-
Notifications
You must be signed in to change notification settings - Fork 9
/
search_listings.py
181 lines (140 loc) · 5.2 KB
/
search_listings.py
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
"""
Download list of rooms
"""
import math
import csv
import json
from typing import Union, Dict, List
from pathlib import Path
from multiprocessing.pool import Pool
import socket
import signal
import tap
from tqdm.auto import tqdm, trange
from helpers import graphql_request, get_slug, save_json
socket.setdefaulttimeout(1)
# Intercept Ctrl-C to exit gracefully
stop = False
def signal_handler(signal_received, frame):
global stop
stop = True
signal.signal(signal.SIGINT, signal_handler)
class Arguments(tap.Tap):
locations: Path # txt files containing one location per line
correspondance: Path = Path("correspondance_listing")
output: Path = Path("results_listing")
num_procs: int = 5
num_splits: int = 40
start: int = 0
def search_page(name: str, offset: int, limit: int):
variables = {
"request": {
"query": name,
"metadataOnly": False,
"version": "1.7.9",
"itemsPerGrid": limit,
"itemsOffset": offset,
"refinementPaths": ["/homes"],
}
}
query = {
"operationName": ["ExploreSearch"],
"locale": ["en"],
"currency": ["EUR"],
"variables": [json.dumps(variables)],
"extensions": [
'{"persistedQuery":{"version":1,"sha256Hash":"1816e0a81cc05d7e63f9a3817dc1e3e8a2e3a9bb5dafdf6dbc212b9aa2880391"}}'
],
}
results = graphql_request("ExploreSearch", query)
return results
def extract_listings(data: Dict) -> List[int]:
for section in data["data"]["dora"]["exploreV3"]["sections"]:
if section["__typename"] == "DoraExploreV3ListingsSection":
return [int(l["listing"]["id"]) for l in section["items"]]
return []
def search_location(name: str, dest: Path, limit: int):
# print(name)
if (dest / "listings.txt").is_file():
return
data = search_page(name, 0, limit)
listings = []
# Sometimes Airbnb says that they are no results, whereas it is going to find some if we retry
for i in range(2):
listings += extract_listings(data)
if listings != []:
break
num_listings = 0
pagination = data["data"]["dora"]["exploreV3"]["metadata"]["paginationMetadata"]
num_listings = (
int(pagination["totalCount"]) if pagination["totalCount"] is not None else 0
)
# assert pagination["pageLimit"] == limit, pagination
for counter in range(limit, int(num_listings), limit):
data = search_page(name, counter, limit)
listings += extract_listings(data)
dest.mkdir(exist_ok=True, parents=True)
with open(dest / "listings.txt", "w") as fid:
fid.write("\n".join([str(l) for l in listings]))
def search_locations(location_file: Union[Path, str], limit: int = 50):
# print(location_file)
with open(location_file, "r") as fid:
num_rows = sum(1 for _ in fid.readlines())
with open(location_file, newline="") as f:
reader = csv.DictReader(f, delimiter="\t", fieldnames=("name", "dest"))
for row in reader:
search_location(row["name"], Path(row["dest"]), limit)
def make_correspondance(args: Arguments):
"""
CSV file: location name\t path/to/location
"""
print(f"Running {args.num_splits} splits")
args.correspondance.mkdir(parents=True, exist_ok=True)
with open(args.locations, "r") as fid:
locations = [l.strip() for l in fid.readlines()]
per_split = math.ceil(len(locations) / args.num_splits)
counter = args.start
for split_id in trange(args.num_splits):
if stop:
break
correspondance_file = args.correspondance / f"listings.part-{split_id}.tsv"
end = min(len(locations), counter + per_split)
with open(correspondance_file, "w") as f:
for i in range(counter, end):
if stop:
break
path_to_location = args.output / get_slug(locations[i])
f.write("\t".join([locations[i], str(path_to_location)]))
f.write("\n")
counter += per_split
def run_downloader(args: Arguments):
"""
Inputs:
process: (int) number of process to run
images_url:(list) list of images url
"""
if args.num_procs == 0:
print(f"Running without parallelization")
correspondances = sorted(list(args.correspondance.iterdir()))[args.start :]
for correspondance in tqdm(correspondances):
search_locations(correspondance)
else:
print(f"Running {args.num_procs} procs")
correspondances = sorted(list(args.correspondance.iterdir()))[args.start :]
with Pool(args.num_procs) as pool:
list(
tqdm(
pool.imap(search_locations, correspondances, chunksize=1),
total=len(correspondances),
)
)
if __name__ == "__main__":
args = Arguments().parse_args()
print(args)
if not args.correspondance.is_dir() or list(args.correspondance.glob("*.tsv")) == []:
print("Making correspondance")
make_correspondance(args)
download = args.output
download.mkdir(exist_ok=True)
# search_location("new york city", Path("test"), 50)
run_downloader(args)