def save_bare_tags(page_size=5000):
print 'Starting {}...'.format(sys._getframe().f_code.co_name)
count = 0
start = datetime.now()
total = MODMTag.find().count()
while count < total:
with transaction.atomic():
tags = []
for modm_tag in MODMTag.find().sort('-_id')[count:count +
page_size]:
tags.append(Tag(_id=modm_tag._id,
lower=modm_tag.lower,
system=False))
count += 1
if count % page_size == 0 or count == total:
then = datetime.now()
print 'Saving tags {} through {}...'.format(
count - page_size, count)
woot = Tag.objects.bulk_create(tags)
now = datetime.now()
print 'Done with {} tags in {} seconds...'.format(
len(woot), (now - then).total_seconds())
tags = None
woot = None
trash = gc.collect()
print 'Took out {} trashes'.format(trash)
print 'MODM Tags: {}'.format(total)
print 'django Tags: {}'.format(Tag.objects.all().count())
print 'Done with {} in {} seconds...'.format(
sys._getframe().f_code.co_name,
(datetime.now() - start).total_seconds())
def save_bare_tags(page_size=5000):
print 'Starting {}...'.format(sys._getframe().f_code.co_name)
count = 0
start = datetime.now()
total = MODMTag.find().count()
while count < total:
with transaction.atomic():
tags = []
for modm_tag in MODMTag.find().sort('-_id')[count:count +
page_size]:
tags.append(
Tag(_id=modm_tag._id, lower=modm_tag.lower, system=False))
count += 1
if count % page_size == 0 or count == total:
then = datetime.now()
print 'Saving tags {} through {}...'.format(
count - page_size, count)
woot = Tag.objects.bulk_create(tags)
now = datetime.now()
print 'Done with {} tags in {} seconds...'.format(
len(woot), (now - then).total_seconds())
tags = None
woot = None
trash = gc.collect()
print 'Took out {} trashes'.format(trash)
print 'MODM Tags: {}'.format(total)
print 'django Tags: {}'.format(Tag.objects.all().count())
print 'Done with {} in {} seconds...'.format(
sys._getframe().f_code.co_name,
(datetime.now() - start).total_seconds())
def test_add_tag(self):
file_ = self.root.append_file('That\'s How Strong My Love Is.mp3')
tag = Tag(_id='Redding')
tag.save()
file_.tags.append(tag)
file_.save()
find = query_tag_file('Redding')['results']
assert_equal(len(find), 1)
def test_add_tag(self):
file_ = self.root.append_file('That\'s How Strong My Love Is.mp3')
tag = Tag(_id='Redding', name='Redding')
tag.save()
file_.tags.add(tag)
file_.save()
find = query_tag_file('Redding')['results']
assert_equal(len(find), 1)
def test_file_remove_tag(self):
file = self.node_settings.get_root().append_file('Champion.mp3')
tag = Tag(name='Graduation')
tag.save()
file.tags.add(tag)
file.save()
assert_in('Graduation', file.tags.values_list('name', flat=True))
url = self.project.api_url_for('osfstorage_remove_tag', fid=file._id)
self.app.delete_json(url, {'tag': 'Graduation'}, auth=self.user.auth)
file.reload()
assert_not_in('Graduation', file.tags.values_list('name', flat=True))
def test_file_add_tag_fail_doesnt_create_log(self, mock_log):
file = self.node_settings.get_root().append_file('UltraLightBeam.mp3')
tag = Tag(_id='The Life of Pablo')
tag.save()
file.tags.append(tag)
file.save()
url = self.project.api_url_for('osfstorage_add_tag', fid=file._id)
res = self.app.post_json(url, {'tag': 'The Life of Pablo'}, auth=self.user.auth, expect_errors=True)
assert_equal(res.status_code, 400)
mock_log.assert_not_called()
def test_tag_the_same_tag(self):
file = self.node_settings.get_root().append_file('Lie,Cheat,Steal.mp3')
tag = Tag(_id='Run_the_Jewels')
tag.save()
file.tags.append(tag)
file.save()
assert_in('Run_the_Jewels', file.tags)
url = self.project.api_url_for('osfstorage_add_tag', fid=file._id)
res = self.app.post_json(url, {'tag': 'Run_the_Jewels'}, auth=self.user.auth, expect_errors=True)
assert_equal(res.status_code, 400)
assert_equal(res.json['status'], 'failure')
def test_file_add_tag_fail_doesnt_create_log(self, mock_log):
