Newsgroups: php.internals
Path: news.php.net
Xref: news.php.net php.internals:129654
Precedence: list
MIME-Version: 1.0
References: <CAH5C8xXUPKgd0wWwLC3G2Ye=kRZN58gK3DY82v5GoduBNNtb5A@mail.gmail.com>
 <83238ad3-c844-4457-dfb3-11321787e022@php.net> <CAH5C8xUOEkfQ6u8AxuV8UGjWU0h7goDsvWa9QQmANhNdU=Dwhw@mail.gmail.com>
 <CAOV5rgZ1LTVf9PBpfQekahmESo=KXE=RVrpeUXrbqxG4OSDMnQ@mail.gmail.com>
 <d6a46bb627a9545e621f37e83c0441e5@glaive.pro> <CAOV5rgZGKJEo+jV7q-qQFB59CXhfNwadYPDbm70e_F+1v=Fgng@mail.gmail.com>
In-Reply-To: <CAOV5rgZGKJEo+jV7q-qQFB59CXhfNwadYPDbm70e_F+1v=Fgng@mail.gmail.com>
Date: Thu, 18 Dec 2025 22:45:59 +0100
Message-ID: <CAH5C8xWwcNC3z=FxQZiN2r1aZ1FgriVNAtdw+YwhaN4S2joyzw@mail.gmail.com>
Subject: Re: [PHP-DEV] [RFC] [Discussion] Followup Improvements for ext/uri
To: ignace nyamagana butera <nyamsprod@gmail.com>
Cc: Juris Evertovskis <juris@glaive.pro>, PHP Internals List <internals@lists.php.net>
Content-Type: multipart/alternative; boundary="000000000000aa6b4d064640e374"
From: kocsismate90@gmail.com (=?UTF-8?B?TcOhdMOpIEtvY3Npcw==?=)

--000000000000aa6b4d064640e374
Content-Type: text/plain; charset="UTF-8"
Content-Transfer-Encoding: quoted-printable

Hi,

The WHATWG URL living standard does the following:
>
> let url =3D new URL('https://example.com?debug&foo=3Dbar&debug=3D');
> console.log(
>     url.searchParams.toString(), //returns debug=3D&foo=3Dbar&debug=3D'
> );
>
> the pair gets converted to ['debug' =3D> '']. The roundtrip does not
> conserve the query string as is but all key/pair (tuples) are present.
>

Yes, confirmed. Unfortunately, WHATWG URL only supports string values, so
there's no way to support
query parameters without a key (e.g. ?debug) in the RFC implementation
either. :(

On the other hand, the RFC 3986 implementation supports this notion, even
uriparser calls this out
in its documentation:
https://uriparser.github.io/doc/api/latest/index.html#querystrings.

However, there are a few other problems which came up when I was updating
my implementation.

1.) Yesterday, I wrote that name mangling of the query params shouldn't
happen. However, as I realized,
it is still needed for non-list arrays, because the [..] suffix must be
added to their name:

$params =3D new Uri\Rfc3986\UriQueryParams()
    ->append("foo", [2 =3D> "bar", 4 =3D> "baz"]);

echo $params->toRfc3986String();                      //
foo%5B2%5D=3Dbar&foo%5B4%5D=3Dbaz

var_dump($params->getFirst("foo"));                   // NULL

Even though I appended params with the name "foo", no items can be returned
when calling getFirst(),
because of name mangling.

2.) I'm not really sure how empty arrays should be represented? PHP doesn't
retain them, and they are
simply skipped. But should we do the same thing? I can't really come up
with any other sensible behavior.

$params =3D new Uri\Rfc3986\UriQueryParams()
    ->append("foo", []);

echo $params->toRfc3986String();                      // ???

3.) We wrote earlier that objects shouldn't be supported when creating
the query string from variables. But what about
backed enums? Support for them in http_build_query() was added not long
ago: https://github.com/php/php-src/pull/15650
Should we support them, right?

