bug#71879: 30.0.60; Docstring error in NOERROR argument part of re-search-forward

bug#71879: 30.0.60; Docstring error in NOERROR argument part of re-search-forward

[Gerard Vermeulen]

Hello,

The last "if" clause in the NOERROR part of the re-search-forward 
docstring
is wrong.

It reads "if it is neither nil nor t, move to the limit of search and 
return nil."

but if NOERROR equals neither nil nor t re-search-forward returns the 
start
position of the match (when found) or nil (when no match found).

See for instance: (info "(Eintr) the-the")

By the way: using the-the finds a duplication "is is" in the NEWS file 
but I
am unsure whether this merits a bug report.

Regards -- Gerard

bug#71879: 30.0.60; Docstring error in NOERROR argument part of re-search-forward

[Eli Zaretskii]

> Date: Mon, 01 Jul 2024 16:13:38 +0000
> From: Gerard Vermeulen <gerard.vermeulen@posteo.net>
> 
> Hello,
> 
> The last "if" clause in the NOERROR part of the re-search-forward 
> docstring
> is wrong.
> 
> It reads "if it is neither nil nor t, move to the limit of search and 
> return nil."
> 
> but if NOERROR equals neither nil nor t re-search-forward returns
> the start position of the match (when found) or nil (when no match
> found).

The full text of that part is as follows:

  The optional third argument NOERROR indicates how errors are handled
    when the search fails.  If it is nil or omitted, emit an error; if
    it is t, simply return nil and do nothing; if it is neither nil nor
    t, move to the limit of search and return nil.

Note the first sentence of the paragraph: this entire paragraph
describes how NOERROR affects the case when the search _fails_.  Thus,
the situation where the search succeeds is not relevant to the above
part of the doc string.

So I think this text is accurate and correct.

> By the way: using the-the finds a duplication "is is" in the NEWS
> file but I am unsure whether this merits a bug report.

Fixed, thanks.

bug#71879: 30.0.60; Docstring error in NOERROR argument part of re-search-forward

[Gerard Vermeulen]

On 01.07.2024 19:30, Eli Zaretskii wrote:
>> Date: Mon, 01 Jul 2024 16:13:38 +0000
>> From: Gerard Vermeulen <gerard.vermeulen@posteo.net>
>> 
>> Hello,
>> 
>> The last "if" clause in the NOERROR part of the re-search-forward
>> docstring
>> is wrong.
>> 
>> It reads "if it is neither nil nor t, move to the limit of search and
>> return nil."
>> 
>> but if NOERROR equals neither nil nor t re-search-forward returns
>> the start position of the match (when found) or nil (when no match
>> found).
> 
> The full text of that part is as follows:
> 
>   The optional third argument NOERROR indicates how errors are handled
>     when the search fails.  If it is nil or omitted, emit an error; if
>     it is t, simply return nil and do nothing; if it is neither nil nor
>     t, move to the limit of search and return nil.
> 
> Note the first sentence of the paragraph: this entire paragraph
> describes how NOERROR affects the case when the search _fails_.  Thus,
> the situation where the search succeeds is not relevant to the above
> part of the doc string.
> 
> So I think this text is accurate and correct.
> 
Yes, I agree that the paragraph can be read like this.

bug#71879: 30.0.60; Docstring error in NOERROR argument part of re-search-forward

[Eli Zaretskii]

> Date: Mon, 01 Jul 2024 17:48:18 +0000
> From: Gerard Vermeulen <gerard.vermeulen@posteo.net>
> Cc: 71879@debbugs.gnu.org
> 
> On 01.07.2024 19:30, Eli Zaretskii wrote:
> > The full text of that part is as follows:
> > 
> >   The optional third argument NOERROR indicates how errors are handled
> >     when the search fails.  If it is nil or omitted, emit an error; if
> >     it is t, simply return nil and do nothing; if it is neither nil nor
> >     t, move to the limit of search and return nil.
> > 
> > Note the first sentence of the paragraph: this entire paragraph
> > describes how NOERROR affects the case when the search _fails_.  Thus,
> > the situation where the search succeeds is not relevant to the above
> > part of the doc string.
> > 
> > So I think this text is accurate and correct.
> > 
> Yes, I agree that the paragraph can be read like this.

I've modified the doc string a bit to make this intent more clear, and
I'm therefore closing the bug.