Better documentation/error messaging when parsing input fails

[dawahs]

With the whole parsing of the log file fiasco yesterday, I think I'll try to comb through process_partis.py and add in some checks and clear messaging/documentation in case we hit a snag like this again.

[dawahs]

A couple of possible issues I can see while thinking toward the future:

• if the option --parameter-dir is not provided in the partis call then partis defaults to printing whatever directory is used to stdout. For now, if no parameter directory is provided then process_partis defaults to searching the IMGT gemlines in the partis/data/germlines directory. We might prefer something else in the long run, though. Should we always assume --parameter-dir will be provided? Raise an error or default to IMGT?
• if the path provided by --parameter-dir is not absolute then we have little recourse since we only have implicit information about the working directory partis was called from. This working directory may have---as it often does in the examples we've been using---relative paths for its /path/to/partis/ argument that I don't quite know how we could reconcile with a nonabsolute path for the --parameter-dir argument. For now process_partis just raises an error if this parameter path is not absolute. Should we require an argument, either in partis or process_partis, to the tune of --path-from-which-partis-was-called?

[psathyrella]

So I was thinking the idea with the code I provided in datascripts was it avoids the whole issue of parsing log files by using the same code that chooses the path structure to start with. Does that make sense?

[dawahs]

I'm a little confused. Which datascripts are you talking about? /fh/fast/matsen_e/processed-data/partis/kate-qrs-2016-09-09/run.sh?

[psathyrella]

the repo I linked to in that issue yesterday, psathyrella/datascripts

On Fri, Dec 9, 2016 at 4:33 PM, D. A. Shaw notifications@github.com wrote:

I'm a little confused. Which datascripts are you talking about? /fh/fast/matsen_e/processed-data/partis/kate-qrs-2016-09-09/run.sh?

— You are receiving this because you commented. Reply to this email directly, view it on GitHub https://github.com/matsengrp/cft/issues/66#issuecomment-266161107, or mute the thread https://github.com/notifications/unsubscribe-auth/AC0rc7zJw59u6AzmE5VY_TS9TAkyQhxrks5rGfNBgaJpZM4LJbjC .