Обсуждение: missing a white space in the last sentence of 51.6. Executor

The following documentation comment has been logged on the website:

Page: https://www.postgresql.org/docs/14/executor.html
Description:

The last sentence of 51.6. Executor appears:
"...feeding that up toModifyTable to perform the insertion."

The below might be correct:
"...feeding that up to ModifyTable to perform the insertion."

PG Doc comments form <noreply@postgresql.org> writes:
> The last sentence of 51.6. Executor appears:
> "...feeding that up toModifyTable to perform the insertion."

Hm, the space is definitely there in the source file:

    A simple <command>INSERT ... VALUES</command> command creates a
    trivial plan tree consisting of a single <literal>Result</literal>
    node, which computes just one result row, feeding that up
    to <literal>ModifyTable</literal> to perform the insertion.

I also see it rendered properly in HTML when looking at a local docs
build.  I think this is some sort of spacing issue in the website
style sheets.  Other <literal> segments on the same page are also
rendered with less-than-usual spacing around them (to my eye
anyway); though this is the only one that looks like it has no
space at all.

            regards, tom lane

On 1/17/22 2:02 PM, Tom Lane wrote:
> PG Doc comments form <noreply@postgresql.org> writes:
>> The last sentence of 51.6. Executor appears:
>> "...feeding that up toModifyTable to perform the insertion."
> 
> Hm, the space is definitely there in the source file:
> 
>      A simple <command>INSERT ... VALUES</command> command creates a
>      trivial plan tree consisting of a single <literal>Result</literal>
>      node, which computes just one result row, feeding that up
>      to <literal>ModifyTable</literal> to perform the insertion.
> 
> I also see it rendered properly in HTML when looking at a local docs
> build.  I think this is some sort of spacing issue in the website
> style sheets.  Other <literal> segments on the same page are also
> rendered with less-than-usual spacing around them (to my eye
> anyway); though this is the only one that looks like it has no
> space at all.

I don't think it's the website stylesheets; those are identical between 
both core "make" and the website itself.

It looks like the whitespace character was stripped out in the generated 
HTML on the website. My thought is that tidylib's parser did that. We 
set indent[1] to "auto", and in the tidylib docs it does indicate that 
it could expose "layout issues" in some browsers. This makes me think 
that this could be the culprit, though I haven't tested it yet.

Is there any reason why we use that setting at all, given the doc HTML 
in the website is generated + stored in the DB, and we rarely, if ever 
need to debug it?

Thanks,

Jonathan

[1] http://tidy.sourceforge.net/docs/quickref.html#indent