Archives/mercury-parser
2
0
Fork 0
mirror of https://github.com/postlight/mercury-parser synced 2024-11-17 03:25:31 +00:00
Code Issues Projects Releases Wiki Activity

feat: add option to allow custom extractors to skip default cleaners

Browse Source
This commit is contained in:
Adam Pash 2016-09-15 14:50:51 -04:00
parent 97a0728ecf
commit f0f216c7b9
2 changed files with 11 additions and 2 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

2
src/extractors/custom/wikipedia.org/index.js
View File

@ -5,6 +5,8 @@ const WikipediaExtractor = {
'#mw-content-text',
],
defaultCleaner: false,
// transform top infobox to an image with caption
transforms: {
'.infobox img': ($node) => {

11
src/extractors/root-extractor.js
View File

@ -51,7 +51,7 @@ export function select(opts) {
// contributors), return the string
if (typeof extractionOpts === 'string') return extractionOpts;
const { selectors } = extractionOpts;
const { selectors, defaultCleaner = true } = extractionOpts;
const matchingSelector = selectors.find(selector => $(selector).length === 1 && $(selector).text().trim() !== '');
@ -88,7 +88,14 @@ export function select(opts) {
// otherwise use the text of the node
result = $(matchingSelector).text();
}
return Cleaners[type](result, opts);
// Allow custom extractor to skip default cleaner
// for this type; defaults to true
if (defaultCleaner) {
return Cleaners[type](result, opts);
}
return result;
}
function extractResult(opts) {