Fabien COELHO writes:
> I notice that buf.data is not freed. I guess that the server memory
> management will recover it.
Yeah, it's in the transient context holding all of the results of
reading the file. I considered pfree'ing it at the end of the
function, but I concluded there's no point.
Hello Tom,
Accordingly, I borrowed some code from that thread and present
the attached revision. I also added some test coverage, since
that was lacking before, and wordsmithed docs and comments slightly.
Hearing no comments, pushed that way.
Thanks for the fixes and improvements!
I
I wrote:
> Accordingly, I borrowed some code from that thread and present
> the attached revision. I also added some test coverage, since
> that was lacking before, and wordsmithed docs and comments slightly.
Hearing no comments, pushed that way.
regards, tom lane
Fabien COELHO writes:
> [ pg-hba-cont-4.patch ]
I looked this over and I think it seems reasonable, but there's
something else we should do. If people are writing lines long
enough that they need to continue them, how long will it be
before they overrun the line length limit? Admittedly,
In the attached v3, I've tried to clarify comments and doc about tokenization
rules relating to comments, strings and continuations.
Attached v4 improves comments & doc as suggested by Justin.
--
Fabien.diff --git a/doc/src/sgml/client-auth.sgml b/doc/src/sgml/client-auth.sgml
index
Hello David,
+Agree. However, it would nice to update the sentence below if I understand
it correctly.
"+ Comments, whitespace and continuations are handled in the same way as
in" pg_hba.conf
In the attached v3, I've tried to clarify comments and doc about
tokenization rules relating
On 2020-04-01 10:25 p.m., Fabien COELHO wrote:
Hello,
FWIW, I don't think so. Generally a trailing backspace is an escape
character for the following newline. And '\ ' is a escaped space,
which is usualy menas a space itself.
In this case escape character doesn't work generally and I think
Hi Justin,
There are no assumption about backslash escaping, quotes and such, which
seems reasonable given the lexing structure of the files, i.e. records of
space-separated words, and # line comments.
Quoting does allow words containing spaces:
Ok.
I meant that the continuation handling
On Thu, Apr 02, 2020 at 07:25:36AM +0200, Fabien COELHO wrote:
>
> Hello,
>
> > FWIW, I don't think so. Generally a trailing backspace is an escape
> > character for the following newline. And '\ ' is a escaped space,
> > which is usualy menas a space itself.
> >
> > In this case escape
Hello,
FWIW, I don't think so. Generally a trailing backspace is an escape
character for the following newline. And '\ ' is a escaped space,
which is usualy menas a space itself.
In this case escape character doesn't work generally and I think it is
natural that a backslash in the middle
At Thu, 02 Apr 2020 00:20:12 +, David Zhang wrote
in
> Hi Fabien,
> Should we consider the case "\ ", i.e. one or more spaces after the backslash?
> For example, if I replace a user map
> "mymap /^(.*)@mydomain\.com$ \1" with
> "mymap /^(.*)@mydomain\.com$ \ "
> "\1"
> by
Hi Fabien,
Should we consider the case "\ ", i.e. one or more spaces after the backslash?
For example, if I replace a user map
"mymap /^(.*)@mydomain\.com$ \1" with
"mymap /^(.*)@mydomain\.com$ \ "
"\1"
by adding one extra space after the backslash, then I got the pg_role="\\"
but
Hello Justin,
thanks for the feedback.
- Records cannot be continued across lines.
+ Records can be backslash-continued across lines.
Maybe say: "lines ending with a backslash are logically continued on the next
line", or similar.
I tried to change it along that.
Since it puts a
Hi,
On Wed, Mar 25, 2020 at 07:09:38PM +0100, Fabien COELHO wrote:
>
> Hello,
>
> After writing an unreadable and stupidly long line for ldap authentification
> in a "pg_hba.conf" file, I figured out that allowing continuations looked
> simple enough and should just be done.
I tried this
Hello,
After writing an unreadable and stupidly long line for ldap
authentification in a "pg_hba.conf" file, I figured out that allowing
continuations looked simple enough and should just be done.
Patch attached.
--
Fabien.diff --git a/doc/src/sgml/client-auth.sgml
15 matches
Mail list logo
