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 # find out our URL, needed for syndication
 206 try:
 207         n = os.environ['SERVER_NAME']
 208         p = os.environ['SERVER_PORT']
 209         s = os.environ['SCRIPT_NAME']
 210         if p == '80': p = ''
 211         else: p = ':' + p
 212         full_url = 'http://%s%s%s' % (n, p, s)
 213 except KeyError:
 214         full_url = 'Not needed'
 215
 216
 217 class Templates (object):
 218         def __init__(self, tpath, db, showyear = None):
 219                 self.tpath = tpath
 220                 self.db = db
 221                 now = datetime.datetime.now()
 222                 if not showyear:
 223                         showyear = now.year
 224
 225                 self.vars = {
 226                         'css_url': css_url,
 227                         'title': title,
 228                         'url': blog_url,
 229                         'fullurl': full_url,
 230                         'year': now.year,
 231                         'month': now.month,
 232                         'day': now.day,
 233                         'showyear': showyear,
 234                         'monthlinks': ' '.join(db.get_month_links(showyear)),
 235                         'yearlinks': ' '.join(db.get_year_links()),
 236                 }
 237
 238         def get_main_header(self):
 239                 p = self.tpath + '/header.html'
 240                 if os.path.isfile(p):
 241                         return open(p).read() % self.vars
 242                 return default_main_header % self.vars
 243
 244         def get_main_footer(self):
 245                 p = self.tpath + '/footer.html'
 246                 if os.path.isfile(p):
 247                         return open(p).read() % self.vars
 248                 return default_main_footer % self.vars
 249
 250         def get_article_header(self, article):
 251                 avars = self.vars.copy()
 252                 avars.update(article.to_vars())
 253
 254                 p = self.tpath + '/art_header.html'
 255                 if os.path.isfile(p):
 256                         return open(p).read() % avars
 257                 return default_article_header % avars
 258
 259         def get_article_footer(self, article):
 260                 avars = self.vars.copy()
 261                 avars.update(article.to_vars())
 262
 263                 p = self.tpath + '/art_footer.html'
 264                 if os.path.isfile(p):
 265                         return open(p).read() % avars
 266                 return default_article_footer % avars
 267
 268
 269 class Article (object):
 270         def __init__(self, path):
 271                 self.path = path
 272                 self.created = None
 273                 self.updated = None
 274                 self.uuid = "%08x" % zlib.crc32(self.path)
 275
 276                 self.loaded = False
 277
 278                 # loaded on demand
 279                 self._title = 'Removed post'
 280                 self._author = author
 281                 self._tags = []
 282                 self._raw_content = ''
 283
 284
 285         def get_title(self):
 286                 if not self.loaded:
 287                         self.load()
 288                 return self._title
 289         title = property(fget = get_title)
 290
 291         def get_author(self):
 292                 if not self.loaded:
 293                         self.load()
 294                 return self._author
 295         author = property(fget = get_author)
 296
 297         def get_tags(self):
 298                 if not self.loaded:
 299                         self.load()
 300                 return self._tags
 301         tags = property(fget = get_tags)
 302
 303         def get_raw_content(self):
 304                 if not self.loaded:
 305                         self.load()
 306                 return self._raw_content
 307         raw_content = property(fget = get_raw_content)
 308
 309
 310         def __cmp__(self, other):
 311                 if self.path == other.path:
 312                         return 0
 313                 if not self.created:
 314                         return 1
 315                 if not other.created:
 316                         return -1
 317                 if self.created < other.created:
 318                         return -1
 319                 return 1
 320
 321         def title_cmp(self, other):
 322                 return cmp(self.title, other.title)
 323
 324
 325         def load(self):
 326                 try:
 327                         raw = open(data_path + '/' + self.path).readlines()
 328                 except:
 329                         return
 330
 331                 count = 0
 332                 for l in raw:
 333                         if ':' in l:
 334                                 name, value = l.split(':', 1)
 335                                 if name.lower() == 'title':
