So, at my sense, this is a bug. Don't know if it is a gnumeric or
libxml bug, but the behavior is clearly not what it should have been.
It most certainly is a bug. Things did use to work -- when we made the
switch to libxml2, old sheets loaded just fine. There must be a change
in the way
For the record, I figured out what is going on:
* Old (1.0.x) gnumeric files are saved without an encoding.
* Non-ascii characters are embedded as #234; but the meaning is
locale dependent.
* Gnumeric can read these old files by converting such characters from
locale to UTF-8
during the load
Morten, it seems that offtraf function has some problems, it will not
give error results on certain circuits and traffic_carried values, for
example, 300 circuit and traffic_carried 230.
It yields 230.000438690186. The earliest code I can find gives the same result.
(That is revision 1.1 in
On 6/7/05, Jay H [EMAIL PROTECTED] wrote:
I think the suggestion about Bugzilla makes a point
about whether Gnumeric (and many other open source
apps) are production ready. Real users don't have
time to help debug their productivity tools. It sort
of defeats the purpose.
Well, compare to
On 6/9/05, Dave Feustel [EMAIL PROTECTED] wrote:
I set data in the File/Page Setup/Headers-Footers/customise dialogs,
but that data does not appear in the print preview or print to pdf file
output. What else do I need to do to make the customized
headers/footers work?
It works fine for me.
Nick,
you have incredibly big feet, metaphorically speaking.
Yes, Gnumeric has lots of modal dialogs, but no that is not a design flaw.
It was a deliberate and correct design decision given the choice between
1. A workable spreadsheet with lots of features and few crash-causing
bugs.
2. A
I've read Morten's blog entry on which he comments about the OASIS
OpenDocument standard, and appreciated it very much from start to end.
Thanks. (That would be http://blogs.gnome.org/view/mortenw/2005/06/16/0
for reference.)
Unfortunately, as it happens with the AbiWord filter, we
In A1, enter 2005-06-21
In A2, enter =A1+1
Dates are just numbers formatted to look like dates.
Morten
___
gnumeric-list mailing list
gnumeric-list@gnome.org
http://mail.gnome.org/mailman/listinfo/gnumeric-list
This works on sheet 1 and fails on sheet 2 in cells with
formatting identical to sheet 1. On sheet 2 I get
I get ###. On sheet 1 the value runs over the next cell
which is blank. I can get the value to display properly in
sheet 2 by reducing the font size to 4, but it's hard to read :-)
On 7/6/05, Ain Vagula [EMAIL PROTECTED] wrote:
In Excel CELL function returns, when second parameter is missing, the
attribute of cell, where cursor is positioned, in OO.o the attribute
of current cell, in Gnumeric this function does not work without
second parameter. Is this possible to
That code is probably post-1.5.2, so in 1.5.2 you can probably look for a
long time and not find it.
M.
___
gnumeric-list mailing list
gnumeric-list@gnome.org
http://mail.gnome.org/mailman/listinfo/gnumeric-list
Chances are that you are missing gtk-devel, a part of gtk+ that many
distributions package separately.
Morten
___
gnumeric-list mailing list
gnumeric-list@gnome.org
http://mail.gnome.org/mailman/listinfo/gnumeric-list
We have a hack that should make texts that are all on one line on screen
also render as one line in print.
It just did not work, i.e., it was incorrectly implemented.
I fixed it up in cvs HEAD and the next 1.5.x release should do better.
Note, that without Pango support for this we cannot fix
Fixed in cvs.
Morten
___
gnumeric-list mailing list
gnumeric-list@gnome.org
http://mail.gnome.org/mailman/listinfo/gnumeric-list
It turns out that there is an ugly, but harmless, error message on saving
to xls in 1.5.3. You will want to apply the following fix.
Morten
Index: workbook-view.c
===
RCS file: /cvs/gnome/gnumeric/src/workbook-view.c,v
retrieving
See http://bugzilla.gnome.org/show_bug.cgi?id=168875
Morten
___
gnumeric-list mailing list
gnumeric-list@gnome.org
http://mail.gnome.org/mailman/listinfo/gnumeric-list
No-one is working on this as far as I can tell, i.e., that position is open.
