Melvin,
* Melvin Davidson (melvin6925@gmail.com) wrote:
> >the query is outright wrong.
> Really? I submit a query to help and all you can do is criticize?
I explained exactly what was wrong with it, and how to write a proper
query to answer this question, with the thought that you'd appreciate
the help and would create a proper query. I'm a bit mystified as to why
you seem to be unhappy with this.
> Yes it is legal to create a table that starts with pg_, but any PostgreSQL
> developer should
> know that is not a good idea.
That's really not an excuse for providing an incorrect query when you
could simply adjust the query to properly filter on nspname instead of
relname, further, while perhaps 'pg_' instead a good prefix to use in a
PG database (note that other databases, of course, wouldn't have any
issue using such a prefix), using a table name starting with
'information' or 'sql_' isn't nearly as questionable, yet your query
also filtered those out.
> If you don't like the query, write a better one yourself, but
> kindly do not try to degrade the efforts of others.
No, pointing out that a query is wrong and doesn't acurately answer the
question is something which should be done, just as someone should call
out a bug in code that's been submitted for inclusion in PostgreSQL.
I've tried to be kind and point out exactly what's wrong, along with
encouraging words of how to adjust the query to be correct. Having
misleading and incorrect information in our list archives isn't being
helpful- those archives are preserved for more-or-less forever, end up
being returned in google search results, and have caused confusion and
even data loss in multiple cases that I'm personally aware of (thank you
to pg_resetxlog...). We should all try to work to minimize that, and to
correct it when necessary as best we can through responses.
Thanks!
Stephen