it's look well, but I still prefer some combination with =
name: = ''
name: = '''
:name = ''
$name = ..
$name = ..
I wonder about name := ''.
:= is used in Pascal/Ada to assign a value. Or would that again be an allowed
operator in pg ?
Andreas
--
Sent via pgsql-hackers mailing list
2008/12/15 Peter Eisentraut pete...@gmx.net:
Pavel Stehule wrote:
Because AS is signal for collecting column (or label) names.
I thing so we should use AS as Tom's proposal, together with SQL/XML
functionality.
Yes, please implement that.
I'll do it - it's better then nothing,
It's
David E. Wheeler wrote:
Perhaps not, but I have to say, looking at Robert's JSON example:
SELECT json(r.foo AS foo, r.bar AS bar, r.baz AS baz, r.bletch AS
quux) FROM rel r;
I would be pretty confused. It looks exactly like the proposed syntax
for named parameters. So while syntactically
2008/12/15 Zeugswetter Andreas OSB sIT andreas.zeugswet...@s-itsolutions.at:
it's look well, but I still prefer some combination with =
name: = ''
name: = '''
:name = ''
$name = ..
$name = ..
I wonder about name := ''.
:= is used in Pascal/Ada to assign a value. Or would that again be
On Dec 15, 2008, at 11:05 AM, Peter Eisentraut wrote:
In my mind, you just have to think about it hard enough to come to
realize that, when viewed from the right angle, the semantic
conflict might not exist after all. It's a bit tricky, but I think
it's possible.
Better for users not to
Pavel Stehule wrote:
Because AS is signal for collecting column (or label) names.
I thing so we should use AS as Tom's proposal, together with SQL/XML
functionality.
Yes, please implement that.
It's only idea: default behave is using as for param name specification,
seconf with flag maybe
On Dec 14, 2008, at 6:55 AM, Tom Lane wrote:
The whole relabeling thing seems like a seriously silly idea.
I wouldn't say that it's silly. What I do say is that it makes no
sense
to imagine that it would be used at the same time as named parameters.
The entire point of something like
2008/12/14 Greg Stark st...@enterprisedb.com:
On Sun, Dec 14, 2008 at 1:42 AM, Robert Haas robertmh...@gmail.com wrote:
What if relabeling support were to spread some more?
The only example I can think of besides XML is JSON. There might be a
few more. Basically, relabelling is a handy
On Fri, Dec 12, 2008 at 8:05 PM, Tom Lane t...@sss.pgh.pa.us wrote:
Michael Meskes mes...@postgresql.org writes:
On Fri, Dec 12, 2008 at 10:06:30AM -0500, Tom Lane wrote:
Hmm ... actually, ecpg might be a problem here anyway. I know it has
special meaning for :name, but does it allow
On Friday 12 December 2008 20:05:57 Tom Lane wrote:
Excellent. I checked that psql's colon-variable feature behaves the
same. So it looks like the proposed name: value syntax would indeed
not break any existing features. Barring better ideas I think we should
go with that one.
I personally
A Dissabte 13 Desembre 2008, Peter Eisentraut va escriure:
On Friday 12 December 2008 20:05:57 Tom Lane wrote:
Excellent. I checked that psql's colon-variable feature behaves the
same. So it looks like the proposed name: value syntax would indeed
not break any existing features. Barring
-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1
Hi,
Le 13 déc. 08 à 11:39, Peter Eisentraut a écrit :
On Friday 12 December 2008 20:05:57 Tom Lane wrote:
Excellent. I checked that psql's colon-variable feature behaves the
same. So it looks like the proposed name: value syntax would
indeed
Dimitri Fontaine dfonta...@hi-media.com writes:
Le 13 déc. 08 à 11:39, Peter Eisentraut a écrit :
I personally thought that AS was a better idea.
It seems some people want to be able to overload some default
parameters (but not others) and at the same time alias them to some
new label.
