X-Git-Url: https://git.phdru.name/?p=phdru.name%2Fphdru.name.git;a=blobdiff_plain;f=reindex_blog.py;h=291e182eee8ddffff655114f347e50b9da54b97a;hp=33a5736dbbc1c022e04a58a65dd9ee9d167ef3e3;hb=HEAD;hpb=eddda9a58e63cc202103cc5266cad173a9e768fd

diff --git a/reindex_blog.py b/reindex_blog.py
index 33a5736..0b7c2a6 100755
--- a/reindex_blog.py
+++ b/reindex_blog.py
@@ -2,32 +2,20 @@
 # -*- coding: koi8-r -*-
 
 __author__ = "Oleg Broytman <phd@phdru.name>"
-__copyright__ = "Copyright (C) 2006-2014 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)]
@@ -54,7 +42,7 @@ for dirpath, dirs, files in os.walk(blog_root):
       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]
 
@@ -86,23 +74,21 @@ for dirpath, dirs, files in os.walk(blog_root):
          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]
@@ -213,7 +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:
@@ -253,7 +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>
 ''')
@@ -381,7 +370,7 @@ linux 
 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:
@@ -446,7 +435,9 @@ for i, (count, tag, links) in enumerate(all_tags):
 #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>
@@ -501,6 +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:
@@ -544,9 +537,11 @@ 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]:
@@ -560,11 +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",