def getData(part):
res = {}
res.update(part)
if "submitted_by" not in res:
res["submitted_by"] = {"year": int(part["part_entered"][:4]), "team_name": "Unknown"}
part_pages = []
for pageid in pageids[part["part_name"]]:
try:
p = pages[str(pageid["pageid"])]
except:
continue
raw_page = partinfo_re.sub("", p["revisions"][0]["*"])
try:
content = remove_html_tags(wikimarkup.parse(raw_page))
except Exception as e: # parse error
content = remove_html_tags(raw_page) # give up parsing
part_pages.append(
{
"content": content,
"title": pageid["title"],
"revid": p["revisions"][0]["revid"],
"timestamp": p["revisions"][0]["timestamp"],
"user": p["revisions"][0]["user"],
"pageid": pageid["pageid"],
}
)
res["pages"] = part_pages
s = score[part["part_name"]]
res["reliability"] = s["reliability"]
res["num_teams_used"] = s["num_teams_used"]
return res
def getData(part):
res = {}
res.update(part)
if 'submitted_by' not in res:
res['submitted_by'] = {
'year': int(part['part_entered'][:4]),
'team_name': 'Unknown',
}
part_pages = []
for pageid in pageids[part['part_name']]:
try:
p = pages[str(pageid['pageid'])]
except:
continue
raw_page = partinfo_re.sub('', p['revisions'][0]['*'])
try:
content = remove_html_tags(wikimarkup.parse(raw_page))
except Exception as e: # parse error
content = remove_html_tags(raw_page) # give up parsing
part_pages.append({
'content': content,
'title': pageid['title'],
'revid': p['revisions'][0]['revid'],
'timestamp': p['revisions'][0]['timestamp'],
'user': p['revisions'][0]['user'],
'pageid': pageid['pageid'],
})
res['pages'] = part_pages
s = score[part['part_name']]
res['reliability'] = s['reliability']
res['num_teams_used'] = s['num_teams_used']
return res
def import_wiki_sessions(sections, article, reply_to, current_task, total_count):
for section in sections:
heading = section.get('heading', None)
if heading:
parsed_text = parse(heading)
comment_author = CommentAuthor.objects.get(disqus_id='anonymous', is_wikipedia=True)
comments = Comment.objects.filter(article=article, author=comment_author, text=parsed_text)
if comments.count() > 0:
comment_wikum = comments[0]
else:
comment_wikum = Comment.objects.create(article = article,
author = comment_author,
text = parsed_text,
reply_to_disqus = reply_to,
text_len = len(parsed_text),
)
comment_wikum.save()
comment_wikum.disqus_id = comment_wikum.id
comment_wikum.save()
total_count += 1
if current_task and total_count % 3 == 0:
current_task.update_state(state='PROGRESS',
meta={'count': total_count})
else:
comment_wikum = reply_to
if len(section['comments']) > 0:
total_count = import_wiki_talk_posts(section['comments'], article, comment_wikum.disqus_id, current_task, total_count)
if len(section['subsections']) > 0:
total_count = import_wiki_sessions(section['subsections'], article, comment_wikum.disqus_id, current_task, total_count)
return total_count
def get_table(title, site='en.wikipedia.org', table_idx=0):
"""
Given an article title, site, and the index of that table within that page
returns a dataframe constructed from that table.
Parameters:
title (str) : article title (not a URL)
site (str) : site base url (i.e. en.wikipedia.org)
table_idx (int) : index of table on page (zero-indexed, so first table is idx=0)
Returns:
a pandas.DataFrame with the column names set to the column headers and a default
index
"""
assert 'http://' not in title, 'dataframe_from_url expects a title, not a full URL'
url = 'http://{0}/w/api.php?action=query&titles={1}&prop=revisions&rvprop=content&format=json'.format(site, title)
logger.info('getting url: %s', url)
res = json.loads(urllib.urlopen(url).read())
wiki = res['query']['pages'].values()[0]['revisions'][0]['*']
html = wikimarkup.parse(wiki)
dom = soup.BeautifulSoup(html)
table = itertools.islice(dom.findAll('table'),table_idx, table_idx + 1).next()
row_tags = table.findAll('tr')
header = [e.text.strip() for e in row_tags[0].findAll('th')]
rows = [[e.text.strip() for e in row.findAll('td')] for row in row_tags[1:]]
df = pd.DataFrame(rows, columns = header)
return df
def result(task_id):
task = processarticle.AsyncResult(task_id)
if task.state == 'SUCCESS':
content = task.result['article']
html = parse(content, False)
return render_template("result.html", content = content, html = html, info = task.result['info'])
else:
return "Be more patient!"
