Bug 106478 - [compiler] warning on unary + operator
Summary: [compiler] warning on unary + operator
Status: VERIFIED WONTFIX
Alias: None
Product: JDT
Classification: Eclipse Project
Component: Core (show other bugs)
Version: 3.1   Edit
Hardware: PC Windows XP
: P5 enhancement (vote)
Target Milestone: 3.6 M3   Edit
Assignee: Olivier Thomann CLA
QA Contact:
URL:
Whiteboard:
Keywords:
Depends on:
Blocks: 291359
  Show dependency tree
 
Reported: 2005-08-09 10:36 EDT by Adam Kiezun CLA
Modified: 2009-10-27 07:21 EDT (History)
3 users (show)

See Also:

Attachments
Proposed fix + regression tests (16.82 KB, patch)
2009-10-02 11:43 EDT, Olivier Thomann CLA 		no flags Details | Diff
View All Add an attachment (proposed patch, testcase, etc.)

Note You need to log in before you can comment on or make changes to this bug.
Description Adam Kiezun CLA 2005-08-09 10:36:39 EDT 
3.1
to prevent bugs like:

int i= ...
i =+ j; //most likely a BUG - should be i += j

eclipse could warn on usages of the unary + operator (or just warn about =+)
Comment 1 Olivier Thomann CLA 2006-10-06 16:55:06 EDT 
Should be trivial to add.
Comment 2 Philipe Mulet CLA 2006-10-08 10:15:41 EDT 
How frequent is it ? i.e. compiler job or some static analysis engine ?
Comment 3 Olivier Thomann CLA 2009-10-02 11:43:48 EDT 
Created attachment 148659 [details]
Proposed fix + regression tests
Comment 4 Dani Megert CLA 2009-10-06 05:35:07 EDT 
>Proposed fix + regression tests
It looks like this is already in HEAD. Please revert it. This diagnose doesn't make much sense and code like:

i=  0;
j= +1;
k= -1;

is most likely intentional. If at all, then we should only warn for "=+<spaces>" and "=-<spaces>" patterns but that's overkill and would be strange as it depends on formatting. I checked all the source in my big workspace and there's not a single occurrence of that pattern.

This is a WONTFIX.
Comment 5 Markus Keller CLA 2009-10-06 05:41:58 EDT 
(In reply to comment #4)
I agree, we should no bloat the list of options with this.
Comment 6 Olivier Thomann CLA 2009-10-06 08:48:14 EDT 
I'll revert the patch.
Comment 7 Olivier Thomann CLA 2009-10-06 09:22:10 EDT 
(In reply to comment #4)
> >Proposed fix + regression tests
> It looks like this is already in HEAD. Please revert it. This diagnose doesn't
> make much sense and code like:
> 
> i=  0;
> j= +1;
> k= -1;
If you write code like this, then you simply keep the warning off.
The warning could have been useful for cases like in comment 0.
Anyway I'll revert the fix.
Comment 8 Jay Arthanareeswaran CLA 2009-10-26 23:33:25 EDT 
Verified for 3.6M3