file = self.node_settings.get_root().append_file('UltraLightBeam.mp3')
tag = Tag(_id='The Life of Pablo')
tag.save()
file.tags.append(tag)
file.save()
url = self.project.api_url_for('osfstorage_add_tag', fid=file._id)
res = self.app.post_json(url, {'tag': 'The Life of Pablo'}, auth=self.user.auth, expect_errors=True)
assert_equal(res.status_code, 400)
mock_log.assert_not_called()
def test_file_remove_tag(self):
file = self.node_settings.get_root().append_file('Champion.mp3')
tag = Tag(_id='Graduation')
tag.save()
file.tags.append(tag)
file.save()
assert_in('Graduation', file.tags)
url = self.project.api_url_for('osfstorage_remove_tag', fid=file._id)
self.app.delete_json(url, {'tag': 'Graduation'}, auth=self.user.auth)
file.reload()
assert_not_in('Graduation', file.tags)
def test_tag_the_same_tag(self):
file = self.node_settings.get_root().append_file('Lie,Cheat,Steal.mp3')
tag = Tag(_id='Run_the_Jewels')
tag.save()
file.tags.append(tag)
file.save()
assert_in('Run_the_Jewels', file.tags)
url = self.project.api_url_for('osfstorage_add_tag', fid=file._id)
res = self.app.post_json(url, {'tag': 'Run_the_Jewels'}, auth=self.user.auth, expect_errors=True)
assert_equal(res.status_code, 400)
assert_equal(res.json['status'], 'failure')
def test_file_remove_tag_creates_log(self):
file = self.node_settings.get_root().append_file('Formation.flac')
tag = Tag(_id='You that when you cause all this conversation')
tag.save()
file.tags.append(tag)
file.save()
url = self.project.api_url_for('osfstorage_remove_tag', fid=file._id)
res = self.app.delete_json(url, {'tag': 'You that when you cause all this conversation'}, auth=self.user.auth)
assert_equal(res.status_code, 200)
self.node.reload()
assert_equal(self.node.logs[-1].action, 'file_tag_removed')
def test_file_remove_tag_creates_log(self):
file = self.node_settings.get_root().append_file('Formation.flac')
tag = Tag(_id='You that when you cause all this conversation')
tag.save()
file.tags.append(tag)
file.save()
url = self.project.api_url_for('osfstorage_remove_tag', fid=file._id)
res = self.app.delete_json(url, {'tag': 'You that when you cause all this conversation'}, auth=self.user.auth)
assert_equal(res.status_code, 200)
self.node.reload()
assert_equal(self.node.logs[-1].action, 'file_tag_removed')
def test_remove_tag(self):
file_ = self.root.append_file('I\'ve Been Loving You Too Long.mp3')
tag = Tag(_id='Blue')
tag.save()
file_.tags.append(tag)
file_.save()
find = query_tag_file('Blue')['results']
assert_equal(len(find), 1)
file_.tags.remove('Blue')
file_.save()
find = query_tag_file('Blue')['results']
assert_equal(len(find), 0)
def test_remove_tag(self):
file_ = self.root.append_file('I\'ve Been Loving You Too Long.mp3')
tag = Tag(_id='Blue', name='Blue')
tag.save()
file_.tags.add(tag)
file_.save()
find = query_tag_file('Blue')['results']
assert_equal(len(find), 1)
file_.tags.remove(tag)
file_.save()
find = query_tag_file('Blue')['results']
assert_equal(len(find), 0)
def add_tag(self, tag, auth, save=True, log=True):
from website.models import Tag, NodeLog # Prevent import error
if tag not in self.tags and not self.node.is_registration:
new_tag = Tag.load(tag)
if not new_tag:
new_tag = Tag(_id=tag)
new_tag.save()
self.tags.append(new_tag)
if log:
self.add_tag_log(NodeLog.FILE_TAG_ADDED, tag, auth)
if save:
self.save()
return True
return False
def add_tag(self, tag, auth, save=True, log=True):
from website.models import Tag, NodeLog # Prevent import error
if tag not in self.tags and not self.node.is_registration:
new_tag = Tag.load(tag)
if not new_tag:
new_tag = Tag(_id=tag)
new_tag.save()
self.tags.append(new_tag)
if log:
self.add_tag_log(NodeLog.FILE_TAG_ADDED, tag, auth)
if save:
self.save()
return True
return False
def conference_submissions(**kwargs):
"""Return data for all OSF4M submissions.
