Debugging File and Directory Operations

From Clutter to Structure

A backup script can work correctly and still be difficult to maintain. If every backup is saved in one flat directory with a long timestamp-based filename, repeated use can produce dozens of similarly named files. Finding a backup from a particular day or checking whether a daily backup exists becomes harder. The maintenance improvement is to organize backups by date first and by time second.

The structure has three levels: a main backup directory, a date-based subdirectory, and a ZIP file whose name represents the backup time. For example, the date directory 20240115 represents January 15, 2024, while 143022.zip represents 2:30:22 PM. Multiple backups from the same day can therefore share one date directory while keeping separate time-stamped archive files.

Tracing the Directory Decision

Before an archive can be saved, the script must know whether the current date directory is already available. This creates a two-branch decision. If the directory exists, the script can use it. If it does not exist, the script creates it before continuing.

if not os.path.exists(date_directory): os.mkdir(date_directory)

Turning Time into Names

The directory and file names are produced from formatted representations of the current date and time. Python's time.strftime function takes a format code and returns a string representation of the current date or time. In this backup design, the date format is %Y%m%d and the time format is %H%M%S.

date_name = time.strftime('%Y%m%d')
time_name = time.strftime('%H%M%S')

From Paths to Archives

After the date directory is available and the time-based filename has been determined, the script constructs a system command. The command includes the full target path for the ZIP file and the source locations to be archived. The sources are joined into one space-separated string, and the resulting command is passed to os.system.

command = 'zip -r ' + target + ' ' + sources
result = os.system(command)

For example, if the target is /Users/swa/backup/20240115/143022.zip and the source is /Users/swa/notes, the constructed command is zip -r /Users/swa/backup/20240115/143022.zip /Users/swa/notes. os.system executes that command as an operating-system command.

Debugging the Expected Path

Debugging this process means tracing each expected state in order. First identify the formatted date. Next confirm that the main backup directory and date directory form the intended location. Then identify the formatted time and append the .zip extension. Finally inspect the constructed command and the return value from os.system. This path-based trace makes it easier to determine whether a problem occurred during naming, directory preparation, command construction, or archive creation.

Mistakes to Catch

• Creating the date directory every time

  A repeated backup on the same day encounters a directory that has already been created.

  Fix: Use os.path.exists first, and call os.mkdir only when the directory is missing.

• Putting the full timestamp into one flat filename

  Repeated daily backups create a cluttered flat list that is harder to search and verify.

  Fix: Use the date as the subdirectory name and the time as the ZIP filename.

• Using the time value for the directory and the date value for the file

  The intended hierarchy is date-based directories containing time-stamped files.

  Fix: Use %Y%m%d for the date directory and %H%M%S for the ZIP filename.

• Ignoring the os.system return value

  Command construction and command success are separate stages.

  Fix: Check whether os.system returned 0 or a non-zero value.

Practice the Trace

Key Takeaways

1. A hierarchical backup uses a main directory, date-based subdirectories, and time-stamped ZIP files.
2. os.path.exists checks whether the date directory is present, while os.mkdir creates it only when necessary.
3. time.strftime('%Y%m%d') produces the date directory name, and time.strftime('%H%M%S') produces the time-based archive name.
4. The ZIP command combines the target path with the source paths and is executed through os.system.
5. An os.system return value of 0 indicates success; a non-zero value indicates failure.