| From: | Quan Zongliang <quanzongliang(at)yeah(dot)net> |
|---|---|
| To: | pgsql-hackers(at)lists(dot)postgresql(dot)org |
| Cc: | quanzl(at)highgo(dot)com |
| Subject: | Re: stored short varlena in array |
| Date: | 2024-12-25 06:32:51 |
| Message-ID: | 36a00e3a-1875-4af1-8216-b4e0820827d0@yeah.net |
| Views: | Raw Message | Whole Thread | Download mbox | Resend email |
| Thread: | |
| Lists: | pgsql-hackers |
On 2024/12/25 14:29, Quan Zongliang wrote:
>
> Hi
>
> Now, the varlena type is stored directly in the array. Did not consider
> short varlena. If it's like fill_val(), using short varlena saves memory
> footprint and disk space.
> In TODO, there is a requirement to be implemented:
> Allow single-byte header storage for array elements
>
> This patch modifies many files.
> Based on 38da053463bef32adf563ddee5277d16d2b6c5af
> Has passed the regression test.
> But it can affect many contribs. The code needs to be adjusted. Like
> hstore and ltree.
>
Sorry I forgot the patch attachment.
| Attachment | Content-Type | Size |
|---|---|---|
| arrayelemhdr.patch | text/plain | 118.9 KB |
| From | Date | Subject | |
|---|---|---|---|
| Next Message | Hayato Kuroda (Fujitsu) | 2024-12-25 07:12:03 | RE: Object identifier types in logical replication binary mode |
| Previous Message | Quan Zongliang | 2024-12-25 06:29:12 | stored short varlena in array |