You would need a gtk+ 2.6 stack, or better, like the Win32 build uses.
Morten
___
gnumeric-list mailing list
gnumeric-list@gnome.org
I don't think gnumeric is involved in the panel's operation at all.
In other words, you will need to talk to whoever made the panel.
Morten
___
gnumeric-list mailing list
gnumeric-list@gnome.org
http://mail.gnome.org/mailman/listinfo/gnumeric-list
The right way to do this is distribution dependent, so you will
have to ask in that direction.
That being said, putting export LANG=en_UK.UTF-8 or
whatever at the end of your ~/.bashrc file ought to do.
Morten
___
gnumeric-list mailing list
I have released Gnumeric 1.5.5, see
http://www.gnome.org/projects/gnumeric/announcements/1.5/gnumeric-1.5.5.html
The solver should work again in this release.
Morten
___
gnumeric-list mailing list
gnumeric-list@gnome.org
The problem here is that the default .gnumeric loader loads the
entire document into memory as an xml tree.
You can try the experimental sax loader. It will not load things
like sheet objects correctly yet, though.
Morten
___
gnumeric-list mailing
Ok, sit down and take a deep breath. Make sure you sharp objects are within
reach.
Excel's numbering which Gnumeric follows incorrectly assumes that
the year 1900 was a leap year.
Evidently, OOo did not follow that crap, but that means that if you import
xls files with dates before 1-Mar-1900,
checking for g++... no
checking for c++... no
checking for gpp... gpp
checking whether we are using the GNU C++ compiler... no
checking whether gpp accepts -g... no
checking dependency style of gpp... none
checking how to run the C++ preprocessor... /lib/cpp
configure: error: C++
=SUM(races!C14:U14)
(just as Excel.)
Morten
___
gnumeric-list mailing list
gnumeric-list@gnome.org
http://mail.gnome.org/mailman/listinfo/gnumeric-list
[Warning for export when format cannot handle certain feature]
Would that be feasible?
Yes, and desireable. If would require an extra scan through the spreadsheet,
but that ought to be fast.
As an extension of this idea, one could in some circumstances work
around the problem, i.e., say this
Jody has released Gnumeric 1.6.1, a new release in the stable 1.6 series.
It requires the newly released libgoffice 0.1.2
Highlights:
* An elusive problem causing crashes when selecting File-Open, was fixed.
This hit non-English users with non-UTF-8 locales and also the Win32 build.
A new
On 11/28/05, Nicholas Lamb [EMAIL PROTECTED] wrote:
Is there any documentation on how to write function docs with the new
format? Compared to what I've seen so far it needs at least:
The test case for the new docs has so far been fn-R. We haven't sorted
about the example issues yet.
Just
AND fails because Gnumeric considers strings to be TRUE
Fixed. Gnumeric converted strings to truth values, not always
to TRUE, though. 0 would be FALSE, for example. Also
fixed the same issue for OR and XOR.
ATAN2 fails because Excel defines ATAN2(0,0) as #DIV/0! while Gnumeric
Have pivot tables been integrated into Gnumeric since the last FAQ
update?
Not yet. Some preliminare work has gone in, but nothing is connected
at the moment.
Morten
___
gnumeric-list mailing list
gnumeric-list@gnome.org
I don't think Gnumeric imposes such a limit (although the text import _preview_
might be truncated).
Various X servers have been known to set such limits. (Or just die when you
try pasting too much.)
M.
___
gnumeric-list mailing list
The problem with picking a random open (or closed, reallyl) software product
for numerical work and using it is that you need to figure out if it
is any good.
A quick look at numpy suggests that the numerical quality is questionable.
Take for example
static double atanh(double x)
{
In principle you just configure with --without-gnome
You are likely to encounter bumps on the way, though. This is not being
tested as actively as the with-gnome build.
M.
___
gnumeric-list mailing list
gnumeric-list@gnome.org
Your solution is probably to write to a temporary file and then stuff that
to the gsf level afterwards. That isn't exactly elegant, but will let you
focus on the important problem at hand.
M.
