Re: PSPP-BUG: [GUI] PSPPire closes unexpectedly when entering text with a length greater than 80 bytes (Utilities | Data File comments)

[Ben Pfaff]

On Sun, Oct 06, 2013 at 06:28:14PM +0200, John Darrington wrote:
> I don't seem to be able to reproduce this crash.
> 
> It looks as if the code you deleted was designed to move the cursor to the 
> end of the text,
> after something has been pasted.  Which is probably what the user wants.  

I don't think this is necessary, because GtkTextView maintains its
cursor as a mark.  Inserting a newline in its buffer prior to the cursor
will push along onto the next line anyway, without extra help.

> I don't know why is is crashing.  Does it fix the problem if you call  
> gtk_text_buffer_place_cursor
> only if gtk_text_iter_forward_to_line_end returns TRUE ?

The problem is that the GtkTextIter gets invalidated by the
gtk_text_buffer_insert() call.  On my system, at least, I get a
Gtk-WARNING about this:

    (psppire:13161): Gtk-WARNING **: Invalid text buffer iterator: either the 
iterator is uninitialized, or the characters/pixbufs/widgets in the buffer have 
been modified since the iterator was created.
    You must use marks, character numbers, or line numbers to preserve a 
position across buffer modifications.
    You can apply tags and insert marks without invalidating your iterators,
    but any mutation that affects 'indexable' buffer contents (contents that 
can be referred to by character offset)
    will invalidate all outstanding iterators

Running under valgrind shows a subsequent bad memory access that I
imagine to be related to the previous warning:

    Invalid read of size 4
       at 0x44C9B4F: gtk_text_iter_make_real (gtktextiter.c:202)
       by 0x44CB9CC: gtk_text_iter_forward_line (gtktextiter.c:2490)
       by 0x44CBF0F: gtk_text_iter_forward_to_line_end (gtktextiter.c:4052)
       by 0x808B902: wrap_line (comments-dialog.c:84)
       by 0x4416777: _gtk_marshal_VOID__BOXED_STRING_INT (gtkmarshalers.c:1425)
       by 0x4C12DC1: g_closure_invoke (gclosure.c:777)
       by 0x4C24C95: signal_emit_unlocked_R (gsignal.c:3621)
       by 0x4C2CB55: g_signal_emit_valist (gsignal.c:3300)
       by 0x4C2CCD2: g_signal_emit (gsignal.c:3356)
       by 0xD9E75B7: ???
     Address 0x1c is not stack'd, malloc'd or (recently) free'd


    Process terminating with default action of signal 11 (SIGSEGV): dumping core
     Access not within mapped region at address 0x1C
       at 0x44C9B4F: gtk_text_iter_make_real (gtktextiter.c:202)
       by 0x44CB9CC: gtk_text_iter_forward_line (gtktextiter.c:2490)
       by 0x44CBF0F: gtk_text_iter_forward_to_line_end (gtktextiter.c:4052)
       by 0x808B902: wrap_line (comments-dialog.c:84)
       by 0x4416777: _gtk_marshal_VOID__BOXED_STRING_INT (gtkmarshalers.c:1425)
       by 0x4C12DC1: g_closure_invoke (gclosure.c:777)
       by 0x4C24C95: signal_emit_unlocked_R (gsignal.c:3621)
       by 0x4C2CB55: g_signal_emit_valist (gsignal.c:3300)
       by 0x4C2CCD2: g_signal_emit (gsignal.c:3356)
       by 0xD9E75B7: ???

The patch I provided avoids the warning and the bad memory accesses.