[ 
https://issues.apache.org/jira/browse/DRILL-805?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14005366#comment-14005366
 ] 

Chun Chang commented on DRILL-805:
----------------------------------

same thing. when we cast string to time, we add date component (1970-01-01T) 
and timezone.

0: jdbc:drill:schema=dfs> select c_row, cast(c_time as varchar(20)), 
cast(c_time as time) from data;
+------------+------------+------------+
|   c_row    |   EXPR$1   |   EXPR$2   |
+------------+------------+------------+
| 1          | 00:00:00   | 1970-01-01T00:00:00.000-08:00 |
| 2          | 01:00:00   | 1970-01-01T01:00:00.000-08:00 |
| 3          | 02:03:00   | 1970-01-01T02:03:00.000-08:00 |
| 4          | 11:59:00   | 1970-01-01T11:59:00.000-08:00 |
| 5          | 12:00:00   | 1970-01-01T12:00:00.000-08:00 |
| 6          | 12:01:00   | 1970-01-01T12:01:00.000-08:00 |
| 7          | 23:59:00   | 1970-01-01T23:59:00.000-08:00 |
| 8          | 23:59:59.99 | 1970-01-01T23:59:59.990-08:00 |
| 9          | 15:36:39   | 1970-01-01T15:36:39.000-08:00 |
| 10         | 15:36:39   | 1970-01-01T15:36:39.000-08:00 |
| 11         | 00:01:00   | 1970-01-01T00:01:00.000-08:00 |
| 12         | 01:10:00   | 1970-01-01T01:10:00.000-08:00 |
| 13         | 02:33:00   | 1970-01-01T02:33:00.000-08:00 |
| 14         | 23:59:00   | 1970-01-01T23:59:00.000-08:00 |
| 15         | 12:03:00   | 1970-01-01T12:03:00.000-08:00 |
| 16         | 12:31:00   | 1970-01-01T12:31:00.000-08:00 |
| 17         | 19:59:00   | 1970-01-01T19:59:00.000-08:00 |
| 18         | 11:59:59.99 | 1970-01-01T11:59:59.990-08:00 |
| 19         | 15:37:39   | 1970-01-01T15:37:39.000-08:00 |
| 20         | 15:36:39   | 1970-01-01T15:36:39.000-08:00 |
| 21         | 11:59:59.99 | 1970-01-01T11:59:59.990-08:00 |
| 22         | 15:37:39   | 1970-01-01T15:37:39.000-08:00 |
| 23         | 15:36:39   | 1970-01-01T15:36:39.000-08:00 |

