Turns a document into something a test can compare or pin as an inline
snapshot. Two projects that look the same to a user give the same result,
whatever ids they carry and whatever order their rows were created in.
Sort:
media assets by filename, then id
tracks back to front by the stacking rule (compareStacking)
clips by track, then start_time, then id
effects by clip, then order, then id
keyframes by effect, then param_name, then time, then id
transitions by outgoing clip, then incoming clip, then id
"By track" and the like mean the position of that row in its own sorted list.
Label: every id becomes asset#n, track#n, clip#n, effect#n,
keyframe#n or transition#n, counting from 1 in sorted order. Every
foreign key becomes the label of the row it points at.
Drop: file_path, file_size and proxy_path on media assets (the file is
named by filename). The document already has no created_at / updated_at.
Numbers are rounded to 6 decimals so float noise never breaks a snapshot.
Turns a document into something a test can compare or pin as an inline snapshot. Two projects that look the same to a user give the same result, whatever ids they carry and whatever order their rows were created in.
Sort: media assets by filename, then id tracks back to front by the stacking rule (
compareStacking) clips by track, then start_time, then id effects by clip, then order, then id keyframes by effect, then param_name, then time, then id transitions by outgoing clip, then incoming clip, then id "By track" and the like mean the position of that row in its own sorted list.Label: every
idbecomesasset#n,track#n,clip#n,effect#n,keyframe#nortransition#n, counting from 1 in sorted order. Every foreign key becomes the label of the row it points at.Drop:
file_path,file_sizeandproxy_pathon media assets (the file is named byfilename). The document already has nocreated_at/updated_at.Numbers are rounded to 6 decimals so float noise never breaks a snapshot.