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                 try:
 339                         raw = open(data_path + '/' + self.path).readlines()
 340                 except:
 341                         return
 342
 343                 count = 0
 344                 for l in raw:
 345                         if ':' in l:
 346                                 name, value = l.split(':', 1)
 347                                 if name.lower() == 'title':
 348                                         self._title = value.strip()
 349                                 elif name.lower() == 'author':
 350                                         self._author = value.strip()
 351                                 elif name.lower() == 'tags':
 352                                         ts = value.split(',')
 353                                         ts = [t.strip() for t in ts]
 354                                         self._tags = set(ts)
 355                         elif l == '\n':
 356                                 # end of header
 357                                 break
 358                         count += 1
 359                 self._raw_content = ''.join(raw[count + 1:])
 360                 self.loaded = True
 361
 362         def to_html(self):
 363                 return rst_to_html(self.raw_content)
 364
 365         def to_vars(self):
 366                 return {
 367                         'arttitle': sanitize(self.title),
 368                         'author': sanitize(self.author),
 369                         'date': self.created.isoformat(' '),
 370                         'uuid': self.uuid,
 371                         'tags': self.get_tags_links(),
 372
 373                         'created': self.created.isoformat(' '),
 374                         'ciso': self.created.isoformat(),
 375                         'cyear': self.created.year,
 376                         'cmonth': self.created.month,
 377                         'cday': self.created.day,
 378                         'chour': self.created.hour,
 379                         'cminute': self.created.minute,
 380                         'csecond': self.created.second,
 381
 382                         'updated': self.updated.isoformat(' '),
 383                         'uiso': self.updated.isoformat(),
 384                         'uyear': self.updated.year,
 385                         'umonth': self.updated.month,
 386                         'uday': self.updated.day,
 387                         'uhour': self.updated.hour,
 388                         'uminute': self.updated.minute,
 389                         'usecond': self.updated.second,
 390                 }
 391
 392         def get_tags_links(self):
 393                 l = []
 394                 tags = list(self.tags)
 395                 tags.sort()
 396                 for t in tags:
 397                         l.append('<a class="tag" href="%s/tag/%s">%s</a>' % \
 398                                 (blog_url, urllib.quote(t), sanitize(t) ))
 399                 return ', '.join(l)
 400
 401
 402 class DB (object):
 403         def __init__(self, dbpath):
 404                 self.dbpath = dbpath
 405                 self.articles = []
 406                 self.uuids = {}
 407                 self.actyears = set()
 408                 self.actmonths = set()
 409                 self.load()
 410
 411         def get_articles(self, year = 0, month = 0, day = 0, tags = None):
 412                 l = []
 413                 for a in self.articles:
 414                         if year and a.created.year != year: continue
 415                         if month and a.created.month != month: continue
 416                         if day and a.created.day != day: continue
 417                         if tags and not tags.issubset(a.tags): continue
 418
 419                         l.append(a)
 420
 421                 return l
 422
 423         def get_article(self, uuid):
 424                 return self.uuids[uuid]
 425
 426         def load(self):
 427                 try:
 428                         f = open(self.dbpath)
 429                 except:
 430                         return
 431
 432                 for l in f:
 433                         # Each line has the following comma separated format:
 434                         # path (relative to data_path), \
 435                         #       created (epoch), \
 436                         #       updated (epoch)
 437                         try:
 438                                 l = l.split(',')
 439                         except:
 440                                 continue
 441
 442                         a = Article(l[0],
 443                                 datetime.datetime.fromtimestamp(float(l[1])),
 444                                 datetime.datetime.fromtimestamp(float(l[2])))