___
gnumeric-list mailing list
gnumeric-list@gnome.org
If your Gnumeric is new enough, this will work:
=cell(sheetname,B1)
M.
___
gnumeric-list mailing list
gnumeric-list@gnome.org
http://mail.gnome.org/mailman/listinfo/gnumeric-list
It's a pretty good idea and, in fact, already report as
http://bugzilla.gnome.org/show_bug.cgi?id=161291
On the other hand, no-one has actually sat down and solved the problem
yet.
Morten
___
gnumeric-list mailing list
gnumeric-list@gnome.org
There currently is no such function.
I am reluctant to include functions that access arbitrary filenames. It feels
like the first step in a security problem down the road. However, creating
such a function yourself should be trivial.
Morten
___
I don't know where you got a link to RedHat, but you can always find the
functions at
http://www.gnome.org/projects/gnumeric/functions.shtml
Depending on the details of what you want, you can use COUNTA(A1:A42)
which counts anything but empties, or something like
=SUM(0+ISTEXT(A1:A42))
OK, I see, it is quite easy to implement. Just two more questions:
- should I use the same x_i values when evaluating the smoothed ones (I
guess the answer is yes).
You can use any x value for m(x), including the x_i values and anything
in between. In fact there is no need to you splines for
While I understand the explanation I am surprised that even if I ask for
25 decimal digits it doesn't show that it is slightly smaller than 22
but if I take its difference from 22 (the integer) it shows a small
difference in the 7th or 8th digit.
The difference is about 3.55e-14. That's no
Also, after doing
the gunzip, the file is now completely empty. I guess it is completely lost.
The gunzip command I showed does not change the source
file. It creates a new. (*.xml)
Does gnumeric store a backup of the files somewhere?
It creates new file as .gsf* and and completion
There is, at present, no GUI to do what you want. The underlying support
is present, though, but you would have to enter those formats (conditional
formats) using (say) Excel.
Morten
___
gnumeric-list mailing list
gnumeric-list@gnome.org
Very interesting. Some comments:
1. Bugzilla is a far better place. Please open a bug and
attach the patch.
2. The guts of this ought to live in goffice, probably go-file.c
3. I don't think the mechanism to lock should be O_EXCL
because it does not work with NFS. Symlinks would
Looping over all cells in a area would be using sheet_foreach_cell_in_range.
Getting all the cell references could be done using using
gnm_expr_top_get_ranges.
Hope that helps,
M.
___
gnumeric-list mailing list
gnumeric-list@gnome.org
One more quite important thing:
You are testing things that should be equal and not
things that should be different.
Some spreadsheets like OO and Excel cheat in the - and
= operators: things that are mostly equal will be considered
equal, i.e., - will return exactly zero.
Such spreadsheets are
This is probably not a major thing.
Please open a bugzilla report for it, and change applix_read_cells to call
g_print (line=[%s]\n, ptr);
right inside the while loop. That should tell us what line it is complaining
about.
M.
___
gnumeric-list
This, and a few other autofilter bugs, were fixed a month or two ago in
the development tree.
Morten
___
gnumeric-list mailing list
gnumeric-list@gnome.org
http://mail.gnome.org/mailman/listinfo/gnumeric-list
Judging from the NEWS file I fixed that around 1.5.3. Current stable release
is 1.6.3
Morten
___
gnumeric-list mailing list
gnumeric-list@gnome.org
http://mail.gnome.org/mailman/listinfo/gnumeric-list
I think the way to go would be to have a gpg filter in libgsf,
not unlike the current encoding filter. That way we could build
a filter chain like
(gnumeric) - gzip - gpg - (file)
easily. The GUI would probably be the hardest part.
Currently you have to use external tools. (And forget
There is nothing fundamentally that keeps it from working, but...
The python plugin is only built if the required libraries are available at build
time. It does not appear that our regular win32 build master has that.
M.
___
gnumeric-list mailing
1.4.3 is quite old by now, but I don't think we have a final good
answer even in 1.6.x, i.e., I think we still use a bone-headed
heuristic for dates.
Please file a bug at bugzilla.gnome.org -- it is far better than
the mailing list in the sense that entries never get forgotten.
M.
Go for it. Eventually we will require Cairo, just not yet.
M.
___
gnumeric-list mailing list
gnumeric-list@gnome.org
http://mail.gnome.org/mailman/listinfo/gnumeric-list
://www.gnome.org/gnumeric/
Morten Welinder
On behalf of the Gnumeric Team
-
Gnumeric 1.7.0
PACKAGERS :
If you store supported mimetypes anywhere other than gnumeric.desktop you
will need to resync.
Andreas
I haven't actually tried it, but from just looking at it...
1. We don't want g_assert calls. It seems this should be able to fail
gracefully by just not doing anything.
2. If container_props ends up as NULL, I think you will end up calling
g_value_array_free on it. That's not allowed.
M.
is it possible to convert files using the command line?
Yes.
~/private/gnome/gnumeric/src ./ssconvert --help
Usage:
ssconvert [OPTION...] INFILE [OUTFILE]
Help Options:
-?, --help Show help options
Application Options:
-v, --version Display
Looks like you need to install libart-devel.
M.
___
gnumeric-list mailing list
gnumeric-list@gnome.org
http://mail.gnome.org/mailman/listinfo/gnumeric-list
gnumeric considers it an inaccessible file format. any possibility, pointers
to
recover data?
gunzip .gsf-... file.xml
Then start your faviourite editor on file.xml
Morten
___
gnumeric-list mailing list
gnumeric-list@gnome.org
unfortunately, this doesn't work: gunzip says not in gzip format. the
size of the file is within less than 0.1 percent of the original .xls
file, which leads me to think that it's in that format (or close to it).
but neither gnumeric nor ooffice can read it (unsupported format,
general i/o
Have a look at the excel plugin -- it knows how to ask for a password
if the file is protected.
M.
___
gnumeric-list mailing list
gnumeric-list@gnome.org
http://mail.gnome.org/mailman/listinfo/gnumeric-list
That was bug 345477.
M.
___
gnumeric-list mailing list
gnumeric-list@gnome.org
http://mail.gnome.org/mailman/listinfo/gnumeric-list
csv should pretty much work.
tsv ought to work reasonably well with 1.7.x, but if you have a case
where it does not, please file a bug.
M.
___
gnumeric-list mailing list
gnumeric-list@gnome.org
http://mail.gnome.org/mailman/listinfo/gnumeric-list
Looks like a glib header problem.
Agreed.
It would be useful if you could put up the preprocessed file somewhere.
Morten
___
gnumeric-list mailing list
gnumeric-list@gnome.org
http://mail.gnome.org/mailman/listinfo/gnumeric-list
In this simple case there is nothing that keeps you from just making a list:
A1=1, B1=1 # parameters
C1=O1 # result
H1=A1 # calculations
I1=B1
J1=H1+I1
K1=I1+J1
...
O1=M1+N1
A2=1, B2=3
C2...O2 as above
etc.
If you have something more
This effect could be achieved with validations, but it would be easier to
tell if you let us know what happens when you try to enter new values.
M.
___
gnumeric-list mailing list
gnumeric-list@gnome.org
Try something like
go_combo_box_set_relief (combo, GTK_RELIEF_NONE);
although that ought to be done already if you use the action combos.
M.
___
gnumeric-list mailing list
gnumeric-list@gnome.org
http://mail.gnome.org/mailman/listinfo/gnumeric-list
If you are using the text importer, be sure to set the format to Text for
any columns you do not want to be interpreted as numbers, dates, etc.
Morten
___
gnumeric-list mailing list
gnumeric-list@gnome.org
I agree that there ought to be a Completely No interpretation option
somewhere in the Text import dialog.
There is.
If you select Text as format for the columns in question, there will be
no interpretation.
M.
___
gnumeric-list mailing list
1) The macro/scripting capabilities within Gnumeric?
2) How to create and run macros within Gnumeric?
3) How to access individual cells within a spreadsheet (reading from and
writing to each cell)?
Nowhere. We haven't gotten to creating that yet.
You can create plugins (in C, or perhaps
See plugins/sample_datasource/ for the general principle.
