Skip to content

chore: do not deobfuscate known top level domains - #1369

Merged
skylot merged 1 commit into
skylot:masterfrom
jpstotz:tld
Feb 6, 2022
Merged

skylot merged 1 commit into
skylot:masterfrom
jpstotz:tld

Conversation

@jpstotz

@jpstotz jpstotz commented Feb 6, 2022

Copy link
Copy Markdown
Collaborator

Do not deobfuscate known top level domains with 2 or 3 characters:

I noticed that common packages like ch.qos.logback, de.greenrobot, io.jsonwebtoken... are always renamed because of length of their first level package name. As they are known top level domains I added a check to exclude all current tlds that have 2 or 3 characters.

@skylot

skylot commented Feb 6, 2022

Copy link
Copy Markdown
Owner

@jpstotz great idea!
One thing: I think it will be better to init that list only once, so please move that parsing to a separate class and to private static final field with a static method for check. Also add a small doc to that class, because tld is a very short name 😄

@jpstotz

jpstotz commented Feb 6, 2022

Copy link
Copy Markdown
Collaborator Author

@skylot As deobfuscation is optional and only used in a short period while loading I thought it is not necessary to keep the tld list in memory all the time. But as the list is very short it does not really make a difference, so I have adapted my PR.

@skylot

skylot commented Feb 6, 2022

Copy link
Copy Markdown
Owner

@jpstotz thanks!
Actually, class will not be loaded until first use, also in jadx-gui full reload happen on every settings change, so improve performance a little is always nice.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

2 participants