def get(self, art_id, kind):
self.kind = kind
self.art = self.session.query(Article).get(art_id)
if self.art is None:
raise tornado.web.HTTPError(404, "Article not found")
if self.kind == SourceKind.PAPER:
self.form = AddPaperForm(TornadoMultiDict(self))
self.render('add_paper.html', art=self.art, form=self.form)
elif self.kind == SourceKind.PR:
self.form = AddPRForm(TornadoMultiDict(self))
self.render('add_pr.html', art=self.art, form=self.form)
elif self.kind == SourceKind.OTHER:
self.form = AddOtherForm(TornadoMultiDict(self))
self.render('add_other.html', art=self.art, form=self.form)
def get(self, art_id, kind):
self.kind = kind
self.art = self.session.query(Article).get(art_id)
if self.art is None:
raise tornado.web.HTTPError(404, "Article not found")
if self.kind == SourceKind.PAPER:
self.form = AddPaperForm(TornadoMultiDict(self))
self.render('add_paper.html', art=self.art, form=self.form)
elif self.kind == SourceKind.PR:
self.form = AddPRForm(TornadoMultiDict(self))
self.render('add_pr.html', art=self.art, form=self.form)
elif self.kind == SourceKind.OTHER:
self.form = AddOtherForm(TornadoMultiDict(self))
self.render('add_other.html', art=self.art, form=self.form)
def post(self, art_id, kind):
self.kind = kind
self.art = self.session.query(Article).get(art_id)
if self.art is None:
raise tornado.web.HTTPError(404, "Article not found")
if self.kind == SourceKind.PAPER:
self.form = AddPaperForm(TornadoMultiDict(self))
elif self.kind == SourceKind.PR:
self.form = AddPRForm(TornadoMultiDict(self))
elif self.kind == SourceKind.OTHER:
self.form = AddOtherForm(TornadoMultiDict(self))
if self.form.validate():
url = self.form.url.data
if self.kind == SourceKind.PAPER:
# if adding a paper, try getting metadata
doi = self.form.get_as_doi()
if doi is not None:
# we were given a doi - yay!
dx_url = 'http://dx.doi.org/' + doi
self.find_doi(dx_url)
else:
self.find_doi(url)
else:
# otherwise, we're all done - add source and finish up
action = self.create_source(self.kind, url=url)
self.wrap_things_up(action)
else:
if self.is_xhr(): # ajax?
# collect up the form errors
errs = {}
for field in self.form:
if field.errors:
errs[field.name] = field.errors
self.write({'success': False, 'errors': errs})
self.finish()
else:
self.render('add_paper.html', art=self.art, form=self.form)
def post(self, art_id, kind):
self.kind = kind
self.art = self.session.query(Article).get(art_id)
if self.art is None:
raise tornado.web.HTTPError(404, "Article not found")
if self.kind == SourceKind.PAPER:
self.form = AddPaperForm(TornadoMultiDict(self))
elif self.kind == SourceKind.PR:
self.form = AddPRForm(TornadoMultiDict(self))
elif self.kind == SourceKind.OTHER:
self.form = AddOtherForm(TornadoMultiDict(self))
if self.form.validate():
url = self.form.url.data
if self.kind == SourceKind.PAPER:
# if adding a paper, try getting metadata
doi = self.form.get_as_doi()
if doi is not None:
# we were given a doi - yay!
dx_url = 'http://dx.doi.org/' + doi
self.find_doi(dx_url)
else:
self.find_doi(url)
else:
# otherwise, we're all done - add source and finish up
action = self.create_source(self.kind, url=url)
self.wrap_things_up(action)
else:
if self.is_xhr(): # ajax?