The total number of submissions for each meeting is calculated and cached
in the Conference.num_submissions field.
"""
conferences = Conference.find(Q('is_meeting', 'ne', False))
# TODO: Revisit this loop, there has to be a way to optimize it
for conf in conferences:
# For efficiency, we filter by tag first, then node
# instead of doing a single Node query
projects = set()
tags = Tag.find(
Q('system', 'eq', False)
& Q('name', 'iexact', conf.endpoint.lower())).values_list(
'pk', flat=True)
nodes = Node.find(
Q('tags', 'in', tags) & Q('is_public', 'eq', True)
& Q('is_deleted', 'ne', True)).include('guids')
projects.update(list(nodes))
num_submissions = len(projects)
# Cache the number of submissions
conf.num_submissions = num_submissions
bulk_update(conferences, update_fields=['num_submissions'])
return {'success': True}
def conference_submissions(**kwargs):
"""Return data for all OSF4M submissions.
The total number of submissions for each meeting is calculated and cached
in the Conference.num_submissions field.
"""
submissions = []
# TODO: Revisit this loop, there has to be a way to optimize it
for conf in Conference.find():
# For efficiency, we filter by tag first, then node
# instead of doing a single Node query
projects = set()
tags = Tag.find(Q('lower', 'eq', conf.endpoint.lower())).get_keys()
nodes = Node.find(
Q('tags', 'in', tags) &
Q('is_public', 'eq', True) &
Q('is_deleted', 'ne', True)
)
projects.update(list(nodes))
for idx, node in enumerate(projects):
submissions.append(_render_conference_node(node, idx, conf))
num_submissions = len(projects)
# Cache the number of submissions
conf.num_submissions = num_submissions
conf.save()
if num_submissions < settings.CONFERENCE_MIN_COUNT:
continue
submissions.sort(key=lambda submission: submission['dateCreated'], reverse=True)
return {'submissions': submissions}
def conference_view(**kwargs):
meetings = []
submissions = []
for conf in Conference.find():
# For efficiency, we filter by tag first, then node
# instead of doing a single Node query
projects = set()
for tag in Tag.find(Q('_id', 'iexact', conf.endpoint)):
for node in tag.node__tagged:
if not node:
continue
if not node.is_public or node.is_deleted:
continue
projects.add(node)
for idx, node in enumerate(projects):
submissions.append(_render_conference_node(node, idx, conf))
num_submissions = len(projects)
if num_submissions < settings.CONFERENCE_MIN_COUNT:
continue
meetings.append({
'name': conf.name,
'active': conf.active,
'url': web_url_for('conference_results', meeting=conf.endpoint),
'count': num_submissions,
})
submissions.sort(key=lambda submission: submission['dateCreated'], reverse=True)
meetings.sort(key=lambda meeting: meeting['count'], reverse=True)
return {'meetings': meetings, 'submissions': submissions}
def conference_submissions(**kwargs):
"""Return data for all OSF4M submissions.
