def query():
''' Process the POST request with the search query. '''
query = request.form["query"]
songs = search.search_query(query)
return jsonify(songs)
def search_query():
global output
try:
query = form['query'].value.decode('utf-8')
output['results'] = search.search_query(query)
output['status'] = 'OK'
except KeyError, e:
output['reason'] = 'nothing-%s' % e
class main:
#vector_space = inverted_index()
# *************************** INPUTTING THE QUERY ***************************
k = input('Enter value of k:') #Accepting query from the user
query = input('Enter your search query:') #Accepting query from the user
search.search_query(query, k)
def on_click(self):
self.statusBar().showMessage('Getting data...')
searchboxValue = self.searchbox.currentText()
#send this search query to kaushik and get back top 10 results in the form of a list or something
books_result = search.search_query(searchboxValue)
if not books_result:
self.incorrect_query.show()
for i in range(0, 6):
self.titles[i].hide()
for i in range(0, 10):
self.serial_no[i].hide()
for i in range(0, 10):
self.rlabel[i].hide()
for i in range(0, 10):
self.author_label[i].hide()
for i in range(0, 10):
self.genre_label[i].hide()
for i in range(0, 10):
self.rating_label[i].hide()
for i in range(0, 10):
self.rbutton[i].hide()
else:
self.incorrect_query.hide()
for i in range(0, 10):
self.rlabel[i].setText(books_result[i].title)
self.author_label[i].setText(books_result[i].author)
self.genre_label[i].setText(books_result[i].genre)
self.rating_label[i].setText(books_result[i].rating)
for i in range(0, 6):
self.titles[i].show()
for i in range(0, 10):
self.serial_no[i].show()
for i in range(0, 10):
self.rlabel[i].show()
for i in range(0, 10):
self.rbutton[i].show()
for i in range(0, 10):
self.author_label[i].show()
for i in range(0, 10):
self.genre_label[i].show()
for i in range(0, 10):
self.rating_label[i].show()
self.statusBar().showMessage('Search Complete')
def unit_test(rootDir, testcase_path, out_file):
''''''
(file_data, file_dirs) = get_filtered_data(rootDir)
tokens_filtered = preprocess(file_data)
token2files, filentoken2occ, token2occ = metrics.get_counts(tokens_filtered)
filentoken2tfidf = metrics.get_tfidf(tokens_filtered, token2files, filentoken2occ)
print('Loading done.')
testcases = read_testcases(testcase_path, file_dirs)
testcase_pos = []
testcase_pos_scored = []
all_results = []
for t in testcases:
results = search_query(t[0], filentoken2tfidf, token2files, debug=True)
all_results.append(results)
targets_pos = []
targets_scores = []
# where are the testcase targets?
for i, r in enumerate(results):
for target in range(len(t) - 1):
if r[1] == t[1 + target]:
# use the average of all positions with the the same score
start, end = range_find(list(zip(*results))[0], r[0])
targets_scores.append((end + start) / 2)
targets_pos.append(i)
# how many not found?
not_found = len(t) - 1 - len(targets_pos)
targets_scores.extend([100] * not_found)
# average target scores
testcase_pos_scored.append(sum(targets_scores) / len(targets_scores))
testcase_pos.append(set(targets_pos))
# if not found, partial error 100
error = sum(testcase_pos_scored)
with open(out_file, 'w') as out:
out.write('Total error: {} (sum of target pos)\nTestcase errors: {}\n'.format(error, ' '.join(map(str, testcase_pos_scored))))
for i, t in enumerate(testcases):
out.write('\n\n\n\nTESTCASE {}\nError: {}\nQuery: {}\n{}\n'.format(i + 1, testcase_pos_scored[i], t[0], '-'*50))
for j, res in enumerate(all_results[i]):
if j in testcase_pos[i]:
out.write('(target) ')
out.write('score: {}, path: {}, explanation: {}\n'.format(*res))
def start(query, page):
click.echo("Please Wait..")
results = search_query(query, page)
click.echo(results)
click.echo("=" * 200)
click.echo("Do You Want Make a Excel File? [y/n]")
a = click.getchar()
click.echo()
if a == "y":
click.echo("Okay")
folder_setup(results)
elif a == "n":
click.echo("Opss..")