 445                         self.uuids[a.uuid] = a
 446                         self.actyears.add(a.created.year)
 447                         self.actmonths.add((a.created.year, a.created.month))
 448                         self.articles.append(a)
 449
 450         def save(self):
 451                 f = open(self.dbpath + '.tmp', 'w')
 452                 for a in self.articles:
 453                         s = ''
 454                         s += a.path + ', '
 455                         s += str(time.mktime(a.created.timetuple())) + ', '
 456                         s += str(time.mktime(a.updated.timetuple())) + '\n'
 457                         f.write(s)
 458                 f.close()
 459                 os.rename(self.dbpath + '.tmp', self.dbpath)
 460
 461         def get_year_links(self):
 462                 yl = list(self.actyears)
 463                 yl.sort(reverse = True)
 464                 return [ '<a href="%s/%d/">%d</a>' % (blog_url, y, y)
 465                                 for y in yl ]
 466
 467         def get_month_links(self, year):
 468                 am = [ i[1] for i in self.actmonths if i[0] == year ]
 469                 ml = []
 470                 for i in range(1, 13):
 471                         name = calendar.month_name[i][:3]
 472                         if i in am:
 473                                 s = '<a href="%s/%d/%d/">%s</a>' % \
 474                                         ( blog_url, year, i, name )
 475                         else:
 476                                 s = name
 477                         ml.append(s)
 478                 return ml
 479
 480 #
 481 # Main
 482 #
 483
 484
 485 def render_html(articles, db, actyear = None):
 486         template = Templates(templates_path, db, actyear)
 487         print 'Content-type: text/html; charset=utf-8\n'
 488         print template.get_main_header()
 489         for a in articles:
 490                 print template.get_article_header(a)
 491                 print a.to_html()
 492                 print template.get_article_footer(a)
 493         print template.get_main_footer()
 494
 495 def render_artlist(articles, db, actyear = None):
 496         template = Templates(templates_path, db, actyear)
 497         print 'Content-type: text/html; charset=utf-8\n'
 498         print template.get_main_header()
 499         print '<h2>Articles</h2>'
 500         for a in articles:
 501                 print '<li><a href="%(url)s/uuid/%(uuid)s">%(title)s</a></li>' \
 502                         % {     'url': blog_url,
 503                                 'uuid': a.uuid,
 504                                 'title': a.title,
 505                                 'author': a.author,
 506                         }
 507         print template.get_main_footer()
 508
 509 def render_atom(articles):
 510         if len(articles) > 0:
 511                 updated = articles[0].updated.isoformat()
 512         else:
 513                 updated = datetime.datetime.now().isoformat()
 514
 515         print 'Content-type: application/atom+xml; charset=utf-8\n'
 516         print """<?xml version="1.0" encoding="utf-8"?>
 517
 518 <feed xmlns="http://www.w3.org/2005/Atom">
 519  <title>%(title)s</title>
 520  <link rel="alternate" type="text/html" href="%(url)s"/>
 521  <link rel="self" type="application/atom+xml" href="%(url)s/atom"/>
 522  <id>%(url)s</id> <!-- TODO: find a better <id>, see RFC 4151 -->
 523  <updated>%(updated)sZ</updated>
 524
 525         """ % {
 526                 'title': title,
 527                 'url': full_url,
 528                 'updated': updated,
 529         }
 530
 531         for a in articles:
 532                 vars = a.to_vars()
 533                 vars.update( {
 534                         'url': full_url,
 535                         'contents': a.to_html(),
 536                 } )
 537                 print """
 538   <entry>
 539     <title>%(arttitle)s</title>
 540     <author><name>%(author)s</name></author>
 541     <link href="%(url)s/post/%(uuid)s" />
 542     <id>%(url)s/post/%(uuid)s</id>
 543     <summary>%(arttitle)s</summary>
 544     <published>%(ciso)sZ</published>
 545     <updated>%(uiso)sZ</updated>
 546     <content type="xhtml">
 547       <div xmlns="http://www.w3.org/1999/xhtml"><p>
 548 %(contents)s
 549       </p></div>
 550     </content>
 551   </entry>
 552                 """ % vars
 553         print "</feed>"
 554
 555
 556 def render_style():
 557         print 'Content-type: text/css\r\n\r\n',
 558         print default_css
 559
 560 def handle_cgi():
 561         import cgitb; cgitb.enable()
 562
 563         form = cgi.FieldStorage()
 564         year = int(form.getfirst("year", 0))
 565         month = int(form.getfirst("month", 0))
 566         day = int(form.getfirst("day", 0))