Dimitri Fontaine wrote:
-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1
Hi,
Le 13 d?c. 08 ? 11:39, Peter Eisentraut a ?crit :
On Friday 12 December 2008 20:05:57 Tom Lane wrote:
Excellent. I checked that psql's colon-variable feature behaves the
same. So it looks like the proposed
On Dec 13, 2008, at 5:05 PM, Tom Lane wrote:
However, after looking at the precedent of XMLELEMENT, it's hard to
deny
that if the SQL committee ever chose to standardize named parameters,
AS is what they would use. The chances that : would become the
standard are negligible --- that's not
David E. Wheeler da...@kineticode.com writes:
On Dec 13, 2008, at 5:05 PM, Tom Lane wrote:
However, after looking at the precedent of XMLELEMENT, it's hard to
deny that if the SQL committee ever chose to standardize named parameters,
AS is what they would use. The chances that : would
On Dec 13, 2008, at 5:19 PM, Tom Lane wrote:
I'm sure it's technically possible, but I see no redeeming social
value
in it ... we should pick one and quit repainting the bike shed.
Well, as I've said, I'm okay with AS, though it's not my favorite. I
can see the argument that it's more
David E. Wheeler da...@kineticode.com writes:
I don't suppose that the position of the label and
the value on either side of AS could be reversible, could it?
No. Consider
SELECT foo(bar AS baz) FROM ...
If the from-clause provides columns named both bar and baz, it would
be
David E. Wheeler wrote:
On Dec 13, 2008, at 5:19 PM, Tom Lane wrote:
I'm sure it's technically possible, but I see no redeeming social
value
in it ... we should pick one and quit repainting the bike shed.
Well, as I've said, I'm okay with AS, though it's not my favorite. I
can see
On 2008-12-13, at 16:19, Tom Lane wrote:
I'm sure it's technically possible, but I see no redeeming social
value
in it ... we should pick one and quit repainting the bike shed.
+1000
--
Sent via pgsql-hackers mailing list (pgsql-hackers@postgresql.org)
To make changes to your
I personally agree that AS seems more SQL-ish, but that's in the eye
of the beholder.
The argument about ambiguity in XMLELEMENT is bogus becase XMLELEMENT
doesn't (and won't) have named parameters. But it is true that
XMLELEMENT is doing something subtly different with the AS clause than
-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1
Le 13 déc. 08 à 17:05, Tom Lane a écrit :
I personally agree that AS seems more SQL-ish, but that's in the eye
of the beholder.
So do I, but I fear it's already taken for another meaning.
The argument about ambiguity in XMLELEMENT is bogus
Dimitri Fontaine dfonta...@hi-media.com writes:
What if relabeling support were to spread some more?
Spread to what? AFAICS the way that XMLELEMENT uses AS is a
single-purpose wart (much like a lot of other stuff the SQL committee
invents :-(). I do not see a need to reserve AS in function
-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1
Le 13 déc. 08 à 22:32, Tom Lane a écrit :
Spread to what? AFAICS the way that XMLELEMENT uses AS is a
single-purpose wart
Ok now that explains.
The common lisp inspired syntax is only nice if we're to avoid using
AS, which I thought was the
What if relabeling support were to spread some more?
The only example I can think of besides XML is JSON. There might be a
few more. Basically, relabelling is a handy shortcut when you are
serializing data and want to avoid specifying a list of columns and an
(almost) identical list of labels.
On Sun, Dec 14, 2008 at 1:42 AM, Robert Haas robertmh...@gmail.com wrote:
What if relabeling support were to spread some more?
The only example I can think of besides XML is JSON. There might be a
few more. Basically, relabelling is a handy shortcut when you are
serializing data and want to
The whole relabeling thing seems like a seriously silly idea. Why is
it at all a shortcut to use AS instead of , ?
Because a lot of times you don't want to relabel, so you omit the AS
label part altogether, and the label is deduced from the expression
itself. For example, I don't need to
Greg Stark st...@enterprisedb.com writes:
On Sun, Dec 14, 2008 at 1:42 AM, Robert Haas robertmh...@gmail.com wrote:
What if relabeling support were to spread some more?
The only example I can think of besides XML is JSON. There might be a
few more. Basically, relabelling is a handy
On Dec 11, 2008, at 3:42 PM, Bruce Momjian wrote:
what do you thing about?
select fce(p1,p2,p3, SET paramname1 = val, paramname2 = val)
example
select dosome(10,20,30, SET flaga = true, flagb = false)
I think AS read more naturally because you expect the parameter to
come
first, not the
2008/12/12 David E. Wheeler da...@kineticode.com:
On Dec 11, 2008, at 3:42 PM, Bruce Momjian wrote:
what do you thing about?
select fce(p1,p2,p3, SET paramname1 = val, paramname2 = val)
example
select dosome(10,20,30, SET flaga = true, flagb = false)
I think AS read more naturally
-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1
Hi,
Le 12 déc. 08 à 14:14, Ian Caulfield a écrit :
unpopular, and '=' et al conflict with operators, would verilog-style
syntax - eg function( .param(value) ) - be an idea?
