These are chat archives for dereneaton/ipyrad

21st
Jan 2016
Isaac Overcast
@isaacovercast
Jan 21 2016 04:08
I just realized step 3 doesn't test for samples in state 2. I think we should be consistent in how we handle requested samples that aren't at or beyond the step requested. I will dig through the code to see how other steps handle but in general for all steps do you think its best to bail out or to warn and continue if requested samples aren't ready for the step requested?
I know for step 7 we bail out hard and report on any samples not in state 6. Do you think this is the right behavior? if so we should make all steps do the same thing.
Isaac Overcast
@isaacovercast
Jan 21 2016 04:15
Dude! You're giving a job talk at columbia! So sick!
Deren Eaton
@dereneaton
Jan 21 2016 04:20
I know!!! Was gonna tell you tomorrow.
Step 7 is the one case where we may want different behavior, since it is dependent on the output cluster&catg files from step 6, rather than input file from any individual samples. It would be more appropriate to test whether the selected samples in step 7 were included in the step6 clustering or not, and raise an error if not.