3:21 p.m.
Being the "weird guy" I am :-), I found some small bugs in a few corner cases:
1)- Please refer to the attached ods file. When I open this data and
export as csv, the y series is written with 12 significant digits
instead of 3.
2)- After ods imports, the "dataset info" is blank. The same window
says "Data imported from Excel file..." (I have'nt tried other import
types.)
3)- Selecting "save as" and pointing to an existing gdt file, gretl
overwrites without warning. Overwriting csv, R, ... opens a
confirmation window.
4)- I think gdt files should not be in the list for export file types
since saving a gdt is technically not exporting.
5)- Adding 4 correlograms to the graph page will not fit in the
portrait paper. Adding 3 creates 3 pages, 2 of which are blank.
6)- Could you please, please add the "Reformat..." function (in the
display data window) to the summary statistics window also? This is
not only useful but also important from a software reliability point
of view.
Cheers
Talha
--
“An expert is a person who has made all the mistakes that can be made
in a very narrow field.” - Niels Bohr (1885-1962)
--
4:44 p.m.
On Wed, 21 Mar 2012, Talha Yalta wrote:
Being the "weird guy" I am :-), I found some small bugs in
a
few corner cases:
Here's a first round of replies, I'll get to others before
long:
1)- Please refer to the attached ods file. When I open this data and
export as csv, the y series is written with 12 significant digits
instead of 3.
That's OK. The data are stored to at least 12 digits in the
ODS file, as you can see if you adjust the display format. We
don't pay attention to the fact that they are _formatted_ to
show only 3 digits.
2)- After ods imports, the "dataset info" is blank. The
same window
says "Data imported from Excel file..." (I have'nt tried other import
types.)
The ODS importer was missing that recording device. I've now
added it.
3)- Selecting "save as" and pointing to an existing gdt
file, gretl
overwrites without warning. Overwriting csv, R, ... opens a
confirmation window.
OK, "Save as" now gives an overwrite warning (unless the
chosen filename is the same as that of the currently loaded
datafile).
4)- I think gdt files should not be in the list for export file
types
since saving a gdt is technically not exporting.
I see your point, but the way things are right now gives you
an extra degree of freedom: if you "save as" you just get a
choice of filename, while if you "export" as gdt you get to
select the particular series you want to export.
Allin
9:40 p.m.
On Wed, 21 Mar 2012, Talha Yalta wrote:
Being the "weird guy" I am :-), I found some small bugs in
a few
corner cases:
Part II of my replies
5)- Adding 4 correlograms to the graph page will not fit in the
portrait paper. Adding 3 creates 3 pages, 2 of which are blank.
I can't replicate that, either using US letter-size paper or A4. It
may be something to do with incompatible versions of LaTeX's
geometry.sty (grrr). If you could save the model table as TeX, run
pdflatex on it, and send me the PDF and the .log file, that might be
helpful.
6)- Could you please, please add the "Reformat..." function
(in the
display data window) to the summary statistics window also?
OK, that's now done for the single-series summary statistics
window. It's more complicated for the multiple-series stats window,
given the existing (internationalized) formatting.
Allin
6:52 a.m.
Thanks very much Prof. Cottrell for the rapid fixes.
That's OK. The data are stored to at least 12 digits in the
ODS file, as you can see if you adjust the display format. We
don't pay attention to the fact that they are _formatted_ to
show only 3 digits.
I am confused. Being aware of the above, I had copied these
series
after first using the "=ROUND(x,3) command on the originals. In my
system, in libreoffice, I see the numbers with 3 decimal places even
if I reformat the cell to show 20 digits. Maybe this is a libreoffice
problem? But then again saving from libreoffice creates a correct csv
file.
I can't replicate that, either using US letter-size paper or A4.
It
may be something to do with incompatible versions of LaTeX's
geometry.sty (grrr). If you could save the model table as TeX, run
pdflatex on it, and send me the PDF and the .log file, that might be
helpful.
Saving the modeltable and running pdflatex created an empty page but
running latex replicated the problem. Please see the attached
test.tar.gz file.
OK, that's now done for the single-series summary statistics
window. It's more complicated for the multiple-series stats window,
given the existing (internationalized) formatting.
