Bug 182459 - [compiler] Inconsistent error range for unresolved field
Summary: [compiler] Inconsistent error range for unresolved field
Status: VERIFIED FIXED
Alias: None
Product: JDT
Classification: Eclipse Project
Component: Core (show other bugs)
Version: 3.3   Edit
Hardware: PC Windows XP
: P3 normal (vote)
Target Milestone: 3.6 M7   Edit
Assignee: Ayushman Jain CLA
QA Contact:
URL:
Whiteboard:
Keywords:
: 182319 (view as bug list)
Depends on:
Blocks:
 
Reported: 2007-04-15 10:43 EDT by Martin Aeschlimann CLA
Modified: 2010-03-18 04:55 EDT (History)
5 users (show)

See Also:

Attachments
Proposed fix + updated regression tests (11.89 KB, patch)
2010-03-15 14:17 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 Martin Aeschlimann CLA 2007-04-15 10:43:46 EDT 
20070415

In the following code:
p/X.java -----------------
package p;
public class X {
   static int NEW_FIELD;
}

q/Y.java -----------------
package q;
import p.X;
public class Y {
  void foo() {
     int i = X.OLD_FIELD;
  }
  void bar() {
     int j = X.OLD_FIELD;
  }
}

'OLD_FIELD' can not be resolved. The error range is 'X.OLD_FIELD', but should be only 'OLD_FIELD'.

If you change to
public class Y {
  void foo() {
     int i = X.OLD_FIELD();
  }
  void bar() {
     int j = X.OLD_FIELD();
  }
}
then the error range is only 'OLD_FIELD()'
Comment 1 Ayushman Jain CLA 2010-03-09 06:51:35 EST 
The error message I get is X.OLD_FIELD cannot be resolved to a variable, instead of OLD_FIELD cannot be resolved, as reported here. In view of this error message, I believe that the error range is consistent.

Srikanth, what do you think? Does the error range need any change here?
Comment 2 Srikanth Sankaran CLA 2010-03-10 02:24:34 EST 
(In reply to comment #1)
> The error message I get is X.OLD_FIELD cannot be resolved to a variable,
> instead of OLD_FIELD cannot be resolved, as reported here. In view of this
> error message, I believe that the error range is consistent.
> 
> Srikanth, what do you think? Does the error range need any change here?

This seems to be have been the behavior at least since 3.3.2.
I agree that current message and range are appropriate.
This defect could be closed with no change.
Comment 3 Srikanth Sankaran CLA 2010-03-10 02:25:11 EST 
Verified for 3.6M6.
Comment 4 Dani Megert CLA 2010-03-15 11:49:51 EDT 
Reopening, given we just fixed bug 303830: we should align this one with it and it will also help us fix bug 182319.
Comment 5 Olivier Thomann CLA 2010-03-15 14:17:52 EDT 
Created attachment 162094 [details]
Proposed fix + updated regression tests
Comment 6 Olivier Thomann CLA 2010-03-15 14:19:26 EDT 
Ayushman, please review.
Comment 7 Frederic Fusier CLA 2010-03-16 10:34:35 EDT 
(In reply to comment #5)
> Created an attachment (id=162094) [details]
> Proposed fix + updated regression tests

Patch looks good to me.
Comment 8 Ayushman Jain CLA 2010-03-17 02:51:41 EDT 
(In reply to comment #7)
> Patch looks good to me.

Yup, I agree with Frederic.
Comment 9 Olivier Thomann CLA 2010-03-17 12:13:32 EDT 
Released for 3.6M7.
Updated existing regression tests.
Comment 10 Dani Megert CLA 2010-03-18 04:55:24 EDT 
Verified in HEAD.
Comment 11 Dani Megert CLA 2010-03-18 04:55:48 EDT 
*** Bug 182319 has been marked as a duplicate of this bug. ***