def get_tables_1(wiki):
html = PyQuery(wikimarkup.parse(wiki))
frames = []
for table in html('table'):
data = [[x.text.strip() for x in row]
for row in table.getchildren()]
df = pd.DataFrame(data[1:], columns=data[0])
frames.append(df)
return frames
def get_tables(url):
html = PyQuery(wikimarkup.parse(get_wiki_raw(url)))
frames = pd.DataFrame()
for table in html('table'):
data = [[x.text.strip() for x in row]
for row in table.getchildren()]
df = pd.DataFrame(data[1:], columns=data[0])
if np.prod(df.shape) > np.prod(frames.shape):
frames = df
return frames
def get_most_active_users_wiki_df(url):
res = json.loads(urllib.urlopen(url).read())
wiki = res['query']['pages'].values()[0]['revisions'][0]['*']
html = wikimarkup.parse(wiki)
dom = soup.BeautifulSoup(html)
row_tags = dom.findAll('tr')
header = [e.text.strip() for e in row_tags[0].findAll('th')]
rows = [[e.text.strip() for e in row.findAll('td')] for row in row_tags[1:]]
df = pd.DataFrame(rows, columns = header)
df.User = df.User.apply(extract_username)
return df[~df.User.isnull()]
def show(self, c):
from wikimarkup import parse
f = os.path.join(self.runpath, 'data', '%s.wiki' % c)
if os.path.exists(f):
wiki = open(f).read()
wiki = reg.sub(self._pic, wiki)
wiki = reg1.sub(' (', wiki)
wiki = reg2.sub(')', wiki)
html = parse(wiki, showToc=False)
self.load_html_string(html, 'file:///')
else:
self.open(os.path.join(self.runpath, 'err.html'))
def show(self, c):
from wikimarkup import parse
f = os.path.join(self.runpath, 'data', '%s.wiki' % c)
if os.path.exists(f):
wiki = open(f).read()
wiki = reg.sub(self._pic, wiki)
wiki = reg1.sub(' (', wiki)
wiki = reg2.sub(')', wiki)
html = parse(wiki, showToc=False)
self.load_html_string(html, 'file:///')
else:
self.open(os.path.join(self.runpath, 'err.html'))
def import_wiki_talk_posts(comments, article, reply_to, current_task, total_count):
for comment in comments:
text = ''
for block in comment['text_blocks']:
t = parse(block)
if t.strip() != '':
text += '<P>%s</P>' % t
author = comment.get('author')
if author:
comment_author = import_wiki_authors([author], article)[0]
else:
comment_author = CommentAuthor.objects.get(disqus_id='anonymous', is_wikipedia=True)
comments = Comment.objects.filter(article=article, author=comment_author,text=text)
if comments.count() > 0:
comment_wikum = comments[0]
else:
time = None
if comment.get('time_stamp'):
time = datetime.datetime.strptime(comment['time_stamp'], '%H:%M, %d %B %Y (%Z)')
cosigners = comment['cosigners']
comment_cosigners = import_wiki_authors(cosigners, article)
comment_wikum = Comment.objects.create(article = article,
author = comment_author,
text = text,
reply_to_disqus = reply_to,
text_len = len(text),
)
if time:
comment_wikum.created_at = time
comment_wikum.save()
comment_wikum.disqus_id = comment_wikum.id
comment_wikum.save()
for signer in comment_cosigners:
comment_wikum.cosigners.add(signer)
total_count += 1
if current_task and total_count % 3 == 0:
current_task.update_state(state='PROGRESS',
meta={'count': total_count})
replies = comment['comments']
total_count = import_wiki_talk_posts(replies, article, comment_wikum.disqus_id, current_task, total_count)
return total_count
def wikified_content(self):