Morten
___
gnumeric-list mailing list
gnumeric-list@gnome.org
http://mail.gnome.org/mailman/listinfo/gnumeric-list
Gnumeric 1.7.1 has been released. This is a development version and it requires
the brand-new libgoffice 0.3.1 and libgsf 1.14.2
New in this release is piles of little fixes all other, aimed at
improving your Gnumeric
experience. Notably:
Emmanuel, Jean: Graph improvements.
Jody, Morten:
The repository has Gnumeric 1.4.3. Does that version have scripting
support?
No. And neither does any other version, yet. However, newer versions
can be extended, function-wise, via perl or python.
M.
___
gnumeric-list mailing list
Let's see if I've got this right, you want 4,294,967,296 rows, right? What
sort of computer do you plan to run this on?
There are various places that explicitly use 32 bit integers, so this
many rows probably would not even work in principle. In practice,
going beyond a million is probably
I found some disturbing results for the CORELATION COEFFICIENT. See the
attached Test Cases, where a corelation coeff. of -1.64 was calculated.
This is definitely wrong.
What version are you using? I see the numbers 1 [exact], 0.994..., and
1 [exact]. That seems right to me.
This is with
...and for the sum I see
-1.11097e-5 -1.11106e-5 -1.11097e-5 -1.11097e-5 -1.11097e-5
Which clearly shows that we aren't perfect, but on the other hand, we do not get
zero as you seem to.
Note, that sorting is not the greatest thing to do. For example if you have,
just three numbers
small,
We have discussed summing algorithms a number of times before
(see http://bugzilla.gnome.org/show_bug.cgi?id=131588, for example)
and it is a hard problem.
Sorting works great if and if only all the numbers have the same sign.
If not, I do not currently know what is best. In fact, all the simple
I am using gnumeric a lot for large spreadsheets and have found that its
performance suffers drastically when I copy large areas (20 cols x 1000
rows) to different areas of a spreadsheet. This degradation of performance
continues and/or worsens unless I reset the copy area to a single
Gnumeric 1.7.4 aka *This* is a Release! has been released.[*]
This release fixes lots of long-standing, but mostly hard to provoke, crashes.
Thanks to Sum1 for provoking and reporting them.
Morten Welinder
on behalf of the team
[*] The ill-fated and very appropriately named version 1.7.3 aka
Can gnumeric import meta tag header info from mp3 and ogg files ?
At present, no. And I am having a hard time convincing myself that such
functionality belongs in a spreadsheet.
But it would be fairly easy to write such a plugin if you want to do so.
M.
Without more detail that is hard to tell. You do realize that
the xy type requires two series, right? One for x and one
for y.
Morten
___
gnumeric-list mailing list
gnumeric-list@gnome.org
http://mail.gnome.org/mailman/listinfo/gnumeric-list
On 1/10/07, Richard Nagle [EMAIL PROTECTED] wrote:
wondering if gnumeric has a built-in spell checker?
It does not.
Morten
___
gnumeric-list mailing list
gnumeric-list@gnome.org
http://mail.gnome.org/mailman/listinfo/gnumeric-list
can someone add aspell plugin ? or sort..
Is it doable, but will probably have to wait for a volunteer. There isn't
enough time in the day for all the things that would be cool to have.
Morten
___
gnumeric-list mailing list
gnumeric-list@gnome.org
what is the best way to automate this process?
We haven't gotten around to macros yet, so at this point the only way
to do this is to write a plugin.
Morten
___
gnumeric-list mailing list
gnumeric-list@gnome.org
How do we edit individual icons (functions) in toolbars. For example, I
want to get rid of Print, Print preview from the standard tool bar
(to open a space for Background properties).
GNOME_Gnumeric-gtk.xml
Also is there a way to decrease the size of the icons in the toolbars?
~/.gtkrc
There is a small amount of MS Excel test files at
http://www.softmaker.com/english/ofwcomp_en.htm. Obviously, they are
heavily filtered in favor of Softmaker Office.
We are aware of these. Note, that last we looked at them, the only reason that
Softmaker did well was that they cheated:
I call bullshit unless you unless you can corroborate that. ;-)
Well, not really. My evaluation copy of softmaker expired _ages_ ago.
I shouldn't complain too much, of course, as I didn't pay.