# collect up the form errors
errs = {}
for field in self.form:
if field.errors:
errs[field.name] = field.errors
self.write({'success':False, 'errors':errs})
self.finish()
else:
self.render('add_paper.html', art=self.art, form=self.form)
class AddSourceHandler(BaseHandler):
def get(self, art_id, kind):
self.kind = kind
self.art = self.session.query(Article).get(art_id)
if self.art is None:
raise tornado.web.HTTPError(404, "Article not found")
if self.kind == SourceKind.PAPER:
self.form = AddPaperForm(TornadoMultiDict(self))
self.render('add_paper.html', art=self.art, form=self.form)
elif self.kind == SourceKind.PR:
self.form = AddPRForm(TornadoMultiDict(self))
self.render('add_pr.html', art=self.art, form=self.form)
elif self.kind == SourceKind.OTHER:
self.form = AddOtherForm(TornadoMultiDict(self))
self.render('add_other.html', art=self.art, form=self.form)
@tornado.web.authenticated
@tornado.web.asynchronous
def post(self, art_id, kind):
self.kind = kind
self.art = self.session.query(Article).get(art_id)
if self.art is None:
raise tornado.web.HTTPError(404, "Article not found")
if self.kind == SourceKind.PAPER:
self.form = AddPaperForm(TornadoMultiDict(self))
elif self.kind == SourceKind.PR:
self.form = AddPRForm(TornadoMultiDict(self))
elif self.kind == SourceKind.OTHER:
self.form = AddOtherForm(TornadoMultiDict(self))
if self.form.validate():
url = self.form.url.data
if self.kind == SourceKind.PAPER:
# if adding a paper, try getting metadata
doi = self.form.get_as_doi()
if doi is not None:
# we were given a doi - yay!
dx_url = 'http://dx.doi.org/' + doi
self.find_doi(dx_url)
else:
self.find_doi(url)
else:
# otherwise, we're all done - add source and finish up
action = self.create_source(self.kind, url=url)
self.wrap_things_up(action)
else:
if self.is_xhr(): # ajax?
# collect up the form errors
errs = {}
for field in self.form:
if field.errors:
errs[field.name] = field.errors
self.write({'success': False, 'errors': errs})
self.finish()
else:
self.render('add_paper.html', art=self.art, form=self.form)
def find_doi(self, url):
""" retreive doi and metadata from url """
self.url = url # save for later
params = {'url': url}
scrapeomat_url = "http://localhost:8889/doi?" + urllib.urlencode(
params)
http = tornado.httpclient.AsyncHTTPClient()
http.fetch(scrapeomat_url, callback=self.on_got_doi_data)
def on_got_doi_data(self, response):
details = {'url': self.url}
if not response.error:
results = json.loads(response.body)
if results['status'] == 0:
#success!
meta = results['metadata']
details['doi'] = meta['doi']
details['title'] = meta['title']
details['publication'] = meta['journal']
try:
details['pubdate'] = datetime.datetime.strptime(
meta['date'], '%Y-%m-%dZ').date()
except ValueError:
# TODO: sometimes there's a year, which we should grab
details['pubdate'] = None
action = self.create_source(self.kind, **details)
self.wrap_things_up(action)
def wrap_things_up(self, action):
if self.is_xhr():
# ajax - just send back the source snippet
m = uimodules.source(self)
html = m.render(action.source, 'li')
self.write({
'success': True,
'new_source': {
'kind': action.source.kind,
'html': html
}
})
self.finish()
else:
self.redirect("/thanks/%d" % (action.id, ))
def create_source(self, kind, **details):
src = Source(article=self.art,
creator=self.current_user,
kind=kind,
**details)
self.session.add(src)
if self.art.needs_sourcing:
self.art.needs_sourcing = False
action = Action('src_add',
self.current_user,
article=self.art,
source=src)
self.art.needs_sourcing = False
self.session.add(action)
self.session.commit()
return action
class AddSourceHandler(BaseHandler):
def get(self, art_id, kind):
self.kind = kind
self.art = self.session.query(Article).get(art_id)
if self.art is None:
raise tornado.web.HTTPError(404, "Article not found")
if self.kind == SourceKind.PAPER:
self.form = AddPaperForm(TornadoMultiDict(self))
self.render('add_paper.html', art=self.art, form=self.form)
elif self.kind == SourceKind.PR:
self.form = AddPRForm(TornadoMultiDict(self))
self.render('add_pr.html', art=self.art, form=self.form)
elif self.kind == SourceKind.OTHER:
self.form = AddOtherForm(TornadoMultiDict(self))
self.render('add_other.html', art=self.art, form=self.form)
@tornado.web.authenticated
@tornado.web.asynchronous
def post(self, art_id, kind):
self.kind = kind
self.art = self.session.query(Article).get(art_id)
if self.art is None:
raise tornado.web.HTTPError(404, "Article not found")
if self.kind == SourceKind.PAPER:
self.form = AddPaperForm(TornadoMultiDict(self))
elif self.kind == SourceKind.PR:
self.form = AddPRForm(TornadoMultiDict(self))
elif self.kind == SourceKind.OTHER:
self.form = AddOtherForm(TornadoMultiDict(self))
if self.form.validate():
url = self.form.url.data
if self.kind == SourceKind.PAPER:
# if adding a paper, try getting metadata
doi = self.form.get_as_doi()
if doi is not None:
