Re: CRASH BUGS

From: Tomas Frydrych (tomas@frydrych.uklinux.net)
Date: Wed Jul 31 2002 - 17:47:15 EDT

  • Next message: Hubert Figuiere: "Re: commit: footnote work"

    Please feel free to hack on this guys, I will not be able to do much
    before Friday.

    Tomas

    > So there are 2 bugs here, most likely unrelated, but definitely both
    > new. I can fix one easily myself, and the other is probably caused by
    > Tomas' isOnScreen methods and is probably some border condition not
    > properly handled.
    >
    > Bug #1 (Tomas): So, I load a 1 page makefile
    > (abiword-plugins/wp/impexp/docbook/xp/Makefile). Document's scrollbars
    > and page count don't update (they actually shrink) so the length
    > calculation is off, potentially by a huge margin depending on how
    > large your doc is. They might resize when you scroll down, thus making
    > them part of the screen, but then again, a lot of times, they don't
    > resize.
    >
    > Bug #2: Scroll down past the end of the same document
    >
    > #6 0x081aba23 in FV_View::_moveInsPtNextPrevLine (this=0x8722f58,
    > bNext=true)
    > at ../../../../../abi/src/text/fmt/xp/fp_Run.h:134
    >
    > #7 0x0819e6d6 in FV_View::warpInsPtNextPrevLine (this=0x8722f58,
    > bNext=true)
    > at ../../../../../abi/src/text/fmt/xp/fv_View.cpp:4117
    >
    > I was able to mask the bug by adding a UT_return_if_fail at
    > fv_View_protected.cpp line 1097. This doesn't seem too bad to do as a
    > sanity check at least. Real bug still needs to be fixed.
    >
    > Now, my 244 page RTF 1.6 specification MSWord document loads fine and
    > doesn't exhibit either bug. It loads much faster now than it did a few
    > days ago (28 seconds versus several minutes). Great job!
    >
    > Just FYI,
    > Dom
    >



    This archive was generated by hypermail 2.1.4 : Wed Jul 31 2002 - 17:56:23 EDT