Loading Data¶
COPY FROM
¶
COPY <table> FROM '<file pattern>' [WITH (<property> = value, ...)];
<file pattern>
must be local on the server. The file pattern can contain
wildcards if you want to load multiple files.
You can import client-side files (\copy
command in mapdql)
but it is significantly slower. For large files, MapD recommends that you first scp the file to the server, and then issue the COPY command.
<property>
in the optional WITH clause can be:
delimiter
: a single-character string for the delimiter between input fields. The default is","
, that is, as a CSV file.nulls
: a string pattern indicating a field is NULL. By default, an empty string or\N
means NULL.header
: can be either'true'
or'false'
indicating whether the input file has a header line in Line 1 that should be skipped. The default is'true'
.escape
: a single-character string for escaping quotes. The default is the quote character itself.quoted
:'true'
or'false'
indicating whether the input file contains quoted fields. The default is'true'
.quote
: a single-character string for quoting a field. The default quote character is double quote"
. All characters inside quotes are imported “as is,” except for line delimiters.line_delimiter
a single-character string for terminating each line. The default is"\n"
.array
: a two-character string consisting of the start and end characters surrounding an array. The default is{}
. For example, data to be inserted into a table with a string array in the second column (e.g.BOOLEAN, STRING[], INTEGER
) can be written astrue,{value1,value2,value3},3
.array_delimiter
: a single-character string for the delimiter between input values contained within an array. The default is","
.threads
number of threads for performing the data import. The default is the number of CPU cores on the system.max_reject
number of records that the COPY statement allows to be rejected before terminating the COPY command. Records can be rejected for a number of reasons: for example, invalid content in a field, or an incorrect number of columns. The details of the rejected records are reported in the ERROR log. COPY returns a message identifying how many records are rejected. The records that are not rejected are inserted into the table, even if the COPY stops due to the max_reject count being reached. The default is 100,000.
Note: by default, the CSV parser assumes one row per line. To import a
file with multiple lines in a single field, specify threads = 1
in
the WITH
clause.
Examples:
COPY tweets from '/tmp/tweets.csv' WITH (nulls = 'NA');
COPY tweets from '/tmp/tweets.tsv' WITH (delimiter = '\t', quoted = 'false');
COPY tweets from '/tmp/*' WITH (header='false');
SQL Importer
¶
java -cp [MapD JDBC driver]:[3rd party JDBC driver]
com.mapd.utility.SQLImporter -t [MapD table name] -su [external source user]
-sp [external source password] -c "jdbc:[external
source]://server:port;DatabaseName=some_database" -ss "[select statement]"
usage: SQLImporter
-b,--bufferSize <arg> transfer buffer size
-c,--jdbcConnect <arg> JDBC Connection string
-d,--driver <arg> JDBC driver class
-db,--database <arg> MapD Database
-f,--fragmentSize <arg> table fragment size
-p,--passwd <arg> MapD Password
--port <arg> MapD Port
-r <arg> Row Load Limit
-s,--server <arg> MapD Server
-sp,--sourcePasswd <arg> Source Password
-ss,--sqlStmt <arg> SQL Select statement
-su,--sourceUser <arg> Source User
-t,--targetTable <arg> MapD Target Table
-tr,--truncate Truncate table if it exists
-u,--user <arg> MapD User
SQL Importer executes a select statement on another database via JDBC and brings the result set into MapD Core.
If the table doesn’t exist it will create the table in MapD Core.
If the truncate flag is set it will truncate the contents in the file.
If the file exists and truncate is not set it will fail if the table does not match the SELECT statements metadata.
It is recommended to use a service account with read-only permissions when accessing data from a remote database.
MySQL Example:
java -cp mapd-1.0-SNAPSHOT-jar-with-dependencies.jar:
mysql/mysql-connector-java-5.1.38/mysql-connector-java-5.1.38-bin.jar
com.mapd.utility.SQLImporter -t test1 -sp mypassword -su myuser
-c jdbc:mysql://localhost -ss "select * from employees.employees"
SQLServer Example:
java -cp
/path/to/mapd/bin/mapd-1.0-SNAPSHOT-jar-with-dependencies.jar:/path/to/sqljdbc4.jar
com.mapd.utility.SQLImporter -d com.microsoft.sqlserver.jdbc.SQLServerDriver -t
mapd_target_table -su source_user -sp source_pwd -c
"jdbc:sqlserver://server:port;DatabaseName=some_database" -ss "select top 10 *
from dbo.some_table"
PostgreSQL Example:
java -cp
/p/to/mapd/bin/mapd-1.0-SNAPSHOT-jar-with-dependencies.jar:
/p/to/postgresql-9.4.1208.jre6.jar
com.mapd.utility.SQLImporter -t mapd_target_table -su source_user -sp
source_pwd -c "jdbc:postgresql://server/database" -ss "select * from some_table
where transaction_date > '2014-01-01'"
StreamInsert
¶
<data stream> | StreamInsert <table> <mapd database> --host <localhost> --port 9091
-u <mapd_user> -p <mapd_pwd> --delim '\t' --batch 1000
Stream data into MapD Core by attaching the StreamInsert program onto the end of a data stream. The data stream could be another program printing to standard out, a Kafka endpoint, or any other real-time stream output. Users may specify the appropriate batch size according to the expected stream rates and desired insert frequency. The target table must already exist before attempting to stream data into the table.
Example:
cat file.tsv | /path/to/mapd/SampleCode/StreamInsert stream_example mapd --host localhost
--port 9091 -u mapd -p MapDRocks! --delim '\t' --batch 1000
HDFS
¶
Consume a CSV or Parquet file residing in HDFS into MapD Core.
Copy the MapD JDBC driver into the sqoop lib, normally /usr/lib/sqoop/lib/
Example:
sqoop-export --table alltypes --export-dir /user/cloudera/ \
--connect "jdbc:mapd:192.168.122.1:9091:mapd" \
--driver com.mapd.jdbc.MapDDriver --username mapd \
--password HyperInteractive --direct --batch
Troubleshooting: How to Avoid Duplicate Rows¶
To detect duplication prior to loading data into MapD Core Database, you can perform the following steps. For this example, the files are labeled A,B,C...Z.
Load file A into table
MYTABLE
.Run the following query.
select t1.uniqueCol from MYTABLE t1 join MYTABLE t2 on t1.uCol = t2.uCol;
There should be no rows returned; if rows are returned, your first A file is not unique.
Load file B into table
TEMPTABLE
.Run the following query.
select t1.uniqueCol from MYTABLE t1 join TEMPTABLE t2 on t1.uCol = t2.uCol;
There should be no rows returned if file B is unique. Fix B if the information is not unique using details from the selection.
Load the fixed B file into
MYFILE
.Drop table
TEMPTABLE
.Repeat steps 3-6 for the rest of the set for each file prior to loading the data to the real
MYTABLE
instance.