 336                                         self._title = value
 337                                 elif name.lower() == 'author':
 338                                         self._author = value
 339                                 elif name.lower() == 'tags':
 340                                         ts = value.split(',')
 341                                         ts = [t.strip() for t in ts]
 342                                         self._tags = set(ts)
 343                         elif l == '\n':
 344                                 # end of header
 345                                 break
 346                         count += 1
 347                 self._raw_content = ''.join(raw[count + 1:])
 348                 self.loaded = True
 349
 350         def to_html(self):
 351                 try:
 352                         raw = open(data_path + '/' + self.path).readlines()
 353                 except:
 354                         return "Can't open post file<p>"
 355                 raw = raw[raw.index('\n'):]
 356
 357                 settings = {
 358                         'input_encoding': encoding,
 359                         'output_encoding': 'utf8',
 360                 }
 361                 parts = publish_parts(self.raw_content,
 362                                 settings_overrides = settings,
 363                                 writer_name = "html")
 364                 return parts['body'].encode('utf8')
 365
 366         def to_vars(self):
 367                 return {
 368                         'arttitle': self.title,
 369                         'author': self.author,
 370                         'date': self.created.isoformat(' '),
 371                         'uuid': self.uuid,
 372                         'tags': self.get_tags_links(),
 373
 374                         'created': self.created.isoformat(' '),
 375                         'ciso': self.created.isoformat(),
 376                         'cyear': self.created.year,
 377                         'cmonth': self.created.month,
 378                         'cday': self.created.day,
 379                         'chour': self.created.hour,
 380                         'cminute': self.created.minute,
 381                         'csecond': self.created.second,
 382
 383                         'updated': self.updated.isoformat(' '),
 384                         'uiso': self.updated.isoformat(),
 385                         'uyear': self.updated.year,
 386                         'umonth': self.updated.month,
 387                         'uday': self.updated.day,
 388                         'uhour': self.updated.hour,
 389                         'uminute': self.updated.minute,
 390                         'usecond': self.updated.second,
 391                 }
 392
 393         def get_tags_links(self):
 394                 l = []
 395                 tags = list(self.tags)
 396                 tags.sort()
 397                 for t in tags:
 398                         l.append('<a class="tag" href="%s/tag/%s">%s</a>' % \
 399                                 (blog_url, urllib.quote(t), t) )
 400                 return ', '.join(l)
 401
 402
 403 class DB (object):
 404         def __init__(self, dbpath):
 405                 self.dbpath = dbpath
 406                 self.articles = []
 407                 self.uuids = {}
 408                 self.actyears = set()
 409                 self.actmonths = set()
 410                 self.load()
 411
 412         def get_articles(self, year = 0, month = 0, day = 0, tags = None):
 413                 l = []
 414                 for a in self.articles:
 415                         if year and a.created.year != year: continue
 416                         if month and a.created.month != month: continue
 417                         if day and a.created.day != day: continue
 418                         if tags and not tags.issubset(a.tags): continue
 419
 420                         l.append(a)
 421
 422                 return l
 423
 424         def get_article(self, uuid):
 425                 return self.uuids[uuid]
 426
 427         def load(self):
 428                 try:
 429                         f = open(self.dbpath)
 430                 except:
 431                         return
 432
 433                 for l in f:
 434                         # Each line has the following comma separated format:
 435                         # path (relative to data_path), \
 436                         #       created (epoch), \
 437                         #       updated (epoch)
 438                         try:
 439                                 l = l.split(',')
 440                         except:
 441                                 continue
 442
 443                         a = Article(l[0])
 444                         a.created = datetime.datetime.fromtimestamp(
 445                                                 float(l[1]) )
