1 # -*- coding: utf-8 -*-
2 # This file is part of Wolnelektury, licensed under GNU Affero GPLv3 or later.
3 # Copyright © Fundacja Nowoczesna Polska. See NOTICE for more information.
5 from __future__ import with_statement
10 from base64 import urlsafe_b64encode
12 from django.http import HttpResponse, HttpResponseRedirect, Http404, HttpResponsePermanentRedirect
13 from django.core.files.uploadedfile import UploadedFile
14 from django.core.files.base import File
15 from django.core.files.storage import DefaultStorage
16 from django.utils.encoding import force_unicode
17 from django.utils.hashcompat import sha_constructor
18 from django.conf import settings
19 from celery.task import task
20 from os import mkdir, path, unlink
21 from errno import EEXIST, ENOENT
22 from fcntl import flock, LOCK_EX
23 from zipfile import ZipFile
24 from traceback import print_exc
26 from reporting.utils import read_chunks
27 from celery.task import task
28 import catalogue.models
30 # Use the system (hardware-based) random number generator if it exists.
31 if hasattr(random, 'SystemRandom'):
32 randrange = random.SystemRandom().randrange
34 randrange = random.randrange
35 MAX_SESSION_KEY = 18446744073709551616L # 2 << 63
38 def get_random_hash(seed):
39 sha_digest = sha_constructor('%s%s%s%s' %
40 (randrange(0, MAX_SESSION_KEY), time.time(), unicode(seed).encode('utf-8', 'replace'),
41 settings.SECRET_KEY)).digest()
42 return urlsafe_b64encode(sha_digest).replace('=', '').replace('_', '-').lower()
48 result.setdefault(tag.category, []).append(tag)
52 class ExistingFile(UploadedFile):
54 def __init__(self, path, *args, **kwargs):
56 super(ExistingFile, self).__init__(*args, **kwargs)
58 def temporary_file_path(self):
65 class LockFile(object):
67 A file lock monitor class; createas an ${objname}.lock
68 file in directory dir, and locks it exclusively.
69 To be used in 'with' construct.
71 def __init__(self, dir, objname):
72 self.lockname = path.join(dir, objname + ".lock")
75 self.lock = open(self.lockname, 'w')
76 flock(self.lock, LOCK_EX)
78 def __exit__(self, *err):
82 if oe.errno != oe.EEXIST:
88 def create_zip(paths, zip_slug):
90 Creates a zip in MEDIA_ROOT/zip directory containing files from path.
91 Resulting archive filename is ${zip_slug}.zip
92 Returns it's path relative to MEDIA_ROOT (no initial slash)
94 # directory to store zip files
95 zip_path = path.join(settings.MEDIA_ROOT, 'zip')
100 if oe.errno != EEXIST:
102 zip_filename = zip_slug + ".zip"
104 with LockFile(zip_path, zip_slug):
105 if not path.exists(path.join(zip_path, zip_filename)):
106 zipf = ZipFile(path.join(zip_path, zip_filename), 'w')
108 for arcname, p in paths:
110 arcname = path.basename(p)
111 zipf.write(p, arcname)
115 return 'zip/' + zip_filename
118 def remove_zip(zip_slug):
120 removes the ${zip_slug}.zip file from zip store.
122 zip_file = path.join(settings.MEDIA_ROOT, 'zip', zip_slug + '.zip')
125 except OSError as oe:
126 if oe.errno != ENOENT:
130 class AttachmentHttpResponse(HttpResponse):
131 """Response serving a file to be downloaded.
133 def __init__ (self, file_path, file_name, mimetype):
134 super(AttachmentHttpResponse, self).__init__(mimetype=mimetype)
135 self['Content-Disposition'] = 'attachment; filename=%s' % file_name
136 self.file_path = file_path
137 self.file_name = file_name
139 with open(DefaultStorage().path(self.file_path)) as f:
140 for chunk in read_chunks(f):
144 def async_build_pdf(book_id, customizations, file_name):
146 A celery task to generate pdf files.
147 Accepts the same args as Book.build_pdf, but with book id as first parameter
148 instead of Book instance
151 book = catalogue.models.Book.objects.get(id=book_id)
152 print "will gen %s" % DefaultStorage().path(file_name)
153 if not DefaultStorage().exists(file_name):
154 book.build_pdf(customizations=customizations, file_name=file_name)
157 print "Error during pdf creation: %s" % e
162 class MultiQuerySet(object):
163 def __init__(self, *args, **kwargs):
164 self.querysets = args
169 self._count = sum(len(qs) for qs in self.querysets)
175 def __getitem__(self, item):
177 indices = (offset, stop, step) = item.indices(self.count())
178 except AttributeError:
179 # it's not a slice - make it one
180 return self[item : item + 1][0]
182 total_len = stop - offset
183 for qs in self.querysets:
187 items += list(qs[offset:stop])
188 if len(items) >= total_len:
192 stop = total_len - len(items)
196 def truncate_html_words(s, num, end_text='...'):
197 """Truncates HTML to a certain number of words (not counting tags and
198 comments). Closes opened tags if they were correctly closed in the given
199 html. Takes an optional argument of what should be used to notify that the
200 string has been truncated, defaulting to ellipsis (...).
202 Newlines in the HTML are preserved.
204 This is just a version of django.utils.text.truncate_html_words with no space before the end_text.
210 html4_singlets = ('br', 'col', 'link', 'base', 'img', 'param', 'area', 'hr', 'input')
211 # Set up regular expressions
212 re_words = re.compile(r'&.*?;|<.*?>|(\w[\w-]*)', re.U)
213 re_tag = re.compile(r'<(/)?([^ ]+?)(?: (/)| .*?)?>')
214 # Count non-HTML words and keep note of open tags
219 while words <= length:
220 m = re_words.search(s, pos)
222 # Checked through whole string
226 # It's an actual non-HTML word
232 tag = re_tag.match(m.group(0))
233 if not tag or end_text_pos:
234 # Don't worry about non tags or tags after our truncate point
236 closing_tag, tagname, self_closing = tag.groups()
237 tagname = tagname.lower() # Element names are always case-insensitive
238 if self_closing or tagname in html4_singlets:
241 # Check for match in open tags list
243 i = open_tags.index(tagname)
247 # SGML: An end tag closes, back to the matching start tag, all unclosed intervening start tags with omitted end tags
248 open_tags = open_tags[i+1:]
250 # Add it to the start of the open tags list
251 open_tags.insert(0, tagname)
253 # Don't try to close tags if we don't need to truncate
255 out = s[:end_text_pos]
258 # Close any tags still open
259 for tag in open_tags: