Appendices
This section contains the following appendices.
OraLoaderMetadata Utility
Use the following syntax to run the OraLoaderMetadata
utility on the Oracle Database system. You must enter the java
command on a single line, although it is shown here on multiple lines for clarity:
java oracle.hadoop.loader.metadata.OraLoaderMetadata -user userName -connection_url connection [-schema schemaName] -table tableName -output fileName.xml
To see the OraLoaderMetadata Help file, use the command with no options.
Options
- -user userName
-
The Oracle Database user who owns the target table. The utility prompts you for the password.
- -connection_url connection
-
The database connection string in the thin-style service name format:
jdbc:oracle:thin:@//hostName:port/serviceName
If you are unsure of the service name, then enter this SQL command as a privileged user:
show parameter service
NAME TYPE VALUE ------------------ ----------- ---------- service_names string orcl
- -schema schemaName
-
The name of the schema containing the target table. Unquoted values are capitalized, and unquoted values are used exactly as entered. If you omit this option, then the utility looks for the target table in the schema specified in the
-user
option. - -table tableName
-
The name of the target table. Unquoted values are capitalized, and unquoted values are used exactly as entered.
- -output fileName.xml
-
The output file name used to store the metadata document.
The following example shows how to store the target table metadata in an XML file.
Example -2 Generating Table Metadata
Run the OraLoaderMetadata utility:
$ java -cp '/tmp/oraloader-<version>-h2/jlib/*' oracle.hadoop.loader.metadata.OraLoaderMetadata -user HR -connection_url jdbc:oracle:thin://@localhost:1521/orcl.example.com -table EMPLOYEES -output employee_metadata.xml
The OraLoaderMetadata utility prompts for the database password.
Oracle Loader for Hadoop Release <version> - Production
Copyright (c) 2011, 2015, Oracle and/or its affiliates. All rights reserved.
[Enter Database Password:] password
OraLoaderMetadata creates the XML file in the same directory as the script.
$ more employee_metadata.xml
<?xml version="1.0" encoding="UTF-8"?>
<!--
Oracle Loader for Hadoop Release <version> - Production
Copyright (c) 2011, 2016, Oracle and/or its affiliates. All rights reserved.
-->
<DATABASE>
<ROWSET><ROW>
<TABLE_T>
<VERS_MAJOR>2</VERS_MAJOR>
<VERS_MINOR>5 </VERS_MINOR>
<OBJ_NUM>78610</OBJ_NUM>
<SCHEMA_OBJ>
<OBJ_NUM>78610</OBJ_NUM>
<DATAOBJ_NUM>78610</DATAOBJ_NUM>
<OWNER_NUM>87</OWNER_NUM>
<OWNER_NAME>HR</OWNER_NAME>
<NAME>EMPLOYEES</NAME>
.
.
.
Oracle Big Data Connectors Accessibility Recommendations
This section provides some tips on using screen readers and screen magnifiers with these tools.
Tips on Using Screen Readers and Braille Displays
The following are some suggestions on using text-to-speech output and braille displays.
-
Use a character mode based terminal such as Putty or Cygwin. Do not use an X-Windows-based VNC.
-
In the settings of the terminal software, set the cursor type to "block" cursor, not blinking or flashing.
-
The output of the certain commands can generate a significant amount of information and might spill off the terminal window, and the virtual window or braille display. In those, cases consider piping the output of a command through
more
in order to break the output into pages. You can then use the space bar key to page through the output. -
A few recommended screen reader settings include the following (JAWS is used here just as an example):
-
Set the JAWS cursor to "All". Use the key combination of
Insert + s
until you hear "All". -
You may need to turn off virtual cursor. If you are using JAWS, you can do this using the key combination of
Insert + z
. -
Use the virtual window to capture text. If you are using JAWS, you can do this using the key combination of
Insert + Alt + w
.
-
Tips on Using Screen Magnifiers
Examples of screen magnifiers include ZoomText, MAGic, and SuperNova.
-
Screen magnifiers can support both character-based terminals and X-Window-based VNC.
-
If you are using the screen reader function of the screen magnifier (ZoomText screen reader), then you should use a character-based terminal as described above.
-
If you are using a VNC, decide your preference for a window display, for example, TWM or ICE. A display setting for ICE can be done with the following:
vncserver -geometry 1600*950 :2
1600*950
specifies the display size, and:2
specifies the VNC display number.
Recent Change History
Changes in Oracle Big Data Connectors Release 4.12
The following are changes in Oracle Big Data Connectors for Release 4.12.
Software Updates in This Release
Connector | Version |
---|---|
Oracle SQL Connector for HDFS (OSCH) |
3.8.2 |
Oracle Loader for Hadoop (OLH) |
3.9.2 |
Oracle Shell for Hadoop Loaders (OHSH) |
1.3.2 |
Oracle XQuery for Hadoop (OXH) |
4.9.1 |
Oracle R Advanced Analytics for Hadoop (ORAAH) |
2.8.0 |
Oracle DataSource for Apache Hadoop (OD4H) |
1.3.1 |
Oracle Data Integrator (ODI) |
12.2.1.2.6 |
New Commands for Monitoring OHSH Jobs
OHSH includes four new commands to monitor jobs and a command to remove jobs.
-
ohsh> show job <job_id>
Shows detailed information about the job specified by ID.
-
ohsh> show job summary <job_id>
Shows the performance of the completed job specified by ID.
-
ohsh> show job abstract <job_id>
Provides a functional description of the job .
-
ohsh> show jobs [failed|running|completed|finished] [extended] [<integer>]
Shows the last
n
jobs of a particular job status.-
The first parameter specifies job status. If the status is not specified, all jobs are shown, regardless of job status.
-
The second parameter specifies whether to show details.
-
The third parameter specifies that the last
n
jobs of the specified status should be shown. Ifn
is not specified, then all jobs of that status are shown.
-
-
ohsh> truncate jobs [<integer>]
Removes the last
n
jobs from the database. If the integer is not specified, the command removes all jobs
See Also:
-
See the OHSH help for descriptions of all OHSH commands:
ohsh> help
-
When OHSH is installed in on-premises environments (outside of Oracle Big Data cloud services), edits to the
smartloader-conf.xml
configuration file are required in order to enable these commands.
Other Changes
In earlier releases, usage examples for each of the Oracle Big Data Connectors are automatically installed into an examples
directory under the home directory for the connector. In this release, the installation zip file for each connector includes an examples.zip
file which you can unpack when you are ready to start working with the examples.