| Bug #49716 | QUAN will not display query if prefix with db_name and table begins with digit | ||
|---|---|---|---|
| Submitted: | 15 Dec 2009 16:08 | Modified: | 5 Nov 2016 10:16 |
| Reporter: | Chris Calender | Email Updates: | |
| Status: | Won't fix | Impact on me: | |
| Category: | MySQL Enterprise Monitor: Agent | Severity: | S3 (Non-critical) |
| Version: | 2.1.0.1096 | OS: | Any |
| Assigned to: | Jan Kneschke | CPU Architecture: | Any |
| Tags: | database name prefix, table beginning with digit numeral | ||
[15 Dec 2009 16:08]
Chris Calender
[12 Jan 2010 16:05]
Enterprise Tools JIRA Robot
Kay Roepke writes: tricky fix, unfortunately. the '.1' gets interpreted as a float. really, the entire d.1t should be tokenized as a 'object name' or 'table name' or something like that. this is likely to cause flex grief, though. test case added, but no fix yet.
[12 Jan 2010 16:08]
Enterprise Tools JIRA Robot
Kay Roepke writes: classdump:mysql-proxy kroepke$ bzr ci -m "Bug#49716/EM-3868: add test case, currently failing, thus commented out. cannot skip individual tests with test cases" Committing to: /Volumes/Storage/Projects/merlin/merlin-trunk/mysql-proxy/ modified tests/unit/check_sql_tokenizer.c Committed revision 950. classdump:mysql-proxy kroepke$ bzr push Using saved push location: bzr+ssh://bazaar.launchpad.net/~mysql-proxy-developers/mysql-proxy/trunk/ Pushed up to revision 950.
[27 May 2010 16:00]
Enterprise Tools JIRA Robot
Jan Kneschke writes:
Just to enhance the test and clarifying how a literal is constructed:
{noformat}
> select 1 + 1;
+-------+
| 1 + 1 |
+-------+
| 2 |
+-------+
1 row in set (0.02 sec)
> select 1+1;
+-----+
| 1+1 |
+-----+
| 2 |
+-----+
1 row in set (0.00 sec)
> select 1e+1;
+------+
| 1e+1 |
+------+
| 10 |
+------+
1 row in set (0.00 sec)
> SELECT 1e + 1;
ERROR 1054 (42S22): Unknown column '1e' in 'field list'
> SELECT 1e+a;
ERROR 1054 (42S22): Unknown column '1e' in 'field list'
{noformat}
[27 May 2010 16:00]
Enterprise Tools JIRA Robot
Jan Kneschke writes:
Pushed to proxy's trunk:
revno: 1099
committer: jan@mysql.com
branch nick: trunk
timestamp: Thu 2010-05-27 16:19:43 +0200
message:
added entry about #49716 - literals starting with digits
modified:
NEWS
------------------------------------------------------------
revno: 1098
committer: jan@mysql.com
branch nick: trunk
timestamp: Thu 2010-05-27 16:17:23 +0200
message:
don't capture the opening parantheses to make sure we count them properly
* regression added in last changeset
modified:
lib/sql-tokenizer.l
------------------------------------------------------------
revno: 1097
committer: jan@mysql.com
branch nick: trunk
timestamp: Thu 2010-05-27 15:58:55 +0200
message:
parse specifiers made up of unquoted_lit.unquoted_lit.unquoted_lit correctly
modified:
lib/sql-tokenizer.l
tests/unit/check_sql_tokenizer.c
------------------------------------------------------------
revno: 1096
committer: jan@mysql.com
branch nick: trunk
timestamp: Thu 2010-05-27 14:54:22 +0200
message:
allow scientific floats to have a [[:digit:]]. part
modified:
lib/sql-tokenizer.l
tests/unit/check_sql_tokenizer.c
------------------------------------------------------------
revno: 1095
committer: jan@mysql.com
branch nick: trunk
timestamp: Thu 2010-05-27 14:49:46 +0200
message:
added support for floats in scientific notation
modified:
lib/sql-tokenizer.l
tests/unit/check_sql_tokenizer.c
------------------------------------------------------------
revno: 1094
fixes bug(s): http://bugs.mysql.com/49716
committer: jan@mysql.com
branch nick: trunk
timestamp: Thu 2010-05-27 14:21:36 +0200
message:
allow literals to start with digits (fixes 49716)
* this is a first round fix, we have sort out how floats in scientific notation
get handled correctly
modified:
lib/sql-tokenizer.l
tests/unit/check_sql_tokenizer.c
[7 Jun 2010 23:43]
Enterprise Tools JIRA Robot
Andy Bang writes: In build 2.2.2.1722.
[8 Jun 2010 19:00]
Enterprise Tools JIRA Robot
Diego Medina writes: I still see SELECT * FROM `d` ? `t`
[5 Nov 2016 10:16]
Mark Leith
This is all now done via the aggregator, not proxy/agent directly, closing this bug.
