From mboxrd@z Thu Jan  1 00:00:00 1970
Return-Path: <v.shpilevoy@tarantool.org>
Received: from smtp50.i.mail.ru (smtp50.i.mail.ru [94.100.177.110])
 (using TLSv1.2 with cipher ECDHE-RSA-AES256-GCM-SHA384 (256/256 bits))
 (No client certificate requested)
 by dev.tarantool.org (Postfix) with ESMTPS id BE021469710
 for <tarantool-patches@dev.tarantool.org>;
 Wed, 10 Jun 2020 01:28:03 +0300 (MSK)
References: <cover.1590671266.git.imeevma@gmail.com>
 <536b434e0e7460b49823f205294a8cadab484877.1590671266.git.imeevma@gmail.com>
 <338cbb21-5268-0e6c-819d-94bf57e5b24d@tarantool.org>
 <e40a6f77-c92e-e66f-62c6-b7d95c339545@tarantool.org>
From: Vladislav Shpilevoy <v.shpilevoy@tarantool.org>
Message-ID: <5c205790-2b11-fbf8-59f0-006b21f2b9a4@tarantool.org>
Date: Wed, 10 Jun 2020 00:28:00 +0200
MIME-Version: 1.0
In-Reply-To: <e40a6f77-c92e-e66f-62c6-b7d95c339545@tarantool.org>
Content-Type: text/plain; charset="utf-8"
Content-Language: en-US
Content-Transfer-Encoding: 8bit
Subject: Re: [Tarantool-patches] [PATCH 1/6] sql: remove implicit cast for
 assignment
List-Id: Tarantool development patches <tarantool-patches.dev.tarantool.org>
List-Unsubscribe: <https://lists.tarantool.org/mailman/options/tarantool-patches>, 
 <mailto:tarantool-patches-request@dev.tarantool.org?subject=unsubscribe>
List-Archive: <https://lists.tarantool.org/pipermail/tarantool-patches/>
List-Post: <mailto:tarantool-patches@dev.tarantool.org>
List-Help: <mailto:tarantool-patches-request@dev.tarantool.org?subject=help>
List-Subscribe: <https://lists.tarantool.org/mailman/listinfo/tarantool-patches>, 
 <mailto:tarantool-patches-request@dev.tarantool.org?subject=subscribe>
To: Mergen Imeev <imeevma@tarantool.org>, tsafin@tarantool.org, tarantool-patches@dev.tarantool.org

Hi! Thanks for the patch!

>>> +    ]], {
>>> +        0
>>> +    })
>>> +
>>> diff --git a/test/sql-tap/numcast.test.lua b/test/sql-tap/numcast.test.lua
>>> index eeac5353a..9bbae5ca4 100755
>>> --- a/test/sql-tap/numcast.test.lua
>>> +++ b/test/sql-tap/numcast.test.lua
>>> @@ -135,7 +135,7 @@ test:do_catchsql_test(
>>>           INSERT INTO t VALUES(20000000000000000000.01);
>>>           SELECT * FROM t;
>>>       ]], {
>>> -        1,"Tuple field 1 type does not match one required by operation: expected integer"
>>> +        1,"Type mismatch: can not convert real to integer"
>> 8. Is it correct, that we can remove all the checks from OP_CheckType
>> except numeric conversions, and everything will work fine, because
>> box makes the type checks anyway? It seems OP_CheckType duplicates
>> box's work.
> This is correct in most cases. We will lose the implicit cast
> between numeric values, but this can be done, for example, during
> MakeRecord. In addition, all type mismatch errors will look the
> same as in Lua.

I didn't meant to remove number casts. Only non-numeric checks.
For example, if mem_check_type is called on a string, and target
is number, we can ignore that, and let box raise an error.

This should become even simpler, when MEM_<type> flags are
replaced with enum field_type.

Please, do it here, if it is easy, or file a new ticket (I mean
error fallthrough down to box, not MEM_<type> removal).