blitiri.cgi

   1 #!/usr/bin/env python
   2 #coding: utf8
   3
   4 # blitiri - A single-file blog engine.
   5 # Alberto Bertogli (albertito@gmail.com)
   6
   7 #
   8 # Configuration section
   9 #
  10 # You can edit these values, or create a file named "config.py" and put them
  11 # there to make updating easier. The ones in config.py take precedence.
  12 #
  13
  14 # Directory where entries are stored
  15 data_path = "/tmp/blog/data"
  16
  17 # Path where templates are stored. Use an empty string for the built-in
  18 # default templates. If they're not found, the built-in ones will be used.
  19 templates_path = "/tmp/blog/templates"
  20
  21 # URL to the blog, including the name. Can be a full URL or just the path.
  22 blog_url = "/blog/blitiri.cgi"
  23
  24 # Style sheet (CSS) URL. Can be relative or absolute. To use the built-in
  25 # default, set it to blog_url + "/style".
  26 css_url = blog_url + "/style"
  27
  28 # Blog title
  29 title = "I don't like blogs"
  30
  31 # Default author
  32 author = "Hartmut Kegan"
  33
  34 # Article encoding
  35 encoding = "utf8"
  36
  37 #
  38 # End of configuration
  39 # DO *NOT* EDIT ANYTHING PAST HERE
  40 #
  41
  42
  43 import sys
  44 import os
  45 import time
  46 import datetime
  47 import calendar
  48 import zlib
  49 import urllib
  50 import cgi
  51 from docutils.core import publish_parts
  52
  53 # Before importing the config, add our cwd to the Python path
  54 sys.path.append(os.getcwd())
  55
  56 # Load the config file, if there is one
  57 try:
  58         from config import *
  59 except:
  60         pass
  61
  62
  63 # Pimp *_path config variables to support relative paths
  64 data_path = os.path.realpath(data_path)
  65 templates_path = os.path.realpath(templates_path)
  66
  67 # Default template
  68
  69 default_main_header = """\
  70 <?xml version="1.0" encoding="utf-8"?>
  71 <!DOCTYPE html PUBLIC "-//W3C//DTD XHTML 1.0 Strict//EN"
  72           "http://www.w3.org/TR/xhtml1/DTD/xhtml1-strict.dtd">
  73
  74 <html xmlns="http://www.w3.org/1999/xhtml" xml:lang="en" lang="en">
  75 <head>
  76 <link rel="alternate" title="%(title)s" href="%(fullurl)s/atom"
  77         type="application/atom+xml" />
  78 <link href="%(css_url)s" rel="stylesheet" type="text/css" />
  79 <title>%(title)s</title>
  80 </head>
  81
  82 <body>
  83
  84 <h1><a href="%(url)s">%(title)s</a></h1>
  85
  86 <div class="content">
  87 """
  88
  89 default_main_footer = """
  90 </div>
  91 <div class="footer">
  92   %(showyear)s: %(monthlinks)s<br/>
  93   years: %(yearlinks)s<br/>
  94   subscribe: <a href="%(url)s/atom">atom</a><br/>
  95   views: <a href="%(url)s/">blog</a> <a href="%(url)s/list">list</a><br/>
  96 </div>
  97
  98 </body>
  99 </html>
 100 """
 101
 102 default_article_header = """
 103 <div class="article">
 104 <h2><a href="%(url)s/post/%(uuid)s">%(arttitle)s</a></h2>
 105 <span class="artinfo">
 106   by %(author)s on <span class="date">
 107
 108 <a class="date" href="%(url)s/%(cyear)d/">%(cyear)04d</a>-\
 109 <a class="date" href="%(url)s/%(cyear)d/%(cmonth)d/">%(cmonth)02d</a>-\
 110 <a class="date" href="%(url)s/%(cyear)d/%(cmonth)d/%(cday)d/">%(cday)02d</a>\
 111     %(chour)02d:%(cminute)02d</span>
 112   (updated on <span class="date">
 113 <a class="date" href="%(url)s/%(uyear)d/">%(uyear)04d</a>-\
 114 <a class="date" href="%(url)s/%(uyear)d/%(umonth)d/">%(umonth)02d</a>-\
 115 <a class="date" href="%(url)s/%(uyear)d/%(umonth)d/%(uday)d/">%(uday)02d</a>\
 116     %(uhour)02d:%(uminute)02d)</span><br/>
 117   <span class="tags">tagged %(tags)s</span>
 118 </span><br/>
 119 <p/>
 120 <div class="artbody">
 121 """
 122
 123 default_article_footer = """
 124 <p/>
 125 </div>
 126 </div>
 127 """
 128
 129 # Default CSS
 130 default_css = """
 131 body {
 132         font-family: sans-serif;
 133         font-size: small;
 134         width: 52em;
 135 }
 136
 137 div.content {
 138         width: 96%;
 139 }
 140
 141 h1 {
 142         font-size: large;
 143         border-bottom: 2px solid #99F;
 144         width: 100%;
 145         margin-bottom: 1em;
 146 }
 147
 148 h2 {
 149         font-size: medium;
 150         font-weigth: none;
 151         margin-bottom: 1pt;
 152         border-bottom: 1px solid #99C;
 153 }
 154
 155 h1 a, h2 a {
 156         text-decoration: none;
 157         color: black;
 158 }
 159
 160 span.artinfo {
 161         font-size: xx-small;
 162 }
 163
 164 span.artinfo a {
 165         text-decoration: none;
 166         color: #339;
 167 }
 168
 169 span.artinfo a:hover {
 170         text-decoration: none;
 171         color: blue;
 172 }
 173
 174 div.artbody {
 175         margin-left: 1em;
 176 }
 177
 178 div.article {
 179         margin-bottom: 2em;
 180 }
 181
 182 div.footer {
 183         margin-top: 1em;
 184         padding-top: 0.4em;
 185         width: 100%;
 186         border-top: 2px solid #99F;
 187         font-size: x-small;
 188 }
 189
 190 div.footer a {
