Lo DE Segmented Packets #1016
Lo DE Segmented Packets #1016
Conversation
sdhoyt
added
Ins: Lo
sdhoyt
requested review from
laspsandoval,
bourque and
tech3371
and removed request for
a team
imap_processing/lo/l0/lo_science.py
Outdated
| # 1 = start of a group of segmented packet | ||
| # 2 = end of a group of segmented packets | ||
| # 3 = unsegmented packet | ||
| seg_starts = np.where((seq_flgs == 1) | (seq_flgs == 3))[0] |
There was a problem hiding this comment.
I nit-pick the use of np.where() a lot. This should really be np.nonzero(). See the Note at the top of the numpy documentation: https://numpy.org/doc/2.0/reference/generated/numpy.where.html
There was a problem hiding this comment.
Ah interesting. I wasn't aware of that. Thanks, I'll make that change
| seq_flgs = dataset.seq_flgs.values | ||
| seq_ctrs = dataset.src_seq_ctr.values | ||
|
|
||
| # Find the start and end of each segment of direct events |
There was a problem hiding this comment.
These short comments were very helpful!
|
I just made an update to this PR. I found out that padding is added to the binary data to make the packet divisible by 8, so I made a small change to help me identify where to look for padding when this is used in processing. The change adds a comma between segments when they're combined into a single binary string. I'm not crazy about that approach, but I should be able to use that with a find for the next comma or end of the string and determine which bits I need to throw way. I also thought about keeping the binary string separated in a list of lists, but that would require another for loop or passing around another variable to keep track of which segment I'm in. Adding the comma seems easier, but the binary string with commas isn't ideal. Hoping to get others' thoughts on this. |
I wish I have ideas for this. This is where Greg is great at :). |
Change Summary
Overview
This PR adds a function to handle Lo direct event segmented packets. This function will soon be added to Lo processing in combination with the event parsing to produce a L1A dataset for Lo direct events.
If there are too many bits of data needed to store Lo direct events for a single Aggregated Science Cycle into a single packet, the data will be segmented across multiple packets. The segments will be indicated by the seq_flag:
seq_flag 1 = start of segment
seq_flag 0 = one of the middle packet of a segment
seq_flag 2 = last packet of a segment
seq_flag 3 = unsegmented packet
The packets after the first will only contain the headers, shcoarse, binary data and checksum. They will not contain the count and passes fields.
Closes #1011