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( {
 253                         'arttitle': article.title,
 254                         'author': article.author,
 255                         'date': article.created.isoformat(' '),
 256                         'uuid': article.uuid,
 257                         'created': article.created.isoformat(' '),
 258                         'updated': article.updated.isoformat(' '),
 259                         'tags': article.get_tags_links(),
 260
 261                         'cyear': article.created.year,
 262                         'cmonth': article.created.month,
 263                         'cday': article.created.day,
 264                         'chour': article.created.hour,
 265                         'cminute': article.created.minute,
 266                         'csecond': article.created.second,
 267
 268                         'uyear': article.updated.year,
 269                         'umonth': article.updated.month,
 270                         'uday': article.updated.day,
 271                         'uhour': article.updated.hour,
 272                         'uminute': article.updated.minute,
 273                         'usecond': article.updated.second,
 274                 } )
 275
 276                 p = self.tpath + '/art_header.html'
 277                 if os.path.isfile(p):
 278                         return open(p).read() % avars
 279                 return default_article_header % avars
 280
 281         def get_article_footer(self, article):
 282                 avars = self.vars.copy()
 283                 avars.update( {
 284                         'arttitle': article.title,
 285                         'author': article.author,
 286                         'date': article.created.isoformat(' '),
 287                         'uuid': article.uuid,
 288                         'created': article.created.isoformat(' '),
 289                         'updated': article.updated.isoformat(' '),
 290                         'tags': article.get_tags_links(),
 291
 292                         'cyear': article.created.year,
 293                         'cmonth': article.created.month,
 294                         'cday': article.created.day,
 295                         'chour': article.created.hour,
 296                         'cminute': article.created.minute,
 297                         'csecond': article.created.second,
 298
 299                         'uyear': article.updated.year,
 300                         'umonth': article.updated.month,
 301                         'uday': article.updated.day,
 302                         'uhour': article.updated.hour,
 303                         'uminute': article.updated.minute,
 304                         'usecond': article.updated.second,
 305                 } )
 306
 307                 p = self.tpath + '/art_footer.html'
 308                 if os.path.isfile(p):
 309                         return open(p).read() % avars
 310                 return default_article_footer % avars
 311
 312
 313 class Article (object):
 314         def __init__(self, path):
 315                 self.path = path
 316                 self.created = None
 317                 self.updated = None
 318                 self.uuid = "%08x" % zlib.crc32(self.path)
 319
 320                 self.loaded = False
 321
 322                 # loaded on demand
 323                 self._title = 'Removed post'
 324                 self._author = author
 325                 self._tags = []
 326                 self._raw_content = ''
 327
 328
 329         def get_title(self):
 330                 if not self.loaded:
 331                         self.load()
 332                 return self._title
 333         title = property(fget = get_title)
 334
 335         def get_author(self):
 336                 if not self.loaded:
 337                         self.load()
 338                 return self._author
 339         author = property(fget = get_author)
 340
 341         def get_tags(self):
 342                 if not self.loaded:
 343                         self.load()
 344                 return self._tags
 345         tags = property(fget = get_tags)
 346
 347         def get_raw_content(self):
 348                 if not self.loaded:
 349                         self.load()
 350                 return self._raw_content
 351         raw_content = property(fget = get_raw_content)
 352
 353
 354         def __cmp__(self, other):
 355                 if self.path == other.path:
 356                         return 0
 357                 if not self.created:
 358                         return 1
 359                 if not other.created:
 360                         return -1
 361                 if self.created < other.created:
 362                         return -1
 363                 return 1
 364
 365         def title_cmp(self, other):
 366                 return cmp(self.title, other.title)
 367
 368
 369         def load(self):
 370                 try:
 371                         raw = open(data_path + '/' + self.path).readlines()
 372                 except:
 373                         return
 374
 375                 count = 0
 376                 for l in raw:
 377                         if ':' in l:
 378                                 name, value = l.split(':', 1)
 379                                 if name.lower() == 'title':
 380                                         self._title = value
 381                                 elif name.lower() == 'author':
 382                                         self._author = value
 383                                 elif name.lower() == 'tags':
 384                                         ts = value.split(',')
 385                                         ts = [t.strip() for t in ts]
 386                                         self._tags = set(ts)
 387                         elif l == '\n':
 388                                 # end of header
 389                                 break
 390                         count += 1
 391                 self._raw_content = ''.join(raw[count + 1:])
 392                 self.loaded = True
 393
 394         def to_html(self):
 395                 try:
 396                         raw = open(data_path + '/' + self.path).readlines()
 397                 except:
