From: "Юрий Соколов" <funny.falcon@...>
Date: 2015-10-22T14:12:00+03:00
Subject: [ruby-core:71151] Re: id_table cannot work for const_tbl

--001a11c3811a6858820522af90c3
Content-Type: text/plain; charset=UTF-8

But, is "class path name resolution" specified somewhere?
In other words:
perhaps it matters more to change tests instead of abandoning this idea.

"class path name resolution" affects only runtime generated classes
(created by `MyClass = Class.new()`, and not `class MyClass`)
(and WEBrick::HTTPStatus::Unauthorized is runtime generated).

And it only may break code which relies on stringified class path name of
such runtime-generated classes (which is very rare in practice).

This particular test can be fixed, if `const_set(err_name, err_class)` will
force
name resolution for `err_class` if its name not already resolved.

2015-10-22 5:47 GMT+03:00 Eric Wong <normalperson@yhbt.net>:

> [Posting to hopefully prevent somebody else from going down
>  the same path I did and wasting time working on a patch]
>
> I started working on the following patch to use id_table for
> const tables, but apparently ordering matters for class path name
> resolution:
>
>   http://80x24.org/spew/20151022023752.9038-1-e%4080x24.org/raw
>
> So unless I'm missing something, I'll reject it.
>

--001a11c3811a6858820522af90c3
Content-Type: text/html; charset=UTF-8
Content-Transfer-Encoding: quoted-printable

<div dir=3D"ltr">But, is &quot;class path name resolution&quot; specified s=
omewhere?<br>In other words:<div>perhaps it matters more to change tests in=
stead of abandoning this idea.</div><div><br>&quot;class path name resoluti=
on&quot; affects only runtime generated classes</div><div>(created by `MyCl=
ass =3D Class.new()`, and not `class MyClass`)</div><div>(and=C2=A0WEBrick:=
:HTTPStatus::Unauthorized is runtime generated).</div><div><br></div><div>A=
nd it only may break code which relies on stringified class path name of</d=
iv><div>such runtime-generated classes=C2=A0(which is very rare in practice=
).</div><div><br></div><div><div>This particular test can be fixed, if `con=
st_set(err_name, err_class)` will force</div><div>name resolution for `err_=
class` if its name not already resolved.</div></div><div><br></div><div>201=
5-10-22 5:47 GMT+03:00 Eric Wong <span dir=3D"ltr">&lt;<a href=3D"mailto:no=
rmalperson@yhbt.net" target=3D"_blank">normalperson@yhbt.net</a>&gt;</span>=
:<br></div><div><div class=3D"gmail_extra"><div class=3D"gmail_quote"><bloc=
kquote class=3D"gmail_quote" style=3D"margin:0px 0px 0px 0.8ex;border-left-=
width:1px;border-left-color:rgb(204,204,204);border-left-style:solid;paddin=
g-left:1ex">[Posting to hopefully prevent somebody else from going down<br>
=C2=A0the same path I did and wasting time working on a patch]<br>
<br>
I started working on the following patch to use id_table for<br>
const tables, but apparently ordering matters for class path name<br>
resolution:<br>
<br>
=C2=A0 <a href=3D"http://80x24.org/spew/20151022023752.9038-1-e%4080x24.org=
/raw" rel=3D"noreferrer" target=3D"_blank">http://80x24.org/spew/2015102202=
3752.9038-1-e%4080x24.org/raw</a><br>
<br>
So unless I&#39;m missing something, I&#39;ll reject it.<br>
</blockquote></div><br></div></div></div>

--001a11c3811a6858820522af90c3--