def run():
"""
如果文件不存在，则创建
:return:
"""
if not os.path.exists('./res'):
os.makedirs('res')
config = get_config()
if not os.path.exists(config['url']) or not os.path.exists(
config['title'] or not os.path.exists(config['content'])):
data_load(config)
if not os.path.exists(config['content_clean']):
data_clean_content(config)
if not os.path.exists(config['content_filter']):
filter_stop_word(config)
if not os.path.exists(config['content_stemming']):
stemming(config)
if not os.path.exists(config['term_list']):
create_term_list(config)
documents = get_content(config)
tf_documents = get_tf(documents)
if not os.path.exists(config['idf']):
create_idf(config, documents)
idf_documents = get_idf(config)
if not os.path.exists(config['tf_idf']):
create_tf_idf(config, tf_documents, idf_documents, documents)
def get(self):
namespace = self.request.get('namespace', 'default-gzip')
# Support 'hash' for compatibility with old links. To remove eventually.
digest = self.request.get('digest', '') or self.request.get('hash', '')
params = {
u'digest': unicode(digest),
u'namespace': unicode(namespace),
}
# Check for existence of element, so we can 400/404
if digest and namespace:
try:
model.get_content(namespace, digest)
except ValueError:
self.abort(400, 'Invalid key')
except LookupError:
self.abort(404, 'Unable to retrieve the entry')
self.response.write(template.render('isolate/browse.html', params))
def GET(self, page, content = None):
c = model.get_content(content)
f = self.content_form()
if c:
f.content.set_value(c.content)
f.title.set_value(c.title)
f.draft.set_value(c.draft)
f.page.set_value(page)
f.c_id.set_value(content)
return render.edit_page_content(page, content, f)
def POST(self, page, c = None):
data = self.content_form()
if data.validates():
d = data.d
cur = model.get_content(d.c_id)
if cur:
model.update_content(d.c_id, d.page, d.title, d.content, int(d.draft))
else:
model.add_page_content(d.page, d.title, d.content, int(d.draft))
raise web.seeother('/page/'+d.page+"/content")
else:
raise web.seeother('/page')
def get(self):
namespace = self.request.get('namespace', 'default-gzip')
digest = self.request.get('digest', '')
content = None
if digest and namespace:
try:
raw_data, entity = model.get_content(namespace, digest)
except ValueError:
self.abort(400, 'Invalid key')
except LookupError:
self.abort(404, 'Unable to retrieve the entry')
if not raw_data:
stream = gcs.read_file(config.settings().gs_bucket, entity.key.id())
else:
stream = [raw_data]
content = ''.join(model.expand_content(namespace, stream))
self.response.headers['X-Frame-Options'] = 'SAMEORIGIN'
# We delete Content-Type before storing to it to avoid having two (yes,
# two) Content-Type headers.
del self.response.headers['Content-Type']
# Apparently, setting the content type to text/plain encourages the
# browser (Chrome, at least) to sniff the mime type and display
# things like images. Images are autowrapped in <img> and text is
# wrapped in <pre>.
self.response.headers['Content-Type'] = 'text/plain; charset=utf-8'
self.response.headers['Content-Disposition'] = str('filename=%s' % digest)
if content.startswith('{'):
# Try to format as JSON.
try:
content = json.dumps(
json.loads(content), sort_keys=True, indent=2,
separators=(',', ': '))
# If we don't wrap this in html, browsers will put content in a pre
# tag which is also styled with monospace/pre-wrap. We can't use
# anchor tags in <pre>, so we force it to be a <div>, which happily
# accepts links.
content = (
'<div style="font-family:monospace;white-space:pre-wrap;">%s</div>'
% content)
# Linkify things that look like hashes
content = re.sub(r'([0-9a-f]{40})',
r'<a target="_blank" href="/browse?namespace=%s' % namespace +
r'&digest=\1">\1</a>',
content)
self.response.headers['Content-Type'] = 'text/html; charset=utf-8'
except ValueError:
pass
self.response.write(content)
def get(self):
namespace = self.request.get('namespace', 'default-gzip')
digest = self.request.get('digest', '')
content = None
if not digest:
self.abort(400, 'Missing digest')
if not namespace:
self.abort(400, 'Missing namespace')
try:
raw_data, entity = model.get_content(namespace, digest)
except ValueError:
self.abort(400, 'Invalid key')
except LookupError:
self.abort(404, 'Unable to retrieve the entry')
logging.info('%s', entity)
if not raw_data:
try:
stream = gcs.read_file(config.settings().gs_bucket,
entity.key.id())
content = ''.join(model.expand_content(namespace, stream))
except cloudstorage.NotFoundError:
logging.error(
'Entity in DB but not in GCS: deleting entity in DB')
entity.key.delete()
self.abort(404, 'Unable to retrieve the file from GCS')
else:
content = ''.join(model.expand_content(namespace, [raw_data]))
self.response.headers['X-Frame-Options'] = 'SAMEORIGIN'
