◆ build()
| Pipeline fiftyone.devicedetection.DeviceDetectionOnPremisePipelineBuilder.build | ( | ) | throws Exception |
Build and return a pipeline that can perform device detection.
- Returns
- the built pipeline
- Exceptions
-
- Exception - on error
◆ setAllowUnmatched()
DeviceDetectionOnPremisePipelineBuilder fiftyone.devicedetection.DeviceDetectionOnPremisePipelineBuilder.setAllowUnmatched ( boolean allow ) If set to false, a non-matching User-Agent will result in properties without set values.
If set to true, a non-matching User-Agent will cause the 'default profiles' to be returned. This means that properties will always have values (i.e. no need to check .hasValue) but some may be inaccurate. By default, this is false.
- Parameters
-
- allow - true if results with no matched hash nodes should be considered valid
- Returns
- this builder
◆ setAutoUpdate()
DeviceDetectionOnPremisePipelineBuilder fiftyone.devicedetection.DeviceDetectionOnPremisePipelineBuilder.setAutoUpdate ( boolean enabled ) Enable/Disable auto update.
Defaults to enabled. If enabled, the auto update system will automatically download and apply new data files for device detection.
- Parameters
-
- enabled - True to enable auto update. False to disable.
- Returns
- This builder instance.
◆ setConcurrency()
DeviceDetectionOnPremisePipelineBuilder fiftyone.devicedetection.DeviceDetectionOnPremisePipelineBuilder.setConcurrency ( int concurrency ) Set the expected number of concurrent operations using the engine.
This sets the concurrency of the internal caches to avoid excessive locking.
- Parameters
-
- concurrency - expected concurrent accesses
- Returns
- this builder
◆ setDataFileSystemWatcher()
DeviceDetectionOnPremisePipelineBuilder fiftyone.devicedetection.DeviceDetectionOnPremisePipelineBuilder.setDataFileSystemWatcher ( boolean enabled ) The DataUpdateService has the ability to watch a file on disk and refresh the engine as soon as that file is updated.
This setting enables/disables that feature.
- Parameters
-
- enabled - True to enable file system watcher. False to disable.
- Returns
- This builder instance.
◆ setDataUpdateLicenseKey()
DeviceDetectionOnPremisePipelineBuilder fiftyone.devicedetection.DeviceDetectionOnPremisePipelineBuilder.setDataUpdateLicenseKey ( String key ) Set the license key used when checking for new device detection data files.
Defaults to null.
- Parameters
-
- key - The license key.
- Returns
- This builder instance.
◆ setDataUpdateOnStartup()
DeviceDetectionOnPremisePipelineBuilder fiftyone.devicedetection.DeviceDetectionOnPremisePipelineBuilder.setDataUpdateOnStartup ( boolean enabled ) Enable/Disable update on startup.
Defaults to enabled. If enabled, the auto update system will be used to check for an update before the device detection engine is created. If an update is available, it will be downloaded and applied before the pipeline is built and returned for use so this may take some time.
- Parameters
-
- enabled - True to enable update on startup. False to disable.
- Returns
- This builder instance.
◆ setDataUpdateService()
DeviceDetectionOnPremisePipelineBuilder fiftyone.devicedetection.DeviceDetectionOnPremisePipelineBuilder.setDataUpdateService ( DataUpdateService dataUpdateService ) Automatic updates require a DataUpdateService.
- Parameters
-
- dataUpdateService - an instance of a dataUpdateService
- Returns
- This builder instance.
◆ setDifference()
DeviceDetectionOnPremisePipelineBuilder fiftyone.devicedetection.DeviceDetectionOnPremisePipelineBuilder.setDifference ( int difference ) Set the maximum difference to allow when processing HTTP headers.
The meaning of difference depends on the Device Detection API being used. The difference is the difference in hash value between the hash that was found, and the hash that is being searched for. By default this is 0.
- Parameters
-
- difference - to allow
- Returns
- this builder
◆ setDrift()
DeviceDetectionOnPremisePipelineBuilder fiftyone.devicedetection.DeviceDetectionOnPremisePipelineBuilder.setDrift ( int drift ) Set the maximum drift to allow when matching hashes.
If the drift is exceeded, the result is considered invalid and values will not be returned. By default this is 0.