Ok, time to revisit the classics then ;)
2008/12/12 Dimitri Fontaine dfonta...@hi-media.com:
-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1
Hi,
Le 12 déc. 08 à 14:14, Ian Caulfield a écrit :
unpopular, and '=' et al conflict with operators, would verilog-style
syntax - eg function( .param(value) ) - be an idea?
Ok, time to
David E. Wheeler da...@kineticode.com writes:
As a Perl hacker, I'm strongly biased toward =, but I guess AS isn't
*too* bad. At least it's the same number of characters. Is - right out?
It's just as bad as = from the perspective of usurping a probable
user-defined operator name.
I think the
On Dec 12, 2008, at 2:33 PM, Dimitri Fontaine wrote:
Ok, time to revisit the classics then ;)
http://www.gigamonkeys.com/book/functions.html#keyword-parameters
That would give us things like this:
SELECT foo(1, :name 'bar', :quantity 10);
As colon character does not appear in the list of
On Dec 12, 2008, at 2:39 PM, Tom Lane wrote:
So I think that really this is never going to fly unless it uses a
keyword-looking reserved word. And we're not going to take some short
word that's not reserved now and suddenly make it so. So, despite
Pavel's objection that the AS syntax proposal
David E. Wheeler da...@kineticode.com writes:
I'm okay with AS if that's the way it has to be, but what about a
colon right at the end of the label?
Hmm ... a colon isn't considered to be an operator name, so this
wouldn't break any existing usage. I'm a little bit worried about
what we
2008/12/12 David E. Wheeler da...@kineticode.com:
On Dec 12, 2008, at 2:39 PM, Tom Lane wrote:
So I think that really this is never going to fly unless it uses a
keyword-looking reserved word. And we're not going to take some short
word that's not reserved now and suddenly make it so. So,
How about IS or INTO?
param_name IS 3
param_name IS 'some string value'
3 INTO param_name
'some string value' INTO param_name
On Fri, Dec 12, 2008 at 8:47 AM, Pavel Stehule pavel.steh...@gmail.com wrote:
2008/12/12 David E. Wheeler da...@kineticode.com:
On Dec 12, 2008, at 2:39 PM, Tom
Rod Taylor rod.tay...@gmail.com writes:
How about IS or INTO?
IS isn't a fully reserved word, and INTO seems pretty weird for this.
(IS is a type_func_name_keyword, so maybe we could make it work anyway,
but it sounds a bit fragile.)
regards, tom lane
--
Sent via
On Dec 12, 2008, at 2:47 PM, Pavel Stehule wrote:
it's look well, but I still prefer some combination with =
name: = ''
name: = '''
:name = ''
$name = ..
$name = ..
Maybe I am too conservative
Given that the colon already indicates This label corresponds to that
value, the other operator
On Fri, Dec 12, 2008 at 09:00:52AM -0500, Rod Taylor wrote:
How about IS or INTO?
param_name IS 3
param_name IS 'some string value'
that wouldn't work with NULL would it? for example is:
a IS NULL
checking if identifier 'a' IS NULL, or if you're giving NULL to
parameter 'a'.
3 INTO
right at the end of the label? A cast is two colons, so no conflict there:
SELECT foo(1, name: 'bar', quantity: 10);
it's look well, but I still prefer some combination with =
name: = ''
name: = '''
:name = ''
$name = ..
$name = ..
hmm :( $name isn't possible
:name is in conflict with
I wrote:
Rod Taylor rod.tay...@gmail.com writes:
How about IS or INTO?
IS isn't a fully reserved word, and INTO seems pretty weird for this.
(IS is a type_func_name_keyword, so maybe we could make it work anyway,
but it sounds a bit fragile.)
Actually, there's an obvious counterexample for
David E. Wheeler wrote:
Coming to this a bit late, but it seems to me that, while it makes sense
to assign a label to a value using AS, it's kind of weird to use it to
assign a value to a label.
