clean-web-scraper 4.3.2 → 4.3.4
This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.
- package/example-usage.js +4 -1
- package/package.json +1 -1
package/example-usage.js
CHANGED
|
@@ -261,7 +261,9 @@ async function bdsmovement ( enable )
|
|
|
261
261
|
"https://bdsmovement.net/contact-us",
|
|
262
262
|
"https://bdsmovement.net/taxonomy",
|
|
263
263
|
"https://bdsmovement.net/news-type",
|
|
264
|
-
"https://bdsmovement.net/cdn-cgi"
|
|
264
|
+
"https://bdsmovement.net/cdn-cgi",
|
|
265
|
+
"https://bdsmovement.net/es/",
|
|
266
|
+
"https://bdsmovement.net/ar/"
|
|
265
267
|
],
|
|
266
268
|
exactExcludeList: [
|
|
267
269
|
"https://bdsmovement.net/",
|
|
@@ -273,6 +275,7 @@ async function bdsmovement ( enable )
|
|
|
273
275
|
/^https:\/\/bdsmovement\.net\/resources\?type=\d+$/,
|
|
274
276
|
/^https:\/\/bdsmovement\.net\/news\?type=\d+$/,
|
|
275
277
|
/^https:\/\/bdsmovement\.net\/news\?campaign=\d+$/,
|
|
278
|
+
/^https:\/\/bdsmovement\.net\/news\?location=\d+$/,
|
|
276
279
|
],
|
|
277
280
|
scrapResultPath: "./dataset/bdsmovement/website",
|
|
278
281
|
jsonlOutputPath: "./dataset/bdsmovement/train.jsonl",
|