Ramp Fitting Errors for Integrations with more than 256 Groups

[stscijgbot-jp]

Issue JP-3583 was created on JIRA by Michael Regan:

The Ramp Fitting step calculates incorrect slopes and uncertainties when an integration has more than 256 integrations. The code uses an 8 bit counter to track the segments between jumps. When the length of the jump-free region is longer than 256 the counter rolls over to zero. The net effect is that the number of groups used for the ramp fitting is mod 256 of the number of groups.

The fix requires changing the data type of three variables from 8 bit to 16 bit.

[stscijgbot-jp]

Comment by Howard Bushouse on JIRA:

Michael Regan Is this something that can be fixed in time for inclusion in Build 10.2 (i.e. in the next couple days)?

[stscijgbot-jp]

Comment by Howard Bushouse on JIRA:

Fixed by spacetelescope/stcal#251

[stscijgbot-jp]

Comment by David Law on JIRA:

Tested and confirmed with jw01864006001, with both python and C ramp fitting code.  Closing.