Re: SQL command to edit postgresql.conf, with comments
От | Darren Duncan |
---|---|
Тема | Re: SQL command to edit postgresql.conf, with comments |
Дата | |
Msg-id | 4CB4FC9E.6080903@darrenduncan.net обсуждение исходный текст |
Ответ на | Re: SQL command to edit postgresql.conf, with comments (Richard Broersma <richard.broersma@gmail.com>) |
Список | pgsql-hackers |
Richard Broersma wrote: > On Tue, Oct 12, 2010 at 3:54 PM, Josh Berkus <josh@agliodbs.com> wrote: >> Oh. Now that's an interesting perspective ... you're suggesting that we >> take the comments and apply them as COMMENTS on the specific pg_settings? > > On a side note regarding comments, I'd like to make a request for a > more comprehensive commenting mechanism. The first though that comes > to my mind would allow for comments to be stored and annotated using > XML or sgml. It'd be nice to be able to generate user documentation > from selected comments taken from application derived database > objects. > > I don't know, maybe this is already possible. When you start going there, you have new issues to consider. (For the record I also prefer plain text for comments.) I should point out that the group making the Perl 6 language has already been looking into such matters extensively of essentially unifying code comments and code documentation into a common metadata both accessible in the source code and programmatically at runtime. I think this approach goes beyond comments as we normally know them, which I just think of plain text strings associated with some code element. But if you want to pursue bringing documentation into this, I suggest looking at what Perl 6, and other languages, have done. While some of the results of the Perl 6 discussion may have just been in the forums, these urls at least are related to it: - http://perlcabal.org/syn/S02.html#Whitespace_and_Comments - http://github.com/perl6/specs/raw/master/S26-documentation.podfor I'm not proposing adopting their syntax, but some features or design may be useful to learn from. I also want to point out that the FoxPro language constitutes some prior art about including comments as data fields in their runtime-accessible objects. -- Darren Duncan
В списке pgsql-hackers по дате отправления: