Dmitry Yemanov wrote:

>"Jim Starkey" <jas@...> wrote:
>
>
>> 1. isc_statement_info makes a available a request "token"
>>
>>
>
>How are tokens generated?
>

Anyway you want. Random number, internal addresss of request block,
counter in attachment block, system time of request compilation, or
Bush's popularity rating (that's volatile enough). Tokens are opaque,
compared only for equality, but must be unique database wide.

>
>
>
>> 2. The "request token" is used in a separate connection to kill a
>> running request. Dpb parameter? New statement? Don't care --
>> anything would work.
>>
>>
>
>Disagreed. It's much better to offer a new (or extended) API than to suggest
>people such an ugly solution.
>
>
>

Ugly is in the eye of the beholder. Someone else might say it's elegant
and simple (my dog, for example, thinks it's almost as good as pizza).
You must admit that it addresses a pressing problem without any impact
on the plumbing or API.