Really, Amazingly Silly Question

Stephen Horne intentionally at blank.co.uk
Sat Mar 8 14:24:11 CET 2003


On Sat, 08 Mar 2003 08:28:43 GMT, "Nilesta"
<nilesta at nospam.sbcglobal.net> wrote:

>First, I realize that I'm asking an incredibly stupid question.  I
>understand it.  I am at one with my stupid question.
>
>Second, the question:  Just how exactly do you compile python?  I'm talking
>write the program, translate it into assembly, smack an .exe on the end of
>the name and give it to someone who's never heard of python and allow them
>to run it on their machine by the incredibly complex process of double
>clicking.  Or the equivilant on any other OS.  I realize there has got to be
>a way to do this, other than manually, otherwise why bother with python at
>all?  I have rumaged through the docs, and found info on all sorts of fun
>things, like zips, tars, and self-extracting magic dohickys.  But my poor,
>work-adled brain seems to have come to the conclusion that the only way to
>distribute it is to pack an interpreter with it, which seems incredibly ..
>unweildy.

I would suggest a method where...

1.  Both your apps files and the Python installer are packed into a
    self-extracting archive.

2.  The archive is set to extract itself, and then run the 'real'
    install program from in the temp folder.

3.  The 'real' install program can detect whether a suitable Python is
    already installed. After putting your application files in their
    real home, you can decide whether to start the Python installer or
    not.

I've not had to worry about this too much with Python programs, but
I've used a similar approach with some plugins I once wrote. I didn't
need to install the main app, but I did need to check if and where it
was installed and to install the plugins in the correct place.

I used WinRAR for the self-extracting archive (which definitely has
the needed options) and <shame> a C++ program to do the main install.

I have no idea how this would be done on non-windows platforms, but I
suspect an equivalent approach could be implemented.

>Third, and entirely off the point:  My first programming language was a
>bastardized version of Forth, MUF (That's sort of an upside-down and
>backwards version of any other programming language, in case you don't know
>it)

I know it from 'white lightening' in the C64 days, and from a
bastardised version which was my first ever attempt at writing my own
language (on the Atari ST).

Few languages manage the level of awkwardness that Forth achieved,
though it certainly had its plus points too. How many languages can
boast an interactive compiler (yes - compiler - not interpreter).

-- 
steve at ninereeds dot fsnet dot co dot uk




More information about the Python-list mailing list