kennethreitz/rhymepad.org
1
0
Fork 0
mirror of https://github.com/kennethreitz/rhymepad.org.git synced 2026-06-11 17:08:33 +00:00
Code Issues Packages Projects Releases Wiki Activity

Google search-page dumps extract to just the lyrics

Browse Source 
Third extractor: the knowledge panel's standalone "Lyrics" header
through the Source:/Songwriters: footer — everything after (result
snippets, videos, People-also-search-for) drops. Requires both
markers, so a poem with a "Lyrics" line passes through whole.

Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
This commit is contained in:
Kenneth Reitz
2026-06-07 12:56:17 -04:00
parent 56ebb2046b
commit ff7a6b5790
1 changed files with 20 additions and 1 deletions
Download Patch File Download Diff File Expand all files Collapse all files
static/index.html
+20 -1
View File
@@ -997,13 +997,32 @@ function extractAZBlock(t){
const block = lines.slice(start + 1, end).join('\n').trim();
return block ? block + '\n' : null;
}
function extractGoogleBlock(t){
// Google's knowledge panel: a standalone "Lyrics" header, the verse,
// then a Source:/Songwriters: footer — search-result noise follows
const lines = t.split('\n');
let start = -1;
for(let i = 0; i < lines.length; i++){
if(lines[i].trim() === 'Lyrics'){ start = i; break; }
}
if(start < 0) return null;
while(start + 1 < lines.length &&
(lines[start + 1].trim() === 'Lyrics' || !lines[start + 1].trim())) start++;
let end = -1;
for(let i = start + 1; i < lines.length; i++){
if(/^(Source:\s|Songwriters?:)/i.test(lines[i].trim())){ end = i; break; }
}
if(end < 0) return null;
const block = lines.slice(start + 1, end).join('\n').trim();
return block ? block + '\n' : null;
}
function cleanPaste(t){
t = t.replace(/\r\n?/g, '\n')
.replace(/[\u2028\u2029\u0085\u000B\u000C]/g, '\n') // Google's lyrics box
.replace(/[\u2018\u2019\u02BC]/g, "'")
.replace(/[\u201C\u201D]/g, '"')
.replace(/\u00A0/g, ' ');
const extracted = extractLyricBlock(t) || extractAZBlock(t);
const extracted = extractLyricBlock(t) || extractGoogleBlock(t) || extractAZBlock(t);
if(extracted) return extracted;
const out = [];
t.split('\n').forEach((l, i)=>{