cd to workflow dir during file parsing.

[hjoliver]

Allow the template processor to access workflow files - in the source dir for a source workflow, or the run dir for installed workflows.

This is the most straightforward way to cater to this kind of use case

(I can't think of any downside...)

Check List

• I have read CONTRIBUTING.md and added my name as a Code Contributor.
• Contains logically grouped changes (else tidy your branch by rebase).
• Does not contain off-topic changes (use other PRs for other changes).
• Applied any dependency changes to both setup.cfg (and conda-environment.yml if present).
• Tests are included (or explain why tests are not needed).
• CHANGES.md entry included if this is a change that can affect users
• Cylc-Doc issue opened if required at Document Jinja2 access to workflow dir (source or installed) cylc-doc#622
• If this is a bug fix, PR should be raised against the relevant ?.?.x branch.

[wxtim]

Might be safer

try:
  os.chdir(fdir)
finally:
  os.chdir(odir)

or even:

@contextmanager
def work_in(path):
  before = os.getcwd()
  try:
    os.chdir(path)
  finally:
    os.chdir(before)

with work_in(fdir):
  # ... stuff ...

Kind of surprised that the latter isn't part of Pathlib.

[oliver-sanders]

It doesn't actually matter if we don't os.chdir back afterwards, this only applies to the Python session itself not the parent shell.

[hjoliver]

Note my change does os.chdir back afterwards. I agree it should not matter, but to explicitly avoid any side-effects of the dir change that I might not be aware of.

[wxtim]

• Read code
• Tried code

I'm generally happy, but I've made some suggestions about the test.

[hjoliver]

@wxtim - I tweaked the new test as you suggested. A final commit just cleans up the file to pass flake8.

[wxtim]

I'll hit merge as soon as the tests finish./