metascraper-defuddle 5.50.6 → 5.50.7
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/package.json +3 -3
- package/src/index.js +3 -15
package/package.json
CHANGED
|
@@ -2,7 +2,7 @@
|
|
|
2
2
|
"name": "metascraper-defuddle",
|
|
3
3
|
"description": "A Defuddle connector for metascraper",
|
|
4
4
|
"homepage": "https://github.com/microlinkhq/metascraper/packages/metascraper-defuddle",
|
|
5
|
-
"version": "5.50.
|
|
5
|
+
"version": "5.50.7",
|
|
6
6
|
"types": "src/index.d.ts",
|
|
7
7
|
"main": "src/index.js",
|
|
8
8
|
"author": {
|
|
@@ -27,7 +27,7 @@
|
|
|
27
27
|
"async-memoize-one": "~1.2.1",
|
|
28
28
|
"debug-logfmt": "~1.4.12",
|
|
29
29
|
"defuddle": "~0.18.1",
|
|
30
|
-
"
|
|
30
|
+
"linkedom": "~0.18.12"
|
|
31
31
|
},
|
|
32
32
|
"devDependencies": {
|
|
33
33
|
"ava": "5"
|
|
@@ -45,5 +45,5 @@
|
|
|
45
45
|
"ava": {
|
|
46
46
|
"timeout": "15s"
|
|
47
47
|
},
|
|
48
|
-
"gitHead": "
|
|
48
|
+
"gitHead": "ff358c29107b437528930e31d76bb9d29fead3e2"
|
|
49
49
|
}
|
package/src/index.js
CHANGED
|
@@ -2,30 +2,18 @@
|
|
|
2
2
|
|
|
3
3
|
const { memoizeOne, composeRule, getHtml } = require('@metascraper/helpers')
|
|
4
4
|
const asyncMemoizeOne = require('async-memoize-one')
|
|
5
|
-
const {
|
|
5
|
+
const { parseHTML } = require('linkedom')
|
|
6
6
|
|
|
7
7
|
const debug = require('debug-logfmt')('metascraper-defuddle')
|
|
8
8
|
|
|
9
|
-
const DOCUMENT_SETTINGS = {
|
|
10
|
-
disableComputedStyleRendering: true,
|
|
11
|
-
disableCSSFileLoading: true,
|
|
12
|
-
disableIframePageLoading: true,
|
|
13
|
-
disableJavaScriptEvaluation: true,
|
|
14
|
-
disableJavaScriptFileLoading: true
|
|
15
|
-
}
|
|
16
|
-
|
|
17
9
|
const defuddleExtract = asyncMemoizeOne(async (url, html) => {
|
|
18
10
|
const { Defuddle } = await import('defuddle/node')
|
|
19
|
-
const window = new Window({ url, settings: DOCUMENT_SETTINGS })
|
|
20
|
-
window.document.documentElement.innerHTML = html
|
|
21
11
|
try {
|
|
22
|
-
const
|
|
23
|
-
return
|
|
12
|
+
const { document } = parseHTML(html)
|
|
13
|
+
return await Defuddle(document, url, { useAsync: false })
|
|
24
14
|
} catch (err) {
|
|
25
15
|
debug('Defuddle failed, fallback to next rule', { message: err.message })
|
|
26
16
|
return undefined
|
|
27
|
-
} finally {
|
|
28
|
-
await window.happyDOM.close()
|
|
29
17
|
}
|
|
30
18
|
}, memoizeOne.EqualityFirstArgument)
|
|
31
19
|
|