Re: Varlena with recursive data structures? - Mailing list pgsql-general

From Michel Pelletier
Subject Re: Varlena with recursive data structures?
Date
Msg-id CACxu=vJZEcY_R3hnea9XNg+019gTvE9=wmkMV5b2_zi6UEz9Nw@mail.gmail.com
Whole thread Raw
In response to Varlena with recursive data structures?  (Sam Patterson <katoriasdev@gmail.com>)
List pgsql-general
Hi Karl,

I'm going down this road myself.  In addition to the files Tom Lane pointed out there is also some helpful documentation here:


On Wed, Jan 16, 2019 at 2:09 PM Sam Patterson <katoriasdev@gmail.com> wrote:
Hi all,

I've recently started developing an extension for Postgres for which I'll need to create a new variable-length base type. The type will require a tree-like structure in order to parse sufficiently, which of course probably means having some sort of recursive data structure, like a struct that has members which are pointers to itself for child nodes. After doing some research, specifically looking at how other variable-length data types store their data, it seems almost all of them store the data in a binary representation, using bit masks and offsets etc in order to store/access the data whilst having an in-memory representation that's used to manipulate the data.

I presume the purpose for using this approach is because all the data in a varlena type has to be contiguous, and the moment you start using pointers this is no longer possible. So my question is, given a structure that looks something like this,

typedef struct Node 
{
    char *data;
    Node *left;
    Node *right;
} Node;

am I right in saying that I wouldn't be able to store that representation on-disk, but instead I'd have to transform it into some binary representation and back again when writing/reading respectively, are there any alternatives?

Regards,

Karl

pgsql-general by date:

Previous
From: Laurenz Albe
Date:
Subject: Re: strange slow query performance
Next
From: "David G. Johnston"
Date:
Subject: Re: strange slow query performance