Update app.py
Browse files
app.py
CHANGED
@@ -1,1140 +1,133 @@
|
|
1 |
-
import asyncio
|
2 |
-
import logging
|
3 |
-
import warnings
|
4 |
-
from typing import Dict, Generator, Optional
|
5 |
-
import nest_asyncio
|
6 |
-
import json
|
7 |
-
import logging
|
8 |
-
import sys
|
9 |
-
from collections import deque
|
10 |
-
from datetime import datetime, timezone
|
11 |
-
from decimal import Decimal
|
12 |
-
from itertools import cycle
|
13 |
-
from typing import AsyncGenerator, Deque, Dict, Optional, Set, Tuple
|
14 |
-
from curl_cffi import requests
|
15 |
-
from docstring_inheritance import GoogleDocstringInheritanceMeta
|
16 |
-
from lxml import html
|
17 |
-
import json
|
18 |
-
import re
|
19 |
-
from html import unescape
|
20 |
-
from typing import Optional
|
21 |
-
from urllib.parse import unquote
|
22 |
-
from dataclasses import dataclass
|
23 |
-
from typing import Dict, Optional
|
24 |
-
from random import randint
|
25 |
-
class DuckDuckGoSearchException(Exception):
|
26 |
-
"""Base exception class for webscout."""
|
27 |
-
|
28 |
-
|
29 |
-
|
30 |
-
@dataclass
|
31 |
-
class MapsResult:
|
32 |
-
"""Represents a result from the maps search."""
|
33 |
-
|
34 |
-
title: Optional[str] = None
|
35 |
-
address: Optional[str] = None
|
36 |
-
country_code: Optional[str] = None
|
37 |
-
latitude: Optional[str] = None
|
38 |
-
longitude: Optional[str] = None
|
39 |
-
url: Optional[str] = None
|
40 |
-
desc: Optional[str] = None
|
41 |
-
phone: Optional[str] = None
|
42 |
-
image: Optional[str] = None
|
43 |
-
source: Optional[str] = None
|
44 |
-
hours: Optional[Dict[str, str]] = None
|
45 |
-
category: Optional[str] = None
|
46 |
-
facebook: Optional[str] = None
|
47 |
-
instagram: Optional[str] = None
|
48 |
-
twitter: Optional[str] = None
|
49 |
-
|
50 |
-
|
51 |
-
|
52 |
-
REGEX_500_IN_URL = re.compile(r"(?:\d{3}-\d{2}\.js)")
|
53 |
-
REGEX_STRIP_TAGS = re.compile("<.*?>")
|
54 |
-
REGEX_VQD = re.compile(rb"""vqd=['"]?([^&"']+)""")
|
55 |
-
|
56 |
-
|
57 |
-
def _extract_vqd(html_bytes: bytes, keywords: str) -> Optional[str]:
|
58 |
-
"""Extract vqd from html using a regular expression."""
|
59 |
-
try:
|
60 |
-
match = REGEX_VQD.search(html_bytes)
|
61 |
-
if match:
|
62 |
-
return match.group(1).decode()
|
63 |
-
except Exception:
|
64 |
-
pass
|
65 |
-
raise DuckDuckGoSearchException(f"_extract_vqd() {keywords=} Could not extract vqd.")
|
66 |
-
|
67 |
-
|
68 |
-
def _text_extract_json(html_bytes: bytes, keywords: str) -> Optional[str]:
|
69 |
-
"""text(backend="api") -> extract json from html."""
|
70 |
-
try:
|
71 |
-
start = html_bytes.index(b"DDG.pageLayout.load('d',") + 24
|
72 |
-
end = html_bytes.index(b");DDG.duckbar.load(", start)
|
73 |
-
data = html_bytes[start:end]
|
74 |
-
return json.loads(data)
|
75 |
-
except Exception as ex:
|
76 |
-
raise DuckDuckGoSearchException(f"_text_extract_json() {keywords=} {type(ex).__name__}: {ex}") from ex
|
77 |
-
|
78 |
-
|
79 |
-
def _is_500_in_url(url: str) -> bool:
|
80 |
-
"""Something like '506-00.js' inside the url."""
|
81 |
-
return bool(REGEX_500_IN_URL.search(url))
|
82 |
-
|
83 |
-
|
84 |
-
def _normalize(raw_html: str) -> str:
|
85 |
-
"""Strip HTML tags from the raw_html string."""
|
86 |
-
return unescape(REGEX_STRIP_TAGS.sub("", raw_html)) if raw_html else ""
|
87 |
-
|
88 |
-
|
89 |
-
def _normalize_url(url: str) -> str:
|
90 |
-
"""Unquote URL and replace spaces with '+'."""
|
91 |
-
return unquote(url.replace(" ", "+")) if url else ""
|
92 |
-
|
93 |
-
logger = logging.getLogger("duckduckgo_search.AsyncWEBS")
|
94 |
-
# Not working on Windows, NotImplementedError (https://curl-cffi.readthedocs.io/en/latest/faq/)
|
95 |
-
if sys.platform.lower().startswith("win"):
|
96 |
-
asyncio.set_event_loop_policy(asyncio.WindowsSelectorEventLoopPolicy())
|
97 |
-
|
98 |
-
|
99 |
-
class AsyncWEBS(metaclass=GoogleDocstringInheritanceMeta):
|
100 |
-
"""webscout_search async class to get search results from duckduckgo.com."""
|
101 |
-
|
102 |
-
def __init__(self, headers=None, proxies=None, timeout=10) -> None:
|
103 |
-
"""Initialize the AsyncWEBS object.
|
104 |
-
|
105 |
-
Args:
|
106 |
-
headers (dict, optional): Dictionary of headers for the HTTP client. Defaults to None.
|
107 |
-
proxies (Union[dict, str], optional): Proxies for the HTTP client (can be dict or str). Defaults to None.
|
108 |
-
timeout (int, optional): Timeout value for the HTTP client. Defaults to 10.
|
109 |
-
"""
|
110 |
-
useragent=f'{randint(0, 1000000)}'
|
111 |
-
headers = {'User-Agent': useragent}
|
112 |
-
self.proxies = proxies if proxies and isinstance(proxies, dict) else {"http": proxies, "https": proxies}
|
113 |
-
self._asession = requests.AsyncSession(
|
114 |
-
headers=headers, proxies=self.proxies, timeout=timeout, impersonate="chrome"
|
115 |
-
)
|
116 |
-
self._asession.headers["Referer"] = "https://duckduckgo.com/"
|
117 |
-
|
118 |
-
async def __aenter__(self) -> "AsyncWEBS":
|
119 |
-
"""A context manager method that is called when entering the 'with' statement."""
|
120 |
-
return self
|
121 |
-
|
122 |
-
async def __aexit__(self, exc_type, exc_val, exc_tb) -> None:
|
123 |
-
"""Closes the session."""
|
124 |
-
return self._asession.close()
|
125 |
-
|
126 |
-
async def _aget_url(self, method: str, url: str, **kwargs) -> Optional[requests.Response]:
|
127 |
-
try:
|
128 |
-
useragent=f'{randint(0, 1000000)}'
|
129 |
-
headers = {'User-Agent': useragent}
|
130 |
-
resp = await self._asession.request(method, url, stream=True, **kwargs,headers=headers)
|
131 |
-
resp.raise_for_status()
|
132 |
-
resp_content = await resp.acontent()
|
133 |
-
logger.debug(f"_aget_url() {url} {resp.status_code} {resp.http_version} {resp.elapsed} {len(resp_content)}")
|
134 |
-
if _is_500_in_url(str(resp.url)) or resp.status_code == 202:
|
135 |
-
raise DuckDuckGoSearchException("Ratelimit")
|
136 |
-
if resp.status_code == 200:
|
137 |
-
return resp_content
|
138 |
-
except Exception as ex:
|
139 |
-
raise DuckDuckGoSearchException(f"_aget_url() {url} {type(ex).__name__}: {ex}") from ex
|
140 |
-
|
141 |
-
async def _aget_vqd(self, keywords: str) -> Optional[str]:
|
142 |
-
"""Get vqd value for a search query."""