Thanks very much. Of course this
would be really useful if it worked
for multiple series. I think this basic feature needs to be eventually
implemented (hopefully before 2.0) but I also know there are other
things. Meanwhile could you add this to the correlation matrix window
also?
Finally, the tab labels for new scripts show "Untitled" although I
remember translating an "Untitled" string. Is this something else that
will be marked for translation later?
Sincerely
Talha
--
“An expert is a person who has made all the mistakes that can be made
in a very narrow field.” - Niels Bohr (1885-1962)
--
8:56 p.m.
On Thu, 22 Mar 2012, Talha Yalta wrote:
Thanks very much Prof. Cottrell for the rapid fixes.
> That's OK. The data are stored to at least 12 digits in the
> ODS file, as you can see if you adjust the display format. We
> don't pay attention to the fact that they are _formatted_ to
> show only 3 digits.
I am confused. Being aware of the above, I had copied these series
after first using the "=ROUND(x,3) command on the originals. In my
system, in libreoffice, I see the numbers with 3 decimal places even
if I reformat the cell to show 20 digits.
Ah, you're right, the numbers are stored to only 3 digits in
the ods file. I was looking at the file using gnumeric, which
showed extra stuff when I reformatted the cells. But the real
point is that the "extra stuff" is in fact there in memory.
For example, the number 30976.746 cannot be represented in a C
double: the closest you get is 30976.745999999999185.
LibreOffice is just sanitizing what you're allowed to see
based on your "ROUND(x,3)" instruction, it's not magically
producing a floating point number with an infinite number of
trailing zeros.
That said, I've slightly modified gretl's CSV printing
apparatus so it's a bit less inclined to print trailing junk.
Thanks for the correlogram graph page test case. This should
now be OK in CVS and snapshots. The problem was specific to
the case where EPS rather than PDF graphs are produced.
Finally, the tab labels for new scripts show "Untitled"
although I
remember translating an "Untitled" string. Is this something else that
will be marked for translation later?
It's now marked for translation.
Allin
6:11 a.m.
Thanks very much for these. I have noticed a few small things:
1)- The reformat button in the "display data" window does not work anymore.
2)- I think it would be much better if opened scripts are added to the
tabbed editor instead of opening in a separate window each time.
3)- Here is an interesting one: In the model window, click on say
"Latex" and, without selecting a menu item, move the pointer to other
menu titles. When you come to "Windows" the title is not highlighted
and you cannot move to to other titles anymore.
4)- This is actually a bit related to the parallel discussion on
automatically created series. In a scatter plot with a regression line
with inverse fit, choosing "ols estimates" creates a new i_x series.
Now this series is shown in the model specification window and can be
used for later regressions (which I like in this particular case),
however, it is never shown in the main window.
Best regards
Talha
On Thu, Mar 22, 2012 at 10:56 PM, Allin Cottrell <cottrell(a)wfu.edu> wrote:
On Thu, 22 Mar 2012, Talha Yalta wrote:
> Thanks very much Prof. Cottrell for the rapid fixes.
>
>> That's OK. The data are stored to at least 12 digits in the
>> ODS file, as you can see if you adjust the display format. We
>> don't pay attention to the fact that they are _formatted_ to
>> show only 3 digits.
> I am confused. Being aware of the above, I had copied these series
> after first using the "=ROUND(x,3) command on the originals. In my
> system, in libreoffice, I see the numbers with 3 decimal places even
> if I reformat the cell to show 20 digits.
Ah, you're right, the numbers are stored to only 3 digits in
the ods file. I was looking at the file using gnumeric, which
showed extra stuff when I reformatted the cells. But the real
point is that the "extra stuff" is in fact there in memory.
For example, the number 30976.746 cannot be represented in a C
double: the closest you get is 30976.745999999999185.
LibreOffice is just sanitizing what you're allowed to see
based on your "ROUND(x,3)" instruction, it's not magically
producing a floating point number with an infinite number of
trailing zeros.
That said, I've slightly modified gretl's CSV printing
apparatus so it's a bit less inclined to print trailing junk.
Thanks for the correlogram graph page test case. This should
now be OK in CVS and snapshots. The problem was specific to
the case where EPS rather than PDF graphs are produced.
> Finally, the tab labels for new scripts show "Untitled" although I
> remember translating an "Untitled" string. Is this something else that
> will be marked for translation later?
It's now marked for translation.