> when we cast a string to date, we add time component and timezone too, which 
> is wrong
> -------------------------------------------------------------------------------------
>
>                 Key: DRILL-805
>                 URL: https://issues.apache.org/jira/browse/DRILL-805
>             Project: Apache Drill
>          Issue Type: Bug
>          Components: Execution - Operators
>            Reporter: Chun Chang
>            Assignee: Jinfeng Ni
>
> tested with git.commit.id.abbrev=2fad21d
> c_date column holds string of date information. when cast it into date type, 
> we add time component, as well as timezone. and the timezone value seems to 
> be random.
> 0: jdbc:drill:schema=dfs> select c_row, cast(c_date as varchar(20)), 
> cast(c_date as date) from data where c_row < 23;
> +------------+------------+------------+
> |   c_row    |   EXPR$1   |   EXPR$2   |
> +------------+------------+------------+
> | 1          | 1957-04-09 | 1957-04-09T00:00:00.000-08:00 |
> | 2          | 1957-06-13 | 1957-06-13T00:00:00.000-07:00 |
> | 3          | 1996-02-28 | 1996-02-28T00:00:00.000-08:00 |
> | 4          | 1996-01-29 | 1996-01-29T00:00:00.000-08:00 |
> | 5          | 1996-03-01 | 1996-03-01T00:00:00.000-08:00 |
> | 6          | 1996-03-02 | 1996-03-02T00:00:00.000-08:00 |
> | 7          | 1997-02-28 | 1997-02-28T00:00:00.000-08:00 |
> | 8          | 1997-01-29 | 1997-01-29T00:00:00.000-08:00 |
> | 9          | 1997-03-01 | 1997-03-01T00:00:00.000-08:00 |
> | 10         | 1997-03-02 | 1997-03-02T00:00:00.000-08:00 |
> | 11         | 2000-04-01 | 2000-04-01T00:00:00.000-08:00 |
> | 12         | 2000-04-02 | 2000-04-02T00:00:00.000-08:00 |
> | 13         | 2000-04-03 | 2000-04-03T00:00:00.000-07:00 |
> | 14         | 2038-04-08 | 2038-04-08T00:00:00.000-07:00 |
> | 15         | 2039-04-09 | 2039-04-09T00:00:00.000-07:00 |
> | 16         | 2040-04-10 | 2040-04-10T00:00:00.000-07:00 |
> | 17         | 1999-01-08 | 1999-01-08T00:00:00.000-08:00 |
> | 18         | 1999-02-08 | 1999-02-08T00:00:00.000-08:00 |
> | 19         | 1999-03-08 | 1999-03-08T00:00:00.000-08:00 |
> | 20         | 1999-01-18 | 1999-01-18T00:00:00.000-08:00 |
> | 21         | 2003-01-02 | 2003-01-02T00:00:00.000-08:00 |
> | 22         | 1999-04-08 | 1999-04-08T00:00:00.000-07:00 |
> +------------+------------+------------+
> 22 rows selected (0.1 seconds)
> physical plan for the query:
> 0: jdbc:drill:schema=dfs> explain plan for select c_row, cast(c_date as 
> varchar(20)), cast(c_date as date) from data where c_row < 23;
> +------------+------------+
> |    text    |    json    |
> +------------+------------+
> | ScreenPrel
>   ProjectPrel(c_row=[$1], EXPR$1=[CAST($2):VARCHAR(20) CHARACTER SET 
> "ISO-8859-1" COLLATE "ISO-8859-1$en_US$primary"], EXPR$2=[CAST($2):DATE])
>     FilterPrel(condition=[<($1, 23)])
>       ScanPrel(groupscan=[ParquetGroupScan [entries=[ReadEntryWithPath 
> [path=maprfs:/user/root/mondrian/data]], 
> selectionRoot=/user/root/mondrian/data, columns=[SchemaPath [`c_row`], 
> SchemaPath [`c_date`]]]])
>  | {
>   "head" : {
>     "version" : 1,
>     "generator" : {
>       "type" : "ExplainHandler",
>       "info" : ""
>     },
>     "type" : "APACHE_DRILL_PHYSICAL",
>     "options" : [ ],
>     "resultMode" : "EXEC"
>   },
>   "graph" : [ {
>     "pop" : "parquet-scan",
>     "@id" : 1,
>     "entries" : [ {
>       "path" : "maprfs:/user/root/mondrian/data"
>     } ],
>     "storage" : {
>       "type" : "file",
>       "connection" : "maprfs:///",
>       "workspaces" : {
>         "default" : {
>           "location" : "/user/root/mondrian/",
>           "writable" : false,
>           "storageformat" : null
>         },
>         "home" : {
>           "location" : "/",
>           "writable" : false,
>           "storageformat" : null
>         },
>         "root" : {
>           "location" : "/",
>           "writable" : false,
>           "storageformat" : null
>         },
>         "tmp" : {
>           "location" : "/tmp",
>           "writable" : true,
>           "storageformat" : "csv"
>         }
>       },
>       "formats" : {
>         "psv" : {
>           "type" : "text",
>           "extensions" : [ "tbl" ],
>           "delimiter" : "|"
>         },
>         "csv" : {
>           "type" : "text",
>           "extensions" : [ "csv" ],
>           "delimiter" : ","
>         },
>         "tsv" : {
>           "type" : "text",
>           "extensions" : [ "tsv" ],
>           "delimiter" : "\t"
>         },
>         "parquet" : {
>           "type" : "parquet"
>         },
>         "json" : {
>           "type" : "json"
>         }
>       }
>     },
>     "format" : {
>       "type" : "parquet"
>     },
>     "columns" : [ "`c_row`", "`c_date`" ],
>     "selectionRoot" : "/user/root/mondrian/data"
>   }, {
>     "pop" : "filter",
>     "@id" : 2,
>     "child" : 1,
>     "expr" : "less_than(`c_row`, 23) ",
>     "initialAllocation" : 1000000,
>     "maxAllocation" : 10000000000
>   }, {
>     "pop" : "project",
>     "@id" : 3,
>     "exprs" : [ {
>       "ref" : "`c_row`",
>       "expr" : "`c_row`"
>     }, {
>       "ref" : "`EXPR$1`",
>       "expr" : "cast( (`c_date` ) as VARCHAR(20) )"
>     }, {
>       "ref" : "`EXPR$2`",
>       "expr" : "cast( (`c_date` ) as DATE )"
>     } ],
>     "child" : 2,
>     "initialAllocation" : 1000000,
>     "maxAllocation" : 10000000000
>   }, {
>     "pop" : "screen",
>     "@id" : 4,
>     "child" : 3,
>     "initialAllocation" : 1000000,
>     "maxAllocation" : 10000000000
>   } ]
> } |



--
This message was sent by Atlassian JIRA
(v6.2#6252)

Reply via email to