Marshall Ward

Dropping the end index test inside `FIndex` was more effective that I had expected, but not perfect. The following tests caused hangs: * test_empty_grp_no_end * test_string_grp_no_end * test_file_grp_no_end * test_findex_iteration...

This is not directly related to the question, but I note that the following case can return an error: ``` &example q(1:2) = 1, 2, 3 q(4:5) = 4, 5,...

I am having trouble resolving this case: ``` &test_nml v(1, :) = 1, 2, 3 / ``` Currently, if `i_end` is unset, then I explicitly set it to `i_start +...

Could you post an example? I'm not sure what you are trying to do. ~There is `f90nml.reads(s)` if you want to parse a string.~ (See below.)

Sorry, let me clarify. The parser has a `reads` function, not the top-level module. Something like this: ``` import f90nml parser = f90nml.Parser() nml = parser.reads('&test_nml x = 1 /')...

I'm also seeing a problem with this case. I have this error: ``` Traceback (most recent call last): File "/f90nml/dev.py", line 5, in nml = parser.reads(content) File "/f90nml/f90nml/parser.py", line 300,...

Looks like it was just a bug, and I didn't use `splitlines()` properly. Adding the `keepends` argument seems to work: ``` diff --git a/f90nml/parser.py b/f90nml/parser.py index 971ad4e..8b71bae 100644 --- a/f90nml/parser.py...

Can you test the latest push? a94f45338ccc4e6208a6447b4063fd4b33e43000

There was a Python 2 issue, fixed in dc5a70bcceb43e9b6afe1f8f33487b17e9a176e0.

I count 23 values in `y(76:)`, filling elements from 76 to 76+23-1=98. Element 99 is unfilled. You then start filling from `y(100:)` onward. Is this not correct? (Also, this probably...