<html>
<head>
<meta http-equiv="Content-Type" content="text/html; charset=utf-8">
</head>
<body text="#000000" bgcolor="#FFFFFF">
<p>Umm, which PR was that again?<br>
</p>
<pre class="moz-signature" cols="72">Best Regards,
Brent Fraser</pre>
<div class="moz-cite-prefix">On 11/27/2017 9:14 AM, Dan Little
wrote:<br>
</div>
<blockquote type="cite"
cite="mid:CABqPoBb1DXBbcj3-_L=EnhkVzTzhpjszZQi9hX6gCo1YYqA4HQ@mail.gmail.com">
<div dir="ltr">I think part of it is matching what the dev team is
using. NVM makes a lot of the "hard parts" of different Node
versions go away. I'm +1 at having an apologetic-policy of
using the current LTS. Knowing that occasionally we will be
behind a little bit. That's just due to the limitations of
project resources. We should publish that we strive to support
current LTS in the build-chain.
<div><br>
</div>
<div>Also, Brent, WHERE'S THAT PR???</div>
</div>
<div class="gmail_extra"><br>
<div class="gmail_quote">On Wed, Nov 15, 2017 at 3:54 PM, James
Klassen <span dir="ltr"><<a
href="mailto:klassen.js@gmail.com" target="_blank"
moz-do-not-send="true">klassen.js@gmail.com</a>></span>
wrote:<br>
<blockquote class="gmail_quote" style="margin:0 0 0
.8ex;border-left:1px #ccc solid;padding-left:1ex">
<div dir="auto">
<div dir="auto">Even though if we do a good job, most
people won't ever have to build GeoMoose, my goal is to
establish clear expectations for what is supported so
that people who need to/want to build or debug GeoMoose
can do so without a lot of guessing and without needing
exotic system configurations.</div>
<div dir="auto"><br>
</div>
<div dir="auto">
<div dir="auto">
<div dir="auto">Maybe nvm makes which node version(s)
largely moot. Building on the production server
generally isn't necessary/advised, so that isn't a
huge deal.</div>
<div dir="auto"><br>
</div>
<div dir="auto">It could also potentially matter for
people integrating GeoMoose 3 as a library into an
existing app/page/etc.</div>
</div>
</div>
</div>
<div class="HOEnZb">
<div class="h5">
<div class="gmail_extra"><br>
<div class="gmail_quote">On Nov 14, 2017 10:55 AM,
"Brent Fraser" <<a
href="mailto:bfraser@geoanalytic.com"
target="_blank" moz-do-not-send="true">bfraser@geoanalytic.com</a>>
wrote:<br type="attribution">
<blockquote class="gmail_quote" style="margin:0 0 0
.8ex;border-left:1px #ccc solid;padding-left:1ex"><span
style="font-family:Arial,Helvetica,Sans-Serif;font-size:12px">
<div>Eli, Jim,</div>
<div> </div>
<div> The build toolchain is more important now
to us implementers because it must be used to
debug GeoMoose. In the past it was possible
to use the Geomoose_dev.html file to load
un-minified version of the code and use the
browser's debugger to step through the code.
The new dependency on NodeJS for debugging is
an added burden (but thankfully it is
mitigated by the doc Jim and Dan have
written).</div>
<div> </div>
<div> Since my exposure to NodeJS is minimal (I
just used it to help me add Bing layer types
other than Aerial and Road), I am OK with
choice A).</div>
<div> </div>
<div> I wonder how many people debug Geomoose?</div>
<div> </div>
<div>Best Regards,</div>
<div>Brent </div>
<div> </div>
<div> </div>
<div> </div>
<div> </div>
<hr size="2" align="center" width="100%">
<div><span
style="font-family:tahoma,arial,sans-serif;font-size:10pt"><b>From</b>:
"Eli Adam" <<a
href="mailto:eadam@co.lincoln.or.us"
target="_blank" moz-do-not-send="true">eadam@co.lincoln.or.us</a>><br>
<b>Sent</b>: Tuesday, November 14, 2017 9:23
AM<br>
<b>To</b>: "Jim Klassen" <<a
href="mailto:klassen.js@gmail.com"
target="_blank" moz-do-not-send="true">klassen.js@gmail.com</a>><br>
<b>Cc</b>: "<a
href="mailto:geomoose-psc@lists.osgeo.org"
target="_blank" moz-do-not-send="true">geomoose-psc@lists.osgeo.org</a>"
<<a
href="mailto:geomoose-psc@lists.osgeo.org"