# We delete Content-Type before storing to it to avoid having two (yes,
# two) Content-Type headers.
del self.response.headers['Content-Type']
# Apparently, setting the content type to text/plain encourages the
# browser (Chrome, at least) to sniff the mime type and display
# things like images. Images are autowrapped in <img> and text is
# wrapped in <pre>.
self.response.headers['Content-Type'] = 'text/plain; charset=utf-8'
# App Engine puts a limit of 33554432 bytes on a request, which includes
# headers. Headers are ~150 bytes. If the content + headers might
# exceed that limit, we give the user an option to workround getting
# their file.
if len(content) > 33554000:
host = modules.get_hostname(module='default', version='default')
# host is something like default.default.myisolateserver.appspot.com
host = host.replace('default.default.', '')
sizeInMib = len(content) / (1024.0 * 1024.0)
content = (
'Sorry, your file is %1.1f MiB big, which exceeds the 32 MiB'
' App Engine limit.\nTo work around this, run the following command:\n'
' python isolateserver.py download -I %s --namespace %s -f %s %s'
% (sizeInMib, host, namespace, digest, digest))
else:
self.response.headers['Content-Disposition'] = str(
'filename=%s' % self.request.get('as') or digest)
try:
json_data = json.loads(content)
if self._is_isolated_format(json_data):
self.response.headers[
'Content-Type'] = 'text/html; charset=utf-8'
json_data['files'] = collections.OrderedDict(
sorted(json_data['files'].items(),
key=lambda (filepath, data): filepath))
params = {
'namespace': namespace,
'isolated': json_data,
}
content = template.render('isolate/isolated.html', params)
except ValueError:
pass
self.response.write(content)
def get(self):
namespace = self.request.get('namespace', 'default-gzip')
digest = self.request.get('digest', '')
content = None
if digest and namespace:
try:
raw_data, entity = model.get_content(namespace, digest)
except ValueError:
self.abort(400, 'Invalid key')
except LookupError:
self.abort(404, 'Unable to retrieve the entry')
if not raw_data:
stream = gcs.read_file(config.settings().gs_bucket,
entity.key.id())
else:
stream = [raw_data]
content = ''.join(model.expand_content(namespace, stream))
self.response.headers['X-Frame-Options'] = 'SAMEORIGIN'
# We delete Content-Type before storing to it to avoid having two (yes,
# two) Content-Type headers.
del self.response.headers['Content-Type']
# Apparently, setting the content type to text/plain encourages the
# browser (Chrome, at least) to sniff the mime type and display
# things like images. Images are autowrapped in <img> and text is
# wrapped in <pre>.
self.response.headers['Content-Type'] = 'text/plain; charset=utf-8'
# App Engine puts a limit of 33554432 bytes on a request, which includes
# headers. Headers are ~150 bytes. If the content + headers might
# exceed that limit, we give the user an option to workround getting
# their file.
if len(content) > 33554000:
host = modules.get_hostname(module='default',
version='default')
# host is something like default.default.myisolateserver.appspot.com
host = host.replace('default.default.', '')
sizeInMib = len(content) / (1024.0 * 1024.0)
content = (
'Sorry, your file is %1.1f MiB big, which exceeds the 32 MiB'
' App Engine limit.\nTo work around this, run the following command:\n'
' python isolateserver.py download -I %s --namespace %s -f %s %s'
% (sizeInMib, host, namespace, digest, digest))
else:
self.response.headers['Content-Disposition'] = str(
'filename=%s' % digest)
if content.startswith('{'):
# Try to format as JSON.
try:
content = json.dumps(json.loads(content),
sort_keys=True,
indent=2,
separators=(',', ': '))
# If we don't wrap this in html, browsers will put content in a pre
# tag which is also styled with monospace/pre-wrap. We can't use
# anchor tags in <pre>, so we force it to be a <div>, which happily
# accepts links.
content = (
'<div style="font-family:monospace;white-space:pre-wrap;">%s'
'</div>' % content)
# Linkify things that look like hashes
content = re.sub(
r'([0-9a-f]{40})',
r'<a target="_blank" href="/browse?namespace=%s' %
namespace + r'&digest=\1">\1</a>', content)
self.response.headers[
'Content-Type'] = 'text/html; charset=utf-8'
except ValueError:
pass
self.response.write(content)
def GET(self, id):
post_content = model.get_content(int(id))
post_comments = model.get_comment(int(id))
return render.view(post_content, post_comments)
