diff --git a/science_access/online_app_backend.py b/science_access/online_app_backend.py index ea97ab2371ce3a0297a1314472ca656316e12773..0ee4114bdcdc1216df32072c1ebbf1e6fdfaf23a 100644 --- a/science_access/online_app_backend.py +++ b/science_access/online_app_backend.py @@ -162,26 +162,32 @@ def ar_manipulation(ar): trainingDats.extend(ar) return (ar, trainingDats) +import os +from crossref_commons.iteration import iterate_publications_as_json -from duckduckpy import query -#@st.cache(suppress_st_warning=True) def call_from_front_end(NAME): if not heroku: scholar_link=str('https://scholar.google.com/scholar?hl=en&as_sdt=0%2C3&q=')+str(NAME) + for link in scholar_link: + st.text(link) + _, _, ar = enter_name_here(scholar_link,NAME) if heroku: - scholar_link=str('https://www.researchgate.net/search?q=')+str(NAME)#+str('&sort=relevance') - #query_string = str('!scholar ')+NAME - #response = query(query_string, secure=False, container=u'namedtuple', verbose=False, - #user_agent=u'duckduckpy 0.2', no_redirect=False, no_html=False, - #skip_disambig=False) - #scholar_link = response[-1] - sleep(np.random.uniform(1,3)) - st.write(scholar_link) - _, _, ar = enter_name_here(scholar_link,NAME) - st.write(ar) + filter_ = {'type': 'journal-article'} + queries = {'query.author': NAME} + ar = [] + bi =[p for p in iterate_publications_as_json(max_results=130, filter=filter_, queries=queries)] + for i in bi[0:15] + temp=str('https://unpaywall.org/'+str(p['DOI'])) + st.text(temp) + urlDat = process(temp) + if not isinstance(urlDat,type(None)): + ar.append(urlDat) + st.text(urlDat) + + (ar, trainingDats) = ar_manipulation(ar) ''' with open('data/traingDats.p','rb') as f: diff --git a/try.py b/try.py new file mode 100644 index 0000000000000000000000000000000000000000..5c2da21a542ce5e82bf35259929d4f943a216817 --- /dev/null +++ b/try.py @@ -0,0 +1,10 @@ +from crossref_commons.iteration import iterate_publications_as_json +filter = {'type': 'journal-article'} +queries = {'query.author': 'McGurrin'} +import os +for p in iterate_publications_as_json(max_results=189, filter=filter, queries=queries): + if 'abstract' in p.keys(): + print(p.keys()) + temp='wget https://unpaywall.org/'+str(p['DOI']) + print(temp) + os.system(temp)