SELECT foo( bar = 'ick', baz = 'ack' );
SELECT foo( bar AS 'ick', baz AS 'ack' );
We could do it
I'm okay with AS if that's the way it has to be, but what about a colon
right at the end of the label? A cast is two colons, so no conflict there:
SELECT foo(1, name: 'bar', quantity: 10);
No doubt I'm missing something…
I'd just like to mention that there are two different cases to
2008/12/12 Heikki Linnakangas heikki.linnakan...@enterprisedb.com:
David E. Wheeler wrote:
Coming to this a bit late, but it seems to me that, while it makes sense
to assign a label to a value using AS, it's kind of weird to use it to
assign a value to a label.
SELECT foo( bar = 'ick', baz
On Dec 12, 2008, at 3:38 PM, Pavel Stehule wrote:
I discussed about this form with Tom.
I thing so following should be readable:
name: [ optional = ] value
SELECT foo( bar: 'ick', baz: 'ack' );
SELECT foo( bar: = 'ick', baz: = 'ack' );
or
SELECT foo( bar: = 'ick', baz: = 'ack' );
Pavel Stehule pavel.steh...@gmail.com writes:
2008/12/12 Heikki Linnakangas heikki.linnakan...@enterprisedb.com:
We could do it the other way round:
SELECT foo( 'ick' AS bar, 'ack' AS baz);
I always assumed we were talking about it this way. Everywhere else in SQL AS
is followed by the
2008/12/12 David E. Wheeler da...@kineticode.com:
On Dec 12, 2008, at 3:38 PM, Pavel Stehule wrote:
I discussed about this form with Tom.
I thing so following should be readable:
name: [ optional = ] value
SELECT foo( bar: 'ick', baz: 'ack' );
SELECT foo( bar: = 'ick', baz: = 'ack' );
On Dec 12, 2008, at 3:57 PM, Gregory Stark wrote:
In any case this is all weird. SQL isn't C and doesn't have random
bits of
punctuation involved in syntax. It uses whole words for just about
everything.
Anything you do using punctuation characters is going to look out of
place.
Well,
On Dec 12, 2008, at 3:56 PM, Pavel Stehule wrote:
Hrm. I can see that, I guess. In that case, though, I think I'd
prefer the
colon at the beginning of the parameter label:
SELECT foo( :bar = 'ick', :baz = 'ack' );
this syntax is used yet
David E. Wheeler da...@kineticode.com writes:
Hrm. I can see that, I guess. In that case, though, I think I'd prefer
the colon at the beginning of the parameter label:
SELECT foo( :bar = 'ick', :baz = 'ack' );
That's ugly, and incompatible with ecpg syntax, and what's the redeeming
value
-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1
Le 12 déc. 08 à 15:57, Gregory Stark a écrit :
These don't solve anything. There's nothing stopping you from
defining a unary
prefix operator = or =
That's why I'm preferring the common-lisp syntax of :param value, or
its variant param: value.
Heikki Linnakangas heikki.linnakan...@enterprisedb.com writes:
We could do it the other way round:
SELECT foo( 'ick' AS bar, 'ack' AS baz);
Yeah, that's the direction I had always assumed that we would use,
if AS is the chosen syntax for this.
regards, tom lane
--
On Fri, Dec 12, 2008 at 3:11 PM, Dimitri Fontaine
dfonta...@hi-media.com wrote:
That's why I'm preferring the common-lisp syntax of :param value, or its
variant param: value.
FWIW there is no such common-lisp syntax. Colon is just a regular
symbol character and :param is just a regular symbol
On Fri, Dec 12, 2008 at 10:31 AM, Greg Stark st...@enterprisedb.com wrote:
On Fri, Dec 12, 2008 at 3:11 PM, Dimitri Fontaine
dfonta...@hi-media.com wrote:
That's why I'm preferring the common-lisp syntax of :param value, or its
variant param: value.
FWIW there is no such common-lisp syntax.
