TMY loading improvements

[wholmgren]

Adds ability to open TMY3 files via a url.

Closes #5

Cleans up some other aspects of the module including unneeded imports, slightly improved documentation.

Not yet ready for merging as it still needs new tests and coerce_year is not yet implemented.

[wholmgren]

I added an underscore to the start of all tmy functions except for readtmy2 and readtmy3. As a result, those underscored functions will no longer appear in the docs. I set up rtfd on my branch to preview the change:

http://wholmgren-pvlib-python-new.readthedocs.org/en/tmyurl/pvlib.html#module-pvlib.tmy

I think that I've now added tests for all of the new behavior and I've implemented coerce_year, so it's ready for @Calama-Consulting @bmu and anybody else to review the changes.

Note that the year coercion is disabled by default. I chose this behavior because I think that this follows the principle of least surprise.

[bmu]

Pandas uses to check, if there is a connection to the internet before doing tests that need such a connection (I think these tests are skiped in such a case). As an example see this test for the use of the decorator and here for definition.

Maybe we can direclty use this decorator for our tests.

[wholmgren]

Good point. I added the pandas @network decorator and tested it with and without an internet connection. Here's the output when I turned off my laptop's wifi...

(pvlib-3.4.3) william-holmgrens-macbook-pro:pvlib-python holmgren$ nosetests pvlib/test/test_tmy.py -v
pvlib.test.test_tmy.test_readtmy3 ... ok
pvlib.test.test_tmy.test_readtmy3_recolumn ... ok
pvlib.test.test_tmy.test_readtmy3_norecolumn ... ok
pvlib.test.test_tmy.test_readtmy3_coerce_year ... ok
pvlib.test.test_tmy.test_readtmy3_no_coerce_year ... ok
pvlib.test.test_tmy.test_readtmy2 ... ok
pvlib.test.test_tmy.test_readtmy3_remote ... SKIP: Skipping test due to lack of connectivity and error <urlopen error [Errno 8] nodename nor servname provided, or not known>

----------------------------------------------------------------------
Ran 7 tests in 13.865s

OK (SKIP=1)

[wholmgren]

Updated to fix a documentation issue mentioned in #16.

Anything else to do here before merging?

[bmu]

Looks good. I think this can be merged. Should I merge or wait for other comments?

[wholmgren]

I think you can go ahead and merge this one.

[dacoex]

Shall we add standard readers for typical data providers such as:

• Meteonorm
• SolarGIS, e.g. http://solargis.info/doc/_files/sample_data/SolarGIS-climData/SolarGIS-TMY-hourly-P50-PSA-ES.zip
• other common formats?

OK, maybe this is more a candidate for a separate model like pylib.io

[bmu]

I would like to have these readers too, but I think you should open a new issue and propose this.
Should be something for a later version.

[dacoex]

@bmu done, see the reference to issue #29