 191         text-decoration: none;
 192 }
 193
 194 /* Articles are enclosed in <div class="section"> */
 195 div.section h1 {
 196         font-size: small;
 197         font-weigth: none;
 198         width: 100%;
 199         margin-bottom: 1pt;
 200         border-bottom: 1px dotted #99C;
 201 }
 202
 203 """
 204
 205 # helper functions
 206 def rst_to_html(rst):
 207         settings = {
 208                 'input_encoding': encoding,
 209                 'output_encoding': 'utf8',
 210         }
 211         parts = publish_parts(rst, settings_overrides = settings,
 212                                 writer_name = "html")
 213         return parts['body'].encode('utf8')
 214
 215 def sanitize(obj):
 216         if isinstance(obj, basestring):
 217                 return cgi.escape(obj, True)
 218         return obj
 219
 220
 221 # find out our URL, needed for syndication
 222 try:
 223         n = os.environ['SERVER_NAME']
 224         p = os.environ['SERVER_PORT']
 225         s = os.environ['SCRIPT_NAME']
 226         if p == '80': p = ''
 227         else: p = ':' + p
 228         full_url = 'http://%s%s%s' % (n, p, s)
 229 except KeyError:
 230         full_url = 'Not needed'
 231
 232
 233 class Templates (object):
 234         def __init__(self, tpath, db, showyear = None):
 235                 self.tpath = tpath
 236                 self.db = db
 237                 now = datetime.datetime.now()
 238                 if not showyear:
 239                         showyear = now.year
 240
 241                 self.vars = {
 242                         'css_url': css_url,
 243                         'title': title,
 244                         'url': blog_url,
 245                         'fullurl': full_url,
 246                         'year': now.year,
 247                         'month': now.month,
 248                         'day': now.day,
 249                         'showyear': showyear,
 250                         'monthlinks': ' '.join(db.get_month_links(showyear)),
 251                         'yearlinks': ' '.join(db.get_year_links()),
 252                 }
 253
 254         def get_template(self, page_name, default_template, extra_vars = None):
 255                 if extra_vars is None:
 256                         vars = self.vars
 257                 else:
 258                         vars = self.vars.copy()
 259                         vars.update(extra_vars)
 260
 261                 p = '%s/%s.html' % (self.tpath, page_name)
 262                 if os.path.isfile(p):
 263                         return open(p).read() % vars
 264                 return default_template % vars
 265
 266         def get_main_header(self):
 267                 return self.get_template('header', default_main_header)
 268
 269         def get_main_footer(self):
 270                 return self.get_template('footer', default_main_footer)
 271
 272         def get_article_header(self, article):
 273                 return self.get_template(
 274                         'art_header', default_article_header, article.to_vars())
 275
 276         def get_article_footer(self, article):
 277                 return self.get_template(
 278                         'art_footer', default_article_footer, article.to_vars())
 279
 280
 281 class Article (object):
 282         def __init__(self, path, created = None, updated = None):
 283                 self.path = path
 284                 self.created = created
 285                 self.updated = updated
 286                 self.uuid = "%08x" % zlib.crc32(self.path)
 287
 288                 self.loaded = False
 289
 290                 # loaded on demand
 291                 self._title = 'Removed post'
 292                 self._author = author
 293                 self._tags = []
 294                 self._raw_content = ''
 295
 296
 297         def get_title(self):
 298                 if not self.loaded:
 299                         self.load()
 300                 return self._title
 301         title = property(fget = get_title)
 302
 303         def get_author(self):
 304                 if not self.loaded:
 305                         self.load()
 306                 return self._author
 307         author = property(fget = get_author)
 308
 309         def get_tags(self):
 310                 if not self.loaded:
 311                         self.load()
 312                 return self._tags
 313         tags = property(fget = get_tags)
 314
 315         def get_raw_content(self):
 316                 if not self.loaded:
 317                         self.load()
 318                 return self._raw_content
 319         raw_content = property(fget = get_raw_content)
 320
 321
 322         def __cmp__(self, other):
 323                 if self.path == other.path:
 324                         return 0
 325                 if not self.created:
 326                         return 1
 327                 if not other.created:
 328                         return -1
