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