X-Git-Url: https://git.phdru.name/?a=blobdiff_plain;f=reindex_blog.py;h=92a02272ea79c60a9f79b0ae21fe8733d6fa47f2;hb=eb2a1c0e122dc1f83bd3b220d2d00103f0dcda8c;hp=be27835e9a9140c3988bc4a39009bd816a5a644d;hpb=6aa7ab00b5dc30b6221d0d55fecc1b0b957cdd79;p=phdru.name%2Fphdru.name.git

diff --git a/reindex_blog.py b/reindex_blog.py
index be27835..92a0227 100755
--- a/reindex_blog.py
+++ b/reindex_blog.py
@@ -1,8 +1,18 @@
-#! /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 Broytman <phd@phd.pp.ru>"
+__copyright__ = "Copyright (C) 2006 PhiloSoft Design"
+
+
 import sys, os
-from glob import glob
+
+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
@@ -11,188 +21,267 @@ except ImportError:
 
 from Cheetah.Template import Template
 
-blog_filename = sys.argv[1]
+
+# Load old blog
+
 try:
    blog_file = open(blog_filename, "rb")
 except IOError:
-   blog = {}
+   old_blog = {}
 else:
-   blog = pickle.load(blog_file)
+   old_blog = pickle.load(blog_file)
    blog_file.close()
 
+
+# blog is a dictionary mapping
+# (year, month, day) => [list of (file, title, lead, tags)]
+
+blog = {}
 years = {}
-months = sys.argv[2:]
-isdir = os.path.isdir
-
-if not months:
-   for year in os.listdir(os.curdir):
-      if isdir(year):
-         years[year] = {}
-         for month in os.listdir(year):
-            m = os.path.join(year, month)
-            if isdir(m):
-               months.append(m)
-
-days = []
-for month in months:
-   year, m = month.split(os.sep)
-   if month not in years:
-      years[year] = {}
-   years[year][m] = days_of_month = []
-   for day in os.listdir(month):
-         d = os.path.join(month, day)
-         if isdir(d):
-            days.append(d)
-            days_of_month.append(d)
-
-for day in days:
-   for tmpl in glob(os.path.join(day, "*.tmpl")):
-      template = Template(file=tmpl)
+
+# Walk the directory recursively
+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:
+      if not file.endswith(".tmpl"):
+         continue
+      fullpath = os.path.join(dirpath, file)
+      template = Template(file=fullpath)
       title_parts = template.Title.split()
       title = ' '.join(title_parts[6:])
       lead = getattr(template, "Lead", None)
 
+      tags = template.Tag
+      if isinstance(tags, basestring):
+         tags = (tags,)
+
       if title:
-         day_parts = day.split(os.sep)
-         blog[tuple(day_parts)] = (title, os.path.basename(tmpl), lead)
+         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, tags))
+
+         if year in years:
+            months = years[year]
+         else:
+            months = years[year] = {}
 
-blog_file = open(blog_filename, "wb")
-pickle.dump(blog, blog_file, pickle.HIGHEST_PROTOCOL)
-blog_file.close()
+         if month in months:
+            days = months[month]
+         else:
+            days = months[month] = []
+
+         if day not in days: days.append(day)
 
 
+# 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()
+
+# 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_template(show_year, show_month, year, month, titles, cut=None, show_years=False):
-   index_tmpl = open(os.path.join(year, month, "index.tmpl"), 'w')
-   if show_year:
-      index_tmpl.write("""\
+
+def write_template(level, year, month, day, titles, tags=None):
+   path = [blog_root]
+   if level >= 1:
+      path.append(year)
+   if level >= 2:
+      path.append(month)
+   if level == 3:
+      path.append(day)
+   path.append("index.tmpl")
+   index_name = os.path.join(*path)
+
+   new_text = ["""\
+## THIS FILE IS AUTOMATICALLY GENERATED. DO NOT EDIT.
 #extends phd_pp_ru
 #implements respond
-#attr $Title = "Oleg BroytMann's blog"
-#attr $Description = "BroytMann Russian Blog Index Document"
+"""]
+
+   if level == 0:
+      new_text.append("""\
+#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")
+)
 ##
 #def body_html
-<H1>öÕÒÎÁÌ</H1>
+<h1>öÕÒÎÁÌ</h1>
 """ % {"cyear": year or 2005})
 
-   elif show_month:
-
-      index_tmpl.write("""\
-#extends phd_pp_ru
-#implements respond
-#attr $Title = "Oleg BroytMann's blog: %(year)s"
-#attr $Description = "BroytMann Russian Blog %(year)s Index Document"
+   elif level == 1:
+      new_text.append("""\
+#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>öÕÒÎÁÌ: %(year)s</H1>
+<h1>öÕÒÎÁÌ: %(year)s</h1>
 """ % {"year": year, "cyear": year or 2005})
 