 329                 if self.created < other.created:
 330                         return -1
 331                 return 1
 332
 333         def title_cmp(self, other):
 334                 return cmp(self.title, other.title)
 335
 336
 337         def load(self):
 338                 # os.path.join ignore other paths if one starts with a slash
 339                 filename = os.path.join(data_path, self.path[1:])
 340                 try:
 341                         raw = open(filename).readlines()
 342                 except:
 343                         return
 344
 345                 count = 0
 346                 for l in raw:
 347                         if ':' in l:
 348                                 name, value = l.split(':', 1)
 349                                 if name.lower() == 'title':
 350                                         self._title = value.strip()
 351                                 elif name.lower() == 'author':
 352                                         self._author = value.strip()
 353                                 elif name.lower() == 'tags':
 354                                         ts = value.split(',')
 355                                         ts = [t.strip() for t in ts]
 356                                         self._tags = set(ts)
 357                         elif l == '\n':
 358                                 # end of header
 359                                 break
 360                         count += 1
 361                 self._raw_content = ''.join(raw[count + 1:])
 362                 self.loaded = True
 363
 364         def to_html(self):
 365                 return rst_to_html(self.raw_content)
 366
 367         def to_vars(self):
 368                 return {
 369                         'arttitle': sanitize(self.title),
 370                         'author': sanitize(self.author),
 371                         'date': self.created.isoformat(' '),
 372                         'uuid': self.uuid,
 373                         'tags': self.get_tags_links(),
 374
 375                         'created': self.created.isoformat(' '),
 376                         'ciso': self.created.isoformat(),
 377                         'cyear': self.created.year,
 378                         'cmonth': self.created.month,
 379                         'cday': self.created.day,
 380                         'chour': self.created.hour,
 381                         'cminute': self.created.minute,
 382                         'csecond': self.created.second,
 383
 384                         'updated': self.updated.isoformat(' '),
 385                         'uiso': self.updated.isoformat(),
 386                         'uyear': self.updated.year,
 387                         'umonth': self.updated.month,
 388                         'uday': self.updated.day,
 389                         'uhour': self.updated.hour,
 390                         'uminute': self.updated.minute,
 391                         'usecond': self.updated.second,
 392                 }
 393
 394         def get_tags_links(self):
 395                 l = []
 396                 tags = list(self.tags)
 397                 tags.sort()
 398                 for t in tags:
 399                         l.append('<a class="tag" href="%s/tag/%s">%s</a>' % \
 400                                 (blog_url, urllib.quote(t), sanitize(t) ))
 401                 return ', '.join(l)
 402
 403
 404 class ArticleDB (object):
 405         def __init__(self, dbpath):
 406                 self.dbpath = dbpath
 407                 self.articles = []
 408                 self.uuids = {}
 409                 self.actyears = set()
 410                 self.actmonths = set()
 411                 self.load()
 412
 413         def get_articles(self, year = 0, month = 0, day = 0, tags = None):
 414                 l = []
 415                 for a in self.articles:
 416                         if year and a.created.year != year: continue
 417                         if month and a.created.month != month: continue
 418                         if day and a.created.day != day: continue
 419                         if tags and not tags.issubset(a.tags): continue
 420
 421                         l.append(a)
 422
 423                 return l
 424
 425         def get_article(self, uuid):
 426                 return self.uuids[uuid]
 427
 428         def load(self):
 429                 try:
 430                         f = open(self.dbpath)
 431                 except:
 432                         return
 433
 434                 for l in f:
 435                         # Each line has the following comma separated format:
 436                         # path (relative to data_path), \
 437                         #       created (epoch), \
 438                         #       updated (epoch)
 439                         try:
 440                                 l = l.split(',')
 441                         except:
 442                                 continue
 443
 444                         a = Article(l[0],
 445                                 datetime.datetime.fromtimestamp(float(l[1])),
 446                                 datetime.datetime.fromtimestamp(float(l[2])))
 447                         self.uuids[a.uuid] = a
 448                         self.actyears.add(a.created.year)
 449                         self.actmonths.add((a.created.year, a.created.month))
 450                         self.articles.append(a)
 451
 452         def save(self):
 453                 f = open(self.dbpath + '.tmp', 'w')
 454                 for a in self.articles:
 455                         s = ''
 456                         s += a.path + ', '
 457                         s += str(time.mktime(a.created.timetuple())) + ', '
 458                         s += str(time.mktime(a.updated.timetuple())) + '\n'
 459                         f.write(s)
 460                 f.close()
 461                 os.rename(self.dbpath + '.tmp', self.dbpath)
 462
 463         def get_year_links(self):
 464                 yl = list(self.actyears)
 465                 yl.sort(reverse = True)
 466                 return [ '<a href="%s/%d/">%d</a>' % (blog_url, y, y)
 467                                 for y in yl ]
 468
 469         def get_month_links(self, year):
 470                 am = [ i[1] for i in self.actmonths if i[0] == year ]
 471                 ml = []
 472                 for i in range(1, 13):
 473                         name = calendar.month_name[i][:3]
 474                         if i in am:
 475                                 s = '<a href="%s/%d/%d/">%s</a>' % \
 476                                         ( blog_url, year, i, name )
 477                         else:
 478                                 s = name
 479                         ml.append(s)
 480                 return ml
 481
 482 #
 483 # Main
 484 #
 485
 486
 487 def render_html(articles, db, actyear = None):
 488         template = Templates(templates_path, db, actyear)
 489         print 'Content-type: text/html; charset=utf-8\n'
 490         print template.get_main_header()
 491         for a in articles:
 492                 print template.get_article_header(a)
 493                 print a.to_html()
 494                 print template.get_article_footer(a)
 495         print template.get_main_footer()
 496
 497 def render_artlist(articles, db, actyear = None):
 498         template = Templates(templates_path, db, actyear)
 499         print 'Content-type: text/html; charset=utf-8\n'
 500         print template.get_main_header()
 501         print '<h2>Articles</h2>'
 502         for a in articles:
 503                 print '<li><a href="%(url)s/uuid/%(uuid)s">%(title)s</a></li>' \
 504                         % {     'url': blog_url,
 505                                 'uuid': a.uuid,
 506                                 'title': a.title,
 507                                 'author': a.author,
 508                         }
 509         print template.get_main_footer()
 510
 511 def render_atom(articles):
 512         if len(articles) > 0:
 513                 updated = articles[0].updated.isoformat()
 514         else:
 515                 updated = datetime.datetime.now().isoformat()
 516
 517         print 'Content-type: application/atom+xml; charset=utf-8\n'
 518         print """<?xml version="1.0" encoding="utf-8"?>
 519
 520 <feed xmlns="http://www.w3.org/2005/Atom">
 521  <title>%(title)s</title>
 522  <link rel="alternate" type="text/html" href="%(url)s"/>
 523  <link rel="self" type="application/atom+xml" href="%(url)s/atom"/>
 524  <id>%(url)s</id> <!-- TODO: find a better <id>, see RFC 4151 -->
 525  <updated>%(updated)sZ</updated>
 526
 527         """ % {
 528                 'title': title,
 529                 'url': full_url,
 530                 'updated': updated,
 531         }
 532
 533         for a in articles:
 534                 vars = a.to_vars()
 535                 vars.update( {
 536                         'url': full_url,
 537                         'contents': a.to_html(),
 538                 } )
 539                 print """
 540   <entry>
 541     <title>%(arttitle)s</title>
 542     <author><name>%(author)s</name></author>
 543     <link href="%(url)s/post/%(uuid)s" />
 544     <id>%(url)s/post/%(uuid)s</id>
 545     <summary>%(arttitle)s</summary>
 546     <published>%(ciso)sZ</published>
 547     <updated>%(uiso)sZ</updated>
 548     <content type="xhtml">
 549       <div xmlns="http://www.w3.org/1999/xhtml"><p>
 550 %(contents)s
 551       </p></div>
 552     </content>
 553   </entry>
 554                 """ % vars
 555         print "</feed>"
 556
 557
 558 def render_style():
 559         print 'Content-type: text/css\r\n\r\n',
 560         print default_css
 561
 562 def handle_cgi():
 563         import cgitb; cgitb.enable()
 564
 565         form = cgi.FieldStorage()
 566         year = int(form.getfirst("year", 0))
 567         month = int(form.getfirst("month", 0))
