From: "n.pettik" <korablev@tarantool.org>
To: tarantool-patches@freelists.org
Cc: Imeev Mergen <imeevma@tarantool.org>
Subject: [tarantool-patches] Re: [PATCH v4 3/8] sql: replace rc with is_aborted status in struct Parse
Date: Tue, 19 Mar 2019 12:54:45 +0300 [thread overview]
Message-ID: <61975B3B-0FDC-4117-88F9-4EB3A6EC247B@tarantool.org> (raw)
In-Reply-To: <20190318152804.GA24675@tarantool.org>
[-- Attachment #1: Type: text/plain, Size: 2748 bytes --]
>>>>> diff --git a/test/sql-tap/check.test.lua b/test/sql-tap/check.test.lua
>>>>> index bab6493..0d8bf15 100755
>>>>> --- a/test/sql-tap/check.test.lua
>>>>> +++ b/test/sql-tap/check.test.lua
>>>>> @@ -516,7 +516,7 @@ test:do_catchsql_test(
>>>>> );
>>>>> ]], {
>>>>> -- <check-5.1>
>>>>> - 1, "Wrong space options (field 5): invalid expression specified (SQL error: bindings are not allowed in DDL)"
>>>>> + 1, "Wrong space options (field 5): invalid expression specified (bindings are not allowed in DDL)"
>>>>> -- </check-5.1>
>>>>> })
>>>>
>>>> Why test results have changed if you provided
>>>> non-functional refactoring?
>>> It become this way because now the error in diag instead of being
>>> only in zErrMsg of struct Parse.
>>
>> So, then it should be related to the previous patch, I guess.
>> Otherwise, still don’t understand. Or fix commit message,
>> since now it implies that only refactoring was provided.
>> Or, what is better - move functional changes to separate patch.
>>
> I divided this ptch into two:
> "sql: remove argument pzErrMsg from sqlRunParser()"
> "sql: replace rc with is_aborted status in struct Parse”
Both patches are OK.
>>> index a2937a0..c2e5d6b 100644
>>> --- a/src/box/sql.c
>>> +++ b/src/box/sql.c
>>> @@ -1363,12 +1363,8 @@ sql_checks_resolve_space_def_reference(ExprList *expr_list,
>>>
>>> sql_resolve_self_reference(&parser, def, NC_IsCheck, NULL, expr_list);
>>> int rc = 0;
>>> - if (parser.rc != SQL_OK) {
>>> - /* Tarantool error may be already set with diag. */
>>> - if (parser.rc != SQL_TARANTOOL_ERROR)
>>> - diag_set(ClientError, ER_SQL, parser.zErrMsg);
>>> + if (parser.is_aborted)
>>> rc = -1;
>>> - }
>>> sql_parser_destroy(&parser);
>>> return rc;
>>
>> diff --git a/src/box/sql.c b/src/box/sql.c
>> index c2e5d6bd1..ea71dd101 100644
>> --- a/src/box/sql.c
>> +++ b/src/box/sql.c
>> @@ -1362,9 +1362,6 @@ sql_checks_resolve_space_def_reference(ExprList *expr_list,
>> parser.parse_only = true;
>>
>> sql_resolve_self_reference(&parser, def, NC_IsCheck, NULL, expr_list);
>> - int rc = 0;
>> - if (parser.is_aborted)
>> - rc = -1;
>> sql_parser_destroy(&parser);
>> - return rc;
>> + return parser.is_aborted ? -1 : 0;
>> }
>>
> I am not sure that this should be applied. I think it isn't right
> to look at field is_aborted of parser after parser destruction.
sql_parser_destroy() doesn’t affect is_abort field. On the other
hand, mb you are right, and using object after destructor is called
may seem strange. Anyway, replace pls ‘if' with ternary operator:
3 lines of code fit into one.
[-- Attachment #2: Type: text/html, Size: 9343 bytes --]
next prev parent reply other threads:[~2019-03-19 9:54 UTC|newest]
Thread overview: 43+ messages / expand[flat|nested] mbox.gz Atom feed top
2019-03-13 17:03 [tarantool-patches] [PATCH v4 0/8] sql: use diag_set() for errors in SQL imeevma
2019-03-13 17:03 ` [tarantool-patches] [PATCH v4 1/8] sql: rework syntax errors imeevma
2019-03-14 18:24 ` [tarantool-patches] " n.pettik
2019-03-14 18:28 ` Imeev Mergen
2019-03-15 14:09 ` Kirill Yukhin
2019-03-13 17:03 ` [tarantool-patches] [PATCH v4 2/8] sql: set SQL parser errors via diag_set() imeevma
2019-03-14 19:26 ` [tarantool-patches] " n.pettik
2019-03-14 19:36 ` n.pettik
2019-03-18 15:06 ` Mergen Imeev
2019-03-19 9:41 ` n.pettik
2019-03-19 11:24 ` Kirill Yukhin
2019-03-13 17:03 ` [tarantool-patches] [PATCH v4 3/8] sql: replace rc with is_aborted status in struct Parse imeevma
2019-03-14 19:53 ` [tarantool-patches] " n.pettik
2019-03-18 15:28 ` Mergen Imeev
2019-03-19 9:54 ` n.pettik [this message]
2019-03-19 13:17 ` Kirill Yukhin
2019-03-13 17:03 ` [tarantool-patches] [PATCH v4 4/8] sql: remove field nErr from " imeevma
2019-03-14 19:58 ` [tarantool-patches] " n.pettik
2019-03-19 13:27 ` Kirill Yukhin
2019-03-13 17:03 ` [tarantool-patches] [PATCH v4 5/8] sql: remove field zErrMsg " imeevma
2019-03-14 22:15 ` [tarantool-patches] " n.pettik
2019-03-19 13:20 ` Kirill Yukhin
2019-03-13 17:03 ` [tarantool-patches] [PATCH v4 6/8] sql: rework three errors of "unsupported" type imeevma
2019-03-14 22:15 ` [tarantool-patches] " n.pettik
2019-03-19 13:30 ` Kirill Yukhin
2019-03-13 17:03 ` [tarantool-patches] [PATCH v4 7/8] sql: rework semantic errors imeevma
2019-03-15 15:49 ` [tarantool-patches] " n.pettik
2019-03-22 12:48 ` Mergen Imeev
2019-03-26 14:14 ` n.pettik
2019-03-26 16:56 ` Mergen Imeev
2019-03-26 18:16 ` n.pettik
2019-03-26 19:20 ` Mergen Imeev
2019-03-26 21:36 ` n.pettik
2019-03-27 6:48 ` Kirill Yukhin
2019-03-13 17:03 ` [tarantool-patches] [PATCH v4 8/8] sql: remove sqlErrorMsg() imeevma
2019-03-15 13:36 ` [tarantool-patches] " n.pettik
2019-03-25 18:47 ` Mergen Imeev
2019-03-26 13:34 ` n.pettik
2019-03-26 17:52 ` Mergen Imeev
2019-03-26 18:28 ` n.pettik
2019-03-26 19:21 ` Mergen Imeev
2019-03-26 21:36 ` n.pettik
2019-03-27 6:49 ` Kirill Yukhin
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=61975B3B-0FDC-4117-88F9-4EB3A6EC247B@tarantool.org \
--to=korablev@tarantool.org \
--cc=imeevma@tarantool.org \
--cc=tarantool-patches@freelists.org \
--subject='[tarantool-patches] Re: [PATCH v4 3/8] sql: replace rc with is_aborted status in struct Parse' \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox