How to Import CSV Data from an External Source – Scenario

The UCMDB administrator must model a vehicle catalog that is stored in a CSV file.

This task includes the following steps:

1. Prerequisites

   The admin opens the CSV file and analyzes the data:

   

   The file includes the name, model, year of manufacture, and the date when the car was purchased, that is, there are four columns of data:

   1	Name	string
   2	Model	string
   3	Year of manufacture	integer
   4	Date of purchase	date

   There are three rows to the file, which means that the admin expects three CIs to be created in UCMDB.

2. Create a CIT

   The admin creates a CIT.

   1. The admin creates a CIT named Car to hold the attributes that are to be mapped to the data in the CSV file (name, model, and so on):

   

   For details, see "Create a CI Type" in the Modeling section of the UCMDB Help.

   2. During the creation of the CIT, the admin adds these attributes as follows:

   

   For details, see "Attributes Page" in the Modeling section of the UCMDB Help.

3. Create a mapping file

   The admin uses the template, mapping_template.xml, to create a mapping file that makes the information available to the Import_CSV adapter. The mapping file is located in the following folder: Adapter Management > Resources pane > External_source_import > Configuration Files.

   1. For each attribute, the admin adds a <map> marker:

   <?xml version="1.0" encoding="UTF-8"?>
   <mappings xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance" xsi:noNamespaceSchemaLocation=".\mapping_schema.xsd"
   parserClassName="com.hp.ucmdb.discovery.library.
   communication.downloader.cfgfiles.CiMappingConfigFile">
       <ci type="car">
           <map>
               <attribute>name</attribute>
               <column>0</column>
           </map>
           <map>
               <attribute>model</attribute>
               <column>1</column>
           </map>
           <map>
               <attribute>year_of_manufacture</attribute>
               <column>2</column>
           </map>
           <map>
               <attribute>date_of_purchase</attribute>
               <column>3</column>
           </map>
       </ci>
   </mappings>
   

   2. The admin then adds information about the attribute type:

   <?xml version="1.0" encoding="UTF-8"?>
   <mappings xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance" xsi:noNamespaceSchemaLocation=".\mapping_schema.xsd"
   parserClassName="com.hp.ucmdb.discovery.library.
   communication.downloader.cfgfiles.CiMappingConfigFile">
       <ci type="">
       <map>
           <attribute>name</attribute>
           <column>0</column>
       </map>
       <map>
           <attribute>model</attribute>
           <column>1</column>
       </map>
       <map>
           <attribute>year_of_manufacture</attribute>
           <column>2</column>
           <converter module="import_converters">stringToInt</converter>
       </map>
       <map>
           <attribute>date_of_purchase</attribute>
           <column>3</column>
           <converter module="import_converters">stringToDate</converter>
       </map>
   </mappings>
   

   All conversions between the values in the CSV file and the CI attributes are done by a converter. Several converter types are included in the package by default. For details, see How to Convert Strings to Numbers.

4. Run the job

   In the Integration Studio, create a new integration point.

   • Provide a name and description for the integration point.

   • Under Integration Properties > Adapter, select the Import from CSV adapter.

   • Under Adapter Properties > Data Flow Probe, select the Data Flow Probe.

   • Under Adapter Properties > Trigger CI instance select:

   • Select Existing CI (if you have a valid, existing CI). The Select Existing CI pane appears. Select the CI; or
   • Create New CI (if you need to create a new CI). The Topology CI Creation Wizard appears. Complete the creation of the CI using the Wizard.

   Note For details on the Topology CI Creation Wizard, see "Topology CI Creation Wizard" in the Data Flow Management section of the UCMDB Help.

   • Save the Integration Point.

   • Run the job.

   Note For details on running an integration job, see "Integration Studio" in the Data Flow Management section of the UCMDB Help.

5. Add the discovered Shell CI to the job

   Note This step only applies if using UCMDB 9.03 and earlier.

   After activation, the admin locates the Shell CI (of the machine where the cars.csv file is located) and adds it to the job. For details, see "Choose CIs to Add Dialog Box" in the Data Flow Management section of the UCMDB Help.

6. Result

   The admin accesses the CIT Manager and searches for instances of the Car CIT. UCMDB finds the three instances of the CIT: