Feature: adsorption-file-parser

[pauliacomi]

Implemented new adsorption-file-parser package into the parser. This should be a drop-in replacement, but it does introduce a few changes in the generated AIFs. The tests give the exact changes where this happens.

Summary of changes to code:

• Streamlined tests and removed some boilerplate code. The tests now generate AIF files, then compare them with AIFs from a previous version. If not identical, test fails and then prints out the offending changes from one file to the other. A flag in conftest.py can be turned on to re-make all reference files (if the decision was made that changes are ok).
• Added a new function in the parser module which tweaks adsorption-file-parser outputs to be identical to the format used in this repo.
• Added quotes around most AIF fields, as they may contain nonstandard chars.
• Add possibility for more kinds of input files supported in adsorption-file-parser.
• Removed duplicated data_ in header.

Resulting changes to AIF files:

• Quotes around multiple AIF fields.
• Removed duplicated data_ in header.
• Kept upper/lowercase accurately.
• Temperature given in original units (K/C).
• Quantachrome AIFs, contain converted loading.

Finally a few open questions:

• How to deal with Quantachrome files giving out absolute amounts adsorbed, unlike other reports? Should they be converted or left as-is in the generated AIF?
• Should as much metadata as possible be saved in the AIF file?

[jackevansadl]

Whoops, I think I might have broken this when trying to merge. Can you fix this implementation with the newest version of data names?

[pauliacomi]

I'll have a look over weekend, should be just a few tweaks away.

[pauliacomi]

That was a bit of a longer weekend than planned.

I integrated a few changes to the micromeritics parser. Thanks for adding the example files. The other errors were from discrepancies in old AIF files everything was checked against. I regenerated those, with following differences:

• metadata name changes (e.g. _sample_sample_mass -> _adsnt_sample_mass)
• new metadata _audit_aif_version
• most values are now quote escaped (e.g. Ar -> Ar)
• capitalization of values is retained (e.g. quadrasorb station 1 -> QuadraSorb Station 1)
• units kept "true to source" (e.g. mmHg instead of Torr and degC instead of K in some cases)

Let me know if these changes to the AIF files are acceptable.

[jackevansadl]

thanks for your efforts! I believe these changes are acceptable.