The total number of submissions for each meeting is calculated and cached
in the Conference.num_submissions field.
"""
submissions = []
for conf in Conference.find():
# For efficiency, we filter by tag first, then node
# instead of doing a single Node query
projects = set()
for tag in Tag.find(Q('lower', 'eq', conf.endpoint.lower())):
for node in tag.node__tagged.find(Q('is_public', 'eq', True) & Q('is_deleted', 'eq', False)):
projects.add(node)
for idx, node in enumerate(projects):
submissions.append(_render_conference_node(node, idx, conf))
num_submissions = len(projects)
# Cache the number of submissions
conf.num_submissions = num_submissions
conf.save()
if num_submissions < settings.CONFERENCE_MIN_COUNT:
continue
submissions.sort(key=lambda submission: submission['dateCreated'], reverse=True)
return {'submissions': submissions}
def conference_submissions(**kwargs):
"""Return data for all OSF4M submissions.
The total number of submissions for each meeting is calculated and cached
in the Conference.num_submissions field.
"""
submissions = []
# TODO: Revisit this loop, there has to be a way to optimize it
for conf in Conference.find():
if (hasattr(conf, 'is_meeting') and (conf.is_meeting is False)):
break
# For efficiency, we filter by tag first, then node
# instead of doing a single Node query
projects = set()
tags = Tag.find(Q('lower', 'eq', conf.endpoint.lower())).get_keys()
nodes = Node.find(
Q('tags', 'in', tags) & Q('is_public', 'eq', True)
& Q('is_deleted', 'ne', True))
projects.update(list(nodes))
for idx, node in enumerate(projects):
submissions.append(_render_conference_node(node, idx, conf))
num_submissions = len(projects)
# Cache the number of submissions
conf.num_submissions = num_submissions
conf.save()
if num_submissions < settings.CONFERENCE_MIN_COUNT:
continue
submissions.sort(key=lambda submission: submission['dateCreated'],
reverse=True)
return {'submissions': submissions}
def remove_tag(self, tag, auth, save=True, log=True):
from website.models import Tag, NodeLog # Prevent import error
tag = Tag.load(tag)
if tag and tag in self.tags and not self.node.is_registration:
self.tags.remove(tag)
if log:
self.add_tag_log(NodeLog.FILE_TAG_REMOVED, tag._id, auth)
if save:
self.save()
return True
return False
def remove_tag(self, tag, auth, save=True, log=True):
from website.models import Tag, NodeLog # Prevent import error
tag = Tag.load(tag)
if tag and tag in self.tags and not self.node.is_registration:
self.tags.remove(tag)
if log:
self.add_tag_log(NodeLog.FILE_TAG_REMOVED, tag._id, auth)
if save:
self.save()
return True
return False
def verify_tags(node, modm_node):
modm_tag_keys = [x for x in sorted(set(modm_node.tags._to_primary_keys())) if MODMTag.load(x)]
django_tag_keys = sorted(set(node.tags.filter(
system=False).values_list('_id',
flat=True)))
modm_system_tag_keys = sorted(set(modm_node.system_tags))
django_system_tag_keys = sorted(set(
node.system_tags.values_list('_id',
flat=True)))
assert modm_tag_keys == django_tag_keys, 'Modm tags {} don\'t match django tags {} in node {}:{}'.format(
modm_tag_keys, django_tag_keys, modm_node._id, node._guid.guid)
assert modm_system_tag_keys == django_system_tag_keys, 'Modm system tag keys {} don\'t match django system tags {}'.format(
modm_system_tag_keys, django_system_tag_keys)
def verify_tags(node, modm_node):
modm_tag_keys = [
x for x in sorted(set(modm_node.tags._to_primary_keys()))
if MODMTag.load(x)
]
django_tag_keys = sorted(
set(node.tags.filter(system=False).values_list('_id', flat=True)))