# TODO: check memcache for rendered page?
# replacements here
transforms = [
AutoLink(),
WikiWords(),
HideReferers(),
]
content = self.content
content = wikimarkup.parse(content)
for transform in transforms:
content = transform.run(content, self)
return content
def save_org_description(request):
if request.user.is_anonymous():
raise Http404
text = request.POST['text']
org_id = request.POST['id']
field = request.POST['field']
data_format = request.POST['data_format']
org = models.Organization.objects.get(id = org_id)
setattr(org, field, text)
org.save()
#save the descr
if data_format == 'mediawiki':
data = simplejson.dumps({'text': parse(text), 'id': org_id })
else:
data = simplejson.dumps({'text': text, 'id': org_id })
return http.HttpResponse(data, mimetype="application/json")
def getData(team):
res = {}
res.update(results[team])
##res.update(wikis[team])
res['year'] = YEAR
idx=0
team_pages = []
for pageid in pageids[team]:
# idx += 1
# if(idx > 79):
# break
try:
p = pages[str(pageid['pageid'])]
except:
continue
raw_page = p['revisions'][0]['*']
try:
content = remove_html_tags(wikimarkup.parse(raw_page))
except Exception as e: # parse error
content = remove_html_tags(raw_page) # give up parsing
print ('give up')
content = '<{<{<wikititle>}>}>' + pageid['title'] + '<{<{</wikititle>}>}>' + content # add title to content header
team_pages.append({
'content': content,
'title': pageid['title'],
'revid': p['revisions'][0]['revid'],
'timestamp': p['revisions'][0]['timestamp'],
'user': p['revisions'][0]['user'],
'pageid': pageid['pageid'],
})
# print(str(idx) + ':' + str(pageid))
# pprint.pprint(content)
res['pages'] = team_pages
# res['school'] = teaminfo[team]['school']
# res['description'] = teaminfo[team]['description']
# res['title'] = teaminfo[team]['title']
# res['track'] = teaminfo[team]['track']
# res['abstract'] = teaminfo[team]['abstract']
res.update(teaminfo[team])
return res
def getData(team):
res = {}
res.update(results[team])
res["year"] = YEAR
# idx=0
team_pages = []
for pageid in pageids[team]:
# idx += 1
# if(idx > 79):
# break
try:
p = pages[str(pageid["pageid"])]
except:
continue
raw_page = p["revisions"][0]["*"]
try:
content = remove_html_tags(wikimarkup.parse(raw_page))
except Exception as e: # parse error
content = remove_html_tags(raw_page) # give up parsing
print ("give up")
content = (
"<{<{<wikititle>}>}>" + pageid["title"] + "<{<{</wikititle>}>}>" + content
) # add title to content header
team_pages.append(
{
"content": content,
"title": pageid["title"],
"revid": p["revisions"][0]["revid"],
"timestamp": p["revisions"][0]["timestamp"],
"user": p["revisions"][0]["user"],
"pageid": pageid["pageid"],
}
)
# print(str(idx) + ':' + str(pageid))
# pprint.pprint(content)
res["pages"] = team_pages
# res['school'] = teaminfo[team]['school']
# res['description'] = teaminfo[team]['description']
# res['title'] = teaminfo[team]['title']
# res['track'] = teaminfo[team]['track']
# res['abstract'] = teaminfo[team]['abstract']
res.update(teaminfo[team])
return res
def save_illustrations_page(page):
def insert_images(markup):
def replace_image(match):
image = match.group(1)
image = mwpage.Page.normalize_title(image)
image_jobs.append(gevent.spawn(download_image, image))
return '<img src={0} alt="{0}/>'.format(image)
images = re.compile(r'Image\:(.+?)(?:\n|\|.+)', re.IGNORECASE)
markup = re.sub(images, replace_image, markup)
return markup
markup = wikimarkup.parse(page.edit().encode('utf-8'), showToc=False)
markup = clean_page(markup)
markup = re.sub(r'<gallery>', '', markup)
markup = re.sub(r'</gallery>', '', markup)
markup = insert_images(markup)
return page, markup
def parse_wiki(self):
for index in self._query_hash:
query = self._query_hash[index]
content = self.retrieve(query)
if content:
doc = json.load(content)
## get the content of the markup
## thanks to http://goo.gl/wDPha
text = doc['query']['pages'].itervalues().next(
)['revisions'][0]['*']