|
143 |
-
resp_content = await self._aget_url("POST", "https://duckduckgo.com", data={"q": keywords})
|
144 |
-
if resp_content:
|
145 |
-
return _extract_vqd(resp_content, keywords)
|
146 |
-
|
147 |
-
async def text(
|
148 |
-
self,
|
149 |
-
keywords: str,
|
150 |
-
region: str = "wt-wt",
|
151 |
-
safesearch: str = "moderate",
|
152 |
-
timelimit: Optional[str] = None,
|
153 |
-
backend: str = "api",
|
154 |
-
max_results: Optional[int] = None,
|
155 |
-
) -> AsyncGenerator[Dict[str, Optional[str]], None]:
|
156 |
-
"""DuckDuckGo text search generator. Query params: https://duckduckgo.com/params.
|
157 |
-
|
158 |
-
Args:
|
159 |
-
keywords: keywords for query.
|
160 |
-
region: wt-wt, us-en, uk-en, ru-ru, etc. Defaults to "wt-wt".
|
161 |
-
safesearch: on, moderate, off. Defaults to "moderate".
|
162 |
-
timelimit: d, w, m, y. Defaults to None.
|
163 |
-
backend: api, html, lite. Defaults to api.
|
164 |
-
api - collect data from https://duckduckgo.com,
|
165 |
-
html - collect data from https://html.duckduckgo.com,
|
166 |
-
lite - collect data from https://lite.duckduckgo.com.
|
167 |
-
max_results: max number of results. If None, returns results only from the first response. Defaults to None.
|
168 |
-
|
169 |
-
Yields:
|
170 |
-
dict with search results.
|
171 |
-
|
172 |
-
"""
|
173 |
-
if backend == "api":
|
174 |
-
results = self._text_api(keywords, region, safesearch, timelimit, max_results)
|
175 |
-
elif backend == "html":
|
176 |
-
results = self._text_html(keywords, region, safesearch, timelimit, max_results)
|
177 |
-
elif backend == "lite":
|
178 |
-
results = self._text_lite(keywords, region, timelimit, max_results)
|
179 |
-
|
180 |
-
async for result in results:
|
181 |
-
yield result
|
182 |
-
|
183 |
-
async def _text_api(
|
184 |
-
self,
|
185 |
-
keywords: str,
|
186 |
-
region: str = "wt-wt",
|
187 |
-
safesearch: str = "moderate",
|
188 |
-
timelimit: Optional[str] = None,
|
189 |
-
max_results: Optional[int] = None,
|
190 |
-
) -> AsyncGenerator[Dict[str, Optional[str]], None]:
|
191 |
-
"""webscout text search generator. Query params: https://duckduckgo.com/params.
|
192 |
-
|
193 |
-
Args:
|
194 |
-
keywords: keywords for query.
|
195 |
-
region: wt-wt, us-en, uk-en, ru-ru, etc. Defaults to "wt-wt".
|
196 |
-
safesearch: on, moderate, off. Defaults to "moderate".
|
197 |
-
timelimit: d, w, m, y. Defaults to None.
|
198 |
-
max_results: max number of results. If None, returns results only from the first response. Defaults to None.
|
199 |
-
|
200 |
-
Yields:
|
201 |
-
dict with search results.
|
202 |
-
|
203 |
-
"""
|
204 |
-
assert keywords, "keywords is mandatory"
|
205 |
-
|
206 |
-
vqd = await self._aget_vqd(keywords)
|
207 |
-
|
208 |
-
payload = {
|
209 |
-
"q": keywords,
|
210 |
-
"kl": region,
|
211 |
-
"l": region,
|
212 |
-
"bing_market": region,
|
213 |
-
"s": "0",
|
214 |
-
"df": timelimit,
|
215 |
-
"vqd": vqd,
|
216 |
-
# "o": "json",
|
217 |
-
"sp": "0",
|
218 |
-
}
|
219 |
-
safesearch = safesearch.lower()
|
220 |
-
if safesearch == "moderate":
|
221 |
-
payload["ex"] = "-1"
|
222 |
-
elif safesearch == "off":
|
223 |
-
payload["ex"] = "-2"
|
224 |
-
elif safesearch == "on": # strict
|
225 |
-
payload["p"] = "1"
|
226 |
-
|
227 |
-
cache = set()
|
228 |
-
for _ in range(11):
|
229 |
-
resp_content = await self._aget_url("GET", "https://links.duckduckgo.com/d.js", params=payload)
|
230 |
-
if resp_content is None:
|
231 |
-
return
|
232 |
-
|
233 |
-
page_data = _text_extract_json(resp_content, keywords)
|
234 |
-
if page_data is None:
|
235 |
-
return
|
236 |
-
|
237 |
-
result_exists, next_page_url = False, None
|
238 |
-
for row in page_data:
|
239 |
-
href = row.get("u", None)
|
240 |
-
if href and href not in cache and href != f"http://www.google.com/search?q={keywords}":
|
241 |
-
cache.add(href)
|
242 |
-
body = _normalize(row["a"])
|
243 |
-
if body:
|
244 |
-
result_exists = True
|
245 |
-
yield {
|
246 |
-
"title": _normalize(row["t"]),
|
247 |
-
"href": _normalize_url(href),
|
248 |
-
"body": body,
|
249 |
-
}
|
250 |
-
if max_results and len(cache) >= max_results:
|
251 |
-
return
|
252 |
-
else:
|
253 |
-
next_page_url = row.get("n", None)
|
254 |
-
if max_results is None or result_exists is False or next_page_url is None:
|
255 |
-
return
|
256 |
-
payload["s"] = next_page_url.split("s=")[1].split("&")[0]
|
257 |
-
|
258 |
-
async def _text_html(
|
259 |
-
self,
|
260 |
-
keywords: str,
|
261 |
-
region: str = "wt-wt",
|
262 |
-
safesearch: str = "moderate",
|
263 |
-
timelimit: Optional[str] = None,
|
264 |
-
max_results: Optional[int] = None,
|
265 |
-
) -> AsyncGenerator[Dict[str, Optional[str]], None]:
|
266 |
-
"""webscout text search generator. Query params: https://duckduckgo.com/params.
|
267 |
-
|
268 |
-
Args:
|
269 |
-
keywords: keywords for query.
|
270 |
-
region: wt-wt, us-en, uk-en, ru-ru, etc. Defaults to "wt-wt".
|
271 |
-
safesearch: on, moderate, off. Defaults to "moderate".
|
272 |
-
timelimit: d, w, m, y. Defaults to None.
|
273 |
-
max_results: max number of results. If None, returns results only from the first response. Defaults to None.
|
274 |
-
|
275 |
-
Yields:
|
276 |
-
dict with search results.
|
277 |
-
|
278 |
-
"""
|
279 |
-
assert keywords, "keywords is mandatory"
|
280 |
-
|
281 |
-
self._asession.headers["Referer"] = "https://html.duckduckgo.com/"
|
282 |
-
safesearch_base = {"on": 1, "moderate": -1, "off": -2}
|
283 |
-
payload = {
|
284 |
-
"q": keywords,
|
285 |
-
"s": "0",
|
286 |
-
"kl": region,
|
287 |
-
"p": safesearch_base[safesearch.lower()],
|
288 |
-
"df": timelimit,
|
289 |
-
}
|
290 |
-
cache: Set[str] = set()
|
291 |
-
for _ in range(11):
|
292 |
-
resp_content = await self._aget_url("POST", "https://html.duckduckgo.com/html", data=payload)
|
293 |
-
if resp_content is None:
|
294 |
-
return
|
295 |
-
|
296 |
-
tree = html.fromstring(resp_content)
|
297 |
-
if tree.xpath('//div[@class="no-results"]/text()'):
|
298 |
-
return
|
299 |
-
|
300 |
-
result_exists = False
|
301 |
-
for e in tree.xpath('//div[contains(@class, "results_links")]'):
|
302 |
-
href = e.xpath('.//a[contains(@class, "result__a")]/@href')
|
303 |
-
href = href[0] if href else None
|
304 |
-
if (
|
305 |
-
href
|
306 |
-
and href not in cache
|
307 |
-
and href != f"http://www.google.com/search?q={keywords}"
|
308 |
-
and not href.startswith("https://duckduckgo.com/y.js?ad_domain")
|
309 |
-
):
|
310 |
-
cache.add(href)
|
311 |
-
title = e.xpath('.//a[contains(@class, "result__a")]/text()')
|
312 |
-
body = e.xpath('.//a[contains(@class, "result__snippet")]//text()')
|
313 |
-
result_exists = True
|
314 |
-
yield {
|
315 |
-
"title": _normalize(title[0]) if title else None,
|
316 |
-
"href": _normalize_url(href),
|
317 |
-
"body": _normalize("".join(body)) if body else None,
|
318 |
-
}
|
319 |
-
if max_results and len(cache) >= max_results:
|
320 |
-
return
|
321 |
-
if max_results is None or result_exists is False:
|
322 |
-
return
|
323 |
-
next_page = tree.xpath('.//div[@class="nav-link"]')
|
324 |
-
next_page = next_page[-1] if next_page else None
|
325 |
-
if next_page is None:
|
326 |
-
return
|
327 |
-
|
328 |
-
names = next_page.xpath('.//input[@type="hidden"]/@name')
|
329 |
-
values = next_page.xpath('.//input[@type="hidden"]/@value')
|
330 |
-
payload = {n: v for n, v in zip(names, values)}
|
331 |
-
|
332 |
-
async def _text_lite(
|
333 |
-
self,
|
334 |
-
keywords: str,
|
335 |
-
region: str = "wt-wt",
|
336 |
-
timelimit: Optional[str] = None,
|
337 |
-
max_results: Optional[int] = None,
|
338 |
-
) -> AsyncGenerator[Dict[str, Optional[str]], None]:
|
339 |
-
"""webscout text search generator. Query params: https://duckduckgo.com/params.
|
340 |
-
|
341 |
-
Args:
|
342 |
-
keywords: keywords for query.
|
343 |
-
region: wt-wt, us-en, uk-en, ru-ru, etc. Defaults to "wt-wt".
|
344 |
-
timelimit: d, w, m, y. Defaults to None.
|
345 |
-
max_results: max number of results. If None, returns results only from the first response. Defaults to None.
|
346 |
-
|
347 |
-
Yields:
|
348 |
-
dict with search results.
|
349 |
-
|
350 |
-
"""
|
351 |
-
assert keywords, "keywords is mandatory"
|
352 |
-
|
353 |
-
self._asession.headers["Referer"] = "https://lite.duckduckgo.com/"
|
354 |
-
payload = {
|
355 |
-
"q": keywords,
|
356 |
-
"s": "0",
|
357 |
-
"o": "json",
|
358 |
-
"api": "d.js",
|
359 |
-
"kl": region,
|
360 |
-
"df": timelimit,
|
361 |
-
}
|
362 |
-
cache: Set[str] = set()
|
363 |
-
for _ in range(11):
|
364 |
-
resp_content = await self._aget_url("POST", "https://lite.duckduckgo.com/lite/", data=payload)
|
365 |
-
if resp_content is None:
|
366 |
-
return
|
367 |
-
|
368 |
-
if b"No more results." in resp_content:
|
369 |
-
return
|
370 |
-
|
371 |
-
tree = html.fromstring(resp_content)
|
372 |
-
|
373 |
-
result_exists = False
|
374 |
-
data = zip(cycle(range(1, 5)), tree.xpath("//table[last()]//tr"))
|
375 |
-
for i, e in data:
|
376 |
-
if i == 1:
|
377 |
-
href = e.xpath(".//a//@href")
|
378 |
-
href = href[0] if href else None
|
379 |
-
if (
|
380 |
-
href is None
|
381 |
-
or href in cache
|
382 |
-
or href == f"http://www.google.com/search?q={keywords}"
|
383 |
-
or href.startswith("https://duckduckgo.com/y.js?ad_domain")
|
384 |
-
):
|
385 |
-
[next(data, None) for _ in range(3)] # skip block(i=1,2,3,4)
|
386 |
-
else:
|
387 |
-
cache.add(href)
|
388 |
-
title = e.xpath(".//a//text()")[0]
|
389 |
-
elif i == 2:
|
390 |
-
body = e.xpath(".//td[@class='result-snippet']//text()")
|
391 |
-
body = "".join(body).strip()
|
392 |
-
elif i == 3:
|
393 |
-
result_exists = True
|
394 |
-
yield {
|
395 |
-
"title": _normalize(title),
|
396 |
-
"href": _normalize_url(href),
|
397 |
-
"body": _normalize(body),
|
398 |
-
}
|
399 |
-
if max_results and len(cache) >= max_results:
|
400 |
-
return
|
401 |
-
if max_results is None or result_exists is False:
|
402 |
-
return
|
403 |
-
next_page_s = tree.xpath("//form[./input[contains(@value, 'ext')]]/input[@name='s']/@value")
|
404 |
-
if not next_page_s:
|
405 |
-
return
|
406 |
-
payload["s"] = next_page_s[0]
|
407 |
-
payload["vqd"] = _extract_vqd(resp_content, keywords)
|
408 |
-
|
409 |
-
async def images(
|
410 |
-
self,
|
411 |
-
keywords: str,
|
412 |
-
region: str = "wt-wt",
|
413 |
-
safesearch: str = "moderate",
|
414 |
-
timelimit: Optional[str] = None,
|
415 |
-
size: Optional[str] = None,
|
416 |
-
color: Optional[str] = None,
|
417 |
-
type_image: Optional[str] = None,
|
418 |
-
layout: Optional[str] = None,
|
419 |
-
license_image: Optional[str] = None,
|
420 |
-
max_results: Optional[int] = None,
|
421 |
-
) -> AsyncGenerator[Dict[str, Optional[str]], None]:
|
422 |
-
"""webscout images search. Query params: https://duckduckgo.com/params.
|
423 |
-
|
424 |
-
Args:
|
425 |
-
keywords: keywords for query.
|
426 |
-
region: wt-wt, us-en, uk-en, ru-ru, etc. Defaults to "wt-wt".
|
427 |
-
safesearch: on, moderate, off. Defaults to "moderate".
|
428 |
-
timelimit: Day, Week, Month, Year. Defaults to None.
|
429 |
-
size: Small, Medium, Large, Wallpaper. Defaults to None.
|
430 |
-
color: color, Monochrome, Red, Orange, Yellow, Green, Blue,
|
431 |
-
Purple, Pink, Brown, Black, Gray, Teal, White. Defaults to None.
|
432 |
-
type_image: photo, clipart, gif, transparent, line.
|
433 |
-
Defaults to None.
|
434 |
-
layout: Square, Tall, Wide. Defaults to None.
|
435 |
-
license_image: any (All Creative Commons), Public (PublicDomain),
|
436 |
-
Share (Free to Share and Use), ShareCommercially (Free to Share and Use Commercially),
|
437 |
-
Modify (Free to Modify, Share, and Use), ModifyCommercially (Free to Modify, Share, and
|
438 |
-
Use Commercially). Defaults to None.
|
439 |
-
max_results: max number of results. If None, returns results only from the first response. Defaults to None.
|
440 |
-
|
441 |
-
Yields:
|
442 |
-
dict with image search results.
