EXTERNAL_FORECAST_TIME
What | nameofinstance.xml |
|---|---|
Required | no |
Description | Export data (timeseries) from Delft-Fews to several file formats |
schema location | https://fewsdocs.deltares.nl/schemas/version1.0/timeSeriesExportRun.xsd |
Configuration
The export module can export timeseries for use in other systems. The configuration of the module is split into three sections:
- General: Specify file name, data type etc...
- metadata: Export specific settings
- timeseriesSets: actual data to export
In the sections below the different elements of the configuration are described
General
description
An optional description
exportTypeStandard
This type specifies which writer should be used to write the file. The type must be one from the enumeration. Presently (2007/02) only bfg and pi are included in this list.
exportType
This type specifies which writer should be used to write the file. It may be any string as long as this type is supported by the TimeSeriesExport module. The list of supported types is given here.
folder
Folder (directory) in which to store the exported files.
exportFileName
This elements describes how to construct the filename(s) of the exported file(s).
If only the name element is given a fixed name is used for each export. The prefix and suffix elements describe how to create a filename prefix and/or suffix. The temporaryPrefix is used to generate a prefix for the temporary file as it is being written. After that the file is renamed.
The option useExternalLocationIdAsName (instead of the regular file name) forces the use of the location ID or Name as filename. Note that this option does not work in combination with an idMap that filters the locations to be exported.
validate
Optional element. Only applicable if the data are exported to the xml-file. This option activates the validation of the exported file against a XML schema.
idmapId
Id of IdMap to be used for parameterId and locationId mapping
externalLocationNameFunction (since Delft-FEWS 2023.2)
For export functions that can write location names along side location ID's, one can choose the location attribute that should be parsed to the location name field. It requires mentioning the attribute between @ signs; e.g.: <externalLocationNameFunction>@externalLocationName@</externalLocationNameFunction>
unitConversionsId
Id of UnitConversions to be used for unit mapping
flagConversionsId
Id of flagConversions to be used for flag mapping
exportMissingValue/exportMissingValueString
Missing value definition for this time series. Either a string or a number. Defaults to NaN if not defined.
exportAttribute
Since 2015.02. Location, Parameter or Qualifier attribute that can be used for export. Specification is needed to include it in the data available for export. For parameter attributes it is supported for the following export types:
NetcdfScalarTimeSeriesSerializer
NetcdfMatroosScalarTimeSeriesSerializer
NetcdfScalarNonEquidistantTimeSeriesSerializer
NetcdfVerticalProfileSerializer
NetcdfGridTimeSeriesSerializer
NetcdfZLayersTimeSeriesSerializer
NetcdfDomainTimeSeriesSerializer
exportLocationAttributeAsNetCDFVariable
Since 2021.01. Adds a variable to the NetCdf file. Name of the variable is the value of ncVariable, the value is the configured location attribute of attributeId.
config example:
<exportLocationAttributeAsNetCDFVariable> <ncVariable>LocationAlias</ncVariable> <attributeId>niceName</attributeId> </exportLocationAttributeAsNetCDFVariable>
Example result (in txt format) of using exportLocationAttributeAsNetCDFVariable:
omitMissingValues
If set to true records with missing values are not exported
precision
Available since 2018.02. Optional element to set the number of decimals all values should be displayed with. If set, additional zeros will be appended and/or values will be rounded when necessary.
It is possible to configure a valueResolution for parameters via the parameters.xml. The configured precision for time series with parameters which have a value resolution should never exceed the maximum number of decimals needed to display values with this resolution. If the precision does exceed this, a warning will be given and the configured precision will be ignored.
The precision can never exceed 8 decimals due to limitations on the resolution with which values can be stored in the FEWS database (floating point errors).
exportTimeZone
TimeZone in which to export the data. Can either be a string (timeZoneName) or an offset (timeZoneOffset).
convertDatum
Convert (vertical) datum to local datum during export. The conversion will be done for all parameters which use datum (as configured in Parameters.xml) The local datum is defined in the z element in the locations.xml file.
geoDatum
Convert the geographical coordinate system (horizontal datum and projection) to specified geoDatum during export. Not all serializers support this parameter so please check the documentation for a particular serializer to see if it is supported.
ensembleMemberFormat
Available since 2019.02. Can either have value 'name' or 'index'. If 'name' is configured, the ensemble member Id is written. Otherwise the ensemble member index is written.
forecastSelectionPeriod
If configured all forecasts with a forecast time within the configured period will be exported. Since 2020.01 all forecasts will be exported to the same file.
When also configuring a <timeZeroFormattingString> in the <prefix> of the <exportFileName>, each forecast will be exported to a separate file to easily differentiate between the different forecasts.
