Job-Manager

To get results from the search, you must first create an indexing job. This is done using the The Job Manager. This is available after a user with administrative rights has successfully logged on (default user: admin/admin).

After successful logon, you can open the Job Manager in two ways:

  1. by entering the context /smartfinder/manager in the address line of the browser, or

  2. by clicking on the Job Manager tool in the user interface.

Administration of indexing jobs

After successful login or call you will see the following interface:

jobmanager de

All published jobs are listed here. These are described as follows:

Parameter Explanation

Title

The title of the job

Status

The current status of the indexing. The following statuses are possible:

  • inactive: the job is currently not indexed

  • scheduled: the job will be executed in a scheduling in the future

  • pending: the job is executed at the next iteration

  • executing: the job is currently indexed

Source

The source indexed by this job

status of last execution

successful or failed

Last Success

Date of the last successful indexing.

Execution

If a scheduling is configured, the next execution time is displayed here.

Number of Indexed Items

Number of documents that were included in the index during the last successfully performed indexing.

Name of the index

Name of the Solr Collection in which the data is indexed, see Cores and Indexes
Since the indexing of the jobs is done asynchronously, the number of indexed documents for a job is displayed with a delay.

Create indexing jobs

To create a new job, click on the + symbol in the upper left corner. A selection dialog box appears where you can choose the source to be indexed.

General information

Each source requires a number of parameters. Some are specific, others are general to each source.

The general parameters are:

Parameter Explanation

Title

The title of the job

Send Status Message to

comma-separated list of email addresses to which status changes are sent

Name of Index

The name of the index to which the resources of the source are to be indexed. If no value is selected, the default index is used, which is defined by the property solr.default.core.name

Scheduling

Repeats the execution of the job at specified intervals. See Scheduling Indexing Jobs

The following sections explain the specific parameters for each source.

Indexing source URL

Select this type if you want to index resources that are addressable via a URL. Examples are resources that are located on a Web page or GetCapabilities Request.

Parameter Explanation

URL

URL of the source

Filter

For Web Site Crawling only: regular expression to define link following.

Search Depth

For Web Site Crawling only: Specifies the maximum search depth within the page hierarchy.
Example 1: Harvest all links of a web page ending with .pdf
URL: https://www.example.com/dir
Filter: .*(\.(pdf))$
Search depth: 2

Results:
https://www.example.com/dir/doc.pdf is indexed
https://www.example.com/dir/1/doc2.pdf is indexed
https://www.example.com/dir/1/doc3.xdoc is not indexed
https://www.example.com/dir/1/2/doc3.pdf is not indexed

Example 2: Indexing of a Capabilities URL
Titel: WMS Demo Portal con terra
URL: http://www.example.com/geoserver/wms?Request=GetCapabilities&Service=WMS

Result: The Capabilities XML of the URL is indexed.

Indexing via Data Import Handler

Many applications store its content in a structured data store, such as a relational database. The Data Import Handler (DIH) is a feature of Apache Solr and provides a mechanism for indexing these contents.

The smart.finder allows the configuration file of a Data Import Handler to be read and included in the Job Manager. The configuration for a specific data source can be found in the Apache Solr documentation

Data Import Handler configurations are always created for a specific index. The following conventions apply in smart.finder:

  1. Configuration files must be located in the /conf directory of the relevant index.

  2. By convention, its name begins with dih- and ends with .xml

Under core0/conf/dih-sample.xml you will find the sample configuration of a Data Import Handler This shows the indexing of an ATOM feed.

To include a configuration file in the Job Manager, select the Data Import option. The following specific parameters can be specified:

Parameter Explanation

Configuration File

The dih-*.xml file to be executed on the server via this job.

type of import

  • Complete: the data source is fully indexed

  • Delta: only new data from the data source is indexed. For a delta import the configuration file must meet certain requirements, see Apache Solr documentation

A delta import can only be performed with a database as data source.

Indexing Source OGC CSW Catalog

To index ISO metadata accessible via an OGC CSW 2.0.2 interface, select the OGC CSW Catalog option. Specify the following values:

Parameter Explanation

URL

HTTP POST Endpoint of the catalog GetRecords interface

Index Distributed Catalogs

Click this option if you want to additionally index the ISO metadata accessible through the above catalog in a distributed search.

Search Depth

Only relevant if the option Index Distributed Catalogs is activated and defines the search depth for the distributed catalogs (the so-called hopCount).
Example Indexing OGC CSW Catalog

If you want to index the catalog in the Demo Portal of con terra, enter the following values:

Title: CSW Demo Portal con terra URL: http://www.example.com/soapServices/CSWStartup
Index Distributed Catalogues: Yes
Search depth: 2

The job thus defined indexes the CSW catalog and all connected catalogs up to a search depth of 2.

Indexing source directory

To index resources that exist in a local directory, select the directory option. Specify the following values:

Parameter Explanation

Directory

The base directory to be searched.

directory depth

  • Base: searches only the root directory

  • Direct: searches the root directory and direct subdirectories

  • All: searches the root directory and all subdirectories

File Types

An optional filter to restrict the files to be indexed. This is described using the glob pattern, see: What Is a Glob? .

Examples:

*.shp: only ESRI shape files are indexed

*.{xml,pdf}: only files with the extension xml or pdf are indexed

Example Indexing Directory

The following directory structure is given:

C:\data\1.pdf
C:\data\1.doc
C:\data\sub\1.xml
C:\data\sub\2.pdf
C:\data\sub\subsub\1.tiff

For example, if you want to index all XML files in the root directory and one level lower, enter the following values:

Title: Verzeichnis 'data'
Directory: C:\data
Directory depth: Direkt
File types: *.xml

If you want to index all PDF and Tiff files, enter the following values:

Title: Verzeichnis 'data'
Directory: C:\data Vezeichnistiefe:
File types: *.{pdf,tiff}

Schedule indexing jobs

In addition to starting jobs manually, you have the option of repeating them at specific times and having them run automatically. To do this, you can explicitly define a scheduling for each job. To do this, activate the Scheduling option when creating a job. You can also define a scheduling for a job later.

When?

A predefined list of values that cover specific time periods.

These are:

  • Every full hour (i.e. hourly)

  • Every day at 00:00 (i.e. daily)

  • Every Sunday at 00:00 (i.e. weekly)

  • Every 1st day of the month at 00:00 (i.e. monthly)

Cron Job

Here the temporal pattern is entered in the cron notation.

Status

Here you define whether the scheduling should be activated (scheduled) or paused (inactive).

The predefined notations cover a wide range of applications. Should you still prefer to schedule the job yourself, you can set the setting to User Defined and define your own cron job. The explanation can be found documented in the Quartz framework, which is used on the server side: Quartz Cron Trigger Tutorial

If a scheduling is defined for an indexing job, the following statuses result from this:

Status Meaning

scheduled

This is the normal state: the job is in the queue and it continuously checks whether the interval specified by the Scheduler has been reached.

pending

The interval specified by the scheduler is currently reached. The job is waiting for a free space in the execution chain.

executing

indexing of the job is running. After successful indexing, the scheduled state is set for the job again.

inactive

A scheduling is defined for the indexing job, but is currently paused.

Delete indexing jobs

Select at least one job in the Job Manager using the checkbox. Then click the - symbol in the upper left corner and confirm the deletion.

When you delete a job, all documents associated with this job are deleted from the index and are no longer available for a search.