clean-web-scraper 4.0.1 → 4.0.2
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
|
@@ -51,7 +51,8 @@ async function khameneiIrFreePalestineTag ( enable )
|
|
|
51
51
|
startURL: "https://english.khamenei.ir/page/search.xhtml?topicid=0&period=0&q=FreePalestine&pageSize=100#",
|
|
52
52
|
maxDepth: 1,
|
|
53
53
|
exactExcludeList: [
|
|
54
|
-
"https://english.khamenei.ir/page/search.xhtml?topicid=0&period=0&q=FreePalestine&pageSize=100#"
|
|
54
|
+
"https://english.khamenei.ir/page/search.xhtml?topicid=0&period=0&q=FreePalestine&pageSize=100#",
|
|
55
|
+
"https://english.khamenei.ir/page/search.xhtml?topicid=0&period=0&q=FreePalestine&pageSize=100"
|
|
55
56
|
],
|
|
56
57
|
scrapResultPath: "./dataset/khamenei-ir-free-palestine-tag/website",
|
|
57
58
|
jsonlOutputPath: "./dataset/khamenei-ir-free-palestine-tag/train.jsonl",
|
|
@@ -107,6 +108,8 @@ async function electronicintifada ( enable )
|
|
|
107
108
|
"https://electronicintifada.net/search/site/",
|
|
108
109
|
"https://electronicintifada.net/news",
|
|
109
110
|
"https://electronicintifada.net/opinion",
|
|
111
|
+
"https://electronicintifada.net/about-ei",
|
|
112
|
+
"https://electronicintifada.net/review"
|
|
110
113
|
],
|
|
111
114
|
exactExcludeList: [
|
|
112
115
|
"https://electronicintifada.net",
|