 398                         return "Can't open post file<p>"
 399                 raw = raw[raw.index('\n'):]
 400
 401                 settings = {
 402                         'input_encoding': encoding,
 403                         'output_encoding': 'utf8',
 404                 }
 405                 parts = publish_parts(self.raw_content,
 406                                 settings_overrides = settings,
 407                                 writer_name = "html")
 408                 return parts['body'].encode('utf8')
 409
 410         def get_tags_links(self):
 411                 l = []
 412                 tags = list(self.tags)
 413                 tags.sort()
 414                 for t in tags:
 415                         l.append('<a class="tag" href="%s/tag/%s">%s</a>' % \
 416                                 (blog_url, urllib.quote(t), t) )
 417                 return ', '.join(l)
 418
 419
 420 class DB (object):
 421         def __init__(self, dbpath):
 422                 self.dbpath = dbpath
 423                 self.articles = []
 424                 self.uuids = {}
 425                 self.actyears = set()
 426                 self.actmonths = set()
 427                 self.load()
 428
 429         def get_articles(self, year = 0, month = 0, day = 0, tags = None):
 430                 l = []
 431                 for a in self.articles:
 432                         if year and a.created.year != year: continue
 433                         if month and a.created.month != month: continue
 434                         if day and a.created.day != day: continue
 435                         if tags and not tags.issubset(a.tags): continue
 436
 437                         l.append(a)
 438
 439                 return l
 440
 441         def get_article(self, uuid):
 442                 return self.uuids[uuid]
 443
 444         def load(self):
 445                 try:
 446                         f = open(self.dbpath)
 447                 except:
 448                         return
 449
 450                 for l in f:
 451                         # Each line has the following comma separated format:
 452                         # path (relative to data_path), \
 453                         #       created (epoch), \
 454                         #       updated (epoch)
 455                         try:
 456                                 l = l.split(',')
 457                         except:
 458                                 continue
 459
 460                         a = Article(l[0])
 461                         a.created = datetime.datetime.fromtimestamp(
 462                                                 float(l[1]) )
 463                         a.updated = datetime.datetime.fromtimestamp(
 464                                                 float(l[2]))
 465                         self.uuids[a.uuid] = a
 466                         self.actyears.add(a.created.year)
 467                         self.actmonths.add((a.created.year, a.created.month))
 468                         self.articles.append(a)
 469
 470         def save(self):
 471                 f = open(self.dbpath + '.tmp', 'w')
 472                 for a in self.articles:
 473                         s = ''
 474                         s += a.path + ', '
 475                         s += str(time.mktime(a.created.timetuple())) + ', '
 476                         s += str(time.mktime(a.updated.timetuple())) + '\n'
 477                         f.write(s)
 478                 f.close()
 479                 os.rename(self.dbpath + '.tmp', self.dbpath)
 480
 481         def get_year_links(self):
 482                 yl = list(self.actyears)
 483                 yl.sort(reverse = True)
 484                 return [ '<a href="%s/%d/">%d</a>' % (blog_url, y, y)
 485                                 for y in yl ]
 486
 487         def get_month_links(self, year):
 488                 am = [ i[1] for i in self.actmonths if i[0] == year ]
 489                 ml = []
 490                 for i in range(1, 13):
 491                         name = calendar.month_name[i][:3]
 492                         if i in am:
 493                                 s = '<a href="%s/%d/%d/">%s</a>' % \
 494                                         ( blog_url, year, i, name )
 495                         else:
 496                                 s = name
 497                         ml.append(s)
 498                 return ml
 499
 500 #
 501 # Main
 502 #
 503
 504
 505 def render_html(articles, db, actyear = None):
 506         template = Templates(templates_path, db, actyear)
 507         print 'Content-type: text/html; charset=utf-8\n'
 508         print template.get_main_header()
 509         for a in articles:
 510                 print template.get_article_header(a)
 511                 print a.to_html()
 512                 print template.get_article_footer(a)
 513         print template.get_main_footer()
 514
 515 def render_artlist(articles, db, actyear = None):
 516         template = Templates(templates_path, db, actyear)
 517         print 'Content-type: text/html; charset=utf-8\n'
 518         print template.get_main_header()
 519         print '<h2>Articles</h2>'
 520         for a in articles:
 521                 print '<li><a href="%(url)s/uuid/%(uuid)s">%(title)s</a></li>' \
 522                         % {     'url': blog_url,
 523                                 'uuid': a.uuid,
 524                                 'title': a.title,
 525                                 'author': a.author,
 526                         }
 527         print template.get_main_footer()
 528
 529 def render_atom(articles):
 530         if len(articles) > 0:
 531                 updated = articles[0].updated.isoformat()
 532         else:
 533                 updated = datetime.datetime.now().isoformat()
 534
 535         print 'Content-type: application/atom+xml; charset=utf-8\n'
 536         print """<?xml version="1.0" encoding="utf-8"?>
 537
 538 <feed xmlns="http://www.w3.org/2005/Atom">
 539  <title>%(title)s</title>
 540  <link rel="alternate" type="text/html" href="%(url)s"/>
 541  <link rel="self" type="application/atom+xml" href="%(url)s/atom"/>
 542  <id>%(url)s</id> <!-- TODO: find a better <id>, see RFC 4151 -->
 543  <updated>%(updated)sZ</updated>
 544
 545         """ % {
 546                 'title': title,
 547                 'url': full_url,
