[Tutor] Command line scripts

[David Hutto]

On Sun, Jan 9, 2011 at 9:03 AM, ALAN GAULD <alan.gauld at btinternet.com> wrote:
>
>
>> The following line is what I mean by calling a  command line from within the
>>app
>> using subprocess.
>>
>> self.espeak =  subprocess.Popen(['espeak', word],stdout  =
>> subprocess.PIPE).communicate()[0]
>
> OK, Now I understand.
> You want to call an external application from within your code
> via subprocess. And you want to know if that will work ok in a
> wxPython application?
>
>> >> In other words,  is a command line app different from
>> >> bindings in a compiled  app?
>
> When you call an app using subprocess it spawns an entirely
> separate process in the OS. There is no linkage to your process
> apart from the redirection of stdin/stdout/stderr back to your
> app instead of to a console.
>
>> > So-called exe generators simply bundle the interpreter  witrh
>> > the code and auto run it.
>>
>> So it's basically just installing  a sandbox version of python?
>
> Sort of, its a minimalist version of python with only the modules
> needed to run the app. And you can't run the interpreter on its
> own you can only run the packaged app.
>
>> Which ultimately is, if I have a standalone  application, that doesn't
>> come from a command line terminal launching(which I  haven't gotten to
>> yet), is using the command line calls going through  subprocess going
>> to be called without the terminal, or will it open a  terminal
>> automatically to make those calls?
>
> The app (eSpeak in your example) will see your app as its "console".
> That is, it will send all output to your app and read all input from your
> app

I was thinking that since the app is called from the command line with
python(which is no different, than using any other command line
script-just like espeak 'word' -> python script.py), it would be go
back to the shell window it was spawned/called from through python and
utilize the command line there. Similar to how when I run the app, my
data can be through the app text/label fields, or, through the
terminal window itself.

 (so you need to make sure it gets any input it is expecting!)
> The other thing to watch in a wxPython (or any other GUI framework)
> is that if the app runs for a long time your app may "freeze" from
> the user's perspective, so you may want to run it in the background
> or as a separate thread in your app.
>
>> Or will all of my python  apps need to have a command line terminal
>> open to launch them.
>
> No, your app takes over the job of the terminal. When you launch
> a program from the terninal the terminal app(for it is just an app like
> any other) is doing the same as you, it is spawning a subprocess
> that sends its output back to the teminal for it to display.
>
> In fact you might find it a useful exercise to build a very basic terminal
> app in wxPython first. Read commands from a command input field
> and display the output in a text widget... Execute the commands
> via subprocess. Its a good way to get used to using subprocess
> and experimenting with its various options. If you are very keen
> you can start adding command history and search etc to it too...
>

I'll eventually get to the above(I have a lot of projects, some will
involve command line/subprocess:), but thanks for the explanation. I'm
guessing that something like cxfreezeI haven't looked it over
thoroughly) might be what I was looking for. Although in the end, it's
just for other to see your work, because most of my own apps are for
me personally, which might be a good thing.