Re: Remove Value node struct - Mailing list pgsql-hackers

From Peter Eisentraut
Subject Re: Remove Value node struct
Date
Msg-id af4727ba-9dad-9eb1-fdee-d5e600951c6a@enterprisedb.com
Whole thread Raw
In response to Re: Remove Value node struct  (Kyotaro Horiguchi <horikyota.ntt@gmail.com>)
List pgsql-hackers
On 08.09.21 04:04, Kyotaro Horiguchi wrote:
> At Tue, 7 Sep 2021 11:22:24 +0200, Peter Eisentraut <peter.eisentraut@enterprisedb.com> wrote in
>> On 30.08.21 04:13, Kyotaro Horiguchi wrote:
>>> +    else if (IsA(obj, Integer))
>>> +        _outInteger(str, (Integer *) obj);
>>> +    else if (IsA(obj, Float))
>>> +        _outFloat(str, (Float *) obj);
>>> I felt that the type enames are a bit confusing as they might be too
>>> generic, or too close with the corresponding binary types.
>>> -    Node       *arg;            /* a (Value *) or a (TypeName *) */
>>> +    Node       *arg;
>>> Mmm. It's a bit pity that we lose the generic name for the value
>>> nodes.
>>
>> Not sure what you mean here.
> 
> The member arg loses the information on what kind of nodes are to be
> stored there. Concretely it just removes the comment "a (Value *) or a
> (TypeName *)". If the (Value *) were expanded in a straight way, the
> comment would be "a (Integer *), (Float *), (String *), (BitString *),
> or (TypeName *)". I supposed that the member loses the comment because
> it become too long.

Ok, I added the comment back in in a modified form.

The patches have been committed now.  Thanks.



pgsql-hackers by date:

Previous
From: "Drouvot, Bertrand"
Date:
Subject: Re: Minimal logical decoding on standbys
Next
From: Pavel Stehule
Date:
Subject: Re: [PROPOSAL] new diagnostic items for the dynamic sql