-   else:
+   elif level == 2:
+      imonth = int(month)
+      new_text.append("""\
+#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>öÕÒÎÁÌ: %(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],
+      "month_name_ru0": months_names_ru0[imonth],
+   })
 
-      month = int(month)
-      index_tmpl.write("""\
-#extends phd_pp_ru
-#implements respond
-#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"
+   elif level == 3:
+      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"
+""" % titles[0][3])
+
+      new_text.append("""\
+#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>öÕÒÎÁÌ: %(month_name_ru0)s %(year)s</H1>
+<h1>öÕÒÎÁÌ: %(day)d %(month_name_ru)s %(year)s</h1>
 """ % {
       "year": year, "cyear": year or 2005,
-      "month_abbr_en": months_abbrs_en[month], "month_name_en": months_names_en[month],
-      "month_name_ru0": months_names_ru0[month],
+      "month_abbr_en": months_abbrs_en[imonth], "month_name_en": months_names_en[imonth],
+      "month_name_ru": months_names_ru[imonth],
+      "day": iday
    })
 
    save_titles = titles[:]
    titles.reverse()
-   if cut:
-      titles = titles[:cut]
 
-   save_day = None
-   for key, tmpl, title, lead in titles:
-      year, month, day = key
+   save_date = None
+   for year, month, day, file, title, lead in titles:
       href = []
-      if show_year:
+      if level == 0:
          href.append(year)
-      if show_month:
+      if level <= 1:
          href.append(month)
-      href.append(day)
-      href.append(tmpl)
+      if level <= 2:
+         href.append(day)
+      href.append(file)
       href = '/'.join(href)
       if day[0] == '0': day = day[1:]
-      if save_day <> day:
-         if show_year:
-            index_tmpl.write('\n<h2>%s %s %s</h2>' % (day, months_names_ru[int(month)], year))
+      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:
-            index_tmpl.write('\n<h2>%s %s</h2>' % (day, months_names_ru[int(month)]))
-         save_day = day
+            new_text.append('\n<h2>%s %s</h2>' % (day, months_names_ru[int(month)]))
+         save_date = year, month, day
       if lead:
          lead = lead + ' '
       else:
          lead = ''
-      index_tmpl.write('''
+      new_text.append('''
 <p class="head">
    %s<a href="%s">%s</a>.
 </p>
 ''' % (lead, href, title))
 
-   if show_years:
+   if level == 0:
+      new_text.append("""
+<hr>
+
+<p class="head">îÏ×ÏÓÔÅ×ÁÑ ÌÅÎÔÁ × ÆÏÒÍÁÔÁÈ
+<A HREF="atom_10.xml">Atom 1.0 <img src="../../Graphics/atom_10.jpg" border=0></A>
+É <A HREF="rss_20.xml">RSS 2.0 <img src="../../Graphics/rss_20.jpg" border=0></A>.
+</p>
+""")
+
       years = {}
-      for key, tmpl, title, lead in save_titles:
-         year, month, day = key
+      for year, month, day, file, title, lead in save_titles:
          years[year] = True
-      first_year = True
-      index_tmpl.write('''
-<hr>
+      new_text.append('''
+<p class="head"><a href="tags/">ôÅÇÉ</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>""" % (tag, tag, count))
+      new_text.append('''
+</p>
+''')
+
+      max_year = int(sorted(years.keys())[-1])
+      years = range(2005, max_year+1)
 
-<p class="years">
+      new_text.append('''
+<p class="head">ðÏ ÇÏÄÁÍ:
 ''')
-      for year in sorted(years.keys()):
+      first_year = True
+      for year in years:
          if first_year:
             first_year = False
          else:
-            index_tmpl.write(' - ')
-         index_tmpl.write('<a href="%s/">%s</a>' % (year, year))
-      index_tmpl.write('''
+            new_text.append(' - ')
+         new_text.append('<a href="%s/">%s</a>' % (year, year))
+      new_text.append('''
 </p>
 ''')
 
-   index_tmpl.write("""\
+      new_text.append("""
+<hr>
+<p class="head"><a href="http://phd.livejournal.com/">öö</a>
+""")
+
+   new_text.append("""\
 #end def
 $phd_pp_ru.respond(self)
 """)
-   index_tmpl.close()
 
