def search_for_citing_articles(UID, SID):
directory = "citing articles search results xml"
if not os.path.exists(directory):
os.makedirs(directory)
filename = "citing articles search results xml/" + UID[4:] + ".txt"
counter = 0
# Save file with citing article data if it hasn't been saved yet
if not os.path.exists(filename):
print(UID + " citing articles")
# Search on WOS
results = wok_soap.citingArticles(UID, SID)
[counter, SID] = counter_check(counter, SID)
queryId = results[0]
results_count = results[1]
# Interpret raw search results stored in 5th line of object
results_unicode = results[4]
if results_count > 100:
retrieve_count = (results_count // 100)
if results_count % 100 == 0:
retrieve_count -= 1
for hundred in range(retrieve_count):
start_count = (100 * hundred) + 101
more_results = wok_soap.retrieve(queryId, SID, start_count,
"Fields")
[counter, SID] = counter_check(counter, SID)
more_results_unicode = more_results[0]
results_unicode = results_unicode[:-10] + more_results_unicode[
82:]
root = ET.fromstring(results_unicode)
length = len(root)
if length != results_count:
raise Exception
# Write raw search results to txt file
with open(filename, "w") as f:
f.write(results_unicode)
return [filename, counter]
def search_for_citing_articles(UID, SID):
directory = "citing articles search results xml"
if not os.path.exists(directory):
os.makedirs(directory)
filename = "citing articles search results xml/" + UID[4:] + ".txt"
counter = 0
# Save file with citing article data if it hasn't been saved yet
if not os.path.exists(filename):
print UID + " citing articles"
# Search on WOS
results = wok_soap.citingArticles(UID, SID)
[counter, SID] = counter_check(counter, SID)
queryId = results[0]
results_count = results[1]
print results_count
# Interpret raw search results stored in 5th line of object
results_unicode = results[4].encode('utf-8')
if results_count > 100:
retrieve_count = (results_count // 100)
if results_count % 100 == 0:
retrieve_count -= 1
for hundred in range(retrieve_count):
start_count = (100*hundred) + 101
more_results = wok_soap.retrieve(queryId, SID, start_count, "Fields")
[counter, SID] = counter_check(counter, SID)
more_results_unicode = more_results[0].encode('utf-8')
results_unicode = results_unicode[:-10] + more_results_unicode[82:]
root = ET.fromstring(results_unicode)
length = len(root)
if length != results_count:
raise Exception
# Write raw search results to txt file
with open(filename, "w") as f:
f.write(results_unicode)
return [filename, counter]
def search_by_grant(csv_file, SID):
directory = "grant search results xml"
if not os.path.exists(directory):
os.makedirs(directory)
with open(csv_file, "rb") as h:
text = csv.reader(h)
grant_list = [row[0] for row in text]
file_list = []
counter = 0
for i, cell in enumerate(grant_list):
# Define query
grant_number_full = cell
if grant_number_full[0:2] == "DE":
prefix = grant_number_full[3:5]
grant_number = grant_number_full[5:]
query = "FT = " + prefix + grant_number + " OR FT = " + prefix + " " + grant_number
filename = "grant search results xml/" + query + ".txt"
else:
query = "FT = " + grant_number_full
filename = "grant search results xml/" + query.replace("/","") + ".txt"
file_list.append(filename)
if not os.path.exists(filename):
print query
# Search on WOS
results = wok_soap.search(query, SID)
[counter, SID] = counter_check(counter, SID)
queryId = results[0]
results_count = results[1]
# Interpret raw search results stored in 4th line of object
results_unicode = results[3].encode('utf-8')
if results_count > 100:
retrieve_count = (results_count // 100)
if results_count % 100 == 0:
retrieve_count -= 1
for hundred in range(retrieve_count):
start_count = (100*hundred) + 101
more_results = wok_soap.retrieve(queryId, SID, start_count, "FullRecord")
[counter, SID] = counter_check(counter, SID)
more_results_unicode = more_results[0].encode('utf-8')
results_unicode = results_unicode[:-10] + more_results_unicode[86:]
root = ET.fromstring(results_unicode)
length = len(root)
if length != results_count:
raise
# Write raw search results to txt file
with open(filename, "w") as f:
f.write(results_unicode)
return [grant_list, file_list, counter]
def search_by_grant(csv_file, SID):
directory = "grant search results xml"
if not os.path.exists(directory):
os.makedirs(directory)
with open(csv_file) as h:
text = csv.reader(h)