Note: earlier versions export a separate file for each forecast by default. The forecast time is used to create a file extension (format yyyyMMddHH), unless <timeZeroFormattingString> is configured
Example configuration:
<export> <general> <exportType>SomeValidExportType</exportType> <folder>MyExportFolder</folder> <exportFileName> <name>_MyExportedFile.txt</name> <prefix> <timeZeroFormattingString>yyyyMMddHHmm</timeZeroFormattingString> </prefix> </exportFileName> <idMapId>MyIdMap</idMapId> <forecastSelectionPeriod start="-2" end="0" unit="day"/> </general> ... </export>
exportManualChanges
If used, only manual changed to the data will be exported. Unless the manualDBChangeViewPeriod is used, the relativeViewPeriod from the associated timeSeriesSet is used. Note: the view period is calculated relative to the dispatch time and not T0
exportChanges
If configured, any changes to the data in the configured period will be exported. Unless the dbChangeViewPeriod is configured, the relativeViewPeriod from the associated timeSeriesSet is used . Note: the view period is calculated relative to the dispatch time and not T0
An example:
<general>
<exportType>PI</exportType>
<folder>$EXPORT_FOLDER$</folder>
<exportFileName>
<name>exportedTimeSeries.xml</name>
</exportFileName>
<exportChanges>
<dbChangeViewPeriod unit="day" multiplier="2"/>
</exportChanges>
</general>
columnSeparator and decimalSeparator
Since 2016.01 (so far only implemented for GeneralCsv export type) it is possible to choose from multiple column separators: comma "," or semi-colon ";" or pipe "|" or tab "	" or space " "
When specifying a column separator it is compulsory to also specify the decimal separator as comma "," or point "."
For an example see generalCsv export type.
properties
Here properties for specific serializers can be configured. For example for the NetCDFSerializers the following properties will be taken into account:
<properties> <bool key="includeComments" value="true"/> <bool key="includeFlags" value="true"/> <bool key="includeTSProperties" value="true"/> <bool key="tryCompactingNetCDFData" value="true"/> <string key="netCDFWriteFormat" value="netcdf4"/> <int key="netCDF4DeflateLevel" value="6"/> </properties>
includecomments
Export comment for each time step to NetCDF, default false
includeFlags
Export flag for each time step to NetCDF, default false
includeTSProperties
Export time series properties for each time step to NetCDF, default false
tryCompactingNetCDFData
Depending on the difference between the minimum and maximum and the value resolution of a netcdf variable, try to use smaller sized integer variables like short or byte to compact the data. A scale factor and offset will be used to fit the data in the smaller sized variable and will added to the netCDF variable as attributes. This kind of compression will keep the precision of the value resolution. All standard netCDF viewers will take these attributes into account automatically, but other tools and especially scripts might not. This property will be false by default and only works for scalar and grid data.
netCDFWriteFormat
With this property the netcdf format can be set to netcdf4, default it will be netcdf3. Netcdf4 is needed to write compressed netcdf files which can result in 2 to 100 times smaller files.
netCDF4DeflateLevel
This property only works with netcdf4.
With this property the deflate level for writing compressed netcdf files can be set from 0 to 9. 0 meaning no compression and 9 maximum compression. Default will be 5, this level gives best compression without losing too much time when reading or writing.
metadata
(Meta data export has only been implemented for a limited set of export types. Currrently the NetCDF, grid2shp, LILA and HHRR types export meta data)
Optional metadata that is written in the exported file. The options netcdfMapDPhase and alertMapDPhase are deprecated (do not use these). For the other options it is possible to use the following tags:
%TIME_ZERO% the T0 of this time series export run.
%CURRENT_TIME% the current time.
%COLD_STATE_TIME% the cold state time.
%FORECAST_END_TIME% the forecast time set by the forecast length estimator or the forecast length option in the manual forecast dialog
%MODULE_INSTANCE_ID% the id of this module instance.
%MODULE_INSTANCE_NAME% the name of this module instance.
%MODULE_INSTANCE_DESCRIPTION% the configured description of this module instance.
%MODULE_INSTANCE_ATTRIBUTE(attributeId, moduleInstanceId)% moduleInstanceId is optional. When not specified the module instance of the module itself is used, like for the %MODULE_INSTANCE_ID% tag.
%WORKFLOW_ID% the id of the workflow in which this export runs.
%WORKFLOW_NAME% the name of the workflow in which this export runs.
%WORKFLOW_DESCRIPTION% the configured description of the workflow in which this export runs.
%TASK_DESCRIPTION% user description of the forecast in which the export runs.
%WHAT_IF_NAME% name of the what-if used in the forecast in which the export runs.
%USER_ID% the id of the user by which this export run is executed
%EXTERNAL_FORECAST_TIME% - external analysis time. When configuring it, it needs two parameters: the first is the external forecast id, the second should be the time format. Neither the id nor the time format should contain "," (a comma). The two arguments should be separated by a comma.
