massdesas

Version 1.0.2 | 2026-08-05

massdesas recursively converts SAS .sas7bdat datasets to Stata .dta files, preserving the directory tree and writing each result beside its source. It is for Stata users who need to batch-convert a collection of SAS files with optional lowercase variable names and controlled source-file deletion.

Quick Start

Point directory() at a folder containing at least one .sas7bdat file:

massdesas, directory("/path/to/sas_files") lower
return list

The command writes .dta files beside the SAS sources and reports the conversion counts in r(). Omit lower when imported variable names should retain their original case.

Requirements

• Stata 14 or later
• Stata's built-in import sas command
• The SSC packages filelist and fs

Installation

capture ado uninstall massdesas
net install massdesas, from("https://raw.githubusercontent.com/tpcopeland/Stata-Tools/main/massdesas") replace
ssc install filelist, replace
ssc install fs, replace

Commands

Command	Description
massdesas	Recursively convert .sas7bdat files to sibling .dta files

How It Works

• If directory() is omitted, massdesas uses the current working directory.
• The command uses filelist to search that root and all subdirectories for files matching *.sas7bdat, then uses fs to enumerate the files in each directory.
• Each file is imported with import sas and saved as a .dta file in the same directory with the same base name.
• lower passes case(lower) to import sas, converting imported variable names to lowercase.
• erase deletes each original .sas7bdat file only after its .dta file has been saved successfully.
• The command preserves the caller's data and restores the original current working directory after processing.

Worked Examples

1. Convert files in the current working directory

Change to a directory containing SAS datasets, then omit directory() to use c(pwd):

cd "/path/to/sas_files"
massdesas
return list

2. Convert a project tree with lowercase variable names

The search includes nested directories, and lower standardizes imported variable names:

massdesas, directory("/path/to/project/raw") lower
return list

3. Check the conversion summary before inspecting outputs

Use the returned failure count to stop a workflow if any file could not be converted:

massdesas, directory("/path/to/staging") lower
assert r(n_failed) == 0
display "Converted files: " r(n_converted)

4. Remove source files after validating a copy

Run erase only on a backup or staging copy after confirming that the generated .dta files are usable:

massdesas, directory("/path/to/backup_copy") lower erase
return list

Command Reference

massdesas

massdesas [, directory(directory_name) erase lower]

The command stops with an error if the root directory does not exist or contains no matching SAS files. When an individual file fails, processing continues and the failure is counted in r(n_failed).

Options

Option	Default	Description
directory(directory_name)	Current working directory	Root directory to search recursively for .sas7bdat files
erase	Off	Permanently delete each original SAS file after its conversion succeeds
lower	Off	Pass case(lower) to import sas so imported variable names are lowercase

Stored Results

After a successful run, massdesas stores the following results:

Result	Type	Description
r(n_converted)	Scalar	Number of SAS files successfully converted
r(n_failed)	Scalar	Number of SAS files that failed to convert
r(directory)	Local macro	Root directory used for the conversion

Assumptions and Limits

• Only files matching the literal pattern *.sas7bdat are searched. On case-sensitive filesystems, files ending in .SAS7BDAT are not found.
• Filenames and directory paths containing spaces are supported when the directory() path is quoted.
• A nonexistent directory or a directory with no matching SAS files produces an error rather than an empty result set.
• If filelist or fs is unavailable, the command stops with r(199); install the missing SSC dependency before running it.
• erase is permanent and is applied only after a successful .dta save; keep a backup until the converted files have been checked.
• Existing .dta files with the same base name are overwritten because each converted dataset is saved with replace; preserve a copy if existing outputs must be retained.
• Conversion behavior and supported SAS features follow Stata's built-in import sas command.

References

• StataCorp's SAS importer: see help import sas in Stata.
• The required SSC utilities: see help filelist and help fs after installation.

Version History

• 1.0.2 (2026-08-05): Documented that converted .dta outputs overwrite existing same-named files and standardized package author metadata.
• 1.0.1 (2026-08-05): Fixed SAS filename quoting, preserved the complete basename when it contains .sas7bdat, and ensured the caller's data are restored after unexpected errors.
• 1.0.0 (2026-04-08): Initial Stata-Tools release

Author

Timothy P Copeland, Karolinska Institutet

License

MIT