global g_current_query
g_current_query = query
html = parse(text)
print 'Query processed: %s' % query
## wait for 1 second to avoid unnecessary banning from WikiPedia server
time.sleep(1)
else:
print 'Skipping query %s' % query
def parse_json(json_file):
try:
with open(json_file) as f:
print 'Loading %s' % (json_file)
dump_json = json.load(f)
except IOError as e:
print 'Failed to open file: %s' % json_file
global g_cur_idx
global g_dist_hash
global g_timestamp
global g_rel_ent_dist_db
sorted_index = dump_json.keys()
sorted_index.sort(key=lambda x: int(x))
for index in sorted_index:
g_cur_idx = index
g_dist_hash[g_cur_idx] = {}
g_rev_hash[g_cur_idx] = {}
query = dump_json[index]['query']
g_rel_ent_dist_db.hset(RedisDB.query_ent_hash, index, query)
#print 'Query: %s %s' % (index, query)
#continue
for rev_id in dump_json[index]['revisions']:
revid = dump_json[index]['revisions'][rev_id]['revid']
g_timestamp = dump_json[index]['revisions'][rev_id]['timestamp']
text = dump_json[index]['revisions'][rev_id]['text']
#print '%s %s %s' % (query, revid, timestamp)
g_rev_hash[g_cur_idx][g_timestamp] = revid
try:
html = parse(text)
# catch all other exceptions in the parse process,
# print out the traceback, and move on without interruption
except:
print "Exception in parse_json()"
print '-' * 60
traceback.print_exc(file=sys.stdout)
print '-' * 60
pass
g_rel_ent_dist_db.rpush(RedisDB.query_ent_list, index)
sorted_ts = g_dist_hash[index].keys()
# sort the timestamp in chronic order
sorted_ts.sort(key=lambda x: datetime.datetime.strptime(x,
'%Y-%m-%dT%H:%M:%SZ'))
# for each month, save the number of related entities
# as well as the related entities
last_num = 0
last_revid = 0
last_ts = ''
last_dt_str = datetime.datetime.strptime(sorted_ts[0],
'%Y-%m-%dT%H:%M:%SZ').strftime('%Y-%m')
for ts in sorted_ts:
revid = g_rev_hash[index][ts]
dt_str = datetime.datetime.strptime(ts, '%Y-%m-%dT%H:%M:%SZ').strftime('%Y-%m')
if dt_str != last_dt_str:
hash_key = 'query-%s' % index
# save the number of related entities for last month
g_rel_ent_dist_db.hset(hash_key, last_dt_str, last_num)
print '%s %s %s %s %d' % (index, query, last_dt_str, last_revid, last_num)
# save the related entities for last month
ent_list = []
for ent in g_dist_hash[index][last_ts]:
ent_list.append(ent)
ent_str = '='.join(ent_list)
hash_key = 'query-rel-ent-%s' % index
g_rel_ent_dist_db.hset(hash_key, last_dt_str, ent_str)
last_num = len(g_dist_hash[index][ts].keys())
last_revid = revid
last_dt_str = dt_str
last_ts = ts
def markdown(s):
wikimarkup.registerInternalLinkHook(None, link_hook)
return wikimarkup.parse(s, showToc=True)
def render(content, *args, **kwargs):
return wikimarkup.parse(content, *args, **kwargs)
def recurse_viz(parent, posts, replaced, article, is_collapsed):
children = []
hid_children = []
replace_children = []
pids = [post.disqus_id for post in posts]
if replaced:
num_subtree_children = 0
else:
num_subtree_children = len(pids)
reps = Comment.objects.filter(reply_to_disqus__in=pids, article=article).select_related()
for post in posts:
if post.json_flatten == '':
#if True:
if post.author:
if post.author.anonymous:
author = "Anonymous"
else:
author = post.author.username
else:
author = ""
v1 = {'size': post.points,
'd_id': post.id,
'parent': parent.id if parent else None,
'author': author,
'replace_node': post.is_replacement,
'collapsed': is_collapsed,
'tags': [(tag.text, tag.color) for tag in post.tags.all()]
}
if 'https://en.wikipedia.org/wiki/' in article.url:
v1['name'] = parse(post.text)
v1['wikitext'] = post.text
if post.summary.strip() == '':
v1['summary'] = ''
else:
v1['summary'] = parse(post.summary)
v1['sumwiki'] = post.summary
if post.extra_summary.strip() == '':
v1['extra_summary'] = ''
else:
v1['summary'] = parse(post.extra_summary)
v1['extrasumwiki'] = post.extra_summary
else:
v1['name'] = post.text
v1['summary'] = post.summary
v1['extra_summary'] = post.extra_summary