 567         tags = set(form.getlist("tag"))
 568         uuid = None
 569         atom = False
 570         style = False
 571         post = False
 572         artlist = False
 573
 574         if os.environ.has_key('PATH_INFO'):
 575                 path_info = os.environ['PATH_INFO']
 576                 style = path_info == '/style'
 577                 atom = path_info == '/atom'
 578                 tag = path_info.startswith('/tag/')
 579                 post = path_info.startswith('/post/')
 580                 artlist = path_info.startswith('/list')
 581                 if not style and not atom and not post and not tag \
 582                                 and not artlist:
 583                         date = path_info.split('/')[1:]
 584                         try:
 585                                 if len(date) > 1 and date[0]:
 586                                         year = int(date[0])
 587                                 if len(date) > 2 and date[1]:
 588                                         month = int(date[1])
 589                                 if len(date) > 3 and date[2]:
 590                                         day = int(date[2])
 591                         except ValueError:
 592                                 pass
 593                 elif post:
 594                         uuid = path_info.replace('/post/', '')
 595                         uuid = uuid.replace('/', '')
 596                 elif tag:
 597                         t = path_info.replace('/tag/', '')
 598                         t = t.replace('/', '')
 599                         t = urllib.unquote_plus(t)
 600                         tags = set((t,))
 601
 602         db = DB(data_path + '/db')
 603         if atom:
 604                 articles = db.get_articles(tags = tags)
 605                 articles.sort(reverse = True)
 606                 render_atom(articles[:10])
 607         elif style:
 608                 render_style()
 609         elif post:
 610                 render_html( [db.get_article(uuid)], db, year )
 611         elif artlist:
 612                 articles = db.get_articles()
 613                 articles.sort(cmp = Article.title_cmp)
 614                 render_artlist(articles, db)
 615         else:
 616                 articles = db.get_articles(year, month, day, tags)
 617                 articles.sort(reverse = True)
 618                 if not year and not month and not day and not tags:
 619                         articles = articles[:10]
 620                 render_html(articles, db, year)
 621
 622
 623 def usage():
 624         print 'Usage: %s {add|rm|update} article_path' % sys.argv[0]
 625
 626 def handle_cmd():
 627         if len(sys.argv) != 3:
 628                 usage()
 629                 return 1
 630
 631         cmd = sys.argv[1]
 632         art_path = os.path.realpath(sys.argv[2])
 633
 634         if os.path.commonprefix([data_path, art_path]) != data_path:
 635                 print "Error: article (%s) must be inside data_path (%s)" % \
 636                                 (art_path, data_path)
 637                 return 1
 638         art_path = art_path[len(data_path):]
 639
 640         if not os.path.isfile(data_path + '/db'):
 641                 open(data_path + '/db', 'w').write('')
 642         db = DB(data_path + '/db')
 643
 644         if cmd == 'add':
 645                 article = Article(art_path, datetime.datetime.now(),
 646                                         datetime.datetime.now())
 647                 for a in db.articles:
 648                         if a == article:
 649                                 print 'Error: article already exists'
 650                                 return 1
 651                 db.articles.append(article)
 652                 db.save()
 653         elif cmd == 'rm':
 654                 article = Article(art_path)
 655                 for a in db.articles:
 656                         if a == article:
 657                                 break
 658                 else:
 659                         print "Error: no such article"
 660                         return 1
 661                 db.articles.remove(a)
 662                 db.save()
 663         elif cmd == 'update':
 664                 article = Article(art_path)
 665                 for a in db.articles:
 666                         if a == article:
 667                                 break
 668                 else:
 669                         print "Error: no such article"
 670                         return 1
 671                 a.updated = datetime.datetime.now()
 672                 db.save()
 673         else:
 674                 usage()
 675                 return 1
 676
 677         return 0
 678
 679
 680 if os.environ.has_key('GATEWAY_INTERFACE'):
 681         handle_cgi()
 682 else:
 683         sys.exit(handle_cmd())
 684
 685