MIME-Version: 1.0
References: <a06e51d0-a7ec-741a-8d93-780011b53cec@cmatte.me>
 <0d7d2e12-bdf9-8b7b-4b90-8380eeef7c16@postgresql.org>
In-Reply-To: <0d7d2e12-bdf9-8b7b-4b90-8380eeef7c16@postgresql.org>
From: Magnus Hagander <magnus@hagander.net>
Date: Fri, 22 Oct 2021 18:23:58 +0200
Message-ID: 
 <CABUevEw+gtEVa1sq0H5FMXuYnm1EXU6mxaqLaYn42tindmCAyQ@mail.gmail.com>
Subject: Re: [PATCH] pgweb: Add SEARCH_DSN to example settings.py
To: "Jonathan S. Katz" <jkatz@postgresql.org>
Cc: =?UTF-8?Q?C=C3=A9lestin_Matte?= <celestin.matte@cmatte.me>,
	PostgreSQL WWW <pgsql-www@lists.postgresql.org>
Content-Type: multipart/alternative; boundary="000000000000e51de205cef37012"
Archived-At: 
 <https://www.postgresql.org/message-id/CABUevEw%2BgtEVa1sq0H5FMXuYnm1EXU6mxaqLaYn42tindmCAyQ%40mail.gmail.com>
Precedence: bulk

--000000000000e51de205cef37012
Content-Type: text/plain; charset="UTF-8"
Content-Transfer-Encoding: quoted-printable

On Fri, Oct 22, 2021 at 6:10 PM Jonathan S. Katz <jkatz@postgresql.org>
wrote:

> On 10/22/21 11:08 AM, C=C3=A9lestin Matte wrote:
> > Hello,
> >
> > Defining SEARCH_DSN is necessary to install the search functionality in
> pgweb, although this is not explained anywhere.
> > This patch adds SEARCH_DSN in the example settings.py so that users kno=
w
> that this variable must be defined.
>
> ...this perhaps explains why I've never gotten the search utility to
> work in my local environment.
>
> In general I'm OK with this, but perhaps we leave the string empty with
> instructions to fill the connection string to the DB that is hosting the
> search schema.
>

I think having a default like in this patch is better. Otherwise it could
randomly connect to a database with the same name as the user for example,
which might actually happen in dev scenarios. So if we want it empty we'd
also need to add code to check if it's set or not before trying to use it -
seems easier to set the default and have it fail :)

--=20
 Magnus Hagander
 Me: https://www.hagander.net/ <http://www.hagander.net/>
 Work: https://www.redpill-linpro.com/ <http://www.redpill-linpro.com/>

--000000000000e51de205cef37012
Content-Type: text/html; charset="UTF-8"
Content-Transfer-Encoding: quoted-printable

<div dir=3D"ltr"><div dir=3D"ltr"><br></div><br><div class=3D"gmail_quote">=
<div dir=3D"ltr" class=3D"gmail_attr">On Fri, Oct 22, 2021 at 6:10 PM Jonat=
han S. Katz &lt;<a href=3D"mailto:jkatz@postgresql.org">jkatz@postgresql.or=
g</a>&gt; wrote:<br></div><blockquote class=3D"gmail_quote" style=3D"margin=
:0px 0px 0px 0.8ex;border-left:1px solid rgb(204,204,204);padding-left:1ex"=
>On 10/22/21 11:08 AM, C=C3=A9lestin Matte wrote:<br>
&gt; Hello,<br>
&gt; <br>
&gt; Defining SEARCH_DSN is necessary to install the search functionality i=
n pgweb, although this is not explained anywhere.<br>
&gt; This patch adds SEARCH_DSN in the example settings.py so that users kn=
ow that this variable must be defined.<br>
<br>
...this perhaps explains why I&#39;ve never gotten the search utility to <b=
r>
work in my local environment.<br>
<br>
In general I&#39;m OK with this, but perhaps we leave the string empty with=
 <br>
instructions to fill the connection string to the DB that is hosting the <b=
r>
search schema.<br></blockquote><div><br></div><div>I think having a default=
 like in this patch is better. Otherwise it could randomly connect to a dat=
abase with the same name as the user for example, which might actually happ=
en in dev scenarios. So if we want it empty we&#39;d also need to add code =
to check if it&#39;s set or not before trying to use it - seems easier to s=
et the default and have it fail :)</div></div><div><br></div>-- <br><div di=
r=3D"ltr" class=3D"gmail_signature"><div dir=3D"ltr"><div>=C2=A0Magnus Haga=
nder<br>=C2=A0Me: <a href=3D"http://www.hagander.net/" target=3D"_blank">ht=
tps://www.hagander.net/</a><br>=C2=A0Work: <a href=3D"http://www.redpill-li=
npro.com/" target=3D"_blank">https://www.redpill-linpro.com/</a></div></div=
></div></div>

--000000000000e51de205cef37012--