Re: [Amforth] A very, very minor suggestion for amforth-shell.py

[John V. <joh...@gm...>]

Can you provide a unified diff of your proposed changes?

On Fri, 8 May 2020 at 15:42, Tristan Williams <ho...@tj...> wrote:
>
> A very, very, minor suggestion for amforth-shell.py in the next
> release.
>
> If amforth-shell.py finds multiple matches for a filename whilst
> executing a #include directive then it exits silently. It really has
> no choice but to exit (which one to load?). Adding a
> self.progress_callback line (as per the other filename related tests)
> gives a hint that things are going awry.
>
> I think this would work
>
> self.progress_callback("Error", None,  "Wrong # of file "+ filename+" found in search path") # add this line above the one below
> raise AMForthException("Wrong # of file occurances: " + filename + " ("+str(len(self._filedirs[filename]))+")\n\t"+"\n\t".join(self._filedirs[filename]))
>
> An easy way to demonstrate this is to defeat amforth-shell.py inbuilt
> intelligence and knowledge of mcu type by
>
> export AMFORTH_LIB=/path-to/amforth-6.8
>
> and create a file with the line
>
> #include marker.frt
>
> and send it to the mcu with amforth-shell.py
>
> In the standard distribution there are two marker.frt files
>
> ../amforth-6.8/avr8/lib/forth2012/core-ext/marker.frt
> ../amforth-6.8/msp430/lib/forth-2012/core-ext/marker.frt
>
> so amforth-shell.py will silently exit after the #include marker.frt
> line.
>
> Best wishes,
> Tristan
>
>
>
>
>
>
>
>
>
>
>
>
> _______________________________________________
> Amforth-devel mailing list for http://amforth.sf.net/
> Amf...@li...
> https://lists.sourceforge.net/lists/listinfo/amforth-devel



-- 


John Verne
about.me/jverne