295870 - EM does not reject extensions which are using applications GUID's as own GUID's

Status: RESOLVED WONTFIX

(Toolkit :: Add-ons Manager, defect)

Description

[Przemyslaw Bialik]

User-Agent:       Mozilla/5.0 (Windows; U; Windows NT 5.0; en-US; rv:1.8b2) Gecko/20050527 Firefox/1.0+
Build Identifier: Mozilla/5.0 (Windows; U; Windows NT 5.0; en-US; rv:1.8b2) Gecko/20050527 Firefox/1.0+

I saw at least one case when author used FF GUID for his extension ID (either by
a mistake or lack of knowledge) and imagine what would happen if user would
install 2 such 'buggy' extensions - the first one would be upgraded to the
second one...
So EM should reject extensions which GUID's are the same as GUID's which belong
to the one of mozilla applications (extra: nvu, netscape, 'new' seamonkey...).

Reproducible: Always

Steps to Reproduce:
Modify extension's <em:id> in install.rdf to match e.g. FF GUID
({ec8030f7-c20a-464f-9b0e-13a3a9e97384}) and install it
Actual Results:  
Extension does install.

Expected Results:  
Extension does not install and the alert is showed to the user, sth like:
"Firefox could not install <em:name> because its ID (<em:id>) is used by
Firefox. Please contact the author about this problem."

Comment 1

[Przemyslaw Bialik]

There is already Bug 273550 but it's against UMO.

Comment 2

[Nickolay_Ponomarev]

no duplicates found, confirming.

Comment 3

[Robert Strong (they/them - no direct email)]

Could you please check if there are any errors experienced with this scenario
and provide them in a comment?

Comment 4

[Przemyslaw Bialik]

No errors experienced, EM just treated second extension as updated version of
the first one (tested on same build as before).
The 'error' would only happen when EM would not reject such GUID's and the user
would want to install >1 extensions wich are using same application GUID as own
(low probability but it is).

Comment 5

[Nickolay_Ponomarev]

The point of the bug, as I understood it, is to help authors which do not
understand, what things in install.rdf mean well enough, still create correct
extensions that don't conflict with others. I confirmed the bug as a valid
request without duplicates, but this doesn't mean I also think this should be
implemented. 

In my opinion, this needs to be implemented, only if it's more or less common
mistake among new extensions. Otherwise some kind of workaround for bug 132183
(display error for such extensions) would be a better thing to work at, that one
DOES cause problems to many begginers.

Comment 6

[Robert Strong (they/them - no direct email)]

Not going to add code to the EM for this check especially since there are no significant ill affects when this happens.