def search(query, offset = 0, limit = 10):
results = None
search_type = None
if query:
if len(query) > 10 and lower_hex_regex.match(query):
for d in hash_sizes:
if len(query) == hash_sizes[d]:
search_type = '%s search' % d.upper()
results = File.all().filter('%s =' % d, query)
break
if not search_type:
search_type = 'SHA1 prefix search'
results = File.all().filter('sha1 >= ', query).filter('sha1 < ', query + u'\ufffd')
else: #Last resort: basename matching
match_case = 0
if query.lower() == query:
results = File.all().filter('name_lower >= ', query).filter('name_lower < ', query + u'\ufffd')
else:
match_case = 1
results = File.all().filter('name >= ', query).filter('name < ', query + u'\ufffd')
search_type = 'Filename prefix search%s' % ['', ', matching case,'][match_case]
#db.GqlQuery("SELECT * FROM MyModel WHERE prop >= :1 AND prop < :2", "abc", u"abc" + u"\ufffd")
if results and limit:
results.fetch(limit)
return {'results': results, 'search_type': search_type, 'query': query, 'search_limit': limit, 'search_offset': offset}
def file_path(slug):
f = File.all().filter('abs_path = ', slug).get()
if f is None:
f = File.all().filter('slug = ', slug).get()
if f is None:
return u''
return f.get_absolute_url()
def get(self):
#TODO insert admin logs
#self.response.headers['Content-Type'] = 'text/plain'
#Validate with url-fetch head on a few urls
for f in File.all().filter('available = ', False).fetch(1):
response = None
#self.response.out.write('URL: %s\n' % f.url())
try:
response = urlfetch.fetch(f.url(), method = urlfetch.HEAD)#, headers = {'Referer': BOT_REFERER})
except Exception, e:
f.delete()
#self.response.out.write('DEL: %s\n' % str(e))
continue
#Check response data
if not response.status_code == 200:
f.delete()
#self.response.out.write('DEL: status %i\n' % response.status_code)
continue
#self.response.out.write(str(dir(response)))
#Read content-MD5/SHA1 etc.
#Content-Length, Content-MD5
try:
f.size = int(response.headers['content-length'])
f.mime_type = response.headers['content-type']
#TODO Add support for http://www.ietf.org/rfc/rfc1864.txt MD5 as base64 encoded digest
#md = response.headers['content-md5'].lower()
#if len(md) == 32:
# f.md5 = md
except Exception:
pass
f.available = True
f.put()
def get(self):
"Responds to GET requets with the admin interface"
# query the datastore for images owned by
# the current user. You can't see anyone elses images
# in the admin
test = geturl("http://org-images.appspot.com/remote/upload/image/[valid key]") # add valid realmkey
url = test.get_url()
user = users.get_current_user()
realms = RealmKeys.all()
images = Image.all().filter("user ="******"-date")
cssfilesq = File.all()
jscriptfilesq = File.all()
cssfiles = cssfilesq.filter("content_type =", "text/css").order("-date")
jscriptfiles = jscriptfilesq.filter("content_type =", "text/x-c").order("-date")
blobs = BlobFile.all()
blobs.filter("user ="******"-date")
# we are enforcing loggins so we know we have a user
# we need the logout url for the frontend
logout = users.create_logout_url("/")
# prepare the context for the template
context = {
"testurl": url,
"blobuploadurl": blobstore.create_upload_url("/upload/blob"),
"blobs": blobs,
"cssfiles": cssfiles,
"jscriptfiles": jscriptfiles,
"images": images,
"logout": logout,
"realms": realms,
}
# calculate the template path
path = os.path.join(os.path.dirname(__file__), "templates", "index.html")
# render the template with the provided context
self.response.out.write(template.render(path, context))
def get(self):
files = File.all().order('-inserted').fetch(file_count+1)
more = True if len(files) > file_count else False
vals = {
'files' : files,
'file_count' : file_count if more else len(files),
'more' : more,
}
self.template( 'file-list.html', vals, 'admin' );
def list(request):
files = dict([(f.get_key(), f) for f in File.all()])
root = None
for file in files:
file = files[file]
if file.path == '0':
root = file
break
if root is not None:
root = rec(root, files)
return render_template('app:files/list.html', root=root)
def delete(request, key):
form = FileConfirmDeleteForm(request.form)
if request.method == "POST" and form.validate():
if form.drop.data is True:
File.drop(key)
return redirect(url_for('nut:files/list'), 301)
if form.cascade.data is True:
File.drop(key, cascade=True)
return redirect(url_for('nut:files/list'), 301)
file = File.get(key)
files = dict([(n.get_key(), n) for n in File.all().filter("ancestors = ", key)])
file = rec(file, files)
return render_template('app:files/confirm_delete.html', file=file, form=form)
def get(self):
#Clean up Imported List
[f.delete() for f in ImportedList.all().filter('ctime < ', datetime.datetime.now() - datetime.timedelta(days = 7))]
#Clean up Hit List
#TODO rewrite to make sure we always have a few redirects left
hitCount = Hit.all().count()
if hitCount > 30:
[f.delete() for f in Hit.all().filter('ctime < ', datetime.datetime.now() - datetime.timedelta(days = 30))]
#Remove old files
for f in File.all().filter('available = ', True).filter('mtime < ', datetime.datetime.now() - datetime.timedelta(days = 32)).fetch(100):
f.available = False
f.put()
def get(self, filename):
file = File.all().filter('filename =', filename).get()
if file:
self.send_blob(file.blob, save_as=True)
else:
self.error(404)
def show(request, key, type):
file = File.all().filter('abs_path = ', key)\
.filter('content_type =', file_ext_to_content_type[type]).get()
return Response(file.data, mimetype=file.content_type)
def get(self):
self.response.headers['Content-Type'] = 'text/plain'
devhelp_url = '\n\nPlease visit http://www.dynmirror.net/help/developers/ for more information.'
#If url is given, use that as the digest
digest = os.path.basename(self.request.path)
if len(digest) and not digest.endswith('.metalink'):
return self.not_found(
'404 not found: If you do file look-ups by digest, it needs to end in ".metalink".'
+ devhelp_url)
digest = digest[:-len('.metalink')]
digests = {}
url = self.request.get('url')
fileLink = LinkElement()
if len(url):
#Try to load the digest from the database
url = File.urlparse(url).geturl()
#Add Coral link if wanted
if self.request.get('coral', default_value='not_set') != 'not_set':
cu = urlparse.urlparse(url) #Parse for the coral url
if cu.scheme == 'http' and not cu.netloc.endswith('.nyud.net'):
fileLink.urls.append({
'v':
'http://' + cu.hostname + '.nyud.net' + cu.path,
'a': {
'type': File.urlparse(url).scheme
}
})
Hit(path=self.request.path,
remote_addr=self.request.remote_addr,
referer=self.request.headers.get('referer', ''),
comment=url).save()
f = File.get_by_key_name(url)
if f == None:
#Host a link only metalink
fileLink.name = os.path.basename(url)
fileLink.urls.insert(0, {
'v': url,
'a': {
'type': File.urlparse(url).scheme
}
})
self.response.headers[
'Content-Type'] = 'application/metalink+xml'
self.response.headers[
'Content-Disposition'] = 'attachment; filename="%s.metalink"' % os.path.basename(
url).replace('"', '\\"')
return self.render_to_response(
'metalink.xml', {
'files': [fileLink],
'comment':
'Link only, because no validated link information could be found in the database. Make sure you use the _exact_ url you used to add the metadata.'
})
#inherit all the digest information from the file in db
fileLink.name = f.name
fileLink.size = f.size
fileLink.digests.update(f.digests())
if len(digest) > 0:
return self.response.out.write(
'For security reasons, you can not combine digest and url, because we can not determine which is authorative and what to do in case they do not match in the database.'
+ devhelp_url)
if len(digest) > 0:
#Malformed digest
if not hash_sizes.has_key(len(digest)):
return self.not_found(
'404 Not Found, the digest is considered malformed. Make sure it is lowercase hex representing an MD5, SHA1, SHA256 or SHA512.'
+ devhelp_url)
if not lower_hex_regex.match(digest):
return self.not_found(
'404 Not Found, the digest is considered malformed because it did not match /[0-9a-f]/.'
+ devhelp_url)
#OK, so a digest is given, set the digest in the fileLink to the given value
fileLink.digests[hash_sizes[len(digest)]] = digest
#TODO Unique hosts; hosts = set
#Explode the file using all known digests, both url and digest
names = {}
sizes = {}
cntry_regex = re.compile(
'.*\.(a[cdefgilmnoqrstuwxz]|c[acdfghiklmnoruvxyz]|b[abdefghijmnorstvwyz]|e[ceghrstu]|d[ejkmoz]|g[abdefghilmnpqrstuwy]|f[ijkmor]|i[delmnoqrst]|h[kmnrtu]|k[eghimnprwyz]|j[emop]|m[acdeghklmnopqrstuvwxyz]|l[abcikrstuvy]|o[m]|n[acefgilopruz]|q[a]|p[aefghklmnrstwy]|s[abcdeghijklmnortuvyz]|r[eosuw]|u[agkmsyz]|t[cdfghjklmnoprtvwz]|w[fs]|v[aceginu]|y[etu]|z[amw])$'
)
for digest_type in fileLink.digests:
files = File.all().filter('%s = ' % digest_type,
fileLink.digests[digest_type]).fetch(20)
for f in files:
attr = {
'type': 'http'
} #Optimization, currently only HTTP supported f.scheme
cntry = cntry_regex.match(f.hostname)
if cntry:
attr['location'] = cntry.group(1)
fileLink.urls.append({'v': f.url(), 'a': attr})
names.setdefault(f.name, 0)
names[f.name] += 1
sizes.setdefault(f.size, 0)
sizes[f.size] += 1
#If a name is given, just rename the file
name = self.request.get('name')
if name:
fileLink.name = name
if not fileLink.name:
#Democratic naming
names = [(names[k], k) for k in names]
names.sort()
fileLink.name = names[-1][1]
if not fileLink.size:
#Democratic size
sizes[0] = 0
sizes = [(sizes[k], k) for k in sizes]
sizes.sort()
fileLink.size = sizes[-1][1]
random.shuffle(fileLink.urls)
self.response.headers['Content-Type'] = 'application/metalink+xml'
self.response.headers[
'Content-Disposition'] = 'attachment; filename="%s.metalink"' % os.path.basename(
fileLink.name).replace('"', '\\"')
return self.render_to_response('metalink.xml', {'files': [fileLink]})
def get(self):
self.response.headers['Content-Type'] = 'text/plain'
devhelp_url = '\n\nPlease visit http://www.dynmirror.net/help/developers/ for more information.'