Morten
___
gnumeric-list mailing list
The finding on the k-largest element in an unordered sequence of n elements is
a linear-time and (if memory serves) constant-space problem in terms of n.
See Knuth for details.
The trouble with the algorithm is that it is rather complicated. More
code means
more chances of bugs. Unless you are
Have I reached Gnumeric List?
Yes. We're still here.
Morten
___
gnumeric-list mailing list
gnumeric-list@gnome.org
http://mail.gnome.org/mailman/listinfo/gnumeric-list
At least I missed some command to
recalculate the values of all the cells or something like that.
F9 will do that.
M.
___
gnumeric-list mailing list
gnumeric-list@gnome.org
http://mail.gnome.org/mailman/listinfo/gnumeric-list
gog-radar.c:391: error: static declaration of 'fmin' follows non-
static declaration
That was fixes ages ago. You are trying to compile a very old version
of goffice.
(Just rename the function, if you must.)
Morten
___
gnumeric-list mailing
1.6.3 is indeed the latest stable version of Gnumeric. I am not sure we have
stable vs. non-stable for goffice, but in any case, taking the latest goffice
and trying to use it for an old Gnumeric will not work.
So it is either 1.6.3 + 0.2.2 + hacking, or 1.7.9 + 0.3.8 with or
without hacking.
Free, Fast, Accurate -- Pick Any Three!
The Gnumeric team is proud to announce the release of Gnumeric 1.7.9,
a development release. This release requires the concurrently released
Goffice 0.3.8, see below.
With this release we (well, Andreas) have switched from gnome-print to
GtkPrint and in
We need some kind of rethinking of the graph properties dialog that
will probably
render your concern moot. While the dialog is powerful, it seems that people
(even people who use it often) have problems figuring out where to do what.
Morten
___
2. enter the command =linest(C16:C19, B16:B19, 1, 1)
3.CTRL+SHIFT+ENTER
What you do is the right thing and it works for me (on Linux with the latest
and greatest version).
One possible thing, though: it is possible that you need to use ; to separate
the arguments. That depends on locale.
I assume that would be version 1.7.10?
How about sharing it with us, mere Windows users?
Well, it is not as-if we provide binary for Linux either. We don't.
We have had someone do the occasional Win32 build, but none of the core
developers are in a position to produce one. Hence, they occur
It is not unlikely that our CLEAN isn't fully compatible with Excel's.
If so, it is probably because no-one ever cared. Please file a bug
at bugzilla.gnome.org with examples so we can have a look.
Morten
___
gnumeric-list mailing list
But I don't know that it's a bug or something I'm doing wrong or
That's ok. A non-trivial part of my bug reports also get closed with a
that's not a bug note.
Bugzilla is very intimidating for a non-programer
In that case the best thing to do is to put up a sample spreadsheet somewhere
and
Plugins get disabled if they fail to load. We are not sure why
plugins fail for some
people in connection with upgrades. Maybe beagle runs ssindex when things are
not fully updated.
In any case, Tools-Plugins is the menu to turn things on.
M.
___
I saw that. I could never imagine you would post a tar.gz for Windows
users... I have already unpacked it.
This is only works for geeks like me. You should really release an
installer...
Yes, of course. But Jody is still fighting with his win32-compilation setup
and a proper installer didn't
How do you reference a cell in another file under Windows? I have read the
docs and the online Manual.
...which unfortunately are out of date.
If you have two files, Book1.gnumeric and Book2.gnumeric,
in the same directory, you can enter something like
=[Book2.gnumeric]Sheet1!A2
in
If you have a column of data (nor formulas) and some of them are
blank, you can do
it the following way. Say the column with data is A.
1. In B1 enter =IF(ISEMPTY(A1),NA(),A1)
2. Copy B1 into B2...B42 or as far down as needed.
3. Select the B column and press Ctrl-C
4. Select A1 and right-click
Please file a sample of this in bugzilla so we can make sure.
Thanks,
Morten
___
gnumeric-list mailing list
gnumeric-list@gnome.org
http://mail.gnome.org/mailman/listinfo/gnumeric-list
1 - 100 of 407 matches
Mail list logo
