- From: TAN Kuan Hui <kuanhui@mobileworkspace.com>
- Date: Wed, 6 Feb 2002 10:03:44 +0800
- To: "Per Bothner" <per@bothner.com>, "Michael Rys" <mrys@microsoft.com>
- Cc: <www-ql@w3.org>
Well, XSLT 2.0 allows variable/param typing as an option and there are advantages besides type assertion functionalities. http://lists.w3.org/Archives/Public/www-xml-query-comments/2002Jan/0293.html http://lists.w3.org/Archives/Public/www-xml-query-comments/2002Feb/0002.html I am also thinking along the line that we can put a Query-prolog centralised in an "include file" or "stored procedure" use for setting up a set of query scripts. This will make code maintenance easier. Well, param seems to be sufficient for such a purpose but I would prefer a variable as it will provide greater flexibility as such "environment variables" could be altered by the calling scripts if so desired depending on processing conditions. IMHO, its probably obvious that it would be less confusing to developers if variable/param concepts in XSLT 2.0 and XQuery 1.0 are aligned since the two will probably go hand-in-hand in many coding scenarios. Thanks. Kuan Hui ----- Original Message ----- From: "Per Bothner" <per@bothner.com> To: "Michael Rys" <mrys@microsoft.com> Cc: <www-ql@w3.org> Sent: Wednesday, February 06, 2002 8:53 AM Subject: Re: Variable Declaration in Query Prolog > Michael Rys wrote: > > That's seems right, but the grammar is not the issue. The issue is that > > LET bound variables (and FOR bound variables) are currently typed by the > > expression and not by an explicit type declaration. Asserting types is > > done by an ASSERT. > > However, the Query-prolog does seem to need some way to define a > "parameter" (such as a filename) passed in from the environment, > rather like XSLT's 'param'. This seems rather essential; is there > some reason it's missing, or perhaps I just overlooked it? > -- > --Per Bothner > per@bothner.com http://www.bothner.com/per/ >
Received on Tuesday, 5 February 2002 21:04:58 UTC