What are the selection modes?
The selection modes provide a rich way to select positions for each dimension in a given dataset. This selection will have a direct impact on the positions made available in the dropdowns of the Data Browser when this dataset will be browsed.
The following explanations will be based on the following example:
A fictitious dataset, which contains yearly data
This dataset contains a Time dimension (with a yearly frequency)
This dataset contains, for the Time dimension, the following positions:
Year 1995 to year 2005: positions are present in the dataset
Year 2006 to year 2010: positions are not present in the dataset (for some reason)
From 2011 to now (2019): positions are present in the dataset
The availability of these positions in this dataset can be visually represented as such:
The selection mode determines, for each dimension, which positions are taken into account (and therefore displayed in the Data Browser), and which positions are not taken into account (and therefore hidden from the Data Browser).
In the images present in the explanations below:
The available positions are represented with green circles, and the unavailable positions are represented with red circles
The selection mode is represented with a purple rectangle/arrow
The selected positions (which are the result of both the available positions and the selected selection mode) are represented with blue rectangle(s)
The available selection modes are the following:
Always all positions: In this mode, all the positions made available in the dataset for this dimension will be taken into account. Applied to the provided example, this will select the following positions:
From 1995 to 2005
From 2011 to 2019
Evolution through time: It is important to understand that using this selection mode means that every position added in the future (by the agency which manages this dataset), for this dimension within this dataset, will also be taken into account in the selection. So, for example, if the agency retroactively publishes data for this dataset for years 2009 and 2010, the following positions will be selected, using this selection mode:
From 1995 to 2005
From 2009 to 2019
Static positions: In this mode, only the positions specifically checked will be taken into account in the selection. So, for example, assume that the following positions are checked:
2003
2014
2018
2019
Only those positions will be taken into account in the selection. If other positions are added in the future in this dataset, those new positions will not be present in the selection.
Fixed time (this selection mode is available only for the Time dimension): In this mode, the selected positions for the Time dimension will match a timespan which is manually defined by selecting “Last X time” from the drop-down. The end of the timespan matches the current time and the beginning is automatically set by counting down to the X time set. All the available positions located inside the interval will be present in the selection. So, for example, if selecting “Last 10 Time” with the provided example, the selected positions will be:
2005
From 2011 to 2019
Using this selection mode, the selection might be impacted by retroactive actions on positions availability:
If positions which are inside the selected timespan interval (for example 2009 and 2010) are retroactively added, 2010 will be added and 2005 will be removed from the selection;
If new position is added to the Time dimension (in other words, the most recent data is added to the dataset), the selection will be modified to reflect the change: this new position will be added to the selection and the oldest year will be removed.
The interval of values present in the drop-down selection is defined in the configuration file.
By default, the values depend on the Frequency mode, as follows:
A: 5
M: 6
Q: 4
The minimum value is 1 and the maximum value is the number of time values.
From-To (this selection mode is available only for the Time dimension): In this mode, the selected positions for the Time dimension will match a timespan interval, for which the beginning and the end are manually defined. All the available positions located inside this interval will be present in the selection. So, for example, if selecting “From 2002 to 2014” with the provided example, the selected positions will be:
From 2002 to 2005
From 2011 to 2014
Using this selection mode, the selection might be impacted by retroactive actions on positions availability:
If positions which are inside the selected timespan interval (for example 2009 and 2010) are retroactively added, these positions will be added to the selection;
If the retroactively added positions are located outside of the selected timespan interval, these added positions won’t affect the selection.
Rolling Time (this selection mode is available only for the Time dimension): In this mode, the selected positions for the Time dimension will match a timespan for which the beginning is manually defined. The end of the timespan matches the current time. All the available positions located inside this interval will be present in the selection. For example, if selecting as beginning the 2004 position, the selected positions will be:
2004 and 2005
2011 up to now (“now” has to be understood as the time when the dataset is browsed in the Data Browser)
Using this selection mode on a dataset will likely have an impact on the selected positions over time: indeed, each time a new position is added to the Time dimension (in other words, each time the most recent data is added to the dataset), this new position will be added to the selection.
If the agency which manages a dataset removes one or more position from the dataset, those positions will not be taken into account in the selection anymore, whatever is the selected selection mode for the affected dimension.
If those very same positions are re-added in the future by the agency, they will be present in the selection again.
How to know which selection mode is applied to a dimension in the Data Browser?
When browsing a dataset in the Data Browser, the “Selection” tab displays all the dimensions available in the dataset:
Each dimension in the layout (with a blue upper border) and each dimension in Page section (with a gray upper border) is linked to a selection mode. The selection mode can vary from one dimension to another.
The selection mode applied to a given dimension is represented with a small icon in the upper right corner of the upper border rectangle:
Here is the matching between the selection modes and their respective icons:
Icon | Selection mode |
|---|---|
| Always all positions |
| Static positions |
| Fixed Time (only available on Time dimension) |
| From-To (only available on Time dimension) |
| Rolling time (only available on Time dimension) |
How to change the selection mode applied to a dimension?
If, for a given dimension, the selection mode must be changed (for example, to get more positions in the dropdown list), it is necessary to create a custom dataset.
This process is described in detail on this page.