|
443 |
-
|
444 |
-
"""
|
445 |
-
assert keywords, "keywords is mandatory"
|
446 |
-
|
447 |
-
vqd = await self._aget_vqd(keywords)
|
448 |
-
|
449 |
-
safesearch_base = {"on": 1, "moderate": 1, "off": -1}
|
450 |
-
timelimit = f"time:{timelimit}" if timelimit else ""
|
451 |
-
size = f"size:{size}" if size else ""
|
452 |
-
color = f"color:{color}" if color else ""
|
453 |
-
type_image = f"type:{type_image}" if type_image else ""
|
454 |
-
layout = f"layout:{layout}" if layout else ""
|
455 |
-
license_image = f"license:{license_image}" if license_image else ""
|
456 |
-
payload = {
|
457 |
-
"l": region,
|
458 |
-
"o": "json",
|
459 |
-
"q": keywords,
|
460 |
-
"vqd": vqd,
|
461 |
-
"f": f"{timelimit},{size},{color},{type_image},{layout},{license_image}",
|
462 |
-
"p": safesearch_base[safesearch.lower()],
|
463 |
-
}
|
464 |
-
|
465 |
-
cache = set()
|
466 |
-
for _ in range(10):
|
467 |
-
resp_content = await self._aget_url("GET", "https://duckduckgo.com/i.js", params=payload)
|
468 |
-
if resp_content is None:
|
469 |
-
return
|
470 |
-
try:
|
471 |
-
resp_json = json.loads(resp_content)
|
472 |
-
except Exception:
|
473 |
-
return
|
474 |
-
page_data = resp_json.get("results", None)
|
475 |
-
if page_data is None:
|
476 |
-
return
|
477 |
-
|
478 |
-
result_exists = False
|
479 |
-
for row in page_data:
|
480 |
-
image_url = row.get("image", None)
|
481 |
-
if image_url and image_url not in cache:
|
482 |
-
cache.add(image_url)
|
483 |
-
result_exists = True
|
484 |
-
yield {
|
485 |
-
"title": row["title"],
|
486 |
-
"image": _normalize_url(image_url),
|
487 |
-
"thumbnail": _normalize_url(row["thumbnail"]),
|
488 |
-
"url": _normalize_url(row["url"]),
|
489 |
-
"height": row["height"],
|
490 |
-
"width": row["width"],
|
491 |
-
"source": row["source"],
|
492 |
-
}
|
493 |
-
if max_results and len(cache) >= max_results:
|
494 |
-
return
|
495 |
-
if max_results is None or result_exists is False:
|
496 |
-
return
|
497 |
-
next = resp_json.get("next", None)
|
498 |
-
if next is None:
|
499 |
-
return
|
500 |
-
payload["s"] = next.split("s=")[-1].split("&")[0]
|
501 |
-
|
502 |
-
async def videos(
|
503 |
-
self,
|
504 |
-
keywords: str,
|
505 |
-
region: str = "wt-wt",
|
506 |
-
safesearch: str = "moderate",
|
507 |
-
timelimit: Optional[str] = None,
|
508 |
-
resolution: Optional[str] = None,
|
509 |
-
duration: Optional[str] = None,
|
510 |
-
license_videos: Optional[str] = None,
|
511 |
-
max_results: Optional[int] = None,
|
512 |
-
) -> AsyncGenerator[Dict[str, Optional[str]], None]:
|
513 |
-
"""webscout videos search. Query params: https://duckduckgo.com/params.
|
514 |
-
|
515 |
-
Args:
|
516 |
-
keywords: keywords for query.
|
517 |
-
region: wt-wt, us-en, uk-en, ru-ru, etc. Defaults to "wt-wt".
|
518 |
-
safesearch: on, moderate, off. Defaults to "moderate".
|
519 |
-
timelimit: d, w, m. Defaults to None.
|
520 |
-
resolution: high, standart. Defaults to None.
|
521 |
-
duration: short, medium, long. Defaults to None.
|
522 |
-
license_videos: creativeCommon, youtube. Defaults to None.
|
523 |
-
max_results: max number of results. If None, returns results only from the first response. Defaults to None.
|
524 |
-
|
525 |
-
Yields:
|
526 |
-
dict with videos search results
|
527 |
-
|
528 |
-
"""
|
529 |
-
assert keywords, "keywords is mandatory"
|
530 |
-
|
531 |
-
vqd = await self._aget_vqd(keywords)
|
532 |
-
|
533 |
-
safesearch_base = {"on": 1, "moderate": -1, "off": -2}
|
534 |
-
timelimit = f"publishedAfter:{timelimit}" if timelimit else ""
|
535 |
-
resolution = f"videoDefinition:{resolution}" if resolution else ""
|
536 |
-
duration = f"videoDuration:{duration}" if duration else ""
|
537 |
-
license_videos = f"videoLicense:{license_videos}" if license_videos else ""
|
538 |
-
payload = {
|
539 |
-
"l": region,
|
540 |
-
"o": "json",
|
541 |
-
"s": 0,
|
542 |
-
"q": keywords,
|
543 |
-
"vqd": vqd,
|
544 |
-
"f": f"{timelimit},{resolution},{duration},{license_videos}",
|
545 |
-
"p": safesearch_base[safesearch.lower()],
|
546 |
-
}
|
547 |
-
|
548 |
-
cache = set()
|
549 |
-
for _ in range(10):
|
550 |
-
resp_content = await self._aget_url("GET", "https://duckduckgo.com/v.js", params=payload)
|
551 |
-
if resp_content is None:
|
552 |
-
return
|
553 |
-
try:
|
554 |
-
resp_json = json.loads(resp_content)
|
555 |
-
except Exception:
|
556 |
-
return
|
557 |
-
page_data = resp_json.get("results", None)
|
558 |
-
if page_data is None:
|
559 |
-
return
|
560 |
-
|
561 |
-
result_exists = False
|
562 |
-
for row in page_data:
|
563 |
-
if row["content"] not in cache:
|
564 |
-
cache.add(row["content"])
|
565 |
-
result_exists = True
|
566 |
-
yield row
|
567 |
-
if max_results and len(cache) >= max_results:
|
568 |
-
return
|
569 |
-
if max_results is None or result_exists is False:
|
570 |
-
return
|
571 |
-
next = resp_json.get("next", None)
|
572 |
-
if next is None:
|
573 |
-
return
|
574 |
-
payload["s"] = next.split("s=")[-1].split("&")[0]
|
575 |
-
|
576 |
-
async def news(
|
577 |
-
self,
|
578 |
-
keywords: str,
|
579 |
-
region: str = "wt-wt",
|
580 |
-
safesearch: str = "moderate",
|
581 |
-
timelimit: Optional[str] = None,
|
582 |
-
max_results: Optional[int] = None,
|
583 |
-
) -> AsyncGenerator[Dict[str, Optional[str]], None]:
|
584 |
-
"""webscout news search. Query params: https://duckduckgo.com/params.
|
585 |
-
|
586 |
-
Args:
|
587 |
-
keywords: keywords for query.
|
588 |
-
region: wt-wt, us-en, uk-en, ru-ru, etc. Defaults to "wt-wt".
|
589 |
-
safesearch: on, moderate, off. Defaults to "moderate".
|
590 |
-
timelimit: d, w, m. Defaults to None.
|
591 |
-
max_results: max number of results. If None, returns results only from the first response. Defaults to None.
|
592 |
-
|
593 |
-
Yields:
|
594 |
-
dict with news search results.
|
595 |
-
|
596 |
-
"""
|
597 |
-
assert keywords, "keywords is mandatory"
|
598 |
-
|
599 |
-
vqd = await self._aget_vqd(keywords)
|
600 |
-
|
601 |
-
safesearch_base = {"on": 1, "moderate": -1, "off": -2}
|
602 |
-
payload = {
|
603 |
-
"l": region,
|
604 |
-
"o": "json",
|
605 |
-
"noamp": "1",
|
606 |
-
"q": keywords,
|
607 |
-
"vqd": vqd,
|
608 |
-
"p": safesearch_base[safesearch.lower()],
|
609 |
-
"df": timelimit,
|
610 |
-
"s": 0,
|
611 |
-
}
|
612 |
-
|
613 |
-
cache = set()
|
614 |
-
for _ in range(10):
|
615 |
-
resp_content = await self._aget_url("GET", "https://duckduckgo.com/news.js", params=payload)
|
616 |
-
if resp_content is None:
|
617 |
-
return
|
618 |
-
try:
|
619 |
-
resp_json = json.loads(resp_content)
|
620 |
-
except Exception:
|
621 |
-
return
|
622 |
-
page_data = resp_json.get("results", None)
|
623 |
-
if page_data is None:
|
624 |
-
return
|
625 |
-
|
626 |
-
result_exists = False
|
627 |
-
for row in page_data:
|
628 |
-
if row["url"] not in cache:
|
629 |
-
cache.add(row["url"])
|
630 |
-
image_url = row.get("image", None)
|
631 |
-
result_exists = True
|
632 |
-
yield {
|
633 |
-
"date": datetime.fromtimestamp(row["date"], timezone.utc).isoformat(),
|
634 |
-
"title": row["title"],
|
635 |
-
"body": _normalize(row["excerpt"]),
|
636 |
-
"url": _normalize_url(row["url"]),
|
637 |
-
"image": _normalize_url(image_url) if image_url else None,
|
638 |
-
"source": row["source"],
|
639 |
-
}
|
640 |
-
if max_results and len(cache) >= max_results:
|
641 |
-
return
|
642 |
-
if max_results is None or result_exists is False:
|
643 |
-
return
|
644 |
-
next = resp_json.get("next", None)
|
645 |
-
if next is None:
|
646 |
-
return
|
647 |
-
payload["s"] = next.split("s=")[-1].split("&")[0]
|
648 |
-
|
649 |
-
async def answers(self, keywords: str) -> AsyncGenerator[Dict[str, Optional[str]], None]:
|
650 |
-
"""webscout instant answers. Query params: https://duckduckgo.com/params.
|
651 |
-
|
652 |
-
Args:
|
653 |
-
keywords: keywords for query.
|
654 |
-
|
655 |
-
Yields:
|
656 |
-
dict with instant answers results.
|
657 |
-
|
658 |
-
"""
|
659 |
-
assert keywords, "keywords is mandatory"
|
660 |
-
|
661 |
-
payload = {
|
662 |
-
"q": f"what is {keywords}",
|
663 |
-
"format": "json",
|
664 |
-
}
|
665 |
-
|
666 |
-
resp_content = await self._aget_url("GET", "https://api.duckduckgo.com/", params=payload)
|
667 |
-
if resp_content is None:
|
668 |
-
yield None
|
669 |
-
try:
|
670 |
-
page_data = json.loads(resp_content)
|
671 |
-
except Exception:
|
672 |
-
page_data = None
|
673 |
-
|
674 |
-
if page_data:
|
675 |
-
answer = page_data.get("AbstractText", None)
|
676 |
-
url = page_data.get("AbstractURL", None)
|
677 |
-
if answer:
|
678 |
-
yield {
|
679 |
-
"icon": None,
|
680 |
-
"text": answer,
|
681 |
-
"topic": None,
|
682 |
-
"url": url,
|
683 |
-
}
|
684 |
-
|
685 |
-
# related:
|
686 |
-
payload = {
|
687 |
-
"q": f"{keywords}",
|
688 |
-
"format": "json",
|
689 |
-
}
|
690 |
-
resp_content = await self._aget_url("GET", "https://api.duckduckgo.com/", params=payload)
|
691 |
-
if resp_content is None:
|
692 |
-
yield None
|
693 |
-
try:
|
694 |
-
page_data = json.loads(resp_content).get("RelatedTopics", None)
|
695 |
-
except Exception:
|
696 |
-
page_data = None
|
697 |
-
|
698 |
-
if page_data:
|
699 |
-
for row in page_data:
|
700 |
-
topic = row.get("Name", None)
|
701 |
-
if not topic:
|
702 |
-
icon = row["Icon"].get("URL", None)
|
703 |
-
yield {
|
704 |
-
"icon": f"https://duckduckgo.com{icon}" if icon else None,
|
705 |
-
"text": row["Text"],
|
706 |
-
"topic": None,
|
707 |
-
"url": row["FirstURL"],
|
708 |
-
}
|
709 |
-
else:
|
710 |
-
for subrow in row["Topics"]:
|
711 |
-
icon = subrow["Icon"].get("URL", None)
|
712 |
-
yield {
|
713 |
-
"icon": f"https://duckduckgo.com{icon}" if icon else None,
|
714 |
-
"text": subrow["Text"],
|
715 |
-
"topic": topic,
|
716 |
-
"url": subrow["FirstURL"],
|
717 |
-
}
|
718 |
-
|
719 |
-
async def suggestions(self, keywords: str, region: str = "wt-wt") -> AsyncGenerator[Dict[str, Optional[str]], None]:
|
720 |
-
"""webscout suggestions. Query params: https://duckduckgo.com/params.
|
721 |
-
|
722 |
-
Args:
|
723 |
-
keywords: keywords for query.
|
724 |
-
region: wt-wt, us-en, uk-en, ru-ru, etc. Defaults to "wt-wt".
|
725 |
-
|
726 |
-
Yields:
|
727 |
-
dict with suggestions results.
|
728 |
-
"""
|
729 |
-
assert keywords, "keywords is mandatory"
|
730 |
-
|
731 |
-
payload = {
|
732 |
-
"q": keywords,
|
733 |
-
"kl": region,
|
734 |
-
}
|
735 |
-
resp_content = await self._aget_url("GET", "https://duckduckgo.com/ac", params=payload)
|
736 |
-
if resp_content is None:
|
737 |
-
yield None
|
738 |
-
try:
|
739 |
-
page_data = json.loads(resp_content)
|
740 |
-
for r in page_data:
|
741 |
-
yield r
|
742 |
-
except Exception:
|
743 |
-
pass
|
744 |
-
|
745 |
-
async def maps(
|
746 |
-
self,
|
747 |
-
keywords: str,
|
748 |
-
place: Optional[str] = None,
|
749 |
-
street: Optional[str] = None,
|
750 |
-
city: Optional[str] = None,
|
751 |
-
county: Optional[str] = None,
|
752 |
-
state: Optional[str] = None,
|
753 |
-
country: Optional[str] = None,
|
754 |
-
postalcode: Optional[str] = None,
|
755 |
-
latitude: Optional[str] = None,
|
756 |
-
longitude: Optional[str] = None,
|
757 |
-
radius: int = 0,
|
758 |
-
max_results: Optional[int] = None,
|
759 |
-
) -> AsyncGenerator[Dict[str, Optional[str]], None]:
|
760 |
-
"""webscout maps search. Query params: https://duckduckgo.com/params.
|
761 |
-
|
762 |
-
Args:
|
763 |
-
keywords: keywords for query
|
764 |
-
place: if set, the other parameters are not used. Defaults to None.
|
765 |
-
street: house number/street. Defaults to None.
|
766 |
-
city: city of search. Defaults to None.
|
767 |
-
county: county of search. Defaults to None.
|
768 |
-
state: state of search. Defaults to None.
|
769 |
-
country: country of search. Defaults to None.
|
770 |
-
postalcode: postalcode of search. Defaults to None.
|
771 |
-
latitude: geographic coordinate (north-south position). Defaults to None.
|
772 |
-
longitude: geographic coordinate (east-west position); if latitude and
|
773 |
-
longitude are set, the other parameters are not used. Defaults to None.
|
774 |
-
radius: expand the search square by the distance in kilometers. Defaults to 0.
|
775 |
-
max_results: max number of results. If None, returns results only from the first response. Defaults to None.
|
776 |
-
|
777 |
-
Yields:
|
778 |
-
dict with maps search results
|
779 |
-
"""
|
780 |
-
assert keywords, "keywords is mandatory"
|
781 |
-
|
782 |
-
vqd = await self._aget_vqd(keywords)
|
783 |
-
|
784 |
-
# if longitude and latitude are specified, skip the request about bbox to the nominatim api
|
785 |
-
if latitude and longitude:
|
786 |
-
lat_t = Decimal(latitude.replace(",", "."))