Allin
_______________________________________________
Gretl-devel mailing list
Gretl-devel(a)lists.wfu.edu
http://lists.wfu.edu/mailman/listinfo/gretl-devel
--
“An expert is a person who has made all the mistakes that can be made
in a very narrow field.” - Niels Bohr (1885-1962)
--
3:08 p.m.
On Fri, 23 Mar 2012, Talha Yalta wrote:
1)- The reformat button in the "display data" window does
not work anymore.
(For more than one series.) Now fixed in CVS.
2)- I think it would be much better if opened scripts are added to
the
tabbed editor instead of opening in a separate window each time.
What do you mean by "opened scripts"? Read-only practice
scripts? I'm not sure they belong in the editor, since you're
not editing them.
3)- Here is an interesting one: In the model window, click on say
"Latex" and, without selecting a menu item, move the pointer to other
menu titles. When you come to "Windows" the title is not highlighted
and you cannot move to to other titles anymore.
The "Windows" menu item produces a popup; click away from it
to make it go away. It may be possible to convert this to a
regular drop-down menu, I'm not sure.
4)- This is actually a bit related to the parallel discussion on
automatically created series. In a scatter plot with a regression line
with inverse fit, choosing "ols estimates" creates a new i_x series.
Now this series is shown in the model specification window and can be
used for later regressions (which I like in this particular case),
however, it is never shown in the main window.
It's just that the main window is not auto-refreshing to show
it.
Allin
3:49 p.m.
On Fri, Mar 23, 2012 at 5:08 PM, Allin Cottrell <cottrell(a)wfu.edu> wrote:
What do you mean by "opened scripts"? Read-only practice
scripts? I'm not sure they belong in the editor, since you're
not editing them.
But one can start modifying them after save as. IMHO it would be
best
if these opened in the regular tabbed window with editing disabled.
The "Windows" menu item produces a popup; click away from
it
to make it go away. It may be possible to convert this to a
regular drop-down menu, I'm not sure.
I see. The behavior seemed like a bug to
me.
It's just that the main window is not auto-refreshing to show
it.
Would it be possible that the main window refreshes once after this,
just like when one adds a variable?
One more small bug discovered by yours truly: In the first tab of the
model window, highlighting a part of the output and doing ctrl+c
copies that text to the clipboard directly. Interestingly, in the 2nd
3rd... tabs, a ctrl+c opens up a copy window. I think a choice of
csv... is not necessary in this context.
Sincerely
Talha
--
“An expert is a person who has made all the mistakes that can be made
in a very narrow field.” - Niels Bohr (1885-1962)
--
4:04 p.m.
On Fri, 23 Mar 2012, Talha Yalta wrote:
On Fri, Mar 23, 2012 at 5:08 PM, Allin Cottrell
<cottrell(a)wfu.edu> wrote:
> What do you mean by "opened scripts"? Read-only practice
> scripts? I'm not sure they belong in the editor, since you're
> not editing them.
But one can start modifying them after save as. IMHO it would be best
if these opened in the regular tabbed window with editing disabled.
Maybe. Right now, if you do "save as" and get an editable
script, and you have a tabbed script editor open, you get the
option of docking the newly editable script (right click).
I've now caught various cases where the main window series
listing was not auto-refreshing when series are added
automatically.
One more small bug discovered by yours truly: In the first tab of
the
model window, highlighting a part of the output and doing ctrl+c
copies that text to the clipboard directly. Interestingly, in the 2nd
3rd... tabs, a ctrl+c opens up a copy window. I think a choice of
csv... is not necessary in this context.
I'll look at that.
Allin
2:39 a.m.
On Fri, 23 Mar 2012, Talha Yalta wrote:
One more small bug discovered by yours truly: In the first tab of
the
model window, highlighting a part of the output and doing ctrl+c
copies that text to the clipboard directly. Interestingly, in the 2nd
3rd... tabs, a ctrl+c opens up a copy window.
I'm glad you spotted that. But it turns out that this anomaly is
quite instructive. Correct behavior with regard to Ctrl+c is easy to
arrange when each model is displayed in its own window, but obscure
and difficult to get right when several models are displayed in one
(tabbed) window.
Allin
4973
days inactive
4976
days old
9 comments
2 participants
participants (2)
-
Allin Cottrell -
Talha Yalta