modm_system_tag_keys = sorted(set(modm_node.system_tags))
django_system_tag_keys = sorted(
set(node.system_tags.values_list('_id', flat=True)))
assert modm_tag_keys == django_tag_keys, 'Modm tags {} don\'t match django tags {} in node {}:{}'.format(
modm_tag_keys, django_tag_keys, modm_node._id, node._guid.guid)
assert modm_system_tag_keys == django_system_tag_keys, 'Modm system tag keys {} don\'t match django system tags {}'.format(
modm_system_tag_keys, django_system_tag_keys)
def remove_tag(self, tag, auth, save=True, log=True):
from website.models import Tag, NodeLog # Prevent import error
if self.node.is_registration:
# Can't perform edits on a registration
raise NodeStateError
tag = Tag.load(tag)
if not tag:
raise InvalidTagError
elif tag not in self.tags:
raise TagNotFoundError
else:
self.tags.remove(tag)
if log:
self.add_tag_log(NodeLog.FILE_TAG_REMOVED, tag._id, auth)
if save:
self.save()
return True
def remove_tag(self, tag, auth, save=True, log=True):
from website.models import Tag, NodeLog # Prevent import error
if self.node.is_registration:
# Can't perform edits on a registration
raise NodeStateError
tag = Tag.load(tag)
if not tag:
raise InvalidTagError
elif tag not in self.tags:
raise TagNotFoundError
else:
self.tags.remove(tag)
if log:
self.add_tag_log(NodeLog.FILE_TAG_REMOVED, tag._id, auth)
if save:
self.save()
return True
def parse(self):
raw_post = random.choice(Post.objects.filter(is_raw = True))
page = self.get_html(raw_post.original_url)
soup = BeautifulSoup(page,"html.parser")
content =str(soup.find("div", class_="entry-content"))
if content == "None" or not content:
raw_post.delete()
return
raw_post.html = content
results = soup.findAll("a", {"rel" : "category tag"})
tags = [ ( r['href'].split('/')[-2:-1][0], r.contents[0] ) for r in results]
# for r in results:
# tags.append( (r['href'].split('/')[-2:-1][0], r.contents[0]) )
images = soup.findAll("img")
raw_post.first_image = images[1]['src']
raw_post.first_text = str(soup.findAll("p")[0].contents[0])
raw_post.title = self.reg_find(r"<h1 class=\"entry-title\">(.+?)</h1>", page)[0]
same_post = Post.objects.filter(title = raw_post.title, is_raw = False).first()
if same_post:
same_post.delete()
post_date = self.reg_find(r">.*(\d\d\.\d\d\.\d\d\d\d)</time>", page)[0]
post_date = post_date.replace('.','')
#post_date = datetime.datetime.strptime(post_date, "%dd%mm%YYYY").date()
format_str = '%d%m%Y' # The format
raw_post.post_date = datetime.datetime.strptime(post_date, format_str)
author = self.reg_find(r"class=\"url fn n\">(.+?)</a>", page)[0]
a = Author.objects.filter(name = author)
if a:
a = a[0]
else:
a = Author(name = author)
a.save()
raw_post.author = a
existed_tags = Tag.objects.filter(name__in = [t[0] for t in tags])
for t in tags:
if not t[0] in [et.name for et in existed_tags]:
new_tag = Tag(name =t[0], value = t[1])
new_tag.save()
post_tags = Tag.objects.filter(name__in = [t[0] for t in tags])
# post = Post(
# title = title,
# html = content,
# author =a,
# original_url = url,
# post_date = post_date,
# first_image = first_image,
# first_text=first_p)
# post.save()
# post.tags.set(post_tags)
raw_post.is_raw = False
raw_post.save()
raw_post.tags.set(post_tags)
print("post %s saved" % str(raw_post))
def do_migration():
for t in Tag.find():
logger.info('Migrating tag {!r}'.format(t))
t.lower = t._id.lower()
t.save(force=True)
def do_migration():
for t in Tag.find():
logger.info('Migrating tag {!r}'.format(t))
t.lower = t._id.lower()
t.save(force=True)
