Horton

Re: WIP: a way forward on bootstrap data

Tom Lane <tgl@sss.pgh.pa.us>

View thread View on postgr.es
From: Tom Lane <tgl@sss.pgh.pa.us>
To: John Naylor <jcnaylor@gmail.com>
Cc: pgsql-hackers <pgsql-hackers@postgresql.org>
Date: 2018-03-14T18:09:58Z
Lists: pgsql-hackers 
John Naylor <jcnaylor@gmail.com> writes:
> It didn't take that long to rebase the remaining parts of the
> patchset, so despite what I said above I went ahead and put them in
> version 10 (attached), this time via scripted bulk editing rather than
> as large patches.

Starting to look into this version now, but a small suggestion while
it's still fresh in mind: it might be easier, in future rounds, to
put all the files in a tarball and attach 'em as one big attachment.
At least with my mail setup, it's way easier to save off a tarball
and "tar xf" it than it is to individually save a dozen attachments.
I suspect that way might be easier on your end, too.

There's some value in posting a patchset as separate attachments
when it's possible to just apply the patches in series; Munro's patch
tester knows what to do with that, but not with a tarball AFAIK.
But in this case, there's little hope that the patch tester would
get it right anyhow.

			regards, tom lane

Commits

  1. Clarify handling of special-case values in bootstrap catalog data.

  2. Replace our traditional initial-catalog-data format with a better design.

  3. Faster partition pruning

  4. Minor cleanup in genbki.pl.

  5. Trivial adjustments in preparation for bootstrap data conversion.

  6. Remove hard-coded schema knowledge about pg_attribute from genbki.pl

  7. Minor edits to catalog files and scripts

  8. Hide most variable-length fields from Form_pg_* structs