>>> {k: v for k, v in CODATA.items() if 'fine-structure' in k}
{
'fine-structure constant': fine-structure constant # dimensionless, 0.0072973525693 ± 1.1e-12 , CODATA 2018,
'hyperfine transition frequency of Cs-133': hyperfine transition frequency of Cs-133 # frequency, 9192631770 Hz, CODATA 2018,
'inverse fine-structure constant': inverse fine-structure constant # dimensionless, 137.035999084 ± 2.1e-08 , CODATA 2018
}
>>> fine_structure_constant # or `alpha`
fine_structure_constant # dimensionless, 0.007297352565305213
>>> inverse_fine structure_constant # <- this was auto-completed
The following errors exist, and as the main dev I invoke my right to be lazy and list them all in one issue:
[x] The inverse ɑ should not have a space in its name!!!
[x] CODATA's dictionary should use the actual snake_case key in the dictionary.
[ ] The string ± 1.1e-12 , for dimensionless units, has a dangling space; a .strip() should be done on the display string.
[ ] The constant's uncertainty is seemingly removed when put into the namespace. It is likely a duplicate unit. CODATA should have some common shorthand for common constants such as this.
[x] Anecdotally, repr_measure should be _repr_measure to hide the API.
Recall Hyrum's law:
With a sufficient number of users of an API, it does not matter what you promise in the contract: all observable behaviors of your system will be depended on by somebody.
Best to make sure 1.0 is tidy before you accidentally break backwards compatibility on a hidden feature. Actual features can wait.
The following errors exist, and as the main dev I invoke my right to be lazy and list them all in one issue:
± 1.1e-12 ,for dimensionless units, has a dangling space; a.strip()should be done on the display string.repr_measureshould be_repr_measureto hide the API.Recall Hyrum's law:
Best to make sure 1.0 is tidy before you accidentally break backwards compatibility on a hidden feature. Actual features can wait.