Norman Megill wrote:
> On Monday, April 27, 2020 at 10:46:41 PM UTC-4, David A. Wheeler wrote:
> >
> > On Tue, 28 Apr 2020 11:23:08 +0900, heiphohmia via Metamath wrote:
> > > Norm, thank you for tagging v0.182 as well as merging the pull request.
> > > I am getting the impression that this
On Tue, 28 Apr 2020 11:23:08 +0900, heiphohmia via Metamath
wrote:
> Norm, thank you for tagging v0.182 as well as merging the pull request.
> I am getting the impression that this thread has mostly increased your
> maintenance burden. That's really counter to my intent and needn't be the
>
Norm, thank you for tagging v0.182 as well as merging the pull request.
I am getting the impression that this thread has mostly increased your
maintenance burden. That's really counter to my intent and needn't be the case!
The whole release process on the GitHub side can be made very automatic.
> On Monday, April 27, 2020 at 3:11:43 AM UTC-4, heiphohmia wrote:
> ...
> >Previously, at the 0.181 update Giovanni asked you to incant the following:
> > git commit -m'Release version 0.181.'
> > git tag v0.181
> > git push
> > git push --tags
Yes, that will work just fine
On Monday, April 27, 2020 at 3:11:43 AM UTC-4, heiphohmia wrote:
...
>Previously, at the 0.181 update Giovanni asked you to incant the
following:
>
> git commit -m'Release version 0.181.'
> git tag v0.181
> git push
> git push --tags
>
> It looks like this automatically
metamath-program.zip has never been on the home page. It was originally
added so Travis could download just the program without being slowed down
by unnecessarily including the .mm files, and it is still used for that.
Later it also become used for Linux distribution, and the compiled Windows
Hello Norm,
This is in reply to
https://groups.google.com/forum/#!searchin/metamath/Packaging$20Metamath$20for$20Linux%7Csort:date/metamath/z2kKJYgnz-g/xmRVy0KhCQAJ
For whatever reason my direct replies to the above seem to be falling into
Google's black hole, so I am sending this as a new
David A. Wheeler:
> > I think we can revisit and improve things, but whatever build and
> > distribution process changes are made needs to be something that Norm is
> > willing to accept. Norm has limited time and his own preferences.
On Fri, 20 Dec 2019 04:52:22 -0800 (PST), "'fl' via
> https://m.youtube.com/watch?v=4q_inV9M_us
I have just looked at it. Well, that's good.
--
FL
--
You received this message because you are subscribed to the Google Groups
"Metamath" group.
To unsubscribe from this group and stop receiving emails from it, send an email
to
> I think we can revisit and improve things, but whatever build and
> distribution process changes are made needs to be something that Norm is
> willing to accept. Norm has limited time and his own preferences.
>
Why does he let people waste their time talking if he knows he doesn't want
Just a quick follow-up, I created a three-part YouTube video on how to use the
auto tools. The auto tools are still one of the most common ways to distribute
C source programs on unix-like systems, including Linux. The first part is
especially useful if you just want to get an idea of what
>Just so we understand each other. I am completely convinced of the
>usefulness of "autotools". And your performance measures prove it. But
>I
>maintain that the use of "autoreconf" is the packager's responsibility.
>The
>user should only use "configure".
That is certainly the ideal.
> That's true, but if you allow the system to detect various settings the
> resulting executable is much faster.
>
Just so we understand each other. I am completely convinced of the
usefulness of "autotools". And your performance measures prove it. But I
maintain that the use of
On December 17, 2019 11:46:21 AM EST, Giovanni Mascellani
wrote:
>Hi,
>
>Il 16/12/19 19:05, David A. Wheeler ha scritto:
>>> The
>>> program is
>>> very portable and purposely calls no external libraries or does
>>> anything
>>> that isn't strictly part of ANSI C
>>
>> That's true, but if you
Hi,
Il 16/12/19 19:05, David A. Wheeler ha scritto:
>> The
>> program is
>> very portable and purposely calls no external libraries or does
>> anything
>> that isn't strictly part of ANSI C
>
> That's true, but if you allow the system to detect various settings the
> resulting executable is
> Maybe you can let me know exactly what files you'd like to see in it.
Thank you for engaging so openly.
Personally, my preferences align mostly with FL's. As a Linux user, the default
expectation I have for a source tarball (in this case metamath-program.zip) is
that I can do
$
>
>
> You may want to think about the planet and relieve the network. The
> Internet consumes as much energy as aeronautics because of all the
> advertising and netflix crap that goes through it.
>
And the plane traffic it is that:
https://flight-radar.eu/fr/suivi-des-vols-en-temps-reel/
>I could add this as an alternate build method, but these are
>Linux/Unix-specific and may not be available on all systems.
The generated configure file from autotools should work on absolutely every
system except windows.
> The
>program is
>very portable and purposely calls no external
>
>
>> I do find it curious that the advice for *nix builds is to directly call
>> gcc.
>> Perhaps the build instructions could be updated to this:
>>
>> $ autoreconf -i && configure && make
>>
>>
> I could add this as an alternate build method, but these are
> Linux/Unix-specific and
On Monday, December 16, 2019 at 8:34:02 AM UTC-5, heiphohmia at wilsonb.com
wrote:
>
> > And why not a special Linux source distribution...
>
> The URL that Norm shared pretty much ticks all the boxes you outline. The
> only
> substantive difference being that the metamath.exe is not excluded.
>
>
> The configure is already generated in a source distribution. No autoreconf
has to be done in a source distribution. Perhaps for the good reason
that the autotools on the end user's computer.
--
FL
--
You received this message because you are subscribed to the Google Groups
"Metamath"
On December 16, 2019 8:33:44 AM EST, heiphohmia via Metamath
>The source uses GNU autotools, so the standard build procedure Just
>Works if
>you first generate the necessary files with
>
>$ autoreconf -i
>
>I do find it curious that the advice for *nix builds is to directly
>call gcc.
>Perhaps
> And why not a special Linux source distribution...
The URL that Norm shared pretty much ticks all the boxes you outline. The only
substantive difference being that the metamath.exe is not excluded.
The source uses GNU autotools, so the standard build procedure Just Works if
you first generate
And why not a special Linux source distribution, a gzipped tarball, without
Windows executable,
without database, but with the manual page, the TeX file of the manual, the
Makefile, a configure.ac,
and a "configure" program. In short, the standard source distribution for
Linux like the
On Friday, December 13, 2019 at 3:30:53 AM UTC-5, heiphohmia wilsonb.com
wrote:
>
> Thank you for the response.
>
> > Would it work for you to download the source?
>
> Unfortunately, no. Void Linux has a policy of only packing "stable"
> releases,
> which in practice means no builds off
Thank you for the response.
> Would it work for you to download the source?
Unfortunately, no. Void Linux has a policy of only packing "stable" releases,
which in practice means no builds off repository branches and the like.
Upstream *might* accept a package off a git tag, but it looks like the
Would it work for you to download the source? The metamath executable is
versioned at https://github.com/metamath/metamath-exe , and the main
library is at https://github.com/metamath/set.mm . I presume you have a
method for handling github projects in Void Linux...?
I believe that Norm also
Dear Metamath,
This outlines some of the issues I encounterd when attempting to package up
metamath for Void Linux. Need to make metamath more available!
The content here essentially reiterates that of a previous post of mine, which
just happened to go out right before the DNS debacle, causing
28 matches
Mail list logo
