clean-web-scraper 4.3.4 → 4.3.5
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 -2
- package/package.json +1 -1
package/example-usage.js
CHANGED
|
@@ -263,13 +263,15 @@ async function bdsmovement ( enable )
|
|
|
263
263
|
"https://bdsmovement.net/news-type",
|
|
264
264
|
"https://bdsmovement.net/cdn-cgi",
|
|
265
265
|
"https://bdsmovement.net/es/",
|
|
266
|
-
"https://bdsmovement.net/ar/"
|
|
266
|
+
"https://bdsmovement.net/ar/",
|
|
267
|
+
"https://bdsmovement.net/resource-type/",
|
|
267
268
|
],
|
|
268
269
|
exactExcludeList: [
|
|
269
270
|
"https://bdsmovement.net/",
|
|
270
271
|
"https://bdsmovement.net/shutdownnation",
|
|
271
272
|
"https://bdsmovement.net/campaigns",
|
|
272
273
|
"https://bdsmovement.net/resources",
|
|
274
|
+
"https://bdsmovement.net/news",
|
|
273
275
|
/^https:\/\/bdsmovement\.net\/resources\?page=\d+$/,
|
|
274
276
|
/^https:\/\/bdsmovement\.net\/resources\?campaign=\d+$/,
|
|
275
277
|
/^https:\/\/bdsmovement\.net\/resources\?type=\d+$/,
|
|
@@ -285,7 +287,7 @@ async function bdsmovement ( enable )
|
|
|
285
287
|
metadataFields: ["author", "articleTitle", "pageTitle", "description", "dataScrapedDate", "url"],
|
|
286
288
|
maxArticles: 2000,
|
|
287
289
|
maxDepth: 16,
|
|
288
|
-
batchSize:
|
|
290
|
+
batchSize: 100,
|
|
289
291
|
axiosHeaders: headers,
|
|
290
292
|
axiosMaxRetries: 2,
|
|
291
293
|
axiosRetryDelay: 8000,
|