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