|
787 |
-
lat_b = Decimal(latitude.replace(",", "."))
|
788 |
-
lon_l = Decimal(longitude.replace(",", "."))
|
789 |
-
lon_r = Decimal(longitude.replace(",", "."))
|
790 |
-
if radius == 0:
|
791 |
-
radius = 1
|
792 |
-
# otherwise request about bbox to nominatim api
|
793 |
-
else:
|
794 |
-
if place:
|
795 |
-
params: Dict[str, Optional[str]] = {
|
796 |
-
"q": place,
|
797 |
-
"polygon_geojson": "0",
|
798 |
-
"format": "jsonv2",
|
799 |
-
}
|
800 |
-
else:
|
801 |
-
params = {
|
802 |
-
"street": street,
|
803 |
-
"city": city,
|
804 |
-
"county": county,
|
805 |
-
"state": state,
|
806 |
-
"country": country,
|
807 |
-
"postalcode": postalcode,
|
808 |
-
"polygon_geojson": "0",
|
809 |
-
"format": "jsonv2",
|
810 |
-
}
|
811 |
-
try:
|
812 |
-
resp_content = await self._aget_url(
|
813 |
-
"GET",
|
814 |
-
"https://nominatim.openstreetmap.org/search.php",
|
815 |
-
params=params,
|
816 |
-
)
|
817 |
-
if resp_content is None:
|
818 |
-
yield None
|
819 |
-
|
820 |
-
coordinates = json.loads(resp_content)[0]["boundingbox"]
|
821 |
-
lat_t, lon_l = Decimal(coordinates[1]), Decimal(coordinates[2])
|
822 |
-
lat_b, lon_r = Decimal(coordinates[0]), Decimal(coordinates[3])
|
823 |
-
except Exception as ex:
|
824 |
-
logger.debug(f"ddg_maps() keywords={keywords} {type(ex).__name__} {ex}")
|
825 |
-
return
|
826 |
-
|
827 |
-
# if a radius is specified, expand the search square
|
828 |
-
lat_t += Decimal(radius) * Decimal(0.008983)
|
829 |
-
lat_b -= Decimal(radius) * Decimal(0.008983)
|
830 |
-
lon_l -= Decimal(radius) * Decimal(0.008983)
|
831 |
-
lon_r += Decimal(radius) * Decimal(0.008983)
|
832 |
-
logger.debug(f"bbox coordinates\n{lat_t} {lon_l}\n{lat_b} {lon_r}")
|
833 |
-
|
834 |
-
# сreate a queue of search squares (bboxes)
|
835 |
-
work_bboxes: Deque[Tuple[Decimal, Decimal, Decimal, Decimal]] = deque()
|
836 |
-
work_bboxes.append((lat_t, lon_l, lat_b, lon_r))
|
837 |
-
|
838 |
-
# bbox iterate
|
839 |
-
cache = set()
|
840 |
-
while work_bboxes:
|
841 |
-
lat_t, lon_l, lat_b, lon_r = work_bboxes.pop()
|
842 |
-
params = {
|
843 |
-
"q": keywords,
|
844 |
-
"vqd": vqd,
|
845 |
-
"tg": "maps_places",
|
846 |
-
"rt": "D",
|
847 |
-
"mkexp": "b",
|
848 |
-
"wiki_info": "1",
|
849 |
-
"is_requery": "1",
|
850 |
-
"bbox_tl": f"{lat_t},{lon_l}",
|
851 |
-
"bbox_br": f"{lat_b},{lon_r}",
|
852 |
-
"strict_bbox": "1",
|
853 |
-
}
|
854 |
-
resp_content = await self._aget_url("GET", "https://duckduckgo.com/local.js", params=params)
|
855 |
-
if resp_content is None:
|
856 |
-
return
|
857 |
-
try:
|
858 |
-
page_data = json.loads(resp_content).get("results", [])
|
859 |
-
except Exception:
|
860 |
-
return
|
861 |
-
if page_data is None:
|
862 |
-
return
|
863 |
-
|
864 |
-
for res in page_data:
|
865 |
-
result = MapsResult()
|
866 |
-
result.title = res["name"]
|
867 |
-
result.address = res["address"]
|
868 |
-
if f"{result.title} {result.address}" in cache:
|
869 |
-
continue
|
870 |
-
else:
|
871 |
-
cache.add(f"{result.title} {result.address}")
|
872 |
-
result.country_code = res["country_code"]
|
873 |
-
result.url = _normalize_url(res["website"])
|
874 |
-
result.phone = res["phone"]
|
875 |
-
result.latitude = res["coordinates"]["latitude"]
|
876 |
-
result.longitude = res["coordinates"]["longitude"]
|
877 |
-
result.source = _normalize_url(res["url"])
|
878 |
-
if res["embed"]:
|
879 |
-
result.image = res["embed"].get("image", "")
|
880 |
-
result.desc = res["embed"].get("description", "")
|
881 |
-
result.hours = res["hours"]
|
882 |
-
result.category = res["ddg_category"]
|
883 |
-
result.facebook = f"www.facebook.com/profile.php?id={x}" if (x := res["facebook_id"]) else None
|
884 |
-
result.instagram = f"https://www.instagram.com/{x}" if (x := res["instagram_id"]) else None
|
885 |
-
result.twitter = f"https://twitter.com/{x}" if (x := res["twitter_id"]) else None
|
886 |
-
yield result.__dict__
|
887 |
-
if max_results and len(cache) >= max_results:
|
888 |
-
return
|
889 |
-
if max_results is None:
|
890 |
-
return
|
891 |
-
# divide the square into 4 parts and add to the queue
|
892 |
-
if len(page_data) >= 15:
|
893 |
-
lat_middle = (lat_t + lat_b) / 2
|
894 |
-
lon_middle = (lon_l + lon_r) / 2
|
895 |
-
bbox1 = (lat_t, lon_l, lat_middle, lon_middle)
|
896 |
-
bbox2 = (lat_t, lon_middle, lat_middle, lon_r)
|
897 |
-
bbox3 = (lat_middle, lon_l, lat_b, lon_middle)
|
898 |
-
bbox4 = (lat_middle, lon_middle, lat_b, lon_r)
|
899 |
-
work_bboxes.extendleft([bbox1, bbox2, bbox3, bbox4])
|
900 |
-
|
901 |
-
async def translate(
|
902 |
-
self, keywords: str, from_: Optional[str] = None, to: str = "en"
|
903 |
-
) -> Optional[Dict[str, Optional[str]]]:
|
904 |
-
"""webscout translate.
|
905 |
-
|
906 |
-
Args:
|
907 |
-
keywords: string or a list of strings to translate
|
908 |
-
from_: translate from (defaults automatically). Defaults to None.
|
909 |
-
to: what language to translate. Defaults to "en".
|
910 |
-
|
911 |
-
Returns:
|
912 |
-
dict with translated keywords.