On Dec 12, 2008, at 4:06 PM, Tom Lane wrote:
David E. Wheeler da...@kineticode.com writes:
Hrm. I can see that, I guess. In that case, though, I think I'd
prefer
the colon at the beginning of the parameter label:
SELECT foo( :bar = 'ick', :baz = 'ack' );
That's ugly, and incompatible
On Fri, Dec 12, 2008 at 10:06:30AM -0500, Tom Lane wrote:
Hmm ... actually, ecpg might be a problem here anyway. I know it has
special meaning for :name, but does it allow space between the colon
and the name? If it does then the colon syntax loses. If it doesn't
No. Here's the lexer rule:
Michael Meskes mes...@postgresql.org writes:
On Fri, Dec 12, 2008 at 10:06:30AM -0500, Tom Lane wrote:
Hmm ... actually, ecpg might be a problem here anyway. I know it has
special meaning for :name, but does it allow space between the colon
and the name? If it does then the colon syntax
On Dec 12, 2008, at 7:05 PM, Tom Lane wrote:
Excellent. I checked that psql's colon-variable feature behaves the
same. So it looks like the proposed name: value syntax would indeed
not break any existing features. Barring better ideas I think we
should
go with that one.
+1
Best,
David
Tom Lane wrote:
Michael Meskes mes...@postgresql.org writes:
On Fri, Dec 12, 2008 at 10:06:30AM -0500, Tom Lane wrote:
Hmm ... actually, ecpg might be a problem here anyway. I know it has
special meaning for :name, but does it allow space between the colon
and the name? If it does
Andrew Dunstan and...@dunslane.net writes:
Excellent. I checked that psql's colon-variable feature behaves the
same. So it looks like the proposed name: value syntax would indeed
not break any existing features. Barring better ideas I think we should
go with that one.
Does that mean the
2008/12/10 Pavel Stehule [EMAIL PROTECTED]:
2008/12/10 Tom Lane [EMAIL PROTECTED]:
Pavel Stehule [EMAIL PROTECTED] writes:
next argument - if we accept AS for param names, then we introduce
nonconsistent behave with SQL/XML functions.
select xmlforest(c1, c2 as foo, c3) -- there foo isn't
Pavel Stehule wrote:
2008/12/10 Pavel Stehule [EMAIL PROTECTED]:
2008/12/10 Tom Lane [EMAIL PROTECTED]:
Pavel Stehule [EMAIL PROTECTED] writes:
next argument - if we accept AS for param names, then we introduce
nonconsistent behave with SQL/XML functions.
select xmlforest(c1, c2 as
Pavel Stehule [EMAIL PROTECTED] writes:
what do you thing about?
select fce(p1,p2,p3, SET paramname1 = val, paramname2 = val)
I'm not really seeing any redeeming social value in that. It's more
keystrokes than the other; and if you dislike AS because of possible
confusion with other usages
Tom Lane [EMAIL PROTECTED] wrote:
In any case, I'm not wedded to using AS for this, and am happy to
consider other suggestions. But = isn't acceptable.
How about using a bare equals sign (or the = characters) for
parameter assignment, but require that the parameter name be prefixed
with
2008/12/11 Tom Lane [EMAIL PROTECTED]:
Pavel Stehule [EMAIL PROTECTED] writes:
what do you thing about?
select fce(p1,p2,p3, SET paramname1 = val, paramname2 = val)
I'm not really seeing any redeeming social value in that. It's more
keystrokes than the other; and if you dislike AS because
2008/12/11 Kevin Grittner [EMAIL PROTECTED]:
Tom Lane [EMAIL PROTECTED] wrote:
In any case, I'm not wedded to using AS for this, and am happy to
consider other suggestions. But = isn't acceptable.
How about using a bare equals sign (or the = characters) for
parameter assignment, but require
On Thursday 11 December 2008 17:11:28 Pavel Stehule wrote:
maybe this combination should be safe
$name = or $name - ...
it's not used everywhere
Why don't you actually just implement the whole thing first using a random,
simple, and nonconflicting syntax?
Adjusting the syntax to
2008/12/11 Peter Eisentraut [EMAIL PROTECTED]:
On Thursday 11 December 2008 17:11:28 Pavel Stehule wrote:
maybe this combination should be safe
$name = or $name - ...
it's not used everywhere
Why don't you actually just implement the whole thing first using a random,
simple, and
Pavel Stehule wrote:
How would a user recognise which of these are legal operator names?
Incidentally -- EDB selling Oracle compatibility may put me in a
questionable
position here -- the more Oracle incompatibilities in stock Postgres the
better for us. But afaik we don't emulate =
Pavel Stehule wrote:
PL/pgSQL PL/SQL ADA so using '=' is only consistent and natural.
And it is my goal.
Well, that is interesting, but in SQL we already use 'AS' in most places
where we want to assign a label to a value, so it seems AS is more
logical for SQL at this point.