c1 = reps.filter(reply_to_disqus=post.disqus_id).order_by('-points')
if c1.count() == 0:
vals = []
hid = []
rep = []
num_subchildren = 0
else:
replace_future = replaced or post.is_replacement
vals, hid, rep, num_subchildren = recurse_viz(post, c1, replace_future, article, is_collapsed or post.is_replacement)
v1['children'] = vals
v1['hid'] = hid
v1['replace'] = rep
post.json_flatten = json.dumps(v1)
if not post.is_replacement:
post.num_subchildren = num_subchildren
else:
post.num_subchildren = 0
post.save()
if not post.is_replacement:
num_subtree_children += num_subchildren
else:
v1 = json.loads(post.json_flatten)
if post.hidden:
hid_children.append(v1)
elif parent and parent.is_replacement:
replace_children.append(v1)
else:
children.append(v1)
return children, hid_children, replace_children, num_subtree_children
def parse(self, input_type, output_format, text):
return parse(text)
def parse_json(json_file):
try:
with open(json_file) as f:
print 'Loading %s' % (json_file)
dump_json = json.load(f)
except IOError as e:
print 'Failed to open file: %s' % json_file
global g_cur_idx
global g_dist_hash
global g_timestamp
global g_rel_ent_dist_db
sorted_index = dump_json.keys()
sorted_index.sort(key=lambda x: int(x))
for index in sorted_index:
g_cur_idx = index
g_dist_hash[g_cur_idx] = {}
g_rev_hash[g_cur_idx] = {}
query = dump_json[index]['query']
g_rel_ent_dist_db.rpush(RedisDB.query_ent_list, index)
g_rel_ent_dist_db.hset(RedisDB.query_ent_hash, index, query)
print 'Query: %s %s' % (index, query)
#continue
for rev_id in dump_json[index]['revisions']:
revid = dump_json[index]['revisions'][rev_id]['revid']
g_timestamp = dump_json[index]['revisions'][rev_id]['timestamp']
text = dump_json[index]['revisions'][rev_id]['text']
#print '%s %s %s' % (query, revid, timestamp)
g_rev_hash[g_cur_idx][g_timestamp] = revid
try:
html = parse(text)
# catch all other exceptions in the parse process,
# print out the traceback, and move on without interruption
except:
print "Exception in parse_json()"
print '-' * 60
traceback.print_exc(file=sys.stdout)
print '-' * 60
pass
sorted_ts = g_dist_hash[index].keys()
# sort the timestamp in chronic order
sorted_ts.sort(key=lambda x: datetime.datetime.strptime(x,
'%Y-%m-%dT%H:%M:%SZ'))
# for each month, save the number of related entities
# as well as the related entities
last_revid = 0
last_ts = ''
last_dt_str = datetime.datetime.strptime(sorted_ts[0],
'%Y-%m-%dT%H:%M:%SZ').strftime('%Y-%m')
rel_ent_hash = {}
irrel_ent_hash = {}
for ts in sorted_ts:
revid = g_rev_hash[index][ts]
dt_str = datetime.datetime.strptime(ts, '%Y-%m-%dT%H:%M:%SZ').strftime('%Y-%m')
if dt_str != last_dt_str:
# save the related entities for last month
ent_list = []
for ent in g_dist_hash[index][last_ts]:
# check the cache first
if ent in rel_ent_hash:
ent_list.append(ent)
continue
if ent in irrel_ent_hash:
continue
# update the cache accordingly
if in_rel_doc(query, ent):
ent_list.append(ent)
rel_ent_hash[ent] = 1
else:
irrel_ent_hash[ent] = 1
# save the number of related entities for last month
hash_key = 'query-rel-ent-num-%s' % index
ent_num = len(ent_list)
g_rel_ent_dist_db.hset(hash_key, last_dt_str, ent_num)
print '%s %s %s %s %d' % (index, query, last_dt_str, last_revid, ent_num)
last_revid = revid
last_dt_str = dt_str
last_ts = ts
def summarize_comment(request):
try:
article_id = request.POST['article']
a = Article.objects.get(id=article_id)
id = request.POST['id']
summary = request.POST['comment']
top_summary, bottom_summary = get_summary(summary)
req_user = request.user if request.user.is_authenticated() else None
c = Comment.objects.get(id=id)
from_summary = c.summary + '\n----------\n' + c.extra_summary
c.summary = top_summary
c.extra_summary = bottom_summary
c.save()
if from_summary != '':
action = 'edit_sum'
explanation = 'edit summary'
else :
action = 'sum_comment'
explanation = 'initial summary'
h = History.objects.create(user=req_user,
article=a,
action=action,
