X-Git-Url: https://git.phdru.name/?a=blobdiff_plain;f=reindex_blog.py;h=197c1c6956db3b3972f1a41a01791b607a3fc1b8;hb=dbd103d7a42fb1285230d81c49b95a7f6e3ce921;hp=716605bf2a7a529e55ab5a35050ffbfb038595fc;hpb=2f6754514eb4110b105d64ed944c8a855858c408;p=phdru.name%2Fphdru.name.git
diff --git a/reindex_blog.py b/reindex_blog.py
index 716605b..197c1c6 100755
--- a/reindex_blog.py
+++ b/reindex_blog.py
@@ -1,14 +1,14 @@
-#! /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
%s%s.
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.
+ 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('''
+
+ 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:
@@ -191,30 +275,22 @@ def write_template(year, month, titles):
öö
+""")
+
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()
-
-
-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 = all_titles[-20:]
+all_titles_tags = []
for year in sorted(years.keys()):
year_titles = []
@@ -222,13 +298,281 @@ 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(year, month, month_titles)
- write_template(year, '', year_titles)
-write_template('', '', all_titles)
+ 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.
+#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
+
+
+öÕÒÎÁÌ
+öÕÒÎÁÌ
""" % {"cyear": year or 2005})
- elif show_month:
-
+ elif level == 1:
new_text.append("""\
-#extends phd_pp_ru
-#implements respond
-#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})
- 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
+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],
+ "month_name_ru0": months_names_ru0[imonth],
+ })
+
+ elif level == 3:
+ iday = int(day)
+ imonth = int(month)
+
+ if len(titles) == 1:
+ new_text.append("""\
+#attr $Refresh = "0; URL=%s"
+""" % titles[0][3])
- month = int(month)
new_text.append("""\
-#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"
+#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: %(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[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()
- 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:
+ 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('''
+
+
+
+ôÅÇÉ
+
+
+"""]
+
+for i, (count, tag, links) in enumerate(all_tags):
+ new_text.append("""\
+
%s
+
+
+""" % tag)
+
+ count = 0
+ for year, month, day, filename, title, lead in reversed(links):
+ link = "../%s/%s/%s/%s" % (year, month, day, filename)
+ item_text = """
+
' + + 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)