X-Git-Url: https://git.phdru.name/?a=blobdiff_plain;ds=sidebyside;f=reindex_blog.py;h=b422784b960c85c9721b3eec07fa57f81d4d5057;hb=e505d3703b5149829a3cf984e7c91f33018f83c4;hp=cf830d1cb94ced402bf1a9cf3502f3afe439cfc1;hpb=49d4d1d442fc3b3089af1079b74be14fb25a8809;p=phdru.name%2Fphdru.name.git diff --git a/reindex_blog.py b/reindex_blog.py index cf830d1..b422784 100755 --- a/reindex_blog.py +++ b/reindex_blog.py @@ -10,6 +10,10 @@ __copyright__ = "Copyright (C) 2006 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 except ImportError: @@ -20,7 +24,6 @@ from Cheetah.Template import Template # Load old blog -blog_filename = sys.argv[1] try: blog_file = open(blog_filename, "rb") except IOError: @@ -30,15 +33,19 @@ else: blog_file.close() -# blog is a dictionary mapping (year, month, day) => (filename, title, lead) +# blog is a dictionary mapping +# (year, month, day) => [list of (file, title, lead, tags)] + blog = {} years = {} # 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) @@ -46,13 +53,17 @@ for dirpath, dirs, files in os.walk(os.curdir): title = ' '.join(title_parts[6:]) lead = getattr(template, "Lead", None) + tags = template.Tag + if isinstance(tags, basestring): + tags = (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] @@ -91,9 +102,11 @@ months_names_ru0 = ['', " "ÉÀÌØ", "Á×ÇÕÓÔ", "ÓÅÎÔÑÂÒØ", "ÏËÔÑÂÒØ", "ÎÏÑÂÒØ", "ÄÅËÁÂÒØ" ] +from news import write_if_changed -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: @@ -102,12 +115,6 @@ def write_template(level, year, month, day, titles): 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. @@ -120,6 +127,9 @@ def write_template(level, year, month, day, titles): #attr $Title = "Oleg BroytMann's blog" #attr $Description = "BroytMann 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
îÏ×ÏÓÔÅ×ÁÑ ÌÅÎÔÁ × ÆÏÒÍÁÔÁÈ +Atom 1.0 +É RSS 2.0 . +
+""") + 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(''' -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: ''') + first_year = True for year in sorted(years.keys()): if first_year: first_year = False @@ -219,20 +260,23 @@ def write_template(level, year, month, day, titles):
''') + new_text.append(""" +öö +""") + new_text.append("""\ #end def $phd_pp_ru.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] @@ -242,15 +286,146 @@ for year in sorted(years.keys()): 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. +#extends phd_pp_ru +#implements respond +#attr $Title = "Oleg BroytMann's blog: tags" +#attr $Description = "BroytMann Russian Blog Tags Index Document" +#attr $Copyright = 2006 +## +#def body_html +
+
+