 446                         a.updated = datetime.datetime.fromtimestamp(
 447                                                 float(l[2]))
 448                         self.uuids[a.uuid] = a
 449                         self.actyears.add(a.created.year)
 450                         self.actmonths.add((a.created.year, a.created.month))
 451                         self.articles.append(a)
 452
 453         def save(self):
 454                 f = open(self.dbpath + '.tmp', 'w')
 455                 for a in self.articles:
 456                         s = ''
 457                         s += a.path + ', '
 458                         s += str(time.mktime(a.created.timetuple())) + ', '
 459                         s += str(time.mktime(a.updated.timetuple())) + '\n'
 460                         f.write(s)
 461                 f.close()
 462                 os.rename(self.dbpath + '.tmp', self.dbpath)
 463
 464         def get_year_links(self):
 465                 yl = list(self.actyears)
 466                 yl.sort(reverse = True)
 467                 return [ '<a href="%s/%d/">%d</a>' % (blog_url, y, y)
 468                                 for y in yl ]
 469
 470         def get_month_links(self, year):
 471                 am = [ i[1] for i in self.actmonths if i[0] == year ]
 472                 ml = []
 473                 for i in range(1, 13):
 474                         name = calendar.month_name[i][:3]
 475                         if i in am:
 476                                 s = '<a href="%s/%d/%d/">%s</a>' % \
 477                                         ( blog_url, year, i, name )
 478                         else:
 479                                 s = name
 480                         ml.append(s)
 481                 return ml
 482
 483 #
 484 # Main
 485 #
 486
 487
 488 def render_html(articles, db, actyear = None):
 489         template = Templates(templates_path, db, actyear)
 490         print 'Content-type: text/html; charset=utf-8\n'
 491         print template.get_main_header()
 492         for a in articles:
 493                 print template.get_article_header(a)
 494                 print a.to_html()
 495                 print template.get_article_footer(a)
 496         print template.get_main_footer()
 497
 498 def render_artlist(articles, db, actyear = None):
 499         template = Templates(templates_path, db, actyear)
 500         print 'Content-type: text/html; charset=utf-8\n'
 501         print template.get_main_header()
 502         print '<h2>Articles</h2>'
 503         for a in articles:
 504                 print '<li><a href="%(url)s/uuid/%(uuid)s">%(title)s</a></li>' \
 505                         % {     'url': blog_url,
 506                                 'uuid': a.uuid,
 507                                 'title': a.title,
 508                                 'author': a.author,
 509                         }
 510         print template.get_main_footer()
 511
 512 def render_atom(articles):
 513         if len(articles) > 0:
 514                 updated = articles[0].updated.isoformat()
 515         else:
 516                 updated = datetime.datetime.now().isoformat()
 517
 518         print 'Content-type: application/atom+xml; charset=utf-8\n'
 519         print """<?xml version="1.0" encoding="utf-8"?>
 520
 521 <feed xmlns="http://www.w3.org/2005/Atom">
 522  <title>%(title)s</title>
 523  <link rel="alternate" type="text/html" href="%(url)s"/>
 524  <link rel="self" type="application/atom+xml" href="%(url)s/atom"/>
 525  <id>%(url)s</id> <!-- TODO: find a better <id>, see RFC 4151 -->
 526  <updated>%(updated)sZ</updated>
 527
 528         """ % {
 529                 'title': title,
 530                 'url': full_url,
 531                 'updated': updated,
 532         }
 533
 534         for a in articles:
 535                 vars = a.to_vars()
 536                 vars.update( {
 537                         'url': full_url,
 538                         'contents': a.to_html(),
 539                 } )
 540                 print """
 541   <entry>
 542     <title>%(arttitle)s</title>
 543     <author><name>%(author)s</name></author>
 544     <link href="%(url)s/post/%(uuid)s" />
 545     <id>%(url)s/post/%(uuid)s</id>
 546     <summary>%(arttitle)s</summary>
 547     <published>%(ciso)sZ</published>
 548     <updated>%(uiso)sZ</updated>
 549     <content type="xhtml">
 550       <div xmlns="http://www.w3.org/1999/xhtml"><p>
 551 %(contents)s
 552       </p></div>
 553     </content>
 554   </entry>
 555                 """ % vars
 556         print "</feed>"
 557
 558
 559 def render_style():
 560         print 'Content-type: text/css\r\n\r\n',
 561         print default_css
 562
 563 def handle_cgi():
 564         import cgitb; cgitb.enable()
 565
 566         form = cgi.FieldStorage()
 567         year = int(form.getfirst("year", 0))
 568         month = int(form.getfirst("month", 0))