2008/12/10 Bruce Momjian [EMAIL PROTECTED]:
Pavel Stehule wrote:
How would a user recognise which of these are legal operator names?
Incidentally -- EDB selling Oracle compatibility may put me in a
questionable
position here -- the more Oracle incompatibilities in stock Postgres the
2008/12/10 Bruce Momjian [EMAIL PROTECTED]:
Pavel Stehule wrote:
PL/pgSQL PL/SQL ADA so using '=' is only consistent and natural.
And it is my goal.
Well, that is interesting, but in SQL we already use 'AS' in most places
where we want to assign a label to a value, so it seems AS is
Pavel Stehule [EMAIL PROTECTED] writes:
2008/12/10 Bruce Momjian [EMAIL PROTECTED]:
Well, that is interesting, but in SQL we already use 'AS' in most places
where we want to assign a label to a value, so it seems AS is more
logical for SQL at this point.
Question is - what is label - is it
Pavel Stehule [EMAIL PROTECTED] writes:
look again
select c as foo from tab ...
select fce(c as foo) from tab ...
when you use AS as param names specification, you change meaning of
some construct via used context?
Uh, what's your point? AS changes the meaning too. For example in
select
2008/12/10 Tom Lane [EMAIL PROTECTED]:
Pavel Stehule [EMAIL PROTECTED] writes:
2008/12/10 Bruce Momjian [EMAIL PROTECTED]:
Well, that is interesting, but in SQL we already use 'AS' in most places
where we want to assign a label to a value, so it seems AS is more
logical for SQL at this point.
2008/12/10 Tom Lane [EMAIL PROTECTED]:
Pavel Stehule [EMAIL PROTECTED] writes:
look again
select c as foo from tab ...
select fce(c as foo) from tab ...
when you use AS as param names specification, you change meaning of
some construct via used context?
Uh, what's your point? AS changes
Pavel Stehule [EMAIL PROTECTED] writes:
next argument - if we accept AS for param names, then we introduce
nonconsistent behave with SQL/XML functions.
select xmlforest(c1, c2 as foo, c3) -- there foo isn't doesn't mean
use it as param foo,
It could be read as meaning that, I think.
In any
2008/12/10 Tom Lane [EMAIL PROTECTED]:
Pavel Stehule [EMAIL PROTECTED] writes:
next argument - if we accept AS for param names, then we introduce
nonconsistent behave with SQL/XML functions.
select xmlforest(c1, c2 as foo, c3) -- there foo isn't doesn't mean
use it as param foo,
It could
if I may request one simple change/addition, Probably trivial to add,
but I don't have too much time to give away now to any other project
than one that pays my debts.
The default param that's in the middle. Would it be hard, or do anyone
objects against adding 'default' keyword there, so
2008/12/9 Grzegorz Jaskiewicz [EMAIL PROTECTED]:
if I may request one simple change/addition, Probably trivial to add, but I
don't have too much time to give away now to any other project than one that
pays my debts.
The default param that's in the middle. Would it be hard, or do anyone
Grzegorz Jaskiewicz wrote:
if I may request one simple change/addition, Probably trivial to add,
but I don't have too much time to give away now to any other project
than one that pays my debts.
The default param that's in the middle. Would it be hard, or do anyone
objects against adding
Grzegorz Jaskiewicz wrote:
Ok, how about
CREATE FUNCTION FOO (one int, two float8 default 3.14, three int[]
default '{6,7,8,90}');
and than SELECT FOO( 777, DEFAULT, '{1,2,3,4,5}');
Yeah, that could be a useful feature.
--
Sent via pgsql-hackers mailing list
Grzegorz Jaskiewicz [EMAIL PROTECTED] writes:
The default param that's in the middle. Would it be hard, or do anyone
objects against adding 'default' keyword there, so one doesn't have to
substitute default param 3, when he only wants to override 2nd in
funct(1,2,3) ?
Yes, and yes. We
Decibel! wrote:
On Nov 30, 2008, at 12:04 PM, David E. Wheeler wrote:
Agreed, default values should not be a part of function signatures,
although it might be nice if ALTER FUNCTION to allow default values to
be changed.