from_str=from_summary,
to_str=summary,
explanation=explanation)
h.comments.add(c)
recurse_up_post(c)
if 'wikipedia.org' in a.url:
res = {}
if top_summary.strip() != '':
res['top_summary'] = parse(top_summary)
else:
res['top_summary'] = ''
res['top_summary_wiki'] = top_summary
if bottom_summary.strip() != '':
res['bottom_summary'] = parse(bottom_summary)
else:
res['bottom_summary'] = ''
res['bottom_summary_wiki'] = bottom_summary
return JsonResponse(res)
else:
return JsonResponse({'top_summary': top_summary,
'bottom_summary': bottom_summary})
except Exception, e:
print e
return HttpResponseBadRequest()
def summarize_selected(request):
try:
article_id = request.POST['article']
a = Article.objects.get(id=article_id)
ids = request.POST.getlist('ids[]')
children_ids = request.POST.getlist('children[]')
children_ids = [int(x) for x in children_ids]
child_id = request.POST['child']
delete_nodes = request.POST.getlist('delete_nodes[]')
summary = request.POST['comment']
top_summary, bottom_summary = get_summary(summary)
req_user = request.user if request.user.is_authenticated() else None
comments = Comment.objects.filter(id__in=ids)
children = [c for c in comments if c.id in children_ids]
child = Comment.objects.get(id=child_id)
new_id = random_with_N_digits(10);
new_comment = Comment.objects.create(article=a,
is_replacement=True,
reply_to_disqus=child.reply_to_disqus,
summary=top_summary,
extra_summary=bottom_summary,
disqus_id=new_id,
points=child.points,
text_len=len(summary))
h = History.objects.create(user=req_user,
article=a,
action='sum_selected',
to_str=summary,
explanation='initial summary of group of comments')
for c in children:
c.reply_to_disqus = new_id
c.save()
for c in comments:
h.comments.add(c)
for node in delete_nodes:
delete_node(node)
recurse_up_post(new_comment)
recurse_down_num_subtree(new_comment)
make_vector(new_comment, a)
if 'wikipedia.org' in a.url:
res = {'d_id': new_comment.id}
if top_summary.strip() != '':
res['top_summary'] = parse(top_summary)
else:
res['top_summary'] = ''
res['top_summary_wiki'] = top_summary
if bottom_summary.strip() != '':
res['bottom_summary'] = parse(bottom_summary)
else:
res['bottom_summary'] = ''
res['bottom_summary_wiki'] = bottom_summary
return JsonResponse(res)
else:
return JsonResponse({'d_id': new_comment.id,
'top_summary': top_summary,
'bottom_summary': bottom_summary})
except Exception, e:
print e
return HttpResponseBadRequest()
def summarize_comments(request):
try:
article_id = request.POST['article']
a = Article.objects.get(id=article_id)
id = request.POST['id']
summary = request.POST['comment']
top_summary, bottom_summary = get_summary(summary)
delete_nodes = request.POST.getlist('delete_nodes[]')
req_user = request.user if request.user.is_authenticated() else None
c = Comment.objects.get(id=id)
if not c.is_replacement:
new_id = random_with_N_digits(10);
new_comment = Comment.objects.create(article=a,
is_replacement=True,
reply_to_disqus=c.reply_to_disqus,
summary=top_summary,
extra_summary=bottom_summary,
disqus_id=new_id,
points=c.points,
text_len=len(summary))
c.reply_to_disqus = new_id
c.save()
h = History.objects.create(user=req_user,
article=a,
action='sum_nodes',
to_str=summary,
explanation='initial summary of subtree')
d_id = new_comment.id
recurse_down_num_subtree(new_comment)
else:
from_summary = c.summary + '\n----------\n' + c.extra_summary
c.summary = top_summary
c.extra_summary=bottom_summary
c.save()
h = History.objects.create(user=req_user,
article=a,
action='edit_sum_nodes',
from_str=from_summary,
to_str=summary,
explanation='edit summary of subtree')
d_id = c.id
new_comment = c
for node in delete_nodes:
delete_node(node)
h.comments.add(c)
recurse_up_post(c)
make_vector(new_comment, a)
if 'wikipedia.org' in a.url:
res = {'d_id': d_id}
if top_summary.strip() != '':
res['top_summary'] = parse(top_summary)
else:
res['top_summary'] = ''
res['top_summary_wiki'] = top_summary
