How does it work?
To start filtering click on the 
icon in the toolbar. This brings up the BUFR filter editor. Having defined the options and run the filter two tabs will appear in the main interface:
- the first tab, titled as Original data, contains the original BUFR file
- the second tab, named as Filtered data, contains the results of the filter.
Th Filtered data tab has the same structure as the Original data tab but it only shows the messages matching the filter conditions. Another change is that when data is extracted the Locations tab gets renamed as Extracted values and shows the resulting tabular data.
The filtered messages can be saved as a new BUFR file using the 
icon on the Filtered data tab. Closing the Filtered data tab removes the filter results and the user interface returns to its initial state showing only the original data.
Message splitting
Please note that when filtering with keys form the data section, each matching subset from a multi-subset message is turned into a separate message in the results (i.e. messages are split).
The filtering algorithm
The actual filter algorithm is made up of two consecutive steps:
- first we identify (filter) the messages/subsets to be filtered
- then, optionally, we extract data in a tabular form (e.g. as CSV) from the chosen messages
These options are explained in detail here. A rich set of examples are available for both the message/subset filtering and the data extraction.
The BUFR filter editor
The filter steps (see above) are reflected in the way the options are grouped in the BUFR filter editor.
Filtering messages/subsets
The message/subset filter itself is split into two parts. The first part defines the options purely using BUFR header data:
while the second part features the options that require access to the data section (i.e. the messages have to be expanded) :
Extracting data values
The data extraction options are all grouped into one big section. By default data extraction is disabled. To enable we need to set option Extract values to On.