kavin/NewPipeExtractor
1
0
Fork 0
mirror of https://github.com/TeamNewPipe/NewPipeExtractor.git synced 2024-12-12 21:30:33 +05:30
Code Issues Projects Releases Wiki Activity

[YouTube] Fix JavaScript n parameter decoding function name extraction

Browse Source 
This commits fixes extraction of the function name decoding the n parameter for
HTML5 clients' streaming URLs for YouTube base JavaScript player 3400486c.

Two new regexes have been added to the existing ones. All regexes and what they
extract has been documented.
This commit is contained in:
AudricV 2024-07-23 19:01:58 +02:00
parent 480f5e223e
commit 22f818109f
No known key found for this signature in database
GPG Key ID: DA92EC7905614198
1 changed files with 39 additions and 9 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

48
extractor/src/main/java/org/schabi/newpipe/extractor/services/youtube/YoutubeThrottlingParameterUtils.java
View File

@ -26,24 +26,54 @@ final class YoutubeThrottlingParameterUtils {
private static final String ARRAY_ACCESS_REGEX = "\\[(\\d+)]";
/**
* The first regex matches this, where we want BDa:
* <p>
* (b=String.fromCharCode(110),c=a.get(b))&&(c=<strong>BDa</strong><strong>[0]</strong>(c)
* <p>
* Array access is optional, but needs to be handled, since the actual function is inside the
* array.
*/
// CHECKSTYLE:OFF
private static final Pattern[] DEOBFUSCATION_FUNCTION_NAME_REGEXES = {
/*
* The first regex matches the following text, where we want rDa and the array index
* accessed:
*
* a.D&&(b="nn"[+a.D],c=a.get(b))&&(c=rDa[0](c),a.set(b,c),rDa.length||rma("")
*/
Pattern.compile(SINGLE_CHAR_VARIABLE_REGEX + "+=\"nn\"\\[\\+"
+ SINGLE_CHAR_VARIABLE_REGEX + "+\\." + SINGLE_CHAR_VARIABLE_REGEX + "+],"
+ SINGLE_CHAR_VARIABLE_REGEX + "+=" + SINGLE_CHAR_VARIABLE_REGEX
+ "+\\.get\\(" + SINGLE_CHAR_VARIABLE_REGEX + "+\\)\\)&&\\("
+ SINGLE_CHAR_VARIABLE_REGEX + "+=(" + SINGLE_CHAR_VARIABLE_REGEX
+ "+)\\[(\\d+)]"),
/*
* The second regex matches the following text, where we want rma:
*
* a.D&&(b="nn"[+a.D],c=a.get(b))&&(c=rDa[0](c),a.set(b,c),rDa.length||rma("")
*/
Pattern.compile(SINGLE_CHAR_VARIABLE_REGEX + "+=\"nn\"\\[\\+"
+ SINGLE_CHAR_VARIABLE_REGEX + "+\\." + SINGLE_CHAR_VARIABLE_REGEX + "+],"
+ SINGLE_CHAR_VARIABLE_REGEX + "+=" + SINGLE_CHAR_VARIABLE_REGEX + "+\\.get\\("
+ SINGLE_CHAR_VARIABLE_REGEX + "+\\)\\).+\\|\\|(" + SINGLE_CHAR_VARIABLE_REGEX
+ "+)\\(\"\"\\)"),
/*
* The third regex matches the following text, where we want BDa and the array index
* accessed:
*
* (b=String.fromCharCode(110),c=a.get(b))&&(c=BDa[0](c)
*/
Pattern.compile("\\(" + SINGLE_CHAR_VARIABLE_REGEX + "=String\\.fromCharCode\\(110\\),"
+ SINGLE_CHAR_VARIABLE_REGEX + "=" + SINGLE_CHAR_VARIABLE_REGEX + "\\.get\\("
+ SINGLE_CHAR_VARIABLE_REGEX + "\\)\\)" + "&&\\(" + SINGLE_CHAR_VARIABLE_REGEX
+ "=(" + FUNCTION_NAME_REGEX + ")" + "(?:" + ARRAY_ACCESS_REGEX + ")?\\("
+ SINGLE_CHAR_VARIABLE_REGEX + "\\)"),
/*
* The fourth regex matches the following text, where we want Yva and the array index
* accessed:
*
* .get("n"))&&(b=Yva[0](b)
*/
Pattern.compile("\\.get\\(\"n\"\\)\\)&&\\(" + SINGLE_CHAR_VARIABLE_REGEX
+ "=(" + FUNCTION_NAME_REGEX + ")(?:" + ARRAY_ACCESS_REGEX + ")?\\("
+ SINGLE_CHAR_VARIABLE_REGEX + "\\)"),
+ SINGLE_CHAR_VARIABLE_REGEX + "\\)")
};
// CHECKSTYLE:ON