if bottom_summary.strip() != '':
res['bottom_summary'] = parse(bottom_summary)
else:
res['bottom_summary'] = ''
res['bottom_summary_wiki'] = bottom_summary
return JsonResponse(res)
else:
return JsonResponse({'d_id': d_id,
'top_summary': top_summary,
'bottom_summary': bottom_summary})
except Exception, e:
print e
import traceback
print traceback.format_exc()
return HttpResponseBadRequest()
def markdown(s):
wikimarkup.registerInternalLinkHook(None, link_hook)
return wikimarkup.parse(s, showToc=True)
def galleryTagHook(parser_env, body, attributes={}):
widths = attributes.get('widths')
if widths:
widths = re.sub('px', '', widths)
gal_width = int(widths)
else:
gal_width = 155
heights = attributes.get('heights')
if heights:
heights = re.sub('px', '', heights)
def_image = int(heights)
else:
def_image = 120
start_text = ''
if attributes.get('mode', None) == 'packed':
start_text = '<ul class="gallery mw-gallery-packed">'
files = body.split('\n')
for file in files:
if file.strip() != '':
res = file.split('|')
filename = res[0].strip()
site = wiki.Wiki('https://en.wikipedia.org/w/api.php')
params = {
'action': 'query',
'titles': filename,
'prop': 'imageinfo',
'iiprop': 'url|thumbmime',
'iiurlheight': 131
}
request = api.APIRequest(site, params)
result = request.query()
try:
url = result['query']['pages'].values(
)[0]['imageinfo'][0]['thumburl']
desc_url = result['query']['pages'].values(
)[0]['imageinfo'][0]['descriptionurl']
width = result['query']['pages'].values(
)[0]['imageinfo'][0]['thumbwidth']
height = result['query']['pages'].values(
)[0]['imageinfo'][0]['thumbheight']
except:
continue
text = '<li class="gallerybox" style="width: %spx"><div style="width: %spx">' % (
float(int(width)) + 1.496, float(int(width)) + 1.496)
text += '<div class="thumb" style="width: %spx;"><div style="margin:0px auto;">' % (
float(int(width)) + 0.496)
text += '<a href="%s" class="image"><img src="%s" width="%s" height="%s"></a>' % (
desc_url, url, width, height)
text += '</div></div></div><div class="gallerytext"><p>'
if res[1] == 'thumb':
inner_text = '|'.join(res[2:]).strip()
else:
inner_text = '|'.join(res[1:]).strip()
text += parse(inner_text)
text += '</p></div></li>'
start_text += text
elif attributes.get('mode', None) == 'nolines':
start_text = '<ul class="gallery mw-gallery-nolines">'
if not attributes.get('widths'):
gal_width = 125
files = body.split('\n')
for file in files:
if file.strip() != '':
res = file.split('|')
filename = res[0].strip()
site = wiki.Wiki('https://en.wikipedia.org/w/api.php')
params = {
'action': 'query',
'titles': filename,
'prop': 'imageinfo',
'iiprop': 'url|thumbmime',
'iiurlwidth': gal_width - 5
}
request = api.APIRequest(site, params)
result = request.query()
try:
url = result['query']['pages'].values(
)[0]['imageinfo'][0]['thumburl']
desc_url = result['query']['pages'].values(
)[0]['imageinfo'][0]['descriptionurl']
width = result['query']['pages'].values(
)[0]['imageinfo'][0]['thumbwidth']
height = result['query']['pages'].values(
)[0]['imageinfo'][0]['thumbheight']
except:
continue
ratio = float(float(width) / float(height))
if height > def_image:
height = def_image
width = ratio * def_image
text = '<li class="gallerybox" style="width: %spx"><div style="width: %spx">' % (
gal_width, gal_width)
text += '<div class="thumb" style="width: %spx;"><div style="margin:0px auto;">' % (
gal_width - 5)
text += '<a href="%s" class="image"><img src="%s" width="%s" height="%s"></a>' % (
desc_url, url, width, height)
text += '</div></div></div><div class="gallerytext"><p>'
if res[1] == 'thumb':
inner_text = '|'.join(res[2:]).strip()
else:
inner_text = '|'.join(res[1:]).strip()
text += parse(inner_text)
text += '</p></div></li>'
start_text += text
else:
start_text = '<ul class="gallery mw-gallery-traditional">'
files = body.split('\n')
for file in files:
if file.strip() != '':
res = file.split('|')
filename = res[0].strip()
site = wiki.Wiki('https://en.wikipedia.org/w/api.php')