#If url is given, use that as the digest
digest = os.path.basename(self.request.path)
if len(digest) and not digest.endswith('.metalink'):
return self.not_found('404 not found: If you do file look-ups by digest, it needs to end in ".metalink".' + devhelp_url)
digest = digest[:-len('.metalink')]
digests = {}
url = self.request.get('url')
fileLink = LinkElement()
if len(url):
#Try to load the digest from the database
url = File.urlparse(url).geturl()
#Add Coral link if wanted
if self.request.get('coral', default_value = 'not_set') != 'not_set':
cu = urlparse.urlparse(url) #Parse for the coral url
if cu.scheme == 'http' and not cu.netloc.endswith('.nyud.net'):
fileLink.urls.append({'v': 'http://' + cu.hostname + '.nyud.net' + cu.path, 'a':{'type': File.urlparse(url).scheme}})
Hit(path = self.request.path, remote_addr = self.request.remote_addr, referer = self.request.headers.get('referer', ''), comment = url).save()
f = File.get_by_key_name(url)
if f == None:
#Host a link only metalink
fileLink.name = os.path.basename(url)
fileLink.urls.insert(0, {'v': url, 'a':{'type': File.urlparse(url).scheme}})
self.response.headers['Content-Type'] = 'application/metalink+xml'
self.response.headers['Content-Disposition'] = 'attachment; filename="%s.metalink"' % os.path.basename(url).replace('"', '\\"')
return self.render_to_response('metalink.xml', {'files': [fileLink], 'comment': 'Link only, because no validated link information could be found in the database. Make sure you use the _exact_ url you used to add the metadata.'})
#inherit all the digest information from the file in db
fileLink.name = f.name
fileLink.size = f.size
fileLink.digests.update(f.digests())
if len(digest) > 0:
return self.response.out.write('For security reasons, you can not combine digest and url, because we can not determine which is authorative and what to do in case they do not match in the database.' + devhelp_url)
if len(digest) > 0:
#Malformed digest
if not hash_sizes.has_key(len(digest)):
return self.not_found('404 Not Found, the digest is considered malformed. Make sure it is lowercase hex representing an MD5, SHA1, SHA256 or SHA512.' + devhelp_url)
if not lower_hex_regex.match(digest):
return self.not_found('404 Not Found, the digest is considered malformed because it did not match /[0-9a-f]/.' + devhelp_url)
#OK, so a digest is given, set the digest in the fileLink to the given value
fileLink.digests[hash_sizes[len(digest)]] = digest
#TODO Unique hosts; hosts = set
#Explode the file using all known digests, both url and digest
names = {}
sizes = {}
cntry_regex = re.compile('.*\.(a[cdefgilmnoqrstuwxz]|c[acdfghiklmnoruvxyz]|b[abdefghijmnorstvwyz]|e[ceghrstu]|d[ejkmoz]|g[abdefghilmnpqrstuwy]|f[ijkmor]|i[delmnoqrst]|h[kmnrtu]|k[eghimnprwyz]|j[emop]|m[acdeghklmnopqrstuvwxyz]|l[abcikrstuvy]|o[m]|n[acefgilopruz]|q[a]|p[aefghklmnrstwy]|s[abcdeghijklmnortuvyz]|r[eosuw]|u[agkmsyz]|t[cdfghjklmnoprtvwz]|w[fs]|v[aceginu]|y[etu]|z[amw])$')
for digest_type in fileLink.digests:
files = File.all().filter('%s = ' % digest_type, fileLink.digests[digest_type]).fetch(20)
for f in files:
attr = {'type': 'http'}#Optimization, currently only HTTP supported f.scheme
cntry = cntry_regex.match(f.hostname)
if cntry:
attr['location'] = cntry.group(1)
fileLink.urls.append({'v': f.url(), 'a': attr})
names.setdefault(f.name, 0)
names[f.name] += 1
sizes.setdefault(f.size, 0)
sizes[f.size] += 1
#If a name is given, just rename the file
name = self.request.get('name')
if name:
fileLink.name = name
if not fileLink.name:
#Democratic naming
names = [(names[k], k) for k in names]
names.sort()
fileLink.name = names[-1][1]
if not fileLink.size:
#Democratic size
sizes[0] = 0
sizes = [(sizes[k], k) for k in sizes]
sizes.sort()
fileLink.size = sizes[-1][1]
random.shuffle(fileLink.urls)
self.response.headers['Content-Type'] = 'application/metalink+xml'
self.response.headers['Content-Disposition'] = 'attachment; filename="%s.metalink"' % os.path.basename(fileLink.name).replace('"', '\\"')
return self.render_to_response('metalink.xml', {'files': [fileLink]})
def get(self):
limit = 20
return self.render_to_response('latest.html', {'limit': limit, 'files': File.all().order('-mtime').fetch(limit)})
