Обсуждение: SQL keywords
The section on SQL keywords in the User's Guide needs some updating. I figured that I could generate these various lists of reserved and non-reserved keywords automatically --- and indeed I can! But now I've got 36 lists of keywords (all the set differences between SQL92 reserved/non-reserved, SQL99 reserved/non-reserved, PG reserved/non-reserved (which is a simplification)) -- none of which are empty (some keywords where dropped from reserved to non-reserved between SQL92 and 99) -- but that's more than anyone wants to know. But does anyone know what they do want to know? The list of PostgreSQL reserved words seems to be the only thing I can see as definitely essential. But which is more important: The list of words that are reserved in PG but *not* reserved in SQLxx (i.e., what problems can I expect when porting stuff to PG), or the opposite (i.e., what words should I avoid when writing portable SQL). -- Peter Eisentraut peter_e@gmx.net http://yi.org/peter-e/
On Sat, Dec 16, 2000 at 12:29:16AM +0100, Peter Eisentraut wrote: > > But does anyone know what they do want to know? > > The list of PostgreSQL reserved words seems to be the only thing I can see > as definitely essential. But which is more important: The list of words > that are reserved in PG but *not* reserved in SQLxx (i.e., what problems > can I expect when porting stuff to PG), or the opposite (i.e., what words > should I avoid when writing portable SQL). If anyone ever gets around to writing an SQL Flagger (as required for FIPS 127-2 compliance: the US Federal gov't standard that incorporates SQL92) they'll need the second list: things in PG not in the standard(s) Ross -- Open source code is like a natural resource, it's the result of providing food and sunshine to programmers, and then staying out of their way. [...] [It] is not going away because it has utility for both the developers and users independent of economic motivations. Jim Flynn, Sunnyvale, Calif.
Peter Eisentraut wrote: > > The section on SQL keywords in the User's Guide needs some updating. > > I figured that I could generate these various lists of reserved and > non-reserved keywords automatically --- and indeed I can! > > But now I've got 36 lists of keywords (all the set differences between > SQL92 reserved/non-reserved, SQL99 reserved/non-reserved, PG > reserved/non-reserved (which is a simplification)) -- none of which are > empty (some keywords where dropped from reserved to non-reserved between > SQL92 and 99) -- but that's more than anyone wants to know. > > But does anyone know what they do want to know? > > The list of PostgreSQL reserved words seems to be the only thing I can see > as definitely essential. But which is more important: The list of words > that are reserved in PG but *not* reserved in SQLxx (i.e., what problems > can I expect when porting stuff to PG), or the opposite (i.e., what words > should I avoid when writing portable SQL). It could be a good thing to have a table of the form Keyword | PostgreSQL | SQL 92 | SQL 99 | -----------+--------------+-------------+-------------+ SELECT | Reserved | Reserved | Reserved | WITH | - | - | Reserved | IN | Yes | - | Reserved | ... ----------- Hannu
Hannu Krosing writes: > It could be a good thing to have a table of the form > > Keyword | PostgreSQL | SQL 92 | SQL 99 | > -----------+--------------+-------------+-------------+ > SELECT | Reserved | Reserved | Reserved | > WITH | - | - | Reserved | > IN | Yes | - | Reserved | > ... I like that. -- Peter Eisentraut peter_e@gmx.net http://yi.org/peter-e/
Hi !
>>>>>>>>>>>>>>>>>> Message d'origine <<<<<<<<<<<<<<<<<<
Le 12/16/00, à 9:14:16 AM h, hannu@tm.ee (Hannu Krosing) vous a écrit sur
le sujet suivant Re: [DOCS] SQL keywords:
> Peter Eisentraut wrote:
> >
> > The section on SQL keywords in the User's Guide needs some updating.
> >
> > I figured that I could generate these various lists of reserved and
> > non-reserved keywords automatically --- and indeed I can!
[ bandwidth savings ... ]
> It could be a good thing to have a table of the form
> Keyword | PostgreSQL | SQL 92 | SQL 99 |
> -----------+--------------+-------------+-------------+
> SELECT | Reserved | Reserved | Reserved |
> WITH | - | - | Reserved |
> IN | Yes | - | Reserved |
Hmmm ...
Better make that
Keyword | Standard | Status |
------------+-----------+---------+
SELECT | SQL92 | Reserved|
SELECT | SQL99 | Reserved|
SELECT | PostgreSQL| Reserved|
WITH | SQL92 | Nope |
WITH | SQL99 | Reserved|
WITH | PostgeSQL | Nope |
IN | SQL92 | Nope |
IN | SQL99 | Reserved|
IN | PostgreSQL| Yep |
This form allows for greater flexibility in query/views building. Obvious
drawback : the primary key is composite.
Any thoughts ?
Emmanuel Charpentier