fix: RichTextBox LinkClicked event for "friendly named" hyperlinks
#1745
Conversation
The newer RichEdit controls (v4.1+) in specific cases had troubles detecting "friendly named" hyperlinks as the `Text` property was reportig less characters than the rage of the actual hyperlink. Fixes dotnet#1631 Relates to dotnet#1139
|
The original code in .NET Framework was presumably put in place in response to "VSWhidbey 504502" bug, but neither me nor @Olina-Zhang were able to track it down. |
RussKie
added
the
waiting-for-testing
Codecov Report
@@ Coverage Diff @@
## release/3.0 #1745 +/- ##
=====================================================
- Coverage 25.97551% 25.92135% -0.05416%
=====================================================
Files 804 804
Lines 267706 267706
Branches 37950 37950
=====================================================
- Hits 69538 69393 -145
- Misses 193242 193381 +139
- Partials 4926 4932 +6
|
|
|
||
| if (c.cpMax > Text.Length || c.cpMax - c.cpMin <= 0) | ||
| Debug.Assert((c.cpMax - c.cpMin) > 0, "CHARRANGE was null or negative - can't do it!"); | ||
| if (c.cpMax - c.cpMin <= 0) |
There was a problem hiding this comment.
Yes, it removes a workaround which was causing problems. its not known what it was working around (some comment refers to a bug id which couldn't be tracked down yet)
a long explanation can be found on stackoverflow
The point is that Text.Length can be the wrong value to compare against in presence of "friendly name links", a feature added by upgrading the RTF control. The workaround existed before the control upgrade but was never updated. Since its not know what its working around removing it makes the most sense.
There was a problem hiding this comment.
I see, so if there is a friendly name link, then Text.Length will be longer than is technically allowed (greater than cpMax), but since it is still visually allowed, this appears as a bug.
It seems odd to me that we would ever have opted to make the string empty is this circumstance. Truncation would be more appropriate, but anyways... fix looks good.
Is there any reason we would want to maintain this behavior? I cannot think of anything off the top of my head.
RussKie
added
waiting-approval
and removed
waiting-for-testing
|
Signed off by the CTI |
|
approved by tactics |
ghost
locked as resolved and limited conversation to collaborators
Fixes #1631
Relates to #1139
Proposed changes
The newer RichEdit controls (v4.1+) in specific cases had troubles detecting "friendly named" hyperlinks as the
Textproperty was reportig less characters than the rage of the actual hyperlink.Customer Impact
LinkClickedevents in RichEdit v4.1+.Regression?
Risk
Test methodology
Test environment(s)
Microsoft Reviewers: Open in CodeFlow