params = {
'action': 'query',
'titles': filename,
'prop': 'imageinfo',
'iiprop': 'url|thumbmime',
'iiurlwidth': gal_width - 35
}
request = api.APIRequest(site, params)
result = request.query()
try:
url = result['query']['pages'].values(
)[0]['imageinfo'][0]['thumburl']
desc_url = result['query']['pages'].values(
)[0]['imageinfo'][0]['descriptionurl']
width = result['query']['pages'].values(
)[0]['imageinfo'][0]['thumbwidth']
height = result['query']['pages'].values(
)[0]['imageinfo'][0]['thumbheight']
except:
continue
ratio = float(float(width) / float(height))
if height > def_image:
height = def_image
width = ratio * def_image
text = '<li class="gallerybox" style="width: %spx"><div style="width: %spx">' % (
gal_width, gal_width)
text += '<div class="thumb" style="width: %spx;"><div style="margin:%spx auto;">' % (
gal_width - 5, float(gal_width - height) / 2.0)
text += '<a href="%s" class="image"><img src="%s" width="%s" height="%s"></a>' % (
desc_url, url, width, height)
text += '</div></div></div><div class="gallerytext"><p>'
if res[1] == 'thumb':
inner_text = '|'.join(res[2:]).strip()
else:
inner_text = '|'.join(res[1:]).strip()
text += parse(inner_text)
text += '</p></div></li>'
start_text += text
start_text += '</ul>'
return start_text
longitude_number = float(lng)
new_location = pw.LocationObject(description=pw.NO_DATA_UNICODE,
latitude=latitude_number,
longitude=longitude_number)
new_plant = pw.PowerPlant(idnr,
name,
plant_country=COUNTRY_NAME,
plant_location=new_location,
plant_fuel=set([u'Coal']),
plant_source=SOURCE_NAME,
plant_source_url=SOURCE_URL,
plant_cap_year=2017)
plants_dictionary[idnr] = new_plant
# use wikimarkup to detect the main table and transform to html code
html = str(PyQuery(wikimarkup.parse(wiki)))
# read html code into pandas dataframe
frames = pd.read_html(html, header=0)
df = frames[0]
# make changes to Units column to include only the plant name, not the unit
for i, unit in enumerate(frames[0]['Unit']):
df.set_value(i, 'Unit', unit.strip('[]').split('|')[0])
for plant in plants_dictionary.values():
plant.capacity = df.groupby(['Unit']).sum().loc[plant.name].values[0]
plant.nat_lang = df.loc[df['Unit'] == plant.name, 'Chinese Name'].values[0]
# print list(set(df.loc[df['Unit'] == plant.name, 'Sponsor'].values))
owner = list(set(df.loc[df['Unit'] == plant.name, 'Sponsor'].values))
if len(owner) == 1:
def save_page(page):
markup = wikimarkup.parse(page.edit().encode('utf-8'), showToc=False)
markup = clean_page(markup)
markup = put_images(markup)
return page, markup
p1_score += score
scores[p1_name] = p1_score
total_score[p1_name][0] += p1_score
total_score[p1_name].append(p1_score)
# total_score[p1_name].append("%s <span style='color: gray;'>(%s)</span>" % (p1_score,))
place = 1
for name, score in sorted(scores.items(), key=lambda x: -x[1]):
color = colors[by_name[name]]
wiki_table += u'# <span style="color: %s">%s</span> → <b>%s</b>\n' \
% (color, name, score)
total_score_places[name].append(place)
place += 1
print wiki_table
html = parse(wiki_table)
save_file(stats_folder + '/' + u'word_%s.html' % word, html)
save_file(u'stats/latest/word_%s.html' % word, html)
content = ''
for name, score in sorted(scores.items(), key=lambda x: x[0]):
content += u'%s\t%s\n' % (name, score)
save_file(stats_folder + u'/score_%s.html' % word, content)
save_file(u'stats/latest/score_%s.html' % word, content)
total = u'{| style="text-align: center" \n'
total += u'! rowspan=2 | Место || rowspan=2 | Участник || rowspan=2 | Всего ' \
u'|| colspan=%s | Матчи \n' % len(words)
total += u'|-\n'
total += u'! %s\n' % u' || '.join(["[/balda/%s/ %s]" % (word, word) for word in words])
total += u'|-\n'
# coding: utf8
import json
from wikimarkup import parse
fh = open("ciccio.txt", "r");
content = fh.read()
fh.close()
obj = json.read(content)
obj = obj['query']['pages']['25458']['revisions'][0]['*']
obj = obj[0:12000]
#print(obj)
print parse(obj)