else:
click.echo("Invalid Input, Please Type Yes or Not")
def search():
searchTerm = request.data.decode('utf-8')
# Perform the search query
# Execute the searching script
results = search_query(searchTerm, filentoken2tfidf, token2files)
paths = [
get_real_filepath(rootDir, realFileDir, file_dirs, results[i][1])
for i in range(len(results))
]
inds = [results[i] for i in range(len(results))]
fileNames = [element[element.rfind('/') + 1:] for element in paths]
print(paths)
print(fileNames)
return jsonify(fileNames)
def search():
query = request.args.get("search")
user = request.args.get("profile")
field = request.args.get("field")
topic = request.args.get("topic")
res, should = search_query(
query,
count=10,
user=user,
topic=topic,
field=field,
)
return render_template("search.html",
tweets=res,
query=query,
should=should,
count=10,
user=user,
topic=topic,
field=field)
def index():
global global_search
global global_artists
global global_genre
global global_music
global global_lyrics
if request.method == 'POST':
if 'form_1' in request.form:
if request.form['nm']:
search = request.form['nm']
global_search = search
print(global_search)
else :
search = global_search
list_songs, artists, genres, music, lyrics = search_query(search)
global_artists, global_genre, global_music, global_lyrics = artists, genres, music, lyrics
elif 'form_2' in request.form:
search = global_search
artist_filter = []
genre_filter = []
music_filter = []
lyrics_filter = []
for i in global_artists :
if request.form.get(i["key"]):
artist_filter.append(i["key"])
for i in global_genre :
if request.form.get(i["key"]):
genre_filter.append(i["key"])
for i in global_music:
if request.form.get(i["key"]):
music_filter.append(i["key"])
for i in global_lyrics:
if request.form.get(i["key"]):
lyrics_filter.append(i["key"])
list_songs, artists, genres, music, lyrics = search_query_filtered(search, artist_filter, genre_filter, music_filter, lyrics_filter)
return render_template('index.html', songs = list_songs, artists = artists, genres = genres, music = music, lyrics = lyrics)
return render_template('index.html', songs = '', artists = '', genres = '', music = '', lyrics = '')
ideal_ndcg[str(key)] = sum
def calculate_dcg(query, documents):
count = 1
sum = 0.0
for document in documents:
relevance =0
key = str(query['query number']) + "_" + str(document)
if key in query_document_relevance:
relevance = query_document_relevance[key]
sum = sum + relevance / math.log10(count + 1)
count = count + 1
return sum
if __name__ == '__main__':
ideal_ndcg_queries()
sum = 0.0
all_queries = [query for query in read_queries() if query['query number'] != 0]
for query in all_queries:
documents = search_query(query)
assert len(documents)==len(set(documents)), "Search results should not have duplicates:"+str(documents)
if len(documents) > 0:
print ("Query:{} and Results:{}".format(query, documents))
dcg = calculate_dcg(query, documents)
idcg = ideal_ndcg[str(query['query number'])]
ndcg = dcg / idcg
print ("dcg={}, ideal={}, ndcg={}".format(dcg, idcg, ndcg))
sum = sum + ndcg
print ("Final ncdg for all queries is {}".format(sum / len(all_queries)))
for document in documents:
relevance = 0
key = str(query['query number']) + "_" + str(document)
if key in query_document_relevance:
relevance = query_document_relevance[key]
sum = sum + relevance / math.log10(count + 1)
count = count + 1
return sum
if __name__ == '__main__':
ideal_ndcg_queries()
sum = 0.0
all_queries = [
query for query in read_queries() if query['query number'] != 0
]
for query in all_queries:
document_objects = search_query(query, load_models())
documents = []
for each in document_objects:
documents.append(each['id'])
if len(documents) > 0:
print("Query:{} and Results:{}".format(query, documents))
dcg = calculate_dcg(query, documents)
idcg = ideal_ndcg[str(query['query number'])]
ndcg = dcg / idcg
print("dcg={}, ideal={}, ndcg={}".format(dcg, idcg, ndcg))
sum = sum + ndcg
print("Final ncdg for all queries is {}".format(sum / len(all_queries)))
def result():
if request.method == 'POST':
query = (request.get_data()).decode("utf-8")
query = query[4:]
search.search_query(query, 3)
return render_template('results.html')
