Re: Feature Request: DROP ALL FROM DATABASE database_name

From: Thomas Swan <tswan-lst(at)ics(dot)olemiss(dot)edu>
To: "Marc G(dot) Fournier" <scrappy(at)hub(dot)org>
Cc: Tom Lane <tgl(at)sss(dot)pgh(dot)pa(dot)us>, pgsql-hackers(at)postgresql(dot)org
Subject: Re: Feature Request: DROP ALL FROM DATABASE database_name
Date: 2002-01-09 06:06:02
Message-ID: 3C3BDDCA.8080902@ics.olemiss.edu
Views: Raw Message | Whole Thread | Download mbox | Resend email
Thread:
Lists: pgsql-hackers

<!DOCTYPE html PUBLIC "-//W3C//DTD HTML 4.01 Transitional//EN">
<html>
<head>
<title></title>
</head>
<body>
Marc G. Fournier wrote:<br>
<blockquote type="cite" cite="mid:20020109010848(dot)D21205-100000(at)earth(dot)hub(dot)org">
<pre wrap="">On Tue, 8 Jan 2002, Tom Lane wrote:<br><br></pre>
<blockquote type="cite">
<pre wrap="">Thomas Swan <a class="moz-txt-link-rfc2396E" href="mailto:tswan-lst(at)ics(dot)olemiss(dot)edu">&lt;tswan-lst(at)ics(dot)olemiss(dot)edu&gt;</a> writes:<br></pre>
<blockquote type="cite">
<pre wrap="">This could be useful for cleaning out a database (sequences, triggers,<br>rules) without deleting it and without having to create it, thus<br>preserving permissions/owner information.<br></pre>
</blockquote>
<pre wrap="">What permissions/owner information? There won't be any left if we<br>remove everything in the database.<br><br>Also, given the possibility that the database has been created from<br>a nonempty template, it's less than clear exactly what should be<br>removed.<br><br>I'd say DROP and CREATE DATABASE is a perfectly presentable way of<br>handling this.<br></pre>
</blockquote>
<pre wrap=""><!----><br>&gt;From an ISP standpoint, this is something that I'd definitely love to see<br>... would save getting requests from users that don't have access to<br>anything but their database from having to ask me to DROP/CREATE their<br>database :)</pre>
</blockquote>
Marc, I agree. &nbsp; One of the disadvantages of using&nbsp;PostgreSQL in a multiuser
multidatabase environment is the user and access controls that are currently
available in PostgreSQL.<br>
<br>
I have dreamed of the ability to go GRANT ALL ON DATABASE database_name TO
USER username and then add a table and them be able to have permissions to
access that table. &nbsp; It would almost be worth having a pg_access system table
that had the OID or database_name, uid/gid, and rights as an array that could
be a gateway to the database. <br>
<br>
<pre> pg_access<br>+----------+--------+--------------------------+<br>| database | userid | rights |<br>+----------+--------+--------------------------+<br>| foo | 101 | select,update |<br>| foo | 102 | select,update,create,drop|<br>+----------+--------+--------------------------+<br></pre>
I'm not sure of the actual information to store, i.e. OID or name, but I
think it would useful from a DBA standpoint and from an ISP standpoint as
well. &nbsp; Currently a GRANT ALL ON only does current objects, but will not
include future objects (an added table for example). &nbsp; I do think an&nbsp; overall
database level permission would be most advantageous. <br>
<br>
The proposed drop command would be a step in making this more useable and
hopefully increase the userbase and popularity of PostgreSQL.<br>
<br>
Personally, I think the developer group has done an incredible and commendable
job so far. &nbsp; And, there's no other database that I would prefer to use.
&nbsp; However, the user and rights management is a bit awkward for me. &nbsp; This
is why I mentioned the above ideas.<br>
<br>
<br>
<pre><br></pre>
------------<br>
</body>
</html>

Attachment Content-Type Size
unknown_filename text/html 3.2 KB

In response to

Browse pgsql-hackers by date

  From Date Subject
Next Message Bruce Momjian 2002-01-09 06:07:34 Re: postmaster vs. postgres
Previous Message Tom Lane 2002-01-09 06:03:13 Re: postmaster vs. postgres