lsys / rbstata Goto Github PK

Put boilerplate messages into separate variables/file?

Can be shared by both main and test file

there should be some solution from pandas.read_stata() for this

make it more verbose?

all the -v.dta

• https://black.readthedocs.io/en/stable/usage_and_configuration/the_basics.html
• https://cookiecutter.readthedocs.io/en/stable/cli_options.html

Similar to https://github.com/LSYS/wbStata/blob/main/.github/workflows/cli-pkg.yml.

https://stackoverflow.com/questions/18394147/how-to-do-a-recursive-sub-folder-search-and-return-files-in-a-list

-r
--recursive

.dta file(s) [*]: .dta files to convert [all .dta files in current directory]
Include subdirectories (y/n) [n]: Include subdirectories if * [no]
Target version [13]: version to convert to [Stata v13]
File suffix for saving [-v13]: Suffix for saving [E.g. save auto.dta to auto-v13.dta]
Print messages (y/n) [y]: Print all messages and errors [yes]

seems like target option is more important than include directories, yet the former comes after the latter.

https://www.google.com/search?q=dta+too+modern&oq=dta+too+modern&aqs=chrome..69i57.3591j0j7&client=ms-android-xiaomi-rev1&sourceid=chrome-mobile&ie=UTF-8#ip=1

Seems like a click version problem. Should be using click>=8.*

https://github.com/NickCH-K/causaldata/tree/main/Stata

The files seem to now be compatible with Stata 13.

Need --check

https://google.github.io/styleguide/pyguide.html#383-functions-and-methods

https://numpydoc.readthedocs.io/en/latest/format.html

Not included.

Set iterator to True to get the labels

Using pathlib

• smth smth takes care of unicode compatibility issues
• takes cares of transferring variable labels, value labels, data labels, if any

In the readme:

File suffix for saving [-v13]:

Change the default behavior to -old? or -wbstata?

So that all files can be converted in path in a single line usage

Like cookiecutter

https://click.palletsprojects.com/en/7.x/utils/#showing-progress-bars

lowercase,
check that .dta extension exist

Cannot use wbStata on nickchk's causal data

Got a

raise ValueError("Variable labels must be 80 characters or fewer")
ValueError: Variable labels must be 80 characters or fewer

Can try using a notebook to investigate

Dta should fail when expected using capture.

Need to figure out what is the error code for Stata data too new