I don't want to do that, but I did think that a simpler alternative would be to inhibit pg_restore from attempting to parse FUNCTION entries. I can't see any strong need for it to do so.
I don't like hard-coding stuff based on the TOC tags; but we *might* be able to get away with a more general rule: do not parse if it's an object definition (as opposed to data).
In the longer term I think we will need to continue to parse TOC entries. In playing around with pg_dump(all), I put user definitions in one TOC entry. For those, we will need to add as many users as possible and ignore individual failures (something we can't to if a single multi-statement string is sent to the backend). Other TOC entries may need to be atomic. Not sure.
If the patch is not kosher, then I'd vote for adding a "do not parse" flag on the TOC entries when dumping them. Or a statement count.
----------------------------------------------------------------
Philip Warner | __---_____
Albatross Consulting Pty. Ltd. |----/ - \
(A.B.N. 75 008 659 498) | /(@) ______---_
Tel: (+61) 0500 83 82 81 | _________ \
Fax: (+61) 03 5330 3172 | ___________ |
Http://www.rhyme.com.au | / \|
| --________--
PGP key available upon request, | /
and from pgp.mit.edu:11371 |/
---------------------------(end of broadcast)--------------------------- TIP 2: you can get off all lists at once with the unregister command (send "unregister YourEmailAddressHere" to [EMAIL PROTECTED])
