X-Git-Url: https://git.phdru.name/?p=phdru.name%2Fphdru.name.git;a=blobdiff_plain;f=reindex_blog.py;h=291e182eee8ddffff655114f347e50b9da54b97a;hp=6c2717c3b49695b4f6d7c4f4a6964a7bea210a03;hb=HEAD;hpb=42465be8abac2fdb32c939bc7e5ad0fbca5df2f8

diff --git a/reindex_blog.py b/reindex_blog.py
index 6c2717c..0b7c2a6 100755
--- a/reindex_blog.py
+++ b/reindex_blog.py
@@ -1,37 +1,21 @@
 #! /usr/bin/env python
 # -*- coding: koi8-r -*-
 
-__version__ = "$Revision$"[11:-2]
-__revision__ = "$Id$"[5:-2]
-__date__ = "$Date$"[7:-2]
 __author__ = "Oleg Broytman <phd@phdru.name>"
-__copyright__ = "Copyright (C) 2006-2010 PhiloSoft Design"
-
+__copyright__ = "Copyright (C) 2006-2024 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
+    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
-
-try:
-   blog_file = open(blog_filename, "rb")
-except IOError:
-   old_blog = {}
-else:
-   old_blog = pickle.load(blog_file)
-   blog_file.close()
-
+old_blog = load_blog()
 
 # blog is a dictionary mapping
 # (year, month, day) => [list of (file, title, lead, tags)]
@@ -53,13 +37,14 @@ 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 = template.Lead
+      lead = template.Lead.decode('utf-8').encode('koi8-r')
 
       tags = template.Tag
-      if isinstance(tags, basestring):
+      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[len(blog_root):].split(os.sep)[1:])
@@ -83,26 +68,27 @@ for dirpath, dirs, files in os.walk(blog_root):
 
          file = file[:-len("tmpl")] + "html"
          key = (year, month, day, file)
-         bodies[key] = template.body()
+         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, "ru_RU.KOI8-R")
-from calendar import _localized_day, _localized_month
+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, "ru_RU.KOI8-R")
-#months_names_ru = list(_localized_month('%B'))
+# 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"
@@ -114,6 +100,8 @@ months_names_ru0 = ['', "
 
 from news import write_if_changed
 
+def encode_tag(tag):
+    return quote(tag.replace(' ', '_'))
 
 def write_template(level, year, month, day, titles, tags=None):
    path = [blog_root]
@@ -128,6 +116,7 @@ def write_template(level, year, month, day, titles, tags=None):
 
    new_text = ["""\
 ## THIS FILE IS AUTOMATICALLY GENERATED. DO NOT EDIT.
+#encoding koi8-r
 #extends phd_site
 #implements respond
 """]
@@ -180,7 +169,7 @@ 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("""\
@@ -212,7 +201,7 @@ def write_template(level, year, month, day, titles, tags=None):
       href.append(file)
       href = '/'.join(href)
       if day[0] == '0': day = day[1:]
-      if save_date <> (year, month, 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:
@@ -252,7 +241,8 @@ def write_template(level, year, month, day, titles, tags=None):
             first_tag = False
          else:
             new_text.append(' - ')
-         new_text.append("""<a href="tags/%s.html">%s (%d)</a>""" % (tag, tag, count))
+         new_text.append("""<a href="tags/%s.html">%s (%d)</a>""" % (
+             encode_tag(tag), tag, count))
       new_text.append('''
 </p>
 ''')
@@ -263,13 +253,20 @@ def write_template(level, year, month, day, titles, tags=None):
       new_text.append('''
 <p class="head">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('<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>
 ''')
@@ -334,6 +331,7 @@ 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"
@@ -343,14 +341,36 @@ new_text = ["""\
 #def body_html
 <h1>ôÅÇÉ</h1>
 
