CONTENTANALYZER Service
- Technische Redaktion
This core service is responsible for the analysis of binary content files during their import to yuuvis® Momentum.
Characteristics
| Service Name | contentanalyzer |
|---|---|
| Port Range | 7430-7439 |
| Profiles | prod, docker, kubernetes, metrics |
| Helm Chart | yuuvis |
Function
In the default configuration, each binary content file imported to yuuvis® Momentum passes the CONTENTANALYZER service. Its mime type is determined and the contained text is extracted for the most common file types.
>> Basic Use Case Flows
Text Extraction
The CONTENTANALYZER can extract the text contained in an imported binary content file. The extracted plain text is stored as text rendition in the search index and will be used for the full-text search together with all values of string properties within the metadata.
As of 2023 Summer, the length of extracted full-text is limited by the configurable parameter as described below. Thus, overloads and downtimes of the CONTENTANALYZER service due to huge content files with much text can be avoided.
The file types for which text extraction is available are listed here:
>> Renditions
The text rendition can be retrieved via the following endpoint:
>> GET /api/dms/objects/{objectId}/contents/renditions/text
Mime Type Determination
The CONTENTANALYZER service is responsible for the determination of the mime type of binary content files that are imported to yuuvis® Momentum. The determined mime type is stored in the content stream properties section of the corresponding DMS object.
Configuration
The default behavior of the CONTENTANALYZER service can be changed via serviceConfiguration.json configuration file. The analysis of the content and/or mime type can be requested or not, depending on the defined conditions. If a condition matches during an import process, the content and/or mime type will be analyzed.
>> serviceConfiguration.json
Note: Within each import request body, this configuration can be overwritten by specifying the options parameters accordingly. The analysis of content and/or mime type can be requested or suppressed even if the opposite behavior is configured in the file serviceConfiguration.json.
Furthermore, it is possible to set the following parameters in a service-specific configuration file for the CONTENTANALYZER:
| Parameter | Type | Description | Default |
|---|---|---|---|
extraction.exclusiveOfficeLock | boolean | If you need text extraction for large binary content files of Microsoft Office file types, the CONTENTANALYZER service might need its full memory for each single file to be processed. If Note: Nevertheless, sufficient RAM is required for the CONTENTANALYZER service. | false |
extraction.maxTextLengthInKB | integer | Available as of 2023 Summer. Limit for the length of extracted full-text. If an extraction process reaches the limit, it is stopped and the full-text created till then is stored. | 2048 |
mimetype.extension.redetection | comma-separated list of mime types | The standard calculation is based on the analysis of the binary content itself. In case a determined mime type is wrong, it is possible to reanalyze the file considering the file ending. The mime types for which this second analysis step should be triggered are listed here. | 'image/x-portable-greymap' |