 548                 'updated': updated,
 549         }
 550
 551         for a in articles:
 552                 print """
 553   <entry>
 554     <title>%(arttitle)s</title>
 555     <author><name>%(author)s</name></author>
 556     <link href="%(url)s/post/%(uuid)s" />
 557     <id>%(url)s/post/%(uuid)s</id>
 558     <summary>%(arttitle)s</summary>
 559     <published>%(created)sZ</published>
 560     <updated>%(updated)sZ</updated>
 561     <content type="xhtml">
 562       <div xmlns="http://www.w3.org/1999/xhtml"><p>
 563 %(contents)s
 564       </p></div>
 565     </content>
 566   </entry>
 567                 """ % {
 568                         'arttitle': a.title,
 569                         'author': a.author,
 570                         'uuid': a.uuid,
 571                         'url': full_url,
 572                         'created': a.created.isoformat(),
 573                         'updated': a.updated.isoformat(),
 574                         'contents': a.to_html(),
 575                 }
 576
 577         print "</feed>"
 578
 579
 580 def render_style():
 581         print 'Content-type: text/css\r\n\r\n',
 582         print default_css
 583
 584 def handle_cgi():
 585         import cgitb; cgitb.enable()
 586
 587         form = cgi.FieldStorage()
 588         year = int(form.getfirst("year", 0))
 589         month = int(form.getfirst("month", 0))
 590         day = int(form.getfirst("day", 0))
 591         tags = set(form.getlist("tag"))
 592         uuid = None
 593         atom = False
 594         style = False
 595         post = False
 596         artlist = False
 597
 598         if os.environ.has_key('PATH_INFO'):
 599                 path_info = os.environ['PATH_INFO']
 600                 style = path_info == '/style'
 601                 atom = path_info == '/atom'
 602                 tag = path_info.startswith('/tag/')
 603                 post = path_info.startswith('/post/')
 604                 artlist = path_info.startswith('/list')
 605                 if not style and not atom and not post and not tag \
 606                                 and not artlist:
 607                         date = path_info.split('/')[1:]
 608                         try:
 609                                 if len(date) > 1 and date[0]:
 610                                         year = int(date[0])
 611                                 if len(date) > 2 and date[1]:
 612                                         month = int(date[1])
 613                                 if len(date) > 3 and date[2]:
 614                                         day = int(date[2])
 615                         except ValueError:
 616                                 pass
 617                 elif post:
 618                         uuid = path_info.replace('/post/', '')
 619                         uuid = uuid.replace('/', '')
 620                 elif tag:
 621                         t = path_info.replace('/tag/', '')
 622                         t = t.replace('/', '')
 623                         t = urllib.unquote_plus(t)
 624                         tags = set((t,))
 625
 626         db = DB(data_path + '/db')
 627         if atom:
 628                 articles = db.get_articles(tags = tags)
 629                 articles.sort(reverse = True)
 630                 render_atom(articles[:10])
 631         elif style:
 632                 render_style()
 633         elif post:
 634                 render_html( [db.get_article(uuid)], db, year )
 635         elif artlist:
 636                 articles = db.get_articles()
 637                 articles.sort(cmp = Article.title_cmp)
 638                 render_artlist(articles, db)
 639         else:
 640                 articles = db.get_articles(year, month, day, tags)
 641                 articles.sort(reverse = True)
 642                 if not year and not month and not day and not tags:
 643                         articles = articles[:10]
 644                 render_html(articles, db, year)
 645
 646
 647 def usage():
 648         print 'Usage: %s {add|rm|update} article_path' % sys.argv[0]
 649
 650 def handle_cmd():
 651         if len(sys.argv) != 3:
 652                 usage()
 653                 return 1
 654
 655         cmd = sys.argv[1]
 656         art_path = os.path.realpath(sys.argv[2])
 657
 658         if os.path.commonprefix([data_path, art_path]) != data_path:
 659                 print "Error: article (%s) must be inside data_path (%s)" % \
 660                                 (art_path, data_path)
 661                 return 1
 662         art_path = art_path[len(data_path):]
 663
 664         if not os.path.isfile(data_path + '/db'):
 665                 open(data_path + '/db', 'w').write('')
 666         db = DB(data_path + '/db')
 667
 668         if cmd == 'add':
 669                 article = Article(art_path)
 670                 for a in db.articles:
 671                         if a == article:
 672                                 print 'Error: article already exists'
 673                                 return 1
 674                 db.articles.append(article)
 675                 article.created = datetime.datetime.now()
 676                 article.updated = datetime.datetime.now()
 677                 db.save()
 678         elif cmd == 'rm':
 679                 article = Article(art_path)
 680                 for a in db.articles:
 681                         if a == article:
 682                                 break
 683                 else:
 684                         print "Error: no such article"
 685                         return 1
 686                 db.articles.remove(a)
 687                 db.save()
 688         elif cmd == 'update':
 689                 article = Article(art_path)
 690                 for a in db.articles:
 691                         if a == article:
 692                                 break
 693                 else:
 694                         print "Error: no such article"
 695                         return 1
 696                 a.updated = datetime.datetime.now()
 697                 db.save()
 698         else:
 699                 usage()
 700                 return 1
 701
 702         return 0
 703
 704
 705 if os.environ.has_key('GATEWAY_INTERFACE'):
 706         handle_cgi()
 707 else:
 708         sys.exit(handle_cmd())
 709
 710