Community
Participate
Working Groups
JGS (6/4/01 11:58:37 AM) If I am trying to set a bp on last line of method, but miss and actually click in ruler on the closing {, the bp actually gets set on the next executable line, even though it is 3 lines down, whereas the line I wanted is only 1 line above. Shouldn't the guess at bp location be bidirectional?
Deferred
Making the search bi-directional makes it more difficult for the user to predict where a breakpoint will be placed. If the user clicks on a line that is equidistant from executable lines above and below, the result is ambiguous. Our current behavior is consistent and easy to understand. Move to close.
Breakpoint location verification has problem in general, and there are several related bugs. We need to have one bug report open (feature request) to improve breakpoint location verification.
Re-opening for 3.0. Should consider location verification while addressing external breakpoints.
*** Bug 8473 has been marked as a duplicate of this bug. ***
*** Bug 12696 has been marked as a duplicate of this bug. ***
*** Bug 15067 has been marked as a duplicate of this bug. ***
*** Bug 26268 has been marked as a duplicate of this bug. ***
*** Bug 27150 has been marked as a duplicate of this bug. ***
Fixed. Changed the way the location of the breakpoint is checked, it now uses the JDOM. The tree generated by the JDOM contains more details about the code, which help to determine if a breakpoint can be set at the location. Parsing the source to get the tree can be slow for huge files ( > 2000 lines of code), so the verification is done in a Job. A breakpoint is first created at the given location, then the location is checked. If needed, the breakpoint is moved to a valid location, or removed (with notification to the user). In the same time that a correct location is search, it resolve the name of type which contains the location, to fix bug 13834.
Jared: there is nothing special to verify for this bug, I will give you to verify the other bugs which contains test cases.
Fixed an off-by-one error in ManageBreakpointRulerAction.
Nothing special verified.