 569         day = int(form.getfirst("day", 0))
 570         tags = set(form.getlist("tag"))
 571         uuid = None
 572         atom = False
 573         style = False
 574         post = False
 575         artlist = False
 576
 577         if os.environ.has_key('PATH_INFO'):
 578                 path_info = os.environ['PATH_INFO']
 579                 style = path_info == '/style'
 580                 atom = path_info == '/atom'
 581                 tag = path_info.startswith('/tag/')
 582                 post = path_info.startswith('/post/')
 583                 artlist = path_info.startswith('/list')
 584                 if not style and not atom and not post and not tag \
 585                                 and not artlist:
 586                         date = path_info.split('/')[1:]
 587                         try:
 588                                 if len(date) > 1 and date[0]:
 589                                         year = int(date[0])
 590                                 if len(date) > 2 and date[1]:
 591                                         month = int(date[1])
 592                                 if len(date) > 3 and date[2]:
 593                                         day = int(date[2])
 594                         except ValueError:
 595                                 pass
 596                 elif post:
 597                         uuid = path_info.replace('/post/', '')
 598                         uuid = uuid.replace('/', '')
 599                 elif tag:
 600                         t = path_info.replace('/tag/', '')
 601                         t = t.replace('/', '')
 602                         t = urllib.unquote_plus(t)
 603                         tags = set((t,))
 604
 605         db = DB(data_path + '/db')
 606         if atom:
 607                 articles = db.get_articles(tags = tags)
 608                 articles.sort(reverse = True)
 609                 render_atom(articles[:10])
 610         elif style:
 611                 render_style()
 612         elif post:
 613                 render_html( [db.get_article(uuid)], db, year )
 614         elif artlist:
 615                 articles = db.get_articles()
 616                 articles.sort(cmp = Article.title_cmp)
 617                 render_artlist(articles, db)
 618         else:
 619                 articles = db.get_articles(year, month, day, tags)
 620                 articles.sort(reverse = True)
 621                 if not year and not month and not day and not tags:
 622                         articles = articles[:10]
 623                 render_html(articles, db, year)
 624
 625
 626 def usage():
 627         print 'Usage: %s {add|rm|update} article_path' % sys.argv[0]
 628
 629 def handle_cmd():
 630         if len(sys.argv) != 3:
 631                 usage()
 632                 return 1
 633
 634         cmd = sys.argv[1]
 635         art_path = os.path.realpath(sys.argv[2])
 636
 637         if os.path.commonprefix([data_path, art_path]) != data_path:
 638                 print "Error: article (%s) must be inside data_path (%s)" % \
 639                                 (art_path, data_path)
 640                 return 1
 641         art_path = art_path[len(data_path):]
 642
 643         if not os.path.isfile(data_path + '/db'):
 644                 open(data_path + '/db', 'w').write('')
 645         db = DB(data_path + '/db')
 646
 647         if cmd == 'add':
 648                 article = Article(art_path)
 649                 for a in db.articles:
 650                         if a == article:
 651                                 print 'Error: article already exists'
 652                                 return 1
 653                 db.articles.append(article)
 654                 article.created = datetime.datetime.now()
 655                 article.updated = datetime.datetime.now()
 656                 db.save()
 657         elif cmd == 'rm':
 658                 article = Article(art_path)
 659                 for a in db.articles:
 660                         if a == article:
 661                                 break
 662                 else:
 663                         print "Error: no such article"
 664                         return 1
 665                 db.articles.remove(a)
 666                 db.save()
 667         elif cmd == 'update':
 668                 article = Article(art_path)
 669                 for a in db.articles:
 670                         if a == article:
 671                                 break
 672                 else:
 673                         print "Error: no such article"
 674                         return 1
 675                 a.updated = datetime.datetime.now()
 676                 db.save()
 677         else:
 678                 usage()
 679                 return 1
 680
 681         return 0
 682
 683
 684 if os.environ.has_key('GATEWAY_INTERFACE'):
 685         handle_cgi()
 686 else:
 687         sys.exit(handle_cmd())
 688
 689