<html>
  <head>
    <meta http-equiv="Content-Type" content="text/html; charset=utf-8">
  </head>
  <body text="#000000" bgcolor="#FFFFFF">
    <p>Yes, I see the inclusion of a metadata file which conforms to an
      unaccepted PEP as potentially dangerous.</p>
    <p>Perhaps I should disable it in the next release?<br>
    </p>
    <br>
    <div class="moz-cite-prefix">Daniel Holth kirjoitti 04.09.2017 klo
      17:03:<br>
    </div>
    <blockquote type="cite"
cite="mid:CAG8k2+6W_+HePq5OuJm+UzmT_2WXgL=Zrk+CLjzHO6A2g7w1vA@mail.gmail.com">
      <p dir="ltr">Some people enjoy using metadata.json which tracked
        pep 426 but I have been meaning to take it out, and perhaps keep
        the key/value to json converter as a command.</p>
      <br>
      <div class="gmail_quote">
        <div dir="ltr">On Mon, Sep 4, 2017, 09:33 Nick Coghlan <<a
            href="mailto:ncoghlan@gmail.com" moz-do-not-send="true">ncoghlan@gmail.com</a>>
          wrote:<br>
        </div>
        <blockquote class="gmail_quote" style="margin:0 0 0
          .8ex;border-left:1px #ccc solid;padding-left:1ex">Some time
          ago, I started the process [1] of adjusting how<br>
          distutils-sig uses the PEP process so that the reference<br>
          specifications will live on <a
            href="http://packaging.python.org" rel="noreferrer"
            target="_blank" moz-do-not-send="true">packaging.python.org</a>,
          and we use the PEP<br>
          process to manage *changes* to those specifications, rather
          than<br>
          serving as the specifications themselves (that is, adopting a
          process<br>
          closer to the URL-centric way the Python language reference is<br>
          managed, rather than using the RFCstyle PEP-number-centric
          model the<br>
          way we do now).<br>
          <br>
          I never actually finished that work, and as a result, it's
          currently<br>
          thoroughly unclear [2] that Description-Content-Type and<br>
          Provides-Extra are defined at<br>
          <a
            href="https://packaging.python.org/specifications/#core-metadata"
            rel="noreferrer" target="_blank" moz-do-not-send="true">https://packaging.python.org/specifications/#core-metadata</a>
          rather than<br>
          in a PEP.<br>
          <br>
          I'm currently at the CPython core development sprint in San
          Francisco,<br>
          and I'm thinking that finalising that migration [3] and
          updating the<br>
          affected PEPs accordingly (most notably, PEP 345) is likely to
          be a<br>
          good use of my time.<br>
          <br>
          However, I'm also wondering if it may still be worthwhile
          writing a<br>
          metadata 1.3 PEP that does the following things:<br>
          <br>
          1. Explicitly notes the addition of the two new fields<br>
          2. Describes the process change for packaging interoperability
          specifications<br>
          3. Defines a canonical transformation between the
          human-readable<br>
          key:value format and a more automation friendly JSON format<br>
          <br>
          That PEP would then essentially be the first one to use the
          new<br>
          process: it would supersede PEP 345 as the latest metadata<br>
          specification, but it would *also* redirect readers to the
          relevant<br>
          URL on <a href="http://packaging.python.org" rel="noreferrer"
            target="_blank" moz-do-not-send="true">packaging.python.org</a>
          as the canonical source of the<br>
          specification, rather than being the reference documentation
          in its<br>
          own right.<br>
          <br>
          Cheers,<br>
          Nick.<br>
          <br>
          [1] <a
            href="https://github.com/pypa/pypa.io/issues/11#issuecomment-173833061"
            rel="noreferrer" target="_blank" moz-do-not-send="true">https://github.com/pypa/pypa.io/issues/11#issuecomment-173833061</a><br>
          [2] <a href="https://github.com/python/peps/issues/388"
            rel="noreferrer" target="_blank" moz-do-not-send="true">https://github.com/python/peps/issues/388</a><br>
          <br>
          P.S. Daniel, if you're currently thinking "I proposed defining
          an<br>
          incremental metadata 1.3 tweak years ago!", aye, you did. My<br>
          subsequent additions to PEP 426 were a classic case of
          second-system<br>
          syndrome: <a
            href="https://en.wikipedia.org/wiki/Second-system_effect"
            rel="noreferrer" target="_blank" moz-do-not-send="true">https://en.wikipedia.org/wiki/Second-system_effect</a>
          (which we<br>
          suspected long ago, hence that PEP's original deferral)<br>
          <br>
          Fortunately, the disciplining effect of working with a
          primarily<br>
          volunteer contributor base has prevented my over-engineered<br>
change-for-change's-sake-rather-than-to-solve-actual-user-problems<br>
          version from becoming reality ;)<br>
          <br>
          --<br>
          Nick Coghlan   |   <a href="mailto:ncoghlan@gmail.com"
            target="_blank" moz-do-not-send="true">ncoghlan@gmail.com</a> 
           |   Brisbane, Australia<br>
          _______________________________________________<br>
          Distutils-SIG maillist  -  <a
            href="mailto:Distutils-SIG@python.org" target="_blank"
            moz-do-not-send="true">Distutils-SIG@python.org</a><br>
          <a
            href="https://mail.python.org/mailman/listinfo/distutils-sig"
            rel="noreferrer" target="_blank" moz-do-not-send="true">https://mail.python.org/mailman/listinfo/distutils-sig</a><br>
        </blockquote>
      </div>
      <br>
      <fieldset class="mimeAttachmentHeader"></fieldset>
      <br>
      <pre wrap="">_______________________________________________
Distutils-SIG maillist  -  <a class="moz-txt-link-abbreviated" href="mailto:Distutils-SIG@python.org">Distutils-SIG@python.org</a>
<a class="moz-txt-link-freetext" href="https://mail.python.org/mailman/listinfo/distutils-sig">https://mail.python.org/mailman/listinfo/distutils-sig</a>
</pre>
    </blockquote>
    <br>
  </body>
</html>