From mboxrd@z Thu Jan  1 00:00:00 1970
Return-Path: <tarantool-patches-bounce@freelists.org>
Received: from localhost (localhost [127.0.0.1])
	by turing.freelists.org (Avenir Technologies Mail Multiplex) with ESMTP id 40E8E25596
	for <tarantool-patches@freelists.org>; Tue, 13 Aug 2019 05:45:58 -0400 (EDT)
Received: from turing.freelists.org ([127.0.0.1])
	by localhost (turing.freelists.org [127.0.0.1]) (amavisd-new, port 10024)
	with ESMTP id ko4SbIL-3JeU for <tarantool-patches@freelists.org>;
	Tue, 13 Aug 2019 05:45:58 -0400 (EDT)
Received: from smtpng3.m.smailru.net (smtpng3.m.smailru.net [94.100.177.149])
	(using TLSv1.2 with cipher ECDHE-RSA-AES256-GCM-SHA384 (256/256 bits))
	(No client certificate requested)
	by turing.freelists.org (Avenir Technologies Mail Multiplex) with ESMTPS id 9530A25553
	for <tarantool-patches@freelists.org>; Tue, 13 Aug 2019 05:45:57 -0400 (EDT)
Subject: [tarantool-patches] Re: [PATCH v2 7/8] sql: get rid of FuncDef
 function hash
References: <cover.1565275469.git.kshcherbatov@tarantool.org>
 <29c44d7790720584498ca1763cb28ff98e35c71d.1565275470.git.kshcherbatov@tarantool.org>
 <20190812221150.GU32337@atlas>
 <28e7766a-0374-36ab-d848-a533106dedda@tarantool.org>
 <20190813084246.GA5284@atlas>
From: Kirill Shcherbatov <kshcherbatov@tarantool.org>
Message-ID: <b83316c4-56dd-aac1-80ff-e415581b9585@tarantool.org>
Date: Tue, 13 Aug 2019 12:45:52 +0300
MIME-Version: 1.0
In-Reply-To: <20190813084246.GA5284@atlas>
Content-Type: text/plain; charset=utf-8
Content-Language: en-US
Content-Transfer-Encoding: 7bit
Sender: tarantool-patches-bounce@freelists.org
Errors-to: tarantool-patches-bounce@freelists.org
Reply-To: tarantool-patches@freelists.org
List-Help: <mailto:ecartis@freelists.org?Subject=help>
List-Unsubscribe: <mailto:tarantool-patches-request@freelists.org?Subject=unsubscribe>
List-software: Ecartis version 1.0.0
List-Id: tarantool-patches <tarantool-patches.freelists.org>
List-Subscribe: <mailto:tarantool-patches-request@freelists.org?Subject=subscribe>
List-Owner: <mailto:>
List-post: <mailto:tarantool-patches@freelists.org>
List-Archive: <http://www.freelists.org/archives/tarantool-patches>
To: Konstantin Osipov <kostja@tarantool.org>, Tarantool MailList <tarantool-patches@freelists.org>
Cc: Nikita Pettik <korablev@tarantool.org>

>> I like this concept because it is uniform and all implementation-dependent
>> details (like the completion of the function definition initialization) are hidden
>> in the corresponding module (sql/func.c).
> 
> What does this code do? Why do you need an anonymous structure,
> instead of initializing struct func_sql_builtn object right away?
> What do you need the binary search for?
> Pre-ordering the array for the binary search is very fragile.

This new anonymous structure contain information both to complete
func_sql_builtn construction and func_def initialization.
...
        func->flags = sql_builtins[idx].flags;                                                             
	func->user_data = sql_builtins[idx].user_data;                                                     
	func->call = sql_builtins[idx].call;     
...
        def->is_deterministic = sql_builtins[idx].is_deterministic;                                        
	def->returns = sql_builtins[idx].returns;                                                          
	def->aggregate = sql_builtins[idx].aggregate;                                                      
	def->exports.sql = true;       
...

(def is an func_sql_builtin_new constructor argument)

The binary search is required to find corresponding metadata entry in
func_sql_builtin_new constructor. 

> Can this be coded in Lua, in upgrade.lua?
I mean in current implementation you need to add a snapshoot entry to
introduce a new function AND define a metadata entry in block with
binary search.