Testing/parsing: completeness of data import (FOR REVIEW)

[epogrebnyak]

Affected prvious issues:

• Check for list of imported varaibles #47 SOLVED
• Output: get_dfm() and get_var_list() in various locations #39
• Parsing: test if a variable label is present in dataset and non-empty #38 SOLVED, but no check for non-empty
• Parsing: some variables not imported #35 (PROFIT)

Steps - need list of:

• [ ] m, q, a and all variables in database
• [x] variables from spec file SOLVED
• [x] variables form spec files via cfg files SOLVED
• [ ] duplicate headers

Explicit test:

• are all variables form spec/cfg in database? SOLVED
• are all headers covered by spec dicts?

Also:

• Need (+ 1) variable in sample dataset to set up explicit cfg test
• get capitalised letters VAR_NAME part of VAR_NAME_unit_of_measurement to deal with group of variables SOLVED

Todo: explain where it will be licated, what is already implemented?

[epogrebnyak]

Partly solved by d1b67aa88338c10887c69407e507379b8746c407

[epogrebnyak]

Also #15.

[epogrebnyak]

Need neew outline, e.g.:

• [ ] inspect import of variables by segment
• [ ] test by available spec files as list
• [ ] test cfg file is running ok.

[epogrebnyak]

• [ ] load full dict must be in specification, not inspection

[epogrebnyak]

must change https://github.com/epogrebnyak/rosstat-kep-data/blob/master/kep/test/pytest/remaining/_package/test_import_complete.py