Re: patch: CHECK FUNCTION statement

On 14 November 2011 20:54, Pavel Stehule <pavel(dot)stehule(at)gmail(dot)com> wrote:
> Hello
>
> 2011/11/14 Thom Brown <thom(at)linux(dot)com>:
>> On 6 October 2011 12:52, Pavel Stehule <pavel(dot)stehule(at)gmail(dot)com> wrote:
>>>
>>> Hello
>>>
>>> I am sending a version with regress tests and basic documentation
>>
>> Hi Pavel,
>>
>> I think this sentence needs rewriting:
>>
>> "checkfunction is the name of a previously registered function that
>> will be called when a new function in the language is created, to
>> check the function by statemnt CHECK FUNCTION or CHECK TRIGGER."
>>
>> to something like:
>>
>> "checkfunction is the name of an existing function that will be called
>> whenever a CHECK FUNCTION or CHECK TRIGGER is requested on a function
>> written in the language."
>>
>> And shouldn't this apply to ALTER LANGUAGE too?
>>
>> And there seem to be copy/paste symptoms in
>> doc/src/sgml/ref/check_function.sgml where it shows the definition of
>> CREATE FUNCTION and CREATE TRIGGER instead of CHECK FUNCTION and CHECK
>> TRIGGER.
>>
>> In src/include/nodes/parsenodes.h there's the error message "there are
>> no plan for query:".  This should probably read "there is no plan for
>> query:".  This appears more than once.
>>
>> And "cannot to identify real type for record type variable" doesn't
>> sound right.  Firstly "to" shouldn't be in there, and referring to a
>> "real" type is ambiguous as there is a data type called "real".  This
>> appears at least twice.
>
> I am not native speaker, so please, fix documentation as you like.

Well I wasn't entirely confident my interpretations were correct. I'd
prefer to have a rebased patch I can fully apply first, and then I can
provide a corrective patch as I'd like to test it too.

--
Thom Brown
Twitter: @darkixion
IRC (freenode): dark_ixion
Registered Linux user: #516935

EnterpriseDB UK: http://www.enterprisedb.com
The Enterprise PostgreSQL Company