-
-def translate(tmpl):
-   if tmpl == "index.tmpl": tmpl = ''
-   if tmpl.endswith(".tmpl"): tmpl = tmpl[:-len("tmpl")] + "html"
-   return tmpl
+   write_if_changed(index_name, ''.join(new_text))
 
 
+all_tags = {}
 all_titles = []
-for key in sorted(blog.keys()):
-   title, tmpl, lead = blog[key]
-   all_titles.append((key, translate(tmpl), title, lead))
+all_titles_tags = []
 
 for year in sorted(years.keys()):
    year_titles = []
@@ -200,13 +289,149 @@ for year in sorted(years.keys()):
    for month in sorted(months.keys()):
       month_titles = []
       for day in sorted(months[month]):
-         day_parts = day.split(os.sep)
-         key = tuple(day_parts)
+         day_titles = []
+         key = year, month, day
          if key in blog:
-            title, tmpl, lead = blog[key]
-            tmpl = translate(tmpl)
-            year_titles.append((key, tmpl, title, lead))
-            month_titles.append((key, tmpl, title, lead))
-      write_template(False, False, year, month, month_titles)
-   write_template(False, True, year, '', year_titles)
-write_template(True, True, '', '', all_titles, 5, True)
+            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)
+
+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.
+#extends phd_pp_ru
+#implements respond
+#attr $Title = "Oleg Broytman's blog: tags"
+#attr $Description = "Broytman Russian Blog Tags Index Document"
+#attr $Copyright = 2006
+##
+#def body_html
+<h1>ôÅÇÉ</h1>
+
+<p class="head">
+<dl>
+"""]
+
+for count, tag, links in all_tags:
+   new_text.append("""\
+   <dt><a href="%s.html">%s (%d)</a></dt>
+""" % (tag, tag, count))
+
+   tag_text = ["""\
+## THIS FILE IS AUTOMATICALLY GENERATED. DO NOT EDIT.
+#extends phd_pp_ru
+#implements respond
+#attr $Title = "Oleg Broytman's blog: tag %s"
+#attr $Description = "Broytman Russian Blog Tag %s Index Document"
+#attr $Copyright = 2006
+##
+#def body_html
+<h1>%s</h1>
+
+<p class="head">
+<ul>
+""" % (tag, tag, tag)]
+
+   count = 0
+   for year, month, day, filename, title, lead in reversed(links):
+      if lead:
+         lead = lead + ' '
+      else:
+         lead = ''
+      link = "../%s/%s/%s/%s" % (year, month, day, filename)
+      item_text = """<li><a href="%s">%s/%s/%s: %s%s</a></li>""" % (link, year, month, day, lead, title)
+
+      count += 1
+      if count <= 5:
+         new_text.append("      <dd>%s</dd>\n" % item_text)
+
+      tag_text.append("   %s\n" % item_text)
+
+   tag_text.append("""\
+</ul>
+</p>
+#end def
+$phd_pp_ru.respond(self)
+""")
+   write_if_changed(os.path.join(blog_root, "tags", tag+".tmpl"), ''.join(tag_text))
+
+new_text.append("""\
+</dl>
+</p>
+#end def
+$phd_pp_ru.respond(self)
+""")
+write_if_changed(os.path.join(blog_root, "tags", "index.tmpl"), ''.join(new_text))
+
+
+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
+else:
+   baseURL = "http://phd.pp.ru/"
+
+items = []
+for item in tuple(reversed(all_titles_tags))[:10]:
+   year, month, day, file, title, lead, tags = item
+   if lead:
+      lead = lead + ' '
+   else:
+      lead = ''
+   item = NewsItem(
+      "%s-%s-%s" % (year, month, day),
+      "%s%s" % (lead, title),
+      "%s/%s/%s/%s" % (year, month, day, file)
+   )
+   items.append(item)
+   item.baseURL = baseURL
+   item.categoryList = tags
+
+namespace = {
+   "title": "Oleg Broytman's blog",
+   "baseURL": baseURL,
+   "indexFile": "",
+   "description": "",
+   "lang": "ru",
+   "author": "Oleg Broytman",
+   "email": "phd@phd.pp.ru",
+   "generator": os.path.basename(sys.argv[0]),
+   "posts": items,
+}
+
+# For english dates
+locale.setlocale(locale.LC_TIME, 'C')
+
+atom_tmpl = str(atom_10(searchList=[namespace]))
+write_if_changed(os.path.join(blog_root, "atom_10.xml"), atom_tmpl)
+rss_tmpl = str(rss_20(searchList=[namespace]))
+write_if_changed(os.path.join(blog_root, "rss_20.xml"), rss_tmpl)