-<p class="head">
+<p class="head small">
+æÏÒÍÁ ÐÏÉÓËÁ ÐÏÚ×ÏÌÑÅÔ ÉÓËÁÔØ ÓÏÏÂÝÅÎÉÑ × ÂÌÏÇÅ, ÓÏÏÔ×ÅÔÓÔ×ÕÀÝÉÅ ×ÙÒÁÖÅÎÉÀ.
+óÉÎÔÁËÓÉÓ ×ÙÒÁÖÅÎÉÑ:</p>
+<ul class="small">
+    <li>ôÅÇ - ÅÓÌÉ ÔÁËÏÊ ÔÅÇ ÓÕÝÅÓÔ×ÕÅÔ, ÐÒÏÉÚÏÊÄ£Ô ÐÅÒÅÎÁÐÒÁ×ÌÅÎÉÅ ÎÁ ÓÔÒÁÎÉÃÕ ÔÅÇÁ.</li>
+    <li>ïÐÅÒÁÔÏÒ '!' (NOT, not) - ÉÝÅÔ ÚÁÐÉÓÉ, × ËÏÔÏÒÙÈ ÎÅÔ ÜÔÏÇÏ ÔÅÇÁ.</li>
+    <li>ïÐÅÒÁÔÏÒ '&amp;' (AND, and) - ÉÝÅÔ ÚÁÐÉÓÉ, × ËÏÔÏÒÙÈ ÅÓÔØ ÏÂÁ ÔÅÇÁ.</li>
+    <li>ïÐÅÒÁÔÏÒ '|' (OR, or) - ÉÝÅÔ ÚÁÐÉÓÉ, × ËÏÔÏÒÙÈ ÅÓÔØ ÌÀÂÙÅ ÉÚ ÔÅÇÏ×.</li>
+    <li>ðÒÉÏÒÉÔÅÔ ÏÐÅÒÁÃÉÊ ÓÔÁÎÄÁÒÔÎÙÊ: NOT &gt; AND &gt; OR. óËÏÂËÉ '()' ÐÏÚ×ÏÌÑÀÔ ÏÂßÅÄÉÎÑÔØ ×ÙÒÁÖÅÎÉÑ.</li>
+</ul>
+<p class="small">
+ðÒÉÍÅÒÙ ×ÙÒÁÖÅÎÉÊ: linux - ÐÒÏÉÚÏÊÄ£Ô ÐÅÒÅÎÁÐÒÁ×ÌÅÎÉÅ
+ÎÁ ÓÔÒÁÎÉÃÕ linux.html; linux&amp;!audio - ÉÓËÁÔØ ÚÁÐÉÓÉ × ËÏÔÏÒÙÈ ÅÓÔØ ÔÅÇ
+linux É ÎÅÔ ÔÅÇÁ audio; linux and not audio - ÔÏ ÖÅ ÓÁÍÏÅ.
+</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="éÓËÁÔØ">
+</form>
+</center>
+
 <dl>
 """]
 
 for i, (count, tag, links) in enumerate(all_tags):
    new_text.append("""\
    <dt><a href="%s.html">%s (%d)</a></dt>
-""" % (tag, tag, count))
+""" % (encode_tag(tag), tag, count))
 
    first = all_tags[0][1]
    if i == 0:
@@ -365,6 +385,7 @@ for i, (count, tag, links) in enumerate(all_tags):
 
    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"
@@ -395,7 +416,6 @@ for i, (count, tag, links) in enumerate(all_tags):
 #def body_html
 <h1>%s</h1>
 
-<p class="head">
 <ul>
 """ % tag)
 
@@ -412,15 +432,15 @@ for i, (count, tag, links) in enumerate(all_tags):
 
    tag_text.append("""\
 </ul>
-</p>
 #end def
 $phd_site.respond(self)
 """)
-   write_if_changed(os.path.join(blog_root, "tags", tag+".tmpl"), ''.join(tag_text))
+   write_if_changed(os.path.join(blog_root, "tags",
+                                 tag.replace(' ', '_') + ".tmpl"),
+                    ''.join(tag_text))
 
 new_text.append("""\
 </dl>
-</p>
 #end def
 $phd_site.respond(self)
 """)
@@ -472,6 +492,8 @@ class AbsURLHTMLParser(_HTMLParser):
       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:
@@ -481,6 +503,19 @@ class AbsURLHTMLParser(_HTMLParser):
    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)
 
@@ -502,13 +537,17 @@ from rss_20 import rss_20
 from news import NewsItem
 
 if blog_root:
-   baseURL = "http://phdru.name/%s/" % 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 = "http://phdru.name/"
+   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),
@@ -516,11 +555,20 @@ for item in tuple(reversed(all_titles_tags))[:10]:
       url_path)
    items.append(item)
    item.baseURL = baseURL
-   item.categoryList = tags
+   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
-   item.excerpt = get_first_p(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",
@@ -537,23 +585,23 @@ 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 = 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_titles.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_titles.xml"), rss_tmpl)
 
 for item in items:
     item.content = item.body
 
-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_full.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_full.xml"), rss_tmpl)