selobu / salstat-statistics-package-2 Goto Github PK

Change so it uses MAc / linux default colours on Mac platform

In an Excel spreadsheet I had a column with approx 2100 rows of numbers. I 
copied it to the clipboard and in S2 pasted it into column A. 

S2 silently truncated it at row 256!

The print function doesn't work

Add the option to create a histogram plot

the transform variable dialog don't add new columns if it needed also it don't 
add the new columns in the continuos empty columns

At present if an exception or other error is thrown, because we have the 
inbuilt python logging, S2 catches the exception and the external IDE being 
used (I use PyCharm on the Mac) to develop S2 does not know it has happened, so 
cannot jump to the line of code that has led to the exception.

While I can see that scripts need S2 to know about the error, I believe that it 
needs to be propagated also out to the external Stdout / Stderr. So we need a 
proxy object to catch errors, and then send it to the internal and external 
Stdout / Stderr streams.

This was an issue that was also brought up by Ronald Oussoren the Py2App 
developer which hindered his diagnosis of our Mac Installer issues, and would 
probably hinder other new developers who do not realise what is happening?!

I noticed an interesting thing today: Move with the cursor keys to a
cell that has something in it. Press the "c" key. Move to another cell
with the cursor keys and press the "v" key and it pastes from the
clipboard! Does it do this for you too? It is as if it is not looking
for the Control key to be pressed?! It also means you cannot enter
data starting with these letters unless you try a different letter,
press backspace, then type that keystroke.

The software should import csv data into the Data Entry Panel

add an option to import xlsx files

undo / redo doesn't work with minus simbol

If you load a big Excel file,
if > 500 lines it silently drops last lines

if you select a serie of data to plot then it doesn't show the vertice names

What steps will reproduce the problem?
1. Make some statistical calculations
2. select some cell in the report panel
3. close the generated report
4. select one cell of the data entry panel
5  The software crash without confirmation or error message asociated to him

Dirty flag not set properly (so it says there is data to discard.
when there hasn't been any text entered)

I'm trying to paste some value into the script editor but it is paste into the 
Data Entry Panel

If you have unsaved data and hit the New icon, it deletes all your
data without offering to save

The pareto chart breaks a big problem into smaller pieces (and) identifies the 
biggest contributors 

What steps will reproduce the problem?
1. Open excel file with date format
2. Plot a scatter plot on x-axis date number instead of dates
3.

What is the expected output? What do you see instead?
In the date number colum the possibility to select date format compatible with 
excel date format.
It is wishful to set out a graph (scatter plot) by date.

What version of the product are you using? On what operating system?
V2.1 RC2

Please provide any additional information below.
See attachments

Add a method to generate random data and put it into the Data Entry Panel

the butttons <ok> and <cancel> of the varibles 
frame don't work

The software will be translated under mac os platform and linux platform  by 
Mark Livingstone

Greath tool, I love it but at the download, AVIRA found a virus on both files:

S2 V2.1 beta 4 portable.exe

as well as

S2 V2.1 Beta 4 setup.exe.  

Regards,

Fabrizio

Setup Mac / linux accelerator keys on Mac

Some plots could be select multiple pairs of data

Traceback (most recent call last):
  File "/Users/mlivingstone/Documents/workspace/salstat-statistics-package-2/src/plotFrame.py", line 1044, in _xlabelFontProp
    fontprop= fontDialog(self)
  File "/Users/mlivingstone/Documents/workspace/salstat-statistics-package-2/src/plotFrame.py", line 1415, in fontDialog
    fuente = wx.Font(wx.FONTSIZE_MEDIUM,wx.FONTFAMILY_DEFAULT,wx.FONTSTYLE_NORMAL,wx.FONTWEIGHT_NORMAL)
AttributeError: 'module' object has no attribute 'FONTSIZE_MEDIUM'

If we are eventually going to do outliers, we will need to add Q1 / Q3 / IQR 
which we really should have for Descriptive Statistics anyway.

What steps will reproduce the problem?
1. Open up S2, you get a blank grid on screen
2. Click to open a data file and load it
3. S2 creates a second sheet instead of using existing blank sheet

Also, at top of screen I have "English (U.S.)" selected (but flag doesn't 
initially show?). When it creates second sheet, it is called "Hoja0" when 
initial grid sheet was called "_1" so we need to fix sheet naming / translating?

The software should check for new versions itself and notify to the user

the user can select an idiom in wich the app will show

When opening Xls file with text on first line, it would be good if
it offered to make that the column headers 

What steps will reproduce the problem?
1. input data [1, 2, 2, 34, 6, 5]
2. Go to menu Preparation - Descriptive Statistics
3. Select Median

What is the expected output? What do you see instead?

Expected value: 3.5
Reported value:2.023

What steps will reproduce the problem?
1. Load attached spreadsheet which was created by S2


What is the expected output? What do you see instead?
Should load since S2 created it, however, depending if you select headers or 
not it gives an error message in the logging, or hangs S2 entirely for me. 

It appears that S2 does not like empty columns in between columns with data. 
Can be reproduced even with only one blank column.

Please use labels and text to provide additional information.

There are 12 columns of data, Six starting at column "A", another six starting 
at column "AA"

I tried to do a scatter graph but it didn't offer all the columns.
This was a bug I fixed in my SalStat, so now it is fixed in S2 too!

Now I tried again to do the scatter graph, but it won't let me because
there are missing values. I imagine this is the same for all graphing.
I had a look at how this is done in Excel, and when you go to do the
graph, it gives you the option to just ignore data points that are
missing but plot all the rest of the data. I can see how this may be a
problem if it was a line graph, but for a scatter plot missing data
shouldn't matter?

the software will accept third party scripts to calculate statistical functions 
and to make custom plots

The bar plot fails if the software is executed in a path containing unicode 
characters 

It would be nice to have a function like in spredsheets to
autosize the width of columns depending on what is in them?

If you click the New Data icon by mistake, there is no way to cancel out 
without having to save, or to lose your data. The logic behind this dialog box 
is wrong.

I propose that I will replace it with a more attractive wxWidgets based dialog 
box using the stock wxWidgets icons which will make it operate correctly, and 
also look nicer than the current dialog we use.


I think where possible in the future that we should use wxWidget based dialogs 
where feasible because then wxWidgets will adapt them for each operating system 
instead of us having to. They will also look more native.