LION files - SAF field diffs

[fvankrieken]

We need to document the field = 'special_address_flag' mismatches in qa__lion_dat_individual_diffs for 26a so we can send all issues to GR at once

There were two issues from before that Rodrigo wanted to punt to 26a. In both of these cases, I believe our pipeline is correct.

diffs from before 26a

Shoreline segment 0241972

• In dev, we assign 'N'. Prod has blank
• This segment has a protosegment
• In commonplace, there is a type 'N' record with segmentid 0241972
• However, in production, neither geometry-modeled segment or protosegment with segment id 0241972 has a (non-blank) special address flag. I'm pretty sure the shoreline segment should have it assigned

Centerline protosegment 0174704

• globalid '{96B37DB0-C48A-4AC5-A2AE-E8A411710BFE}'
• In dev, blank. Prod has 'B'
• Same boro as geometry-modeled segment. Altsegdata_type is 'R'
• In prod, both geometry-modeled segment and this protosegment have special address flag 'B'.
• Docs say in general, only one of geometry-modeled segment and protosegment should get the SAF record assigned, and in this case it seems it should be the geometry-modeled one. Certainly not both like the prod etl is spitting out

[damonmcc]

latest LION SAF diffs (qa__lion_dat_individual_diffs where field = 'special_address_flag')

_source_table	borough	face_code	sequence_number	segmentid	field	dev	prod
altsegmentdata	4	9396	00450	0174704	special_address_flag		B
centerline	4	0125	03005	0257930	special_address_flag	X
centerline	5	2063	00031	0290232	special_address_flag		X
shoreline	2	3659	00035	0241972	special_address_flag	N

[damonmcc]

update LION-related doc for GR to review latest SAF column diffs