| From: | Jim Jones <jim(dot)jones(at)uni-muenster(dot)de> |
|---|---|
| To: | Chao Li <li(dot)evan(dot)chao(at)gmail(dot)com> |
| Cc: | PostgreSQL Hackers <pgsql-hackers(at)postgresql(dot)org> |
| Subject: | Re: XML functions can exhaust backend memory with libxml2 < 2.11 |
| Date: | 2026-11-12 08:40:56 |
| Message-ID: | afd5408c-4d85-438b-8d23-7c83fe11d7c1@uni-muenster.de |
| Views: | Whole Thread | Raw Message | Download mbox | Resend email |
| Thread: | |
| Lists: | pgsql-hackers |
Hi Chao
On 12/11/2026 09:17, Chao Li wrote:
> I just have a question. Though libxml2 caches 1000 messages, if PG has to follow that?
Not necessarily. I just thought it would be a good idea to mimic the
limit libxml2 already has in newer versions.
Can we just keep the first relevant error?
In theory, a much shorter limit would suffice. I realise that 100 looks
arbitrary, but somehow deviating from libxml2 here does not feel right
either, since it's the authority in many aspects regarding xml handling.
But you're right, if we agree on PG_XML_MAX_ERRORS the limit should be
open for debate.
Thanks!
Best, Jim
| From | Date | Subject | |
|---|---|---|---|
| Next Message | Pavel Borisov | 2026-11-12 09:05:17 | Re: [PATCH] Halve peak memory allocation for toast_flatten_tuple |
| Previous Message | Chao Li | 2026-11-12 08:17:36 | Re: XML functions can exhaust backend memory with libxml2 < 2.11 |