">Atom 1.0</A> /
+<A HREF="atom_10_full.xml">Atom 1.0 полные тексты</A>
+и <img src="../../Graphics/rss_20.jpg" border=0>
+<A HREF="rss_20_titles.xml">RSS 2.0 только заголовки</A> /
+<A HREF="rss_20.xml">RSS 2.0</A> /
+<A HREF="rss_20_full.xml">RSS 2.0 полные тексты</A>.
+</p>
+""")
+
years = {}
- for year, month, day, filename, title, lead in save_titles:
+ for year, month, day, file, title, lead in save_titles:
years[year] = True
- first_year = True
new_text.append('''
-<hr>
+<p class="head"><a href="tags/">Теги</a>:
+''')
+ first_tag = True
+ for count, tag, links in all_tags:
+ if first_tag:
+ first_tag = False
+ else:
+ new_text.append(' - ')
+ new_text.append("""<a href="tags/%s.html">%s (%d)</a>""" % (
+ encode_tag(tag), tag, count))
+ new_text.append('''
+</p>
+''')
+
+ max_year = int(sorted(years.keys())[-1])
+ years = range(max_year, 2005, -1)
-<p class="years">
+ new_text.append('''
+<p class="head">По годам:
''')
- for year in sorted(years.keys()):
+
+ 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('<a href="%s/">%s</a>' % (year, year))
+ new_text.append('<a href="%s/">%s (%d)</a>' % (year, year, year_counts[str(year)]))
new_text.append('''
</p>
''')
+ new_text.append("""
+<hr>
+<p class="head"><a href="http://phd.livejournal.com/">ЖЖ</a>
+""")
+
new_text.append("""\
#end def
-$phd_pp_ru.respond(self)
+$phd_site.respond(self)
""")
- new_text = ''.join(new_text)
- if old_text <> new_text:
- print "Writing", index_name
- index_tmpl = open(index_name, 'w')
- index_tmpl.write(new_text)
- index_tmpl.close()
+ write_if_changed(index_name, ''.join(new_text))
+all_tags = {}
all_titles = []
+all_titles_tags = []
+
for year in sorted(years.keys()):
year_titles = []
months = years[year]
day_titles = []
key = year, month, day
if key in blog:
- for filename, title, lead in blog[key]:
- if filename.endswith(".tmpl"):
- filename = filename[:-len("tmpl")] + "html"
- value = (year, month, day, filename, title, lead)
+ 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)
day_titles.append(value)
+ for tag in tags:
+ if tag in all_tags:
+ tag_links = all_tags[tag]
+ else:
+ tag_links = all_tags[tag] = []
+ tag_links.append(value)
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(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.
+#encoding koi8-r
+#extends phd_site
+#implements respond
+#attr $Title = "Oleg Broytman's blog: tags"
+#attr $Description = "Broytman Russian Blog Tags Index Document"
+#attr $Copyright = 2006
+##
+#def body_html
+<h1>Теги</h1>
+
+<p class="head small">
+Форма поиска позволяет искать сообщения в блоге, соответствующие выражению.
+Синтаксис выражения:</p>
+<ul class="small">
+ <li>Тег - если такой тег существует, произойдёт перенаправление на страницу тега.</li>
+ <li>Оператор '!' (NOT, not) - ищет записи, в которых нет этого тега.</li>
+ <li>Оператор '&' (AND, and) - ищет записи, в которых есть оба тега.</li>
+ <li>Оператор '|' (OR, or) - ищет записи, в которых есть любые из тегов.</li>
+ <li>Приоритет операций стандартный: NOT > AND > OR. Скобки '()' позволяют объединять выражения.</li>
+</ul>
+<p class="small">
+Примеры выражений: linux - произойдёт перенаправление
+на страницу linux.html; linux&!audio - искать записи в которых есть тег
+linux и нет тега audio; linux and not audio - то же самое.
+</p>
+
+<center>
+<form method=GET action="../../../cgi-bin/blog-ru/search-tags/search-tags.py">
+ <input type=text name=q>
+ <input type=submit name=submit value="Искать">
+</form>
+</center>
+
+<dl>
+"""]
+
+for i, (count, tag, links) in enumerate(all_tags):
+ new_text.append("""\
+ <dt><a href="%s.html">%s (%d)</a></dt>
+""" % (encode_tag(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.
+#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 $Copyright = 2006
+##
+#def body_html
+<h1>%s</h1>
+
+<ul>
+""" % tag)
+
+ count = 0
+ for year, month, day, filename, title, lead in reversed(links):
+ link = "../%s/%s/%s/%s" % (year, month, day, filename)
+ item_text = """<li><a href="%s">%s/%s/%s: %s%s</a></li>""" % (link, year, month, day, lead+' ' if lead else '', title)
+
+ count += 1
+ if count <= 5:
+ new_text.append(" <dd>%s</dd>\n" % item_text)
+
+ tag_text.append(" %s\n" % item_text)
+
+ tag_text.append("""\
+</ul>
+#end def
+$phd_site.respond(self)
+""")
+ write_if_changed(os.path.join(blog_root, "tags",
+ tag.replace(' ', '_') + ".tmpl"),
+ ''.join(tag_text))
+
+new_text.append("""\
+</dl>
+#end def
+$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 = '<p>'
+
+ def end_p(self):
+ self.first_p = self.accumulator + '</p>'
+ 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 += '<a'
+ for attrname, value in attrs:
+ value = cgi.escape(value, True)
+ if isinstance(value, unicode):
+ value = value.encode('koi8-r')
+ if attrname == 'href':
+ self.accumulator += ' href="%s"' % urljoin(self.base, value)
+ else:
+ self.accumulator += ' %s="%s"' % (attrname, value)
+ self.accumulator += '>'
+
+ def end_a(self):
+ self.accumulator += '</a>'
+
+ def start_img(self, attrs):
+ self.accumulator += '<img'
+ for attrname, value in attrs:
+ value = cgi.escape(value, True)
+ if attrname == 'src':
+ self.accumulator += ' src="%s"' % urljoin(self.base, value)
+ else:
+ self.accumulator += ' %s="%s"' % (attrname, value)
+ 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:
+ blog_root_url = blog_root[
+ blog_root.find('/htdocs/phdru.name/') + len('/htdocs/phdru.name/'):]
+ baseURL = "https://phdru.name/%s/" % blog_root_url
+else:
+ baseURL = "https://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 = [t.decode('koi8-r').encode('utf-8') for t in tags]
+ body = bodies[(year, month, day, file)]
+ body = absolute_urls(body, baseURL + url_path)
+ try:
+ body.decode('utf-8')
+ except UnicodeDecodeError:
+ body = body.decode('koi8-r').encode('utf-8')
+ item.body = body
+ excerpt = get_first_p(body)
+ try:
+ excerpt.decode('utf-8')
+ except UnicodeDecodeError:
+ excerpt = excerpt.decode('koi8-r').encode('utf-8')
+ item.excerpt = excerpt
+
+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)