# we were given a doi - yay!
dx_url = 'http://dx.doi.org/' + doi
self.find_doi(dx_url)
else:
self.find_doi(url)
else:
# otherwise, we're all done - add source and finish up
action = self.create_source(self.kind, url=url)
self.wrap_things_up(action)
else:
if self.is_xhr(): # ajax?
# collect up the form errors
errs = {}
for field in self.form:
if field.errors:
errs[field.name] = field.errors
self.write({'success':False, 'errors':errs})
self.finish()
else:
self.render('add_paper.html', art=self.art, form=self.form)
def find_doi(self,url):
""" retreive doi and metadata from url """
self.url = url # save for later
params = {'url': url}
scrapeomat_url = "http://localhost:8889/doi?" + urllib.urlencode(params)
http = tornado.httpclient.AsyncHTTPClient()
http.fetch(scrapeomat_url, callback=self.on_got_doi_data)
def on_got_doi_data(self,response):
details = {'url':self.url}
if not response.error:
results = json.loads(response.body)
if results['status']==0:
#success!
meta = results['metadata']
details['doi'] = meta['doi']
details['title'] = meta['title']
details['publication'] = meta['journal']
try:
details['pubdate'] = datetime.datetime.strptime(meta['date'], '%Y-%m-%dZ').date()
except ValueError:
# TODO: sometimes there's a year, which we should grab
details['pubdate'] = None
action = self.create_source(self.kind, **details)
self.wrap_things_up(action)
def wrap_things_up(self,action):
if self.is_xhr():
# ajax - just send back the source snippet
m = uimodules.source(self)
html = m.render(action.source,'li')
self.write({'success':True, 'new_source': {'kind':action.source.kind, 'html':html} })
self.finish()
else:
self.redirect("/thanks/%d" % (action.id,))
def create_source(self, kind,**details):
src = Source(article=self.art,
creator=self.current_user,
kind=kind,
**details)
self.session.add(src)
if self.art.needs_sourcing:
self.art.needs_sourcing = False
action = Action('src_add', self.current_user,
article=self.art,
source=src)
self.art.needs_sourcing = False
self.session.add(action)
self.session.commit()
return action
def go(self, html, scrape_err=None):
art = self.art
if html is not None:
highlight_spans = self.analyse_text(html)
# mark up the html
html = highlight.html_highlight(html, highlight_spans)
# now find the unique matches for each kind
uniq = collections.defaultdict(set)
for (start, end, kind, name, url) in highlight_spans:
uniq[kind].add((name, url))
rs = uniq['researcher']
institutions = uniq['institution']
journals = uniq['journal']
researchers = []
for (name, url) in rs:
parts = name.split()
initial = parts[0][0]
surname = parts[-1]
researchers.append({
'name':
name,
'search_value':
'"%s %s"' % (initial, surname)
})
else:
# don't have any text available :-(
html = ''
researchers, institutions, journals = [], [], []
n_actions = 6 # show most recent N actions
recent_actions = self.session.query(Action).\
filter(Action.article_id==art.id).\
filter(Action.what.in_(('src_add','src_remove','art_add','tag_add','tag_remove','mark_sourced','mark_unsourced','helpreq_open','helpreq_close','comment','label_add','label_remove'))).\
order_by(Action.performed.desc()).\
slice(0,n_actions+1).\
all()
more_actions = False
if len(recent_actions) > n_actions:
more_actions = True
recent_actions = recent_actions[:n_actions]
recent_actions = reversed(recent_actions)
add_paper_form = AddPaperForm()
add_pr_form = AddPRForm()
add_other_form = AddOtherForm()
self.render(
'article.html',
art=art,
article_content=html,
researchers=researchers,
institutions=institutions,
journals=journals,
scrape_err=scrape_err,
add_paper_form=add_paper_form,
add_pr_form=add_pr_form,
add_other_form=add_other_form,
recent_actions=recent_actions,
more_actions=more_actions,
)
