Talk:SQL compliance

This article was nominated for deletion on 28 October 2022. The result of the discussion was soft delete.

Should we have an indication of what version of the standard this table indicates compliance with? -- Scottedwards2000 (talk)

Maybe. I am not sure how that would look and how it would be stated and if it could be done in easy and clear way. I am mostly interested in seeing less "Unknown" values. -- Frap (talk)

The table contains the mandatory features from the ISO/ANSI SQL standard, a.k.a. Core SQL. This set of features has been the same since SQL-99 until today's SQL-2016. There is no official decision that Core SQL will be fixed forever, but it is still the general understanding among those involved. This means the version of the SQL standard can be left out. Or, use the current version. Fjerdingen (talk) 15:27, 24 March 2021 (UTC)[reply]

Oracle DROP features

I'm no Oracle expert (my experience is primarily DB2 and Postgres, with some Firebird and SQLite on the side), but I'm a bit confused about F031-13 and F031-16 being marked "No" for Oracle. Looking at the DROP TABLE and DROP VIEW references for Oracle 11g it does appear from a glance at the syntax graphs that "CASCADE CONSTRAINTS" is the mandatory default in both cases. However, the definition text below states:

Specify CASCADE CONSTRAINTS to drop all referential integrity constraints that refer to primary and unique keys in the view to be dropped. If you omit this clause, and such constraints exist, then the DROP statement fails.

Which would seem to imply to me that, when "CASCADE CONSTRAINTS" is not included, the default is effectively RESTRICT-like. Anyway, as stated I'm definitely no Oracle expert (nor do I have access to an Oracle instance to try this out) so do correct me if I'm wrong, but it did seem odd to me that a venerable engine like Oracle would lack these features! Dave (talk) 21:32, 21 April 2020 (UTC)[reply]

Hey Dave , try dbfiddle.uk - they have Oracle! Appreciate the help on this. I'm happy to help as well when I can. — Preceding unsigned comment added by Scottedwards2000 (talk • contribs) 23:58, 24 April 2020 (UTC)[reply]

Often databases do have features but they are not compliant with the SQL specifications. So then it is still a "No" even if the database actually has an equivalent or similar feature. -- Frap (talk) 09:21, 7 December 2020 (UTC)[reply]

Removed Features

The features T321-06 (ROUTINES view) and T321-07 (PARAMETERS view) can not be found in the SQL-2016 specification.

Interesting. I do not have access to the standard. Is there any new features? -- Frap (talk) 09:22, 7 December 2020 (UTC)[reply]

Features T321-06 and T321-07 are both listed in Table 2 of Annex F of Part 11, Schemata, in the 2016 edition of the SQL standard. They seem to be absent from the full list of features in Part 2 of that edition. 2A00:23C7:FF81:2C00:7967:33BE:A4AC:D61C (talk) 17:51, 25 December 2020 (UTC)PRB[reply]

SQLite SUBSTRING

A recent edit states that SQLite now supports SUBSTRING, E021-06, which is defined as:

SUBSTRING <left paren> <character value expression> FROM <start position> [ FOR <string length> ] [ USING <char length units> ] <right paren>

However, does SQLite really support this syntax? What I find at https://sqlite.org/lang_corefunc.html#substr is substring(X,Y,Z), which is not standard compliant. Fjerdingen (talk) 08:10, 30 April 2021 (UTC)[reply]

Stub?

Does this article define as a stub if most of the information is found in a diagram? Without this diagram, there would just be 1 paragraph of information, so if I’m wrong, would this not constitute as a stub? Senomo Drines (talk) 16:39, 25 August 2022 (UTC)[reply]