There is a patch below I would like people to look at !
On Sun, Nov 25, 2001 at 09:31:42AM +0100, Michael Schmitt wrote:
> - Load file "scary_eqns.lyx" and try to click at the last subscript "0" in a formula
>called "philinear"
> -> the cursor is positioned past the formula
can I have this file please ?
> - In the paragraph layout dialog, it is impossible to enter negative vertical spaces;
> if you enabled them again, please check whether they are displayed correctly on
>screen
> (in former lyx versions, negative values have been displayed as positive value)
fixed.
> - Loading a document with a very large table (125 rows x 5 columns) causes the
> following console message(s): "Actcell not equal to actual cell!" (document by
>request)
>
> - Juergen V.: There is still a way to provoke infinite table repaintings (at least
> theoretically). But Juergen knows how to fix them.
fixed by Juergen.
> - Insert a minipage into an empty document; insert a few characters in front of the
> (open) minipage -> the minipage is moved past the right border of the LyX main
>window
fixed by Allan.
> - Insert a minipage or footnote into an empty document; select the right border
> of the LyX main window and reduce its size (with KDE 2, the window is updated
> repetitively as you move the mouse); at some threshold, the size of the inset
> is not updated anymore such that it is painted past the right border of the
> main window (virtually, of course :-))
this is #444641 I reckon.
> - "Remove all error boxes" does not remove error boxes from within minipages
> (you can produce them by changing the document class to a class not installed)
I think I've fixed this !
> - Place 10 collapsed footnotes after each other and move the cursor between them
> -> the cursor position becomes more and more incorrect from left to right
fixed by Juergen.
> - Herbert says: insert an url with any text; close it; open it with click on
> the url button; do anything; close it -> you get a second url
can't reproduce - I think it's dead.
> - Even though _a_spell dies, the message "The _i_spell process has died" is printed
> - Cosmetics: If the spell checker dies, an error dialog is shown in which
> the OK button touches the last line of the error description
> - The spell checker ignores table cells! (#229140)
fixed by Juergen.
> - Open a new document; insert a note inset with two pars (e.g., letters "a" and "b");
> cut the pars and place them in front of the note (separate paragraph!);
> cut the note; move the cursor upwards
This now infinite loops for me with :
InsetText is 0x43257ef4
inset_owner is (nil)
> - When selecting plain text with an embedded formula, the formula is not
> highlighted
>
> - If a footnote is used inside a tabular which is 1) outside floats
> 2) not in a minipage 3) no long-tabular, the following two lines have to be
> inserted to the preamble (according to Dekel):
> \usepackage{footnote}
> \makesavenoteenv{tabular}
> Else you get a footnote symbol but no footnote text.
so .... anyone ?
> - An hfill is just drawn as a vertical bar if no text is following it. This is not
> correct for the case where there is more than one hfill in a line.
Is the rule: if there is an hfill aty the end of the row, then it doesn't count, unless
there is /another/ hfill on that line ?
try this (other people: is it right ?)
diff -u -r1.202 text.C
--- text.C 2001/11/27 15:05:32 1.202
+++ text.C 2001/11/28 22:21:31
@@ -1162,8 +1182,16 @@
return false;
// at the end of a row it does not count
- if (pos >= rowLast(row_ptr))
- return false;
+ // unless another hfill exists on the line
+ if (pos >= rowLast(row_ptr)) {
+ pos_type i = row_ptr->pos();
+ while (i < pos && !row_ptr->par()->isHfill(i)) {
+ ++i;
+ }
+ if (i == pos) {
+ return false;
+ }
+ }
// at the beginning of a row it does not count, if it is not
// the first row of a paragaph
> - Create a note inset in a large document and fill it with text until its
> content cannot be fully displayed on screen.
> Then move the cursor in front of the note, press CursorDown (you are now
> scrolling past the note), press CursorUp (you are back in front of the note),
> and press CursorLeft -> you enter the note but again the document is scrolled,
> this time totally unnecessarily. Too much scrolling is really confusing!
>
> - PageDown (only 5 lines downwards) and PageUp (movement to either the last
> or one but last line) do not work in a (large) note inset (#457384)
the first looks like a duplicate of the second here ?
> - Delete the first row of a 5x5 table (where the cells in the first two rows have
> a single letter as content) (#458165?)
>
> Freed memory access! (Backtrace of 2001/11/24)
>
> #0 Paragraph::getLayout (this=0x6f2e8fdc) at paragraph.C:2084
hmm, so it is cpar(bv) that is wrong. The LyXText's cursor has an out of date par...
> - Create a new document; insert 5x5 table; delete (first) column; try to move the
>cursor to the right (#458165)
same bug as above.
> - Add a table to an empty document; insert a few characters into the
> table cells; then move the cursor in front of the table and invoke
> "find" with a search string of your choice. When the end of the table
> is reached, the cursor is placed _in front_of the table again
> (#457396)
I wonder if we could use Dekel's par iterator now ??
> - If I start LyX in my document directory, LyX should offer the current directory
> as default location in the "Open..." dialog. But it always shows the home
> directory instead. (#451293)
This is now bug #486698 instead.
> - Add a note to an empty document; add three lines of text into the note;
> cut some of the text (in all lines); undo the operation -> note is not
> drawn correctly ("past the end of the document") (#451286)
do you just mean te background leaking a line halfway through the undos ??
john
--
"They're talking about a submanifold of space which is a 2-dimensional torus whose
cross-sectional radii
are on the order of a millimeter. Please make some minimal attempt to understand
what's being
discussed before declaring it 'drivel'."
- AC, /.