- Parameters
-
- drift - to set
- Returns
- this builder
◆ setPerformanceProfile()
DeviceDetectionOnPremisePipelineBuilder fiftyone.devicedetection.DeviceDetectionOnPremisePipelineBuilder.setPerformanceProfile ( Constants.PerformanceProfiles profile ) Set the performance profile for the device detection engine.
Defaults to balanced.
- Parameters
-
- profile - The performance profile to use.
- Returns
- This builder instance.
◆ setProperty()
DeviceDetectionOnPremisePipelineBuilder fiftyone.devicedetection.DeviceDetectionOnPremisePipelineBuilder.setProperty ( String property ) Add a property to the list of properties that the engine will populate in the response.
By default all properties will be populated.
- Parameters
-
- property - the property that we want the engine to populate
- Returns
- this builder
◆ setShareUsage()
DeviceDetectionOnPremisePipelineBuilder fiftyone.devicedetection.DeviceDetectionOnPremisePipelineBuilder.setShareUsage ( boolean enabled ) Set share usage enabled/disabled.
Defaults to enabled.
- Parameters
-
- enabled - True to enable usage sharing. False to disable.
- Returns
- This builder instance.
◆ setUpdatePollingInterval()
DeviceDetectionOnPremisePipelineBuilder fiftyone.devicedetection.DeviceDetectionOnPremisePipelineBuilder.setUpdatePollingInterval ( int pollingIntervalSeconds ) Set the time between checks for a new data file made by the DataUpdateService in seconds.
Default = 30 minutes.
- Parameters
-
- pollingIntervalSeconds - The number of seconds between checks.
- Returns
- This builder instance.
◆ setUpdatePollingIntervalMillis()
DeviceDetectionOnPremisePipelineBuilder fiftyone.devicedetection.DeviceDetectionOnPremisePipelineBuilder.setUpdatePollingIntervalMillis ( long pollingIntervalMillis ) Set the time between checks for a new data file made by the DataUpdateService in milliseconds.
- Parameters
-
- pollingIntervalMillis - The number of milliseconds between checks.
- Returns
- This builder instance.
◆ setUpdateRandomisationMax()
DeviceDetectionOnPremisePipelineBuilder fiftyone.devicedetection.DeviceDetectionOnPremisePipelineBuilder.setUpdateRandomisationMax ( int randomisationMaxSeconds ) A random element can be added to the DataUpdateService polling interval.
This option sets the maximum length of this random addition. Default = 10 minutes.
- Parameters
-
- randomisationMaxSeconds - The maximum time added to the data update polling interval in seconds.
- Returns
- This builder instance.
◆ setUpdateRandomisationMaxMillis()
DeviceDetectionOnPremisePipelineBuilder fiftyone.devicedetection.DeviceDetectionOnPremisePipelineBuilder.setUpdateRandomisationMaxMillis ( long randomisationMaxMillis ) A random element can be added to the DataUpdateService polling interval.
This option sets the maximum length of this random addition. Default = 10 minutes.
- Parameters
-
- randomisationMaxMillis - The maximum time added to the data update polling interval in milliseconds.
- Returns
- This builder instance.
◆ setUsePerformanceGraph()
DeviceDetectionOnPremisePipelineBuilder fiftyone.devicedetection.DeviceDetectionOnPremisePipelineBuilder.setUsePerformanceGraph ( boolean use ) Set whether or not the performance optimized graph is used for processing.
When processing evidence, the performance graph is optimised to find an answer as quick as possible. However, this can be at the expense of finding the best match for evidence which was not in the training data. If the predictive graph is also enabled, it will be used next if there was no match in the performance graph.
- See also
- Hash Algorithm
- Parameters
-
- use - true if the performance graph should be used
- Returns
- this builder
◆ setUsePredictiveGraph()
DeviceDetectionOnPremisePipelineBuilder fiftyone.devicedetection.DeviceDetectionOnPremisePipelineBuilder.setUsePredictiveGraph ( boolean use ) Set whether or not the predictive optimized graph is used for processing.
When processing evidence, the predictive graph is optimised to find the best answer for evidence which was not in the training data. However, this is at the expense of processing time, as more possibilities are taken into consideration.
- See also
- Hash Algorithm
- Parameters
-
- use - true if the predictive graph should be used
- Returns
- this builder