Adding an AppData key
The administrator can add AppData keys for all users (globally) or keys can be added by other users depending on their role and permissions.
Procedure
- Display the Edit AppData window:
From Contenta Explorer, click Tools > Options > Tool Preferences.
From Contenta Web, right click on sysadmin, and then click More. From the resulting pop-up menu, click Edit AppData.
- Click the key path in the configuration tree where you want to add the new key.
The key is added at this level in the hierarchy, that is, collection_name and configuration name are keys:
- Click Tools > Options > Tool Preferences > {Global} Collections .
- Click Insert Key.
- Enter the name and value that you want for the key, and then press OK.
Where required, enter AppData information for the keys described in the following table, noting the key path for each.
Key Function AppData Key Path {Global}/Collections MARK_DB_FOR_INDEX This value instructs the Contenta Portal server to queue new, modified, and deleted objects for indexing The database will not be indexed in Solr if this setting is not TRUE.
CAUTION:If you set MARK_DB_FOR_INDEX through the Contenta Explorer (CE) user interface or through the Contenta Web (CW) interface, you and all other CE and CW users connected to the same database must exit and log in again. Some changes made by users to the database will not be indexed if a user does not do so.The upgrade_56 executable and the newdb.pl Perl script set this value (
trueorfalse) based on user input. You run these using the command line from theContenta_home/bindirectory.RETRY Setting RETRY value to trueor 1 instructs the Crawler to retry indexing objects that recently failed to index for this database (see Activity and Error Monitoring).The Crawler will change the value of this field from
truetoQUEUEDonce it has started the indexing process for the failed objects. The upgrade_56 executable and thenewdb.pl Perl script set this value (trueorfalse) based on user input.You run these using the command line from the
Contenta_home/bindirectory.AppData Key Path {Global}/Collections/collection_namewhere collection_name is a unique name for a Solr collection.
There can be only one database per Solr collection. It is named
collectionSDL_Contenta_Sample; however, the administrator can change this name while running the upgrade_56 executable or the newdb.pl Perl script, or later by editing AppData.OBJECT_TYPES This value lists object types that the Crawler will index. S1000D examples include: DModule, PModule. Object types in the list must be comma separated. If this value is empty, objects of all types, that are configurations or descendants of configurations, will be indexed.
The
upgrade_56executable and thenewdbPerl script create an empty value named OBJECT_TYPES under the collection key. The collection key is named eithercollectionSDL_Contenta_Sample, the default, or a name entered when running theupgrade_56executable and thenewdbPerl script. You run these using the command line from theContenta_home/bindirectory.URL The URL to the Solr collection. For example: http/https://localhost:8983/ solr/#/collectionSDL_Contenta_Sample The name of the collection in this URL is collectionSDL_Contenta_Sample If present, the hash and slash (
#/) will be removed by the Crawler.The
upgrade_56executable and thenewdb.plPerl script set the collection name and port number based on user input. You run these using the command line from theContenta_home/bindirectory.LANGUAGE_CODE The language code for the content. During indexing, the Crawler attempts to read the language code for each object from the Language ISO Code property data field. If this fails, the Crawler attempts to read the LANGUAGE_CODE value from this AppData field. If that also fails, the Crawler uses Apache Tika to determine the language of the object. Tika is less than 100% accurate in determining foreign languages. Tip: To improve accuracy and performance, SDL recommends that you use the Language ISO Code property data field or this AppData value to specify a language.The
upgrade_56executable and thenewdb.plPerl script set this value based on user input. You run these using the command line from theContenta_home/bindirectory.This value defaults to
enif not specified.Note: Only the language code, and not the country code, is supported. See the Foreign language support topic for information on supported language codes.DEFAULT If set to true, this collection will be used for Search; otherwise, the Contenta application developer must call thePCMSearch::SetCollection(STRING)API.This value is used by the Search in Contenta Web and in Contenta Explorer (it is not used by the Crawler).
The
upgrade_56executable and thenewdb.plPerl script set this value totrue. You run these using the command line from theContenta_home/bindirectory.AppData Key Path {Global}/Collections/collection_name/configuration_namewhere collection_name is the name of a Solr collection and configuration_name is the name of a configuration within that collection to be indexed.
IDPATH Specifies the exact path of the configuration that requires indexing. The
upgrade_56executable and the newdb.pl Perl script create the All_Configs key, but do not create the IDPATH under it; therefore, the Crawler will know to index the Contenta objects in all Contenta configurations in Solr.If you want specific configurations indexed, rather than all configurations, replace the
All_Configskey with keys having the name of each configuration to index. Under each of these keys, insert the IDPATH value of the appropriate configuration object (see Global AppData Settings).You run these using the command line from the
Contenta_home/bindirectory.STATUS_REPORTING If set to true or 1, the Crawler will write status to the INDEXING_STATUS AppData value while performing the initial indexing of the database. The
upgrade_56executable and thenewdb.plPerl script set this value totrue. You run these using the command line from theContenta_home/bindirectory.INDEXING_STATUS The Crawler creates this value and writes status to it if STATUS_REPORTING is set to trueor1. Status is written only during initial indexing.The Crawler sets this value to indicate how many objects have been indexed. Example:
5172 objects indexed, 3 objects failed.QUEUED_FOR_INDEXING This will be set to trueby the Crawler after it has queued this configuration for indexing.Once set to
true, the Crawler will not index this container again if the Crawler is subsequently restarted.If you set it to
falsethe Crawler will reindex this configuration.MAXDOCS MAXDOCS determines the maximum number of documents that can be retrieved by each search operation. The default value is 1024.You can create your own value to override the default value by adding the MAXDOCS AppData value. This value can be overridden in an advanced search query that includes the parameter:
rows=99, where99is the maximum number of rows that the query will retrieve.This value can also be overridden programmatically by calling the API methodSetSolrMaxRows.Note: This setting applies to the entire collection.AppData Key Path {Global}/Collections/collection_name/All_configswhere collection_name is a Solr collection.
The All_configs configuration contains values to be applied to all configurations within the collection. In addition to the previously described STATUS_REPORTING and QUEUED_FOR_INDEXING values, the All_configs configuration can contain the following optional value.
CONFIGS_ALREADY_INITIALLY_INDEXED When using the All_configs option, the Crawler will skip initial indexing for any configurations specified in this field, which takes as its value a semicolon-delimited list of configuration IDs within the collection, such as the following. 6;932;54;7319;88;397;253;841;9273The
CONFIGS_ALREADY_INITIALLY_INDEXEDfield must be added under theAll_configsAppData key.Because the Crawler skips initial indexing of configurations specified in this value, using it can reduce the time required for initial indexing with the All_configs option.Restriction: The value for this AppData key is limited 20,000 characters. If this maximum field length is insufficient, contact SDL Customer Support for information about an alternate method of specifying configurations to be skipped during initial indexing.