Hi Juergen, Elias and all,
Some time ago we have been discussing the idea to move help from
gnu-apl-mode in Emacs into the GNU APL itself.
Now since we have a help in GNU APL, I've implemented network command in
emacs_mode part of GNU APL which will send requested help as a list of
S-expressions
Thank you for this work. This is something have planned to do for a while
but never got around to.
I merged your changes and tested, and it all seems to work fine. I'll merge
this into gnu-apl-mode, but Jürgen needs to do the same on his side before
this is complete.
Regards,
Elias
On 7 January
Hi
What are you new times after the fix?
why would there be a need for a fix anyway? aren't all 'lines' (10/50)
treated the same and run as fast as possible anyway?
On Fri, 5 Jan 2018 17:05:59 +0100
Hans-Peter Sorge wrote:
> running < ⍎¨ 10 ⍴'0' > takes 2 sec
> running < ⍎¨ 500
Just FYI - I had to update all of my dump'ed workspaces (with VI) because
quad-PS changed from a scalar to a vector. It would have been great for
backward compatibility if the loading of dump'ed workspaces handled the old
quad-PS format.
Thanks!
Blake
Hi,
it seems like Elias' sources in git are slightly out of sync with
the current GNU APL sources.
For example (emacs.hh around line 55):
-
return string((const char *)&utf[0], utf.size());
+ return string((const char *)(utf.g
That's OK. The C++ sources in my repository isn't actually used. I should
probably remove them from the gnu-apl-mode repository.
The code in the main GNU APL repository is the official source.
Regards,
Elias
On 7 January 2018 at 22:53, Juergen Sauermann wrote:
> Hi,
>
> it seems like Elias' so
Hi,
they are actually used (I sync my sources from your git repo). I
suppose that you applied
Alexey's patch to your sources and the result doesn't compile when
I fetch the changes from
your git repo.
I would prefer if the two repos remain in
Hi Blake,
fixed in SVN 1033.
/// Jürgen
On 01/07/2018 01:50 AM, Blake McBride
wrote:
Hi.
I was updating some dump files and noticed a small change
that I'd think was uninte
Hi,
The files I've attached are based on the sources from SVN, not git.
Br
/Alexey
On Jan 7, 2018 4:52 PM, "Juergen Sauermann"
wrote:
> Hi,
>
> they are actually used (I sync my sources from your git repo). I suppose
> that you applied
> Alexey's patch to your sources and the result doesn't co
Hi Blake,
thanks, fixed in SVN 1034.
/// Jürgen
On 01/07/2018 03:42 PM, Blake McBride
wrote:
Just FYI - I had to update all of my
dump'ed workspaces (with VI) because quad-PS changed from a
scalar to a
Hi,
yes, but it is Elias who can OK the files. The proper order should be:
1. sync GNU APL sources to Elias' git
2. apply changes and test Emacs
3. sync Elias' git to my local GNU APL repo
4. commit to public GNU APL repo
If I would apply the changes directly to the GNU APL repo then the
GNU AP
To me, it's a lot better if the primary code stays in the main GNU APL
repository.
Jürgen,
We've discussed in the past the possibility of moving to a public
development repository where all of us can create commit requests, but at
the time you expressed concerns with that model.
Perhaps we can r
> On Jan 7, 2018, at 11:52 AM, Elias Mårtenson wrote:
>
> We've discussed in the past the possibility of moving to a public development
> repository
> where all of us can create commit requests, but at the time you expressed
> concerns with that model.
Hi Elias:
I'm absolutely sure that you
Hi,
my report was just a simple example.
You are correct, all elements are being treated equally well.
All chars turn up as numbers.
It just looks like the program doing the job took it's own way.
I was playing around with pi (pi-10million.txt)
Without the fix it took 11!!hours.
The fix reduces
Hi,
On 01/07/2018 05:52 PM, Elias Mårtenson wrote:
To me, it's a lot better if the primary code stays in the main GNU APL
repository.
Agreed. The current problem is that your git repository is a little
behind the
main GNU APL repository. I made some code clean-ups already some time ago
whic
15 matches
Mail list logo