grant_list = [row[0] for row in text]
file_list = []
counter = 0
for i, cell in enumerate(grant_list):
# Define query
grant_number_full = cell
if grant_number_full[0:2] == "DE":
prefix = grant_number_full[3:5]
grant_number = grant_number_full[5:]
query = "FT = " + prefix + grant_number + " OR FT = " + prefix + " " + grant_number
filename = "grant search results xml/" + query + ".txt"
else:
query = "FT = " + grant_number_full
filename = "grant search results xml/" + query.replace("/",
"") + ".txt"
file_list.append(filename)
if not os.path.exists(filename):
print(query)
# Search on WOS
results = wok_soap.search(query, SID)
[counter, SID] = counter_check(counter, SID)
queryId = results[0]
results_count = results[1]
# Interpret raw search results stored in 4th line of object
results_unicode = results[3]
if results_count > 100:
retrieve_count = (results_count // 100)
if results_count % 100 == 0:
retrieve_count -= 1
for hundred in range(retrieve_count):
start_count = (100 * hundred) + 101
more_results = wok_soap.retrieve(queryId, SID, start_count,
"FullRecord")
[counter, SID] = counter_check(counter, SID)
more_results_unicode = more_results[0].encode('utf-8')
results_unicode = results_unicode[:
-10] + more_results_unicode[
86:]
root = ET.fromstring(results_unicode)
length = len(root)
if length != results_count:
raise
# Write raw search results to txt file
with open(filename, "w") as f:
f.write(results_unicode)
return [grant_list, file_list, counter]
def searchByGrantOrDOI(csv_file, searchType):
directory = "search by grant or doi xml/"
if not os.path.exists(directory): # Check for and create a directory
os.makedirs(directory)
column1List = []
with open(csv_file) as h: # Open a CSV file
text = csv.reader(h)
column1List = [row[0].replace(u'\ufeff','') for row in text] # gets rid of '\ufeff' at beginning of csv
counter = 0
SID = ""
# define queryList, a list of queries
queryList = []
file_list = []
# === Handle second argument, searchType ====
searchType = searchType.lower() # converts the string to lowercase
acceptableSearchTypes = ["grant", "doi"] # later can add author, etc
if searchType not in acceptableSearchTypes: # raise error if grantOrDOI is not a grant or a doi
raise Exception("Second argument of searchByGrantOrDOI must be 'grant' or 'doi'")
# CREATE QUERY
if searchType == "grant": # === Create grant query ===
for fullNumber in column1List:
if fullNumber[0:2] == "DE":
prefix = fullNumber[3:5]
grantNumber = fullNumber[5:]
query = "FT = " + prefix + grantNumber + " OR FT = " + prefix + " " + grantNumber
else:
query = "FT = " + str(fullNumber)
queryList.append(query)
elif searchType == "doi": #=== Create DOI query ===
for result in column1List:
print(result)
ID = result.strip(' \t\n\r').replace(" ","").replace(u'\u200b','') #remove non-printing characters
print("ID is " + ID)
if ID[0:3] == "WOS": # Define query
query = "UT = " + ID
print("query = " + str(query))
else:
query = 'DO = "' + ID + '"'
print("query = " + str(query))
queryList.append(query)
for q in queryList:
# create filename without slashes or quotes
filename = directory + q.replace("/"," ").replace('"',"") + ".txt"
# Add each file to file list
file_list.append(filename)
# Search on WOS
if not os.path.exists(filename):
[counter, SID] = counter_check(counter, SID)
results = wok_soap.search(q, SID)
queryId = results[0]
results_count = results[1]
results_unicode = results[3]
# Handling throttle problems - can't get more than 100 at once
if results_count > 100:
retrieve_count = (results_count // 100)
if results_count % 100 == 0:
retrieve_count -= 1
for hundred in range(retrieve_count):
start_count = (100*hundred) + 101
[counter, SID] = counter_check(counter, SID)
more_results = wok_soap.retrieve(queryId, SID, start_count, "FullRecord")
more_results_unicode = more_results[0].encode('utf-8')
results_unicode = str(results_unicode[:-10]) + str(more_results_unicode[86:])
root = ET.fromstring(results_unicode) # ET = element tree. results_unicode is the object that contains all the search results
length = len(root)
if length != results_count:
raise Exception("length does not equal results_count")# throw error message
# Write raw search results to txt file
with open(filename, "w") as f:
f.write(results_unicode)
print(file_list)
return [column1List, file_list, counter] #subscription allows only 2500 records/session.
