Pyspread - a spreadsheet where each cell can be a Python expression

[Thomas Passin]

I just learned about Pyspread, which is a spreadsheet program where cells contain Python expressions or code.  They can also contain images, which makes it interesting to see how the image data is stored and used.  It's a PyQt program, so there might be some good lessons to learn.

I think I see a study guide in my future...

[Edward K. Ream]

On Mon, May 1, 2023 at 5:51 PM Thomas Passin <tbp1...@gmail.com> wrote:

I just learned about Pyspread, which is a spreadsheet program where cells contain Python expressions or code.  They can also contain images, which makes it interesting to see how the image data is stored and used.  It's a PyQt program, so there might be some good lessons to learn.

I agree.

Edward

[jkn]

There was a much older 'python in a spreadsheet' program, created by Resolver Systems. I followed it with interest, they tried to create a commercial product out of it but failed.

IIRC pyspread is much less ambitious than Resolver, but it might be that it is more successful because of that. I still with the beta version or whatever of the Resolver Systems one was available though.

[David Szent-Györgyi]

On Tuesday, May 2, 2023 at 1:04:27 PM UTC-4 jkn wrote:

There was a much older 'python in a spreadsheet' program, created by Resolver Systems. I followed it with interest, they tried to create a commercial product out of it but failed.

That product used IronPython - an alternate implementation of Python implemented on the Dynamic Language Runtime for .NET. Resolver Systems is long gone, but IronPython is still out there, though development by its tiny team is slow - the current release added Python3 features, and was released early in 2023; it is closest to CPython 3.4. 

Unlike CPython, IronPython has no Global Interpreter Lock ("GIL"), and it used unicode for strings long before that was sorted out in CPython. Differences between IronPython and CPython make IronPython a dialect, but  one well-suited to multi-threaded projects. It is an excellent "glue language" as it is an interpreter with a JIT compiler; it has full access to .NET  as well as to Win32; it also has access to libraries accessible through CTypes. In my day job, I write complex macros in IronPython for an application that controls exotic hardware; I prefer the IronPython read-eval-print-loop to compiling code through a heavyweight IDE and compiler that Get In My Way. 

Michael Foord and Christian Muirhead, who were with Resolver Systems, wrote an excellent book on IronPython. While its content has not been updated to address the current release, its exploration of .NET specifics makes it worth reading for the newcomer to IronPython. 

[jkn]

Yea, I know Michael Foord's (at least) writing and work on Python from quite a while ago. I had forgotten the link with IronPython, thanks.