Data unblocking in decode software
Updated by Ole Hansen 6 months ago
- % Done changed from 0 to 50
This should work now. I updated the 1190 code In Podd given the format mentioned in the above log entry. The 1190 multi block format does not follow what I consider the CODA standard, using block headers/trailer etc. Instead, one gets a simple concatenation of single event data, fortunately in one contiguous block for each module. The lack of block headers and the different format in general prevents a number of sanity checks and makes recovery from hypothetical decoding errors difficult. In particular, the block level of the data needs to be discovered from the data instead of being available up front. There's some room for improvement, but it would be a bit time consuming to implement.
Additionally, I corrected a few general issues with the multi block handling in Podd's decoder. In particular, the event metadata (event number, time stamp, trigger bits, etc.) were not updated for events following the first one in a block (cached events). I also added a check to prevent processing of banks that are defined in the crate map but not present in the current event block. They were mistakenly processed (with errors) for cached events.
These changes are available now in Podd Release 1.7.5.