Re: 17 & 20 November Downloads

[Richard Shann]

On Wed, 2021-11-24 at 08:51 +0000, Joe Wilkinson wrote:
> 
> 
> On 23/11/2021 17:23, Richard Shann wrote:
> > On Mon, 2021-11-22 at 16:51 +0000, Joe Wilkinson wrote:
> > 
> > [...]So, I think you have later versions failing to launch LilyPond
> > even
> > though it is present.
> > It would be interesting to know what the following Scheme does if
> > you
> > execute it.
> > 
> > (d-InfoDialog (format #f "~s" (system "notepad")))
> > 
> > - assuming you have "notepad" in your path(*), but substitute
> > something
> > else even just "dir" if not. Something that does successfully
> > execute
> > should return "0"
> > 
> > Richard
> > (*) if you are trying a full path to an executable double up any
> > backslashes you use...
> As you expected: a command window opens followed by notepad. and when
> I 
> close notepad, the command window also closes and a small popup
> appears 
> with the number 0.

Ah, right. I think that means that the denemo program you are executing
is able to launch some new processes ("helper programs", in this case
notepad). I was starting to wonder if the problem involved Windows not
allowing "untrusted" programs to execute. So I was going to ask if you
were running with or without admin privileges, but as you have been
able to execute the LilyPond outside of Denemo that, as a cause, would
seem implausible (I suppose Windows might have had a category of
"trusted to run but not start other processes").

One thing that is not crystal clear to me is whether it is purely the
launch of LilyPond that is failing or whether the files
../AppData/Temp/Denemo_XXXXXX/denemoprintA.ly, denemoprintB.ly are
being generated. If you were to open windows explorer on that folder
and startup Denemo (with continuous typesetting off so you can control
the creation of the .ly files) and watch for if/when these .ly files
are generated that would be a useful check. 

Beyond that (and short of getting a debugger running on your machine -
possible but stretching the supply lines) I think I'll have to
instrument the code to show what's happening in more detail - I'll look
into that.

Richard