It would be VERY nice. I routinely cut and paste an entire function
Ok, how about
CREATE FUNCTION FOO (one int, two float8 default 3.14, three int[]
default '{6,7,8,90}');
and than SELECT FOO( 777, DEFAULT, '{1,2,3,4,5}');
I have no idea what SQL standard says in that case, all I know is that
keyword DEFAULT exists in it, and is used in queries for
2008/12/9 Grzegorz Jaskiewicz [EMAIL PROTECTED]:
Ok, how about
CREATE FUNCTION FOO (one int, two float8 default 3.14, three int[] default
'{6,7,8,90}');
and than SELECT FOO( 777, DEFAULT, '{1,2,3,4,5}');
I have no idea what SQL standard says in that case, all I know is that
keyword
Pavel Stehule [EMAIL PROTECTED] writes:
select foo(777, three= '{1,2,3,4,5});
it's more safe and more readable.
... and it breaks an operator that's already in use.
I did some test, and I thing so it is implementable. I had to solve
problem with hstore module. There is defined operator =
2008/12/9 Tom Lane [EMAIL PROTECTED]:
Pavel Stehule [EMAIL PROTECTED] writes:
select foo(777, three= '{1,2,3,4,5});
it's more safe and more readable.
... and it breaks an operator that's already in use.
I did some test, and I thing so it is implementable. I had to solve
problem with
Pavel Stehule [EMAIL PROTECTED] writes:
2008/12/9 Tom Lane [EMAIL PROTECTED]:
... and it breaks an operator that's already in use.
what is acceptable workaround? I unhappy, so this symbol was used for
this minor contrib module (for this operator doesn't exists regress
test).
If you could
2008/12/9 Tom Lane [EMAIL PROTECTED]:
Pavel Stehule [EMAIL PROTECTED] writes:
2008/12/9 Tom Lane [EMAIL PROTECTED]:
... and it breaks an operator that's already in use.
what is acceptable workaround? I unhappy, so this symbol was used for
this minor contrib module (for this operator doesn't
Pavel Stehule [EMAIL PROTECTED] writes:
2008/12/9 Tom Lane [EMAIL PROTECTED]:
If you could prove that it were *only* being used by this contrib module
then I might hold still for replacing it. But you can't. The odds are
good that people have custom data types using similarly-named
it means, so we must not implement any new operator?
If the operator were called [EMAIL PROTECTED], I think you could make a good
argument that no one else is likely using that for anything.
Surely the same cannot be said of =
Of course, [EMAIL PROTECTED] is not a very convenient name for an
2008/12/9 Tom Lane [EMAIL PROTECTED]:
Pavel Stehule [EMAIL PROTECTED] writes:
2008/12/9 Tom Lane [EMAIL PROTECTED]:
If you could prove that it were *only* being used by this contrib module
then I might hold still for replacing it. But you can't. The odds are
good that people have custom
Pavel Stehule [EMAIL PROTECTED] writes:
what is problematic on GUC?
Basically, it's a bad idea to have GUCs that silently make significant
changes in the syntactic meaning of a query. We've learned that lesson
the hard way I think. There are places where we've been forced to do
it because of
Pavel Stehule [EMAIL PROTECTED] writes:
2008/12/9 Tom Lane [EMAIL PROTECTED]:
Pavel Stehule [EMAIL PROTECTED] writes:
2008/12/9 Tom Lane [EMAIL PROTECTED]:
... and it breaks an operator that's already in use.
what is acceptable workaround? I unhappy, so this symbol was used for
this minor
How would a user recognise which of these are legal operator names?
Incidentally -- EDB selling Oracle compatibility may put me in a questionable
position here -- the more Oracle incompatibilities in stock Postgres the
better for us. But afaik we don't emulate = anyways so that hardly
Pavel Stehule [EMAIL PROTECTED] writes:
PL/pgSQL PL/SQL ADA so using '=' is only consistent and natural.
And it is my goal.
[ shrug... ] Don't be too surprised when the patch gets rejected.
Oracle compatibility is nice when we can get it, but we aren't going
to break existing behavior for
2008/12/9 Tom Lane [EMAIL PROTECTED]:
Pavel Stehule [EMAIL PROTECTED] writes:
PL/pgSQL PL/SQL ADA so using '=' is only consistent and natural.
And it is my goal.
[ shrug... ] Don't be too surprised when the patch gets rejected.
Oracle compatibility is nice when we can get it, but we
1 - 100 of 122 matches
Mail list logo
