Customizing settings for the search index
Full-text searching uses an external search index, which returns an object key as result. The object key is used to run a search query in the database. This database search query takes the permissions of the logged in user into account during the search.
You can change the following configuration settings.
-
Common | Indexing | BatchSize: Use this configuration parameter to specify the maximum number of objects to be indexed when indexing runs. The default value is 50000.
-
Common | Indexing | DefaultResultLimit: Use this configuration parameter to specify the maximum number of objects to be returned by the search index. The default value is 5000.
During the search, the search strings are broken down into tokens by the search index in use. The search terms are compared with these tokens. The following configuration parameters affect how the text is parsed.
-
Common | Indexing | IndexNonTokenChars: This configuration parameter contains the delimiters to be used for separating out the text. You can extend the configuration parameter if certain characters in the search text affect how it is divided up. Default delimiters are ().[].
-
Common | Indexing | IndexUseLegacyAnalyzer: If the configuration parameter is set, an alternative tokenizing is also performed. The alternative method of tokenizing is preferable for long tokens. For example, if the string Department_01 is a token, the partial string Department is not considered to be a token.
The following tokens are recognized by the alternative tokenizing.
Table 29: Tokens for alternative tokenizing
Words |
Sequence of letters and/or numbers |
Enumeration |
Words linked by punctuation marks (_-/.,) of which at least every second one contains a number.
An example is Department_01.
Sequences are also decimal numbers and IP addresses. |
Email addresses |
An email address is often made up of first name, last name, company name and generic top-level domain (for example .com). The order or spelling of the first and last names may vary (for example, use of initials). The special character @ and the punctuation mark (.) not only separate each part of the email address but also links them so that the email address is recognized as a token.
Examples of email addresses are Alex.Miller@example.com or A.Meyer@example.com. |
Host names |
For example website.example.com. |
Acronym |
For example U. S. A. |
Apostrophe |
For example O'Name. |
@, & surrounded by letters |
For example Me&you. |
Umlauts such as ä, ö, ü |
For example, Alex Müller. |
Configuring columns for full-text search
Certain important columns are already indexed for full-text search in the default installation. You configure more columns for full-text searching if you require.
To configure a column for full-text search
-
In the Designer, select the One Identity Manager Schema category.
-
Select the table and start the Schema Editor with the Show table definition task.
-
Select the column and then the Column properties view.
-
Select the Column tab and edit the Index weighting property.
-
If the value is less than or equal to 0, no indexing takes place.
-
If the value is greater than 0, the data value is indexed.
-
Select the Database > Save to database and click Save.
Configuring tables for full-text search
The standard installation indexes all tables that have the XDateInserted, XDateUpdated, and XObjectKey columns. Customize tables indexing using the following configuration parameters.
-
Common | Indexing | PriorityTables: Use this configuration parameter to influence the order in which the search index processes the database tables. The value entered in this configuration parameter is a comma-delimited list of table names. The search index processes the given tables in this sequence before all other tables to be indexed. The default value is Person,AccProduct.
-
Common | Indexing | ExcludeTables: Use this configuration parameter to exclude certain tables from indexing. The value entered in this configuration parameter is a comma-delimited list of table names. The listed tables are not processed by the search index. Existing data is remains intact.
Localization in One Identity Manager
One Identity Manager requires country information at different stages, for example, identity country and state assignments are accessed when email notifications are created or IT Shop workflows are being determined. Languages, time zones, public holidays, and working hours are mapped as well as countries and states. The basis data is loaded into the database during schema installation.
One Identity Manager supports language-dependent representation of data. You can use this feature to edit display text in different languages for the One Identity Manager tool user interfaces. You can also create multi-language text for process information output, script processing as well as processing messages.
The default One Identity Manager installation is supplied in the English - United States [en-US] and German - Germany [de-DE] language. You can add other languages to the user interface and display text if required. In this instance, you must translate the text before One Identity Manager goes live. There is a Language Editor in the Designer to help you do this. A special control is provided in the One Identity Manager tools that aids multi-language input.
To help you translate One Identity Manager, other languages are made available with the Web Portal Language Pack.
Detailed information about this topic