For #81 - fix undefined variable in get_type_by_name()

[arporter]

Small code fix and added new test for the associated routine.

[TeranIvy]

Happy to review it.

[TeranIvy]

The bug fix looks good. Please see the inline comments and pylint report below for the proposed changes.

• All tests pass.
• x-failing tests are unaffected by the current issue.
• Modified/added tests do not use nose.
• pycodestyle on added/modified code passes.
• pylint on test_block_stmts.py is 10/10.
• The interface to get_type_by_name(self, name) in modified class HasImplicitStmt(object) is now documented.
• pylint returns some warnings and errors for the modified class (below). I am aware not all of them are for the modified lines, however I think would be good to improve the code quality for the entire class.

C: 99, 0: Missing class docstring (missing-docstring)
C:101, 4: Invalid class attribute name "a" (invalid-name)
E:129,56: Instance of 'HasImplicitStmt' has no 'item' member (no-member)
E:132,53: Instance of 'HasImplicitStmt' has no 'item' member (no-member)
C:135, 4: Missing method docstring (missing-docstring)
C:140,12: Invalid variable name "c" (invalid-name)
C:140,15: Invalid variable name "t" (invalid-name)
C:143,12: Invalid variable name "st" (invalid-name)
C:150, 8: Invalid variable name "s" (invalid-name)
C:151, 8: Invalid variable name "ls" (invalid-name)
C:152,12: Invalid variable name "st" (invalid-name)
C:152,16: Invalid variable name "l" (invalid-name)
C:155, 8: Invalid variable name "s" (invalid-name)

• All modified/new code is covered, but not all of the existing code in get_type_by_name(self, name) (see inline comment).
• Documentation builds correctly and it is up-to-date.
• Please bring the branch up-to-date before the next review if required.

[TeranIvy]

Perhaps The variable name as other descriptions are capitalised after :?

[arporter]

Good point, done.

[TeranIvy]

Please consider adding tests to cover lines 116 and 120 so the entire function is covered. I am aware the missing coverage is not the result of this PR so this is up to the developer's good will :)

[coveralls]

Coverage increased (+0.4%) to 89.163% when pulling 7d50bfa on undev_var_fix into f5ad578 on master.

[codecov-io]

Codecov Report

Merging #88 into master will increase coverage by 0.72%.
The diff coverage is 100%.

@@            Coverage Diff             @@
##           master      #88      +/-   ##
==========================================
+ Coverage   88.73%   89.46%   +0.72%     
==========================================
  Files          23       24       +1     
  Lines       10513    10575      +62     
==========================================
+ Hits         9329     9461     +132     
+ Misses       1184     1114      -70

Impacted Files	Coverage Δ
src/fparser/one/typedecl_statements.py	85.71% <100%> (+1.61%)	⬆️
src/fparser/one/tests/test_block_stmts.py	100% <100%> (ø)
src/fparser/one/block_statements.py	77.68% <100%> (+7.55%)	⬆️

---

Continue to review full report at Codecov.

Legend - Click here to learn more
Δ = absolute <relative> (impact), ø = not affected, ? = missing data
Powered by Codecov. Last update f5ad578...7d50bfa. Read the comment docs.

[arporter]

I've updated for pylint as requested and added further tests to hit the missed lines. They then revealed a Python 3 problem which I have also fixed.

[arporter]

I've rebased this branch onto the head of master. Ready for second review.

[arporter]

(Forgot to say that I haven't updated the class attribute named "a" because, even though it's a rubbish name, it permeates a lot of the code. Something for a separate issue.)

[TeranIvy]

Thank you for pylint clean-up, updating tests and fixing Python 3 issue in typedecl_statements.py. The last revealed some pycodestyle cleanup is required there, but that can be left for another day.
Proceeding to merge.