1. 19 5月, 2016 1 次提交
  2. 13 5月, 2016 4 次提交
  3. 06 5月, 2016 1 次提交
  4. 04 5月, 2016 1 次提交
  5. 30 4月, 2016 1 次提交
  6. 19 4月, 2016 1 次提交
  7. 13 4月, 2016 1 次提交
  8. 09 4月, 2016 1 次提交
    • R
      Call ITextUndoHistoryRegistry.TryGetHistory · 2d41e4da
      Ravi Chande 提交于
      Completion sometimes runs in buffer that haven't been registered for undo
      with the editor, causing ITextUndoHistoryRegistry.GetHistory to throw an
      exception. Call TryGetHistory to avoid the exception.
      2d41e4da
  9. 29 3月, 2016 2 次提交
  10. 25 3月, 2016 1 次提交
  11. 15 3月, 2016 1 次提交
  12. 01 3月, 2016 1 次提交
  13. 20 2月, 2016 1 次提交
  14. 19 2月, 2016 1 次提交
  15. 18 2月, 2016 1 次提交
  16. 17 2月, 2016 1 次提交
  17. 13 2月, 2016 2 次提交
  18. 10 2月, 2016 1 次提交
  19. 09 2月, 2016 1 次提交
    • B
      Change how we format on certain commit.. · a3e699a5
      Balaji Krishnan 提交于
      ..chars from completion. We used to call the formatting service from the
      completion list with the span equal to that of the text that we
      inserted, in all cases. This may not be the right approach in all cases
      and that is manifested when we are formatting for a typed char that
      happens to be the trigger character for the formatter as well. In such
      cases we must find the right span to give to the formatter, or in other
      words, the span the formatter would have come up with if it were
      triggered first for the typed char. We do this by simply invoking the
      appropriate formatting service API and giving it the typed char and its
      position, which lets the formatting service to compute the appropriate
      span.
      We do this only when the completion's commit char is also
      a trigger for formatting. In other cases, we continue taking the
      existing code path of formatting by giving it a known span.
      
      Fixes #7768, #8228
      a3e699a5
  20. 19 1月, 2016 1 次提交
  21. 18 12月, 2015 1 次提交
  22. 15 12月, 2015 1 次提交
  23. 30 11月, 2015 2 次提交
  24. 06 11月, 2015 2 次提交
  25. 16 10月, 2015 1 次提交
  26. 12 10月, 2015 1 次提交
  27. 10 10月, 2015 2 次提交
    • J
      Harden the tests · 918b2c35
      Jared Parsons 提交于
      918b2c35
    • J
      Make foreground thread scheduling predictable · 1c480f57
      Jared Parsons 提交于
      This change makes foreground scheduling code predictable in the IDE
      tests.  Instead of using a static to hold the data we use instance
      values.  The static data the instance values are populated from are not
      100% predictable but far more so than they used to be.
      1c480f57
  28. 09 10月, 2015 1 次提交
  29. 07 10月, 2015 1 次提交
    • D
      Don't explicitly focus the editor when our ICompletionSession is dismissed · d1b7e813
      Dustin Campbell 提交于
      Years ago, we added a line of code to explicitly focus the editor when our ICompletionSession is dismissed, to handle a specific problem with VB completion. Essentially, through a subtle confluence of focus issues and events firing, VB line commit would run when double-clicking an item in the completion list to commit it. However, this line is causing numerous other problems in Visual Studio. (There's a particularly bad one when clicking VS toolbar item when the completion list is up.) After removing this line, I can no longer reproduce the original problem. So, I'm removing this to address the other problems. If it turns out that the original problem still occurs in certain circumstances, we'll work with the VS editor team to add a proper fix.
      d1b7e813
  30. 26 9月, 2015 1 次提交
  31. 15 9月, 2015 2 次提交