EVOgeek
/
Vmeda.Online
1
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity

MOBILE-3698 core: Change word count as in LMS

main
Dani Palou 2021-04-30 09:17:12 +02:00
parent af7037d5e6
commit e6d364d688
1 changed files with 22 additions and 16 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

38
src/core/services/utils/text.ts
View File

@ -304,6 +304,7 @@ export class CoreTextUtilsProvider {
/** /**
* Count words in a text. * Count words in a text.
* This function is based on Moodle's count_words.
* *
* @param text Text to count. * @param text Text to count.
* @return Number of words. * @return Number of words.
@ -312,27 +313,32 @@ export class CoreTextUtilsProvider {
if (!text || typeof text != 'string') { if (!text || typeof text != 'string') {
return 0; return 0;
} }
const blockTags = ['address', 'article', 'aside', 'blockquote', 'br', ' details', 'dialog', 'dd', 'div', 'dl', 'dt',
'fieldset', 'figcaption', 'figure', 'footer', 'form', 'h1', 'h2', 'h3', 'h4', 'h5', 'h6', 'header', 'hgroup', 'hr',
'li', 'main', 'nav', 'ol', 'p', 'pre', 'section', 'table', 'ul'];
// Clean HTML scripts and tags. // Before stripping tags, add a space after the close tag of anything that is not obviously inline.
text = text.replace(/<script[^>]*>([\S\s]*?)<\/script>/gmi, ''); // Also, br is a special case because it definitely delimits a word, but has no close tag.
// Replace block tags by space to get word count aware of line break and remove inline tags. text = text.replace(/(<\/(?!a>|b>|del>|em>|i>|ins>|s>|small>|strong>|sub>|sup>|u>)\w+>|<br>|<br\s*\/>)/ig, '$1 ');
text = text.replace(/<(\/[ ]*)?([a-zA-Z0-9]+)[^>]*>/gi, (str, p1, match) => {
if (blockTags.indexOf(match) >= 0) {
return ' ';
}
return ''; // Now remove HTML tags.
}); text = text.replace(/(<([^>]+)>)/ig, '');
// Decode HTML entities. // Decode HTML entities.
text = this.decodeHTMLEntities(text); text = this.decodeHTMLEntities(text);
// Replace underscores (which are classed as word characters) with spaces.
text = text.replace(/_/gi, ' ');
// This RegEx will detect any word change including Unicode chars. Some languages without spaces won't be counted fine. // Now, the word count is the number of blocks of characters separated
return text.match(/\S+/gi)?.length || 0; // by any sort of space. That seems to be the definition used by all other systems.
// To be precise about what is considered to separate words:
// * Anything that Unicode considers a 'Separator'
// * Anything that Unicode considers a 'Control character'
// * An em- or en- dash.
let words: string[];
try {
words = text.split(/[\p{Z}\p{Cc}—–]+/u);
} catch {
// Unicode-aware flag not supported.
words = text.split(/\s+/);
}
// Filter empty words.
return words.filter(word => word).length;
} }
/** /**