 568         day = int(form.getfirst("day", 0))
 569         tags = set(form.getlist("tag"))
 570         uuid = None
 571         atom = False
 572         style = False
 573         post = False
 574         artlist = False
 575
 576         if os.environ.has_key('PATH_INFO'):
 577                 path_info = os.environ['PATH_INFO']
 578                 style = path_info == '/style'
 579                 atom = path_info == '/atom'
 580                 tag = path_info.startswith('/tag/')
 581                 post = path_info.startswith('/post/')
 582                 artlist = path_info.startswith('/list')
 583                 if not style and not atom and not post and not tag \
 584                                 and not artlist:
 585                         date = path_info.split('/')[1:]
 586                         try:
 587                                 if len(date) > 1 and date[0]:
 588                                         year = int(date[0])
 589                                 if len(date) > 2 and date[1]:
 590                                         month = int(date[1])
 591                                 if len(date) > 3 and date[2]:
 592                                         day = int(date[2])
 593                         except ValueError:
 594                                 pass
 595                 elif post:
 596                         uuid = path_info.replace('/post/', '')
 597                         uuid = uuid.replace('/', '')
 598                 elif tag:
 599                         t = path_info.replace('/tag/', '')
 600                         t = t.replace('/', '')
 601                         t = urllib.unquote_plus(t)
 602                         tags = set((t,))
 603
 604         db = ArticleDB(os.path.join(data_path, 'db'))
 605         if atom:
 606                 articles = db.get_articles(tags = tags)
 607                 articles.sort(reverse = True)
 608                 render_atom(articles[:10])
 609         elif style:
 610                 render_style()
 611         elif post:
 612                 render_html( [db.get_article(uuid)], db, year )
 613         elif artlist:
 614                 articles = db.get_articles()
 615                 articles.sort(cmp = Article.title_cmp)
 616                 render_artlist(articles, db)
 617         else:
 618                 articles = db.get_articles(year, month, day, tags)
 619                 articles.sort(reverse = True)
 620                 if not year and not month and not day and not tags:
 621                         articles = articles[:10]
 622                 render_html(articles, db, year)
 623
 624
 625 def usage():
 626         print 'Usage: %s {add|rm|update} article_path' % sys.argv[0]
 627
 628 def handle_cmd():
 629         if len(sys.argv) != 3:
 630                 usage()
 631                 return 1
 632
 633         cmd = sys.argv[1]
 634         art_path = os.path.realpath(sys.argv[2])
 635
 636         if os.path.commonprefix([data_path, art_path]) != data_path:
 637                 print "Error: article (%s) must be inside data_path (%s)" % \
 638                                 (art_path, data_path)
 639                 return 1
 640         art_path = art_path[len(data_path):]
 641
 642         db_filename = os.path.join(data_path, 'db')
 643         if not os.path.isfile(db_filename):
 644                 open(db_filename, 'w').write('')
 645         db = ArticleDB(db_filename)
 646
 647         if cmd == 'add':
 648                 article = Article(art_path, datetime.datetime.now(),
 649                                         datetime.datetime.now())
 650                 for a in db.articles:
 651                         if a == article:
 652                                 print 'Error: article already exists'
 653                                 return 1
 654                 db.articles.append(article)
 655                 db.save()
 656         elif cmd == 'rm':
 657                 article = Article(art_path)
 658                 for a in db.articles:
 659                         if a == article:
 660                                 break
 661                 else:
 662                         print "Error: no such article"
 663                         return 1
 664                 db.articles.remove(a)
 665                 db.save()
 666         elif cmd == 'update':
 667                 article = Article(art_path)
 668                 for a in db.articles:
 669                         if a == article:
 670                                 break
 671                 else:
 672                         print "Error: no such article"
 673                         return 1
 674                 a.updated = datetime.datetime.now()
 675                 db.save()
 676         else:
 677                 usage()
 678                 return 1
 679
 680         return 0
 681
 682
 683 if os.environ.has_key('GATEWAY_INTERFACE'):
 684         handle_cgi()
 685 else:
 686         sys.exit(handle_cmd())
 687
 688