| From: | Chapman Flack <chap(at)anastigmatix(dot)net> |
|---|---|
| To: | Tom Lane <tgl(at)sss(dot)pgh(dot)pa(dot)us>, Andres Freund <andres(at)anarazel(dot)de> |
| Cc: | pgsql-hackers(at)lists(dot)postgresql(dot)org, Peter Eisentraut <peter(dot)eisentraut(at)enterprisedb(dot)com>, pgsql-hackers <pgsql-hackers(at)postgresql(dot)org> |
| Subject: | Re: dfmgr additional ABI version fields |
| Date: | 2021-10-07 16:56:22 |
| Message-ID: | 615F26B6.1020805@anastigmatix.net |
| Views: | Raw Message | Whole Thread | Download mbox | Resend email |
| Thread: | |
| Lists: | pgsql-hackers |
On 10/07/21 12:42, Tom Lane wrote:
> Can we make the addition be a string not a number, so that we
> could include something more useful than "1234" in the error
> message?
I was wondering the same thing, just to sidestep the "who hands out IDs"
question.
Just using a string like "EDB v" + something would probably rule out
collisions in practice. To be more formal about it, something like
the tag URI scheme [0] could be recommended. Nothing at runtime would
have to know or care about tag URI syntax; it would just match a string
with a fixed opaque prefix and some suffix. The scheme gives the developer
an easy way to construct a meaningful and reliably non-colliding string.
Surely loading libraries isn't a hot enough operation to begrudge
a strcmp.
Regards,
-Chap
| From | Date | Subject | |
|---|---|---|---|
| Next Message | Stephen Frost | 2021-10-07 16:57:12 | Re: storing an explicit nonce |
| Previous Message | Bruce Momjian | 2021-10-07 16:55:33 | Re: storing an explicit nonce |