prototype sort code crashes on trying to print output

[jeftreece]

Reproduce: set kitlimit to 50 in config.yaml, drop tables redux.py -t redux.py -o redux.py -mp redux.py -m

observe error: Beginning run [00:49:07] beg MatrixData restore: 0.99071 end MatrixData restore: 1.002341 beg stdout-matrix: 1.00277

Traceback (most recent call last): File "./redux.py", line 198, in vt.matrix(args.rest) File "./sort.py", line 134, in matrix self.sort.stdout_matrix(vix=vix,kix=kix) File "./sort.py", line 1456, in stdout_matrix print(table) File "/u/treece/.local/lib/python3.6/site-packages/beautifultable/beautifultable.py", line 498, in str return self.get_string() File "/u/treece/.local/lib/python3.6/site-packages/beautifultable/beautifultable.py", line 1071, in get_string content = str(row) File "/u/treece/.local/lib/python3.6/site-packages/beautifultable/rows.py", line 103, in str row[i] = '{:{sign}}'.format(row[i], sign=sign.value) TypeError: unsupported format string passed to NoneType.format

[jazdrv]

you're running 50 kits?

I've only been doing 16. That's all I can get my hands on.

I haven't stress tested beautifultable with lots of data; I wasn't sure it would be something that could carry a big load like the project will need in its full scope. As I saw it, it was something so I could visually work out bugs and such with smaller sets of data. But sure if the idea continued to work with larger sets of data in a practical way ... then that would be great too.

Stuff like this ... I would need to be working with a similar data construct you are to be at all helpful. Because I can't recreate your error otherwise.

Zak

[jeftreece]

I have seen this with different number of kits, and it does not require 50 kits to reproduce. However, I agree it's input dependent. I'll narrow it down to a smaller repro, but as I noted in a separate thread, if row[i] is None, it's easy to reproduce outside of bt

[jeftreece]

closing this issue with merge #38