|
913 |
-
"""
|
914 |
-
assert keywords, "keywords is mandatory"
|
915 |
-
|
916 |
-
vqd = await self._aget_vqd("translate")
|
917 |
-
|
918 |
-
payload = {
|
919 |
-
"vqd": vqd,
|
920 |
-
"query": "translate",
|
921 |
-
"to": to,
|
922 |
-
}
|
923 |
-
if from_:
|
924 |
-
payload["from"] = from_
|
925 |
-
|
926 |
-
resp_content = await self._aget_url(
|
927 |
-
"POST",
|
928 |
-
"https://duckduckgo.com/translation.js",
|
929 |
-
params=payload,
|
930 |
-
data=keywords.encode(),
|
931 |
-
)
|
932 |
-
if resp_content is None:
|
933 |
-
return None
|
934 |
-
try:
|
935 |
-
page_data = json.loads(resp_content)
|
936 |
-
page_data["original"] = keywords
|
937 |
-
except Exception:
|
938 |
-
page_data = None
|
939 |
-
return page_data
|
940 |
-
|
941 |
-
logger = logging.getLogger("duckduckgo_search.WEBS")
|
942 |
-
nest_asyncio.apply()
|
943 |
-
|
944 |
-
|
945 |
-
class WEBS(AsyncWEBS):
|
946 |
-
def __init__(self, headers=None, proxies=None, timeout=10):
|
947 |
-
if asyncio.get_event_loop().is_running():
|
948 |
-
warnings.warn("WEBS running in an async loop. This may cause errors. Use AsyncWEBS instead.", stacklevel=2)
|
949 |
-
super().__init__(headers, proxies, timeout)
|
950 |
-
self._loop = asyncio.get_event_loop()
|
951 |
-
|
952 |
-
def __enter__(self) -> "WEBS":
|
953 |
-
return self
|
954 |
-
|
955 |
-
def __exit__(self, exc_type, exc_val, exc_tb) -> None:
|
956 |
-
self._loop.create_task(self.__aexit__(exc_type, exc_val, exc_tb))
|
957 |
-
|
958 |
-
def _iter_over_async(self, async_gen):
|
959 |
-
"""Iterate over an async generator."""
|
960 |
-
while True:
|
961 |
-
try:
|
962 |
-
yield self._loop.run_until_complete(async_gen.__anext__())
|
963 |
-
except StopAsyncIteration:
|
964 |
-
break
|
965 |
-
|
966 |
-
def text(self, *args, **kwargs) -> Generator[Dict[str, Optional[str]], None, None]:
|
967 |
-
async_gen = super().text(*args, **kwargs)
|
968 |
-
return self._iter_over_async(async_gen)
|
969 |
-
|
970 |
-
def images(self, *args, **kwargs) -> Generator[Dict[str, Optional[str]], None, None]:
|
971 |
-
async_gen = super().images(*args, **kwargs)
|
972 |
-
return self._iter_over_async(async_gen)
|
973 |
-
|
974 |
-
def videos(self, *args, **kwargs) -> Generator[Dict[str, Optional[str]], None, None]:
|
975 |
-
async_gen = super().videos(*args, **kwargs)
|
976 |
-
return self._iter_over_async(async_gen)
|
977 |
-
|
978 |
-
def news(self, *args, **kwargs) -> Generator[Dict[str, Optional[str]], None, None]:
|
979 |
-
async_gen = super().news(*args, **kwargs)
|
980 |
-
return self._iter_over_async(async_gen)
|
981 |
-
|
982 |
-
def answers(self, *args, **kwargs) -> Generator[Dict[str, Optional[str]], None, None]:
|
983 |
-
async_gen = super().answers(*args, **kwargs)
|
984 |
-
return self._iter_over_async(async_gen)
|
985 |
-
|
986 |
-
def suggestions(self, *args, **kwargs) -> Generator[Dict[str, Optional[str]], None, None]:
|
987 |
-
async_gen = super().suggestions(*args, **kwargs)
|
988 |
-
return self._iter_over_async(async_gen)
|
989 |
-
|
990 |
-
def maps(self, *args, **kwargs) -> Generator[Dict[str, Optional[str]], None, None]:
|
991 |
-
async_gen = super().maps(*args, **kwargs)
|
992 |
-
return self._iter_over_async(async_gen)
|
993 |
-
|
994 |
-
def translate(self, *args, **kwargs) -> Optional[Dict[str, Optional[str]]]:
|
995 |
-
async_coro = super().translate(*args, **kwargs)
|
996 |
-
return self._loop.run_until_complete(async_coro)
|
997 |
-
|
998 |
import g4f
|
999 |
from webscout import WEBS
|
1000 |
-
from time import time as t
|
1001 |
from flask import Flask, jsonify, request
|
|
|
1002 |
app = Flask(__name__)
|
1003 |
|
1004 |
@app.route('/search', methods=['POST'])
|
1005 |
-
def
|
1006 |
data = request.get_json()
|
1007 |
if 'query' not in data:
|
1008 |
-
return jsonify({'error': 'Query parameter missing'})
|
1009 |
|
1010 |
query = data['query']
|
1011 |
-
|
1012 |
-
|
1013 |
-
|
1014 |
-
|
1015 |
-
|
1016 |
-
|
1017 |
-
|
1018 |
-
return jsonify(responses)
|
1019 |
|
1020 |
@app.route('/health', methods=['GET'])
|
1021 |
def health_check():
|
1022 |
-
# You can add more sophisticated checks here if needed
|
1023 |
return jsonify({'status': 'ok'})
|
1024 |
-
|
1025 |
@app.route('/video', methods=['GET'])
|
1026 |
-
def
|
1027 |
params = request.args
|
1028 |
if 'keywords' not in params:
|
1029 |
-
return jsonify({'error': 'Keywords parameter missing'})
|
1030 |
|
1031 |
keywords = params['keywords']
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1032 |
|
1033 |
-
with WEBS() as WEBS:
|
1034 |
-
responses = []
|
1035 |
-
for r in WEBS.videos(
|
1036 |
-
keywords,
|
1037 |
-
region="wt-wt",
|
1038 |
-
safesearch="off",
|
1039 |
-
timelimit="w",
|
1040 |
-
resolution="high",
|
1041 |
-
duration="medium",
|
1042 |
-
max_results=10,
|
1043 |
-
):
|
1044 |
-
responses.append(r)
|
1045 |
-
return jsonify(responses)
|
1046 |
-
|
1047 |
@app.route('/img', methods=['GET'])
|
1048 |
-
def
|
1049 |
params = request.args
|
1050 |
if 'keywords' not in params:
|
1051 |
-
return jsonify({'error': 'Keywords parameter missing'})
|
1052 |
|
1053 |
keywords = params['keywords']
|
1054 |
-
|
1055 |
-
|
1056 |
-
|
1057 |
-
|
1058 |
-
|
1059 |
-
|
1060 |
-
|
1061 |
-
|
1062 |
-
|
1063 |
-
|
1064 |
-
|
1065 |
-
|
1066 |
-
|
1067 |
-
|
1068 |
-
responses.append(r)
|
1069 |
-
return jsonify(responses)
|
1070 |
|
1071 |
@app.route('/news', methods=['GET'])
|
1072 |
-
def
|
1073 |
params = request.args
|
1074 |
if 'keywords' not in params:
|
1075 |
-
return jsonify({'error': 'Keywords parameter missing'})
|
1076 |
|
1077 |
keywords = params['keywords']
|
1078 |
-
|
1079 |
-
|
1080 |
-
|
1081 |
-
|
1082 |
-
|
1083 |
-
|
1084 |
-
|
1085 |
-
|
1086 |
-
|
1087 |
-
)
|
1088 |
-
|
1089 |
-
return jsonify(responses)
|
1090 |
|
1091 |
@app.route('/int', methods=['GET'])
|
1092 |
-
def
|
1093 |
query = request.args.get('query')
|
1094 |
if not query:
|
1095 |
-
return jsonify({'error': 'Query parameter missing'})
|
1096 |
-
|
1097 |
-
#
|
1098 |
-
|
1099 |
-
|
1100 |
-
|
1101 |
-
|
1102 |
-
|
1103 |
-
responses.append(r)
|
1104 |
-
|
1105 |
-
# You can process the search results further or customize the response
|
1106 |
-
# For now, let's return the search results as-is
|
1107 |
return jsonify(responses)
|
1108 |
-
app = Flask(__name__)
|
1109 |
|
1110 |
@app.route('/translate', methods=['GET'])
|
1111 |
-
def
|
1112 |
params = request.args
|
1113 |
if 'keywords' not in params or 'to' not in params:
|
1114 |
-
return jsonify({'error': 'Keywords or target language parameter missing'})
|
1115 |
|
1116 |
keywords = params['keywords']
|
1117 |
target_language = params['to']
|
|
|
|
|
|
|
1118 |
|
1119 |
-
with WEBS() as WEBS:
|
1120 |
-
translation = WEBS.translate(keywords, to=target_language)
|
1121 |
-
return jsonify(translation)
|
1122 |
-
|
1123 |
@app.route('/chat', methods=['POST'])
|
1124 |
def chat_gpt():
|
1125 |
user_input = request.json.get('message')
|
|
|
|
|
1126 |
|
1127 |
messages = [
|
1128 |
{
|
1129 |
"role": "system",
|
1130 |
"content": "Hello! I'm your virtual assistant. How can I help you?"