Regards,
M=C3=A1t=C3=A9

--000000000000aa6b4d064640e374
Content-Type: text/html; charset="UTF-8"
Content-Transfer-Encoding: quoted-printable

<div dir=3D"ltr"><div>Hi,</div><div><br></div><div class=3D"gmail_quote gma=
il_quote_container"><blockquote class=3D"gmail_quote" style=3D"margin:0px 0=
px 0px 0.8ex;border-left:1px solid rgb(204,204,204);padding-left:1ex"><div =
dir=3D"ltr"><div class=3D"gmail_quote"><div><span style=3D"font-family:Verd=
ana,Geneva,sans-serif;font-size:13.3333px">The WHATWG URL living standard d=
oes the following:</span></div><div><div style=3D"color:rgb(8,8,8)"><pre st=
yle=3D"font-family:&quot;JetBrains Mono&quot;,monospace"><span style=3D"col=
or:rgb(0,51,179)">let </span><span style=3D"color:rgb(131,0,145)">url </spa=
n>=3D <span style=3D"color:rgb(0,51,179)">new </span><span style=3D"font-st=
yle:italic">URL</span>(<span style=3D"color:rgb(6,125,23)">&#39;<a href=3D"=
https://example.com?debug&amp;foo=3Dbar&amp;debug=3D" target=3D"_blank">htt=
ps://example.com?debug&amp;foo=3Dbar&amp;debug=3D</a>&#39;</span>);<br><spa=
n style=3D"color:rgb(131,0,145)">console</span>.<span style=3D"color:rgb(14=
5,76,7)">log</span>(<br>    <span style=3D"color:rgb(131,0,145)">url</span>=
.<span style=3D"color:rgb(135,16,148)">searchParams</span>.<span style=3D"c=
olor:rgb(145,76,7)">toString</span>(), //returns debug=3D&amp;foo=3Dbar&amp=
;debug=3D&#39;<br>);</pre></div></div><div><span style=3D"font-family:Verda=
na,Geneva,sans-serif;font-size:13.3333px">the pair gets converted to [&#39;=
debug&#39; =3D&gt; &#39;&#39;]. The roundtrip does not conserve the query s=
tring as is but all key/pair (tuples) are present.</span></div></div></div>=
</blockquote><div><br></div><div>Yes, confirmed. Unfortunately, WHATWG URL =
only supports string values, so there&#39;s no way to support</div><div>que=
ry parameters without a key (e.g. ?debug) in the RFC implementation either.=
 :(</div><div><br></div><div>On the other hand, the RFC 3986 implementation=
 supports this notion,=C2=A0even uriparser calls this out</div><div>in its =
documentation: <a href=3D"https://uriparser.github.io/doc/api/latest/index.=
html#querystrings">https://uriparser.github.io/doc/api/latest/index.html#qu=
erystrings</a>.</div><div>=C2=A0</div><div>However, there are a few other p=
roblems which came up when I was updating my implementation.</div><div><br>=
</div><div>1.) Yesterday, I wrote that name mangling of the query params sh=
ouldn&#39;t happen. However, as I realized,</div><div>it is still needed fo=
r non-list arrays, because the [..] suffix must be added to their name:</di=
v><div><br></div><div>$params =3D new Uri\Rfc3986\UriQueryParams()<br>=C2=
=A0 =C2=A0 -&gt;append(&quot;foo&quot;, [2 =3D&gt; &quot;bar&quot;, 4 =3D&g=
t; &quot;baz&quot;]);<br><br>echo $params-&gt;toRfc3986String();=C2=A0 =C2=
=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 // foo%5=
B2%5D=3Dbar&amp;foo%5B4%5D=3Dbaz</div><div><br></div><div>var_dump($params-=
&gt;getFirst(&quot;foo&quot;));=C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =
=C2=A0 =C2=A0 =C2=A0 =C2=A0// NULL</div><div><br></div><div>Even though I a=
ppended params with the name &quot;foo&quot;, no items can be returned when=
 calling getFirst(),</div><div>because of name mangling.</div><div><br></di=
v><div>2.) I&#39;m not really sure how empty arrays should=C2=A0be represen=
ted? PHP doesn&#39;t retain them, and they are</div><div>simply skipped. Bu=
t should we do the same thing? I can&#39;t really come up with any other se=
nsible behavior.</div><div><br></div><div>$params =3D new Uri\Rfc3986\UriQu=
eryParams()<br>=C2=A0 =C2=A0 -&gt;append(&quot;foo&quot;, []);<br><br>echo =
$params-&gt;toRfc3986String();=C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=
=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 // ???</div><div><br></div><div>3.) We wrot=
e earlier that objects shouldn&#39;t be supported when creating the=C2=A0qu=
ery string from variables. But what about</div><div>backed enums? Support f=
or them in http_build_query() was added not long ago: <a href=3D"https://gi=
thub.com/php/php-src/pull/15650">https://github.com/php/php-src/pull/15650<=
/a></div><div>Should we support them, right?</div><div><br></div><div>Regar=
ds,</div><div>M=C3=A1t=C3=A9</div><div><br></div></div></div>

--000000000000aa6b4d064640e374--