Isolating programming environments in Stata

These are instructions to make use of Stata’s ado folders to create blank-slate programming environments for reproducibility packages.

If a main script is present

1. Create a new folder named ado inside the code folder.
2. If the README mentions a Stata version, add the version line at the top of the main script:

   version 18.0 // Replace with the version mentioned in the README
   

3. After the path globals are defined, add the following line to redirect the PLUS ado folder to the ado folder you created:

   sysdir set PLUS "${path_to_code}/ado" // Replace with the correct global
   

4. After the sysdir line, add the following block to install required packages into the ado folder. Replace xxx with the required commands. Once installed, you can comment this block out — the packages will persist in the ado folder and do not need to be reinstalled on subsequent runs. ```stata

   • Install once, then comment out local user_commands xxx xxx // Replace with required user-written commands foreach command of local user_commands { cap which command' if _rc == 111 { ssc install command’ } }

   • For packages not available on SSC, use net install, e.g.:
   • net install xxx, from(“https://raw.githubusercontent.com/org/repo/main/”) ```
5. Check the ado folder and verify that all packages have been installed there.

If a main script is not present

1. If there are only a few do-files and the order in which they should be run is clear from the folder structure or README, create a main script using the World Bank main.do template and follow all steps above.
2. If there are many do-files and the correct order is not immediately clear, ask the authors to create a main script before proceeding.

---

If you followed these instructions correctly, the ado folder should contain all packages needed to run the code, and the main script should redirect the PLUS ado folder to that location before any sub-scripts are run. There is no need to install any additional commands manually.