|
|
|
|
|
|
|
|
|
1131 |
}
|
1132 |
]
|
1133 |
|
1134 |
-
assert user_input is not None
|
1135 |
-
|
1136 |
-
messages.append({"role": "user", "content": user_input})
|
1137 |
-
|
1138 |
response = g4f.ChatCompletion.create(
|
1139 |
model="gpt-4-32k-0613",
|
1140 |
provider=g4f.Provider.GPTalk,
|
@@ -1142,12 +135,9 @@ def chat_gpt():
|
|
1142 |
stream=True,
|
1143 |
)
|
1144 |
|
1145 |
-
ms = ""
|
1146 |
-
for message in response:
|
1147 |
-
ms += message
|
1148 |
-
|
1149 |
messages.append({"role": "assistant", "content": ms})
|
1150 |
return jsonify({"response": ms})
|
1151 |
-
if __name__ == '__main__':
|
1152 |
-
app.run()
|
1153 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
import g4f
|
2 |
from webscout import WEBS
|
|
|
3 |
from flask import Flask, jsonify, request
|
4 |
+
|
5 |
app = Flask(__name__)
|
6 |
|
7 |
@app.route('/search', methods=['POST'])
|
8 |
+
def WEBScout2_search():
|
9 |
data = request.get_json()
|
10 |
if 'query' not in data:
|
11 |
+
return jsonify({'error': 'Query parameter missing'})
|
12 |
|
13 |
query = data['query']
|
14 |
+
WEBS_instance = WEBS() # Instantiate WEBS without context manager
|
15 |
+
responses = []
|
16 |
+
for i, r in enumerate(WEBS_instance.text(query, region='wt-wt', safesearch='off', timelimit='y')):
|
17 |
+
if i == 10: # Limiting the results to 10
|
18 |
+
break
|
19 |
+
responses.append(r)
|
20 |
+
return jsonify(responses)
|
|
|
21 |
|
22 |
@app.route('/health', methods=['GET'])
|
23 |
def health_check():
|
|
|
24 |
return jsonify({'status': 'ok'})
|
25 |
+
|
26 |
@app.route('/video', methods=['GET'])
|
27 |
+
def WEBScout_videos():
|
28 |
params = request.args
|
29 |
if 'keywords' not in params:
|
30 |
+
return jsonify({'error': 'Keywords parameter missing'})
|
31 |
|
32 |
keywords = params['keywords']
|
33 |
+
WEBS_instance = WEBS() # Instantiate WEBS without context manager
|
34 |
+
responses = []
|
35 |
+
for r in WEBS_instance.videos(
|
36 |
+
keywords,
|
37 |
+
region="wt-wt",
|
38 |
+
safesearch="off",
|
39 |
+
timelimit="w",
|
40 |
+
resolution="high",
|
41 |
+
duration="medium",
|
42 |
+
max_results=10,
|
43 |
+
):
|
44 |
+
responses.append(r)
|
45 |
+
return jsonify(responses)
|
46 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
47 |
@app.route('/img', methods=['GET'])
|
48 |
+
def WEBScout2_images():
|
49 |
params = request.args
|
50 |
if 'keywords' not in params:
|
51 |
+
return jsonify({'error': 'Keywords parameter missing'})
|
52 |
|
53 |
keywords = params['keywords']
|
54 |
+
WEBS_instance = WEBS() # Instantiate WEBS without context manager
|
55 |
+
responses = []
|
56 |
+
for r in WEBS_instance.images(
|
57 |
+
keywords,
|
58 |
+
region="wt-wt",
|
59 |
+
safesearch="off",
|
60 |
+
size=None,
|
61 |
+
type_image=None,
|
62 |
+
layout=None,
|
63 |
+
license_image=None,
|
64 |
+
max_results=10,
|
65 |
+
):
|
66 |
+
responses.append(r)
|
67 |
+
return jsonify(responses)
|
|
|
|
|
68 |
|
69 |
@app.route('/news', methods=['GET'])
|
70 |
+
def WEBScout_news():
|
71 |
params = request.args
|
72 |
if 'keywords' not in params:
|
73 |
+
return jsonify({'error': 'Keywords parameter missing'})
|
74 |
|
75 |
keywords = params['keywords']
|
76 |
+
WEBS_instance = WEBS() # Instantiate WEBS without context manager
|
77 |
+
responses = []
|
78 |
+
for r in WEBS_instance.news(
|
79 |
+
keywords,
|
80 |
+
region="wt-wt",
|
81 |
+
safesearch="off",
|
82 |
+
timelimit="m",
|
83 |
+
max_results=10
|
84 |
+
):
|
85 |
+
responses.append(r)
|
86 |
+
return jsonify(responses)
|
|
|
87 |
|
88 |
@app.route('/int', methods=['GET'])
|
89 |
+
def WEBScout3_search():
|
90 |
query = request.args.get('query')
|
91 |
if not query:
|
92 |
+
return jsonify({'error': 'Query parameter missing'})
|
93 |
+
|
94 |
+
WEBS_instance = WEBS() # Instantiate WEBS without context manager
|
95 |
+
responses = []
|
96 |
+
for i, r in enumerate(WEBS_instance.text(query, region='wt-wt', safesearch='off', timelimit='y')):
|
97 |
+
if i == 2: # Limiting the results to 10
|
98 |
+
break
|
99 |
+
responses.append(r)
|
|
|
|
|
|
|
|
|
100 |
return jsonify(responses)
|
|
|
101 |
|
102 |
@app.route('/translate', methods=['GET'])
|
103 |
+
def WEBScout_translate():
|
104 |
params = request.args
|
105 |
if 'keywords' not in params or 'to' not in params:
|
106 |
+
return jsonify({'error': 'Keywords or target language parameter missing'})
|
107 |
|
108 |
keywords = params['keywords']
|
109 |
target_language = params['to']
|
110 |
+
WEBS_instance = WEBS() # Instantiate WEBS without context manager
|
111 |
+
translation = WEBS_instance.translate(keywords, to=target_language)
|
112 |
+
return jsonify(translation)
|
113 |
|
|
|
|
|
|
|
|
|
114 |
@app.route('/chat', methods=['POST'])
|
115 |
def chat_gpt():
|
116 |
user_input = request.json.get('message')
|
117 |
+
if user_input is None:
|
118 |
+
return jsonify({'error': 'Message parameter missing'})
|
119 |
|
120 |
messages = [
|
121 |
{
|
122 |
"role": "system",
|
123 |
"content": "Hello! I'm your virtual assistant. How can I help you?"
|
124 |
+
},
|
125 |
+
{
|
126 |
+
"role": "user",
|
127 |
+
"content": user_input
|
128 |
}
|
129 |
]
|
130 |
|
|
|
|
|
|
|
|
|
131 |
response = g4f.ChatCompletion.create(
|
132 |
model="gpt-4-32k-0613",
|
133 |
provider=g4f.Provider.GPTalk,
|
|
|
135 |
stream=True,
|
136 |
)
|
137 |
|
138 |
+
ms = "".join([str(message) for message in response])
|
|
|
|
|
|
|
139 |
messages.append({"role": "assistant", "content": ms})
|
140 |
return jsonify({"response": ms})
|
|
|
|
|
141 |
|
142 |
+
if __name__ == '__main__':
|
143 |
+
app.run(debug=True)
|