1450435 - Some URI components or Punycode domain names shouldn't have been decoded into Unicode, they should just be the original.

Status: NEW

(DevTools :: Netmonitor, defect, P3)

Description

[張俊芝(ZHANG Junzhi)]

User Agent: Mozilla/5.0 (X11; Linux x86_64; rv:58.0) Gecko/20100101 Firefox/58.0
Build ID: 20180123231643

Steps to reproduce:

This bug actually does the oppisite work of what bug 1448553 is going to do.

bug 1448553 is going to decode those human-unreadable URLs or paths, which should have been decoded  but are not decoded.

On the contrary, there are cases where URI components or Punycode domain names should not have been decoded into Unicode but are decoded, so they should be restored to the original, for example, the request URL in the Headers panel of the Netmonitor should have displayed the raw header info, it doesn't make sense to display a modified one, even if it is readable than the raw one.

Comment 1

[張俊芝(ZHANG Junzhi)]

Attachment: The filename should be raw, that is, being URI-encoded.png

Comment 2

[Jan Honza Odvarko [:Honza] (always need-info? me)]

Sounds like you'd like to have a new UI (button, link, etc.) allowing the user to switch between original/decoded URLs?
Any suggestions how the UI could look like?


Honza

Comment 3

[張俊芝(ZHANG Junzhi)]

I love the idea of having the choice of switching between original/decoded URLs, so that we don't miss the raw data while we are able to straightforwardly correspond the URLs.

Comment 4

[張俊芝(ZHANG Junzhi)]

Currently, I don't have a strong opinion on how the UI should look like. But I am interested in it.

Are there any folk who typically look into the UI issues?

Comment 5

[張俊芝(ZHANG Junzhi)]

I came up with a simple solution:

Maybe we can list both the URLs, with the decoded one embraced by brackets([, ]), if the original URL is encoded?