Example: %EXTERNAL_FORECAST_TIME(thisIsTheId),(yyyy/MM/dd HH:mm:ss z)%
%COLD_STATE_TIME(yyyy/MM/dd HH:mm:ss z)% - the cold state start time. If data is unavailable it will be filled as "Unknown".
Since 2022.02 location/parameter/qualifier/moduleInstance attributes between @ are also recognized.
Configuration example of metadata:
<metadata> <title>title</title> <institution> institution </institution> <source>source</source> <history>Exported at time zero = %TIME_ZERO(yyyy/MM/dd HH:mm:ss z)% in module instance %MODULE_INSTANCE_ID% as part of workflow %WORKFLOW_NAME% by user %USER_ID%.</history> <references>references</references> <comment>The actual time of writing was %CURRENT_TIME(yyyy-MM-dd HH:mm:ss z)%</comment> <summary>A summary of the data for @ATTRIBUTEID@</summary> <keyword>keyword1</keyword> <keyword> keyword with lots of spaces </keyword> <keyword>keyword 3</keyword> <customAttributes> <string key="emptyAttribute" value=" "/> <int key=" custom2 " value="123456"/> <string key="custom_3" value="This is a custom attribute with 'quotes' in it."/> <string key=" " value="attribute with empty key specified is not written"/> <float key="just_another_float" value="3.5"/> <bool key="truth" value="true"/> </customAttributes> </metadata>
title
A short description of the dataset. Its value will be used by THREDDS opendap servers as the name of the dataset. It therefore should be human readable and reasonable to display in a list of such names.
institution
Specifies where the original data was produced.
source
The method of production of the original data. If it was model-generated, source should name the model and its version, as specifically as could be useful. If it is observational, source should characterize it (e.g. "surface observation" or "radiosonde").
history
Provides an audit trail for modifications to the original data. It should contain a separate line for each modification with each line including a timestamp, user name, modification name, and modification arguments. Its value will be used by THREDDS opendap servers as a history-type documentation. It is recommended that each line begins with a timestamp indicating the date and time of day at which the modification was performed.
references
Published or web-based references that describe the data or methods used to produce it.
comment
Miscellaneous information about the data or methods used to produce it.
summary
The "summary" attribute gives a longer description of the dataset. In many discovery systems, the title and the summary will be displayed in the results list from a search. It should therefore capture the essence of the dataset it describes. For instance, include information on the type of data contained in the dataset, how the data was created (e.g. instrument X or model X, run Y), the creator of the dataset, the project for which the data was created, the geospatial coverage of the data, and the temporal coverage of the data.
keyword
Optional one or more key words or phrases that are relevant to the dataset. The values in this list may be taken from a controlled list of keywords (e.g. the AGU Index list or the GCMD Science Keywords).
customAttributes
If you want to add an attribute that is not predefined in the schema, then you can add it as a custom attribute here.
timeseriesSet
Define the timeseriesset to be exported. Please note that not all exports support all timeseriestypes (e.g. csv only supports scalar type).
filterId
Since 2018.01 it is possible to configure a filter id that refers to a filter from Filters.xml in the RegionConfigFiles.
</general> <filterId>AllQualifiersFilter</filterId> </export>
This way time series can be exported based on all options present in a filter like location, parameter and qualifier constraints:
<filter id="AllQualifiersFilter">
<timeSeries>
<moduleInstanceId>ExportRunMultipleTimeSeries</moduleInstanceId>
</timeSeries>
<relativeViewPeriod unit="day" start="-7" end="0"/>
<locationConstraints>
<idContains contains="12965"/>
</locationConstraints>
<parameterConstraints>
<idContains contains="H.m"/>
</parameterConstraints>
</filter>
Annotation location set id
Since 2021.01 it is possible to export annotations via the generalCsv type.
It will export all annotations that exist for the configured <annotationLocationSetId>annotationLocationSet</annotationLocationSetId>
The value colum will be used for the text of the annotation itself.
<timeSeriesExportRun xmlns="http://www.wldelft.nl/fews" xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance" xsi:schemaLocation="http://www.wldelft.nl/fews http://fews.wldelft.nl/schemas/version1.0/timeSeriesExportRun.xsd">
<export>
<general>
<exportTypeStandard>generalCsv</exportTypeStandard>
<folder>export</folder>
<exportFileName>
<name>ExportGeneralCsvAnnotations.csv</name>
</exportFileName>
<table>
<dateTimeColumn name="DATE" pattern="dd-MM-yy HH:mm"/>
<startDateTimeColumn name="START" pattern="dd-MM-yy HH:mm"/>
<endDateTimeColumn name="END" pattern="dd-MM-yy HH:mm"/>
<locationColumn name="locatie"/>
<propertyColumn name="firstProperty" key="firstProperty"/>
<propertyColumn name="secondProperty" key="secondProperty"/>
<valueColumn name="annotatie"/>
</table>
</general>
<annotationLocationSetId>annotationLocationSet</annotationLocationSetId>
</export>
</timeSeriesExportRun>