target="_blank" moz-do-not-send="true">geomoose-psc@lists.osgeo.org</a>><br>
<b>Subject</b>: Re: [geomoose-psc] GeoMoose
Supported NodeJS Vesion(s)?</span>
<div> </div>
On Mon, Nov 13, 2017 at 1:53 PM, Jim Klassen
<<a href="mailto:klassen.js@gmail.com"
target="_blank" moz-do-not-send="true">klassen.js@gmail.com</a>>
wrote:<br>
> I am trying to come up with a policy for
which nodejs versions we<br>
> officially support for the *build*
toolchain. My general thoughts are<br>
<br>
Do we need a policy for the *build* toolchain?<br>
<br>
> either current LTS (8.9.1 now) or current
and previous LTS (now 6.9.x).<br>
> The benefits to two versions is it gives
us and others more leeway when<br>
> to update things. The benefit to one
version is that is likely what we<br>
> are testing in practice and it can
simplify the build scripts.<br>
<br>
More than one version seems infeasible. We'll
actively be working<br>
with one. Maybe additional versions will work
but that could be<br>
happenstance, not design.<br>
<br>
><br>
> The nodejs release schedule is here [1]:<br>
> Node 8.x became LTS on 2017-10-31 and
will be supported until 2019-12.<br>
> Node 6.x became LTS on 2016-10-18 and
will be supported until 2019-4.<br>
> Node 4.x became LTS on 2015-10-1 and will
be supported until 2018-4.<br>
> (We made some efforts to support this
version at one point, but it has<br>
> not been tested in a long time by me and
probably doesn't currently work<br>
> with GeoMoose).<br>
><br>
> I also point out that 8.9.1 doesn't build
out of the box on Ubuntu 14.04<br>
> because the gcc version is not new
enough. I'm pretty sure Jessie and<br>
> 16.04 are ok with 8.9.1.<br>
><br>
> So how do people feel about supporting:<br>
><br>
> A) Current LTS version of nodejs only.<br>
><br>
> or<br>
><br>
> B) Current and previous LTS versions of
nodejs.<br>
><br>
><br>
> (Note: the demo server is still on Ubuntu
14.04 due to MapScript not<br>
> being in Ubuntu 16.04. Of course GeoMoose
3 doesn't care about<br>
> MapScript, but we are still hosting
GeoMoose 2.9 demo which does. This<br>
> could easily be split up so it isn't the
end of the world, just where<br>
> we're at now.)<br>
<br>
<br>
C) Whatever one version is convenient based on
a variety of factors<br>
like MapScript running a previous version of
the demo. Not to be<br>
flippant, but since this an internal facing
build tool, I'm not<br>
convinced that we need a policy.<br>
<br>
Best regards, Eli<br>
<br>
><br>
> [1] <a
href="https://github.com/nodejs/Release"
target="_blank" moz-do-not-send="true">https://github.com/nodejs/Rele<wbr>ase</a><br>
><br>
> ______________________________<wbr>_________________<br>
> geomoose-psc mailing list<br>
> <a
href="mailto:geomoose-psc@lists.osgeo.org"
target="_blank" moz-do-not-send="true">geomoose-psc@lists.osgeo.org</a><br>
> <a
href="https://lists.osgeo.org/mailman/listinfo/geomoose-psc"
target="_blank" moz-do-not-send="true">https://lists.osgeo.org/mailma<wbr>n/listinfo/geomoose-psc</a><br>
______________________________<wbr>_________________<br>
geomoose-psc mailing list<br>
<a href="mailto:geomoose-psc@lists.osgeo.org"
target="_blank" moz-do-not-send="true">geomoose-psc@lists.osgeo.org</a><br>
<a
href="https://lists.osgeo.org/mailman/listinfo/geomoose-psc"
target="_blank" moz-do-not-send="true">https://lists.osgeo.org/mailma<wbr>n/listinfo/geomoose-psc</a></div>
</span>
</blockquote>
</div>
</div>
</div>
</div>
<br>
______________________________<wbr>_________________<br>
geomoose-psc mailing list<br>
<a href="mailto:geomoose-psc@lists.osgeo.org"
moz-do-not-send="true">geomoose-psc@lists.osgeo.org</a><br>
<a
href="https://lists.osgeo.org/mailman/listinfo/geomoose-psc"
rel="noreferrer" target="_blank" moz-do-not-send="true">https://lists.osgeo.org/<wbr>mailman/listinfo/geomoose-psc</a><br>
</blockquote>
</div>
<br>
</div>
</blockquote>
<br>
</body>
</html>