Fix increment_path() explicit file vs dir handling

[glenn-jocher]

Resolves #5341 (comment)

Tests:

import glob
import re
from pathlib import Path


def increment_path(path, exist_ok=False, sep='', mkdir=False):
    # Increment file or directory path, i.e. runs/exp --> runs/exp{sep}2, runs/exp{sep}3, ... etc.
    path = Path(path)  # os-agnostic
    if path.exists() and not exist_ok:
        path, suffix = (path.with_suffix(''), path.suffix) if path.is_file() else (path, '')
        dirs = glob.glob(f"{path}{sep}*")  # similar paths
        matches = [re.search(rf"%s{sep}(\d+)" % path.stem, d) for d in dirs]
        i = [int(m.groups()[0]) for m in matches if m]  # indices
        n = max(i) + 1 if i else 2  # increment number
        path = Path(f"{path}{sep}{n}{suffix}")  # increment path
    if mkdir:
        path.mkdir(parents=True, exist_ok=True)  # make directory
    return path


print(increment_path('runs'))
print(increment_path('export.py'))
print(increment_path('abc.def.dir'))
print(increment_path('abc.def.file'))

# Returns
# runs2
# export2.py
# abc.def.dir2
# abc.def2.file

🛠️ PR Summary

_{Made with ❤️ by Ultralytics Actions}

🌟 Summary

Optimized path increment function for better handling of file and directory names.

📊 Key Changes

• Simplified path and suffix extraction into a single line.
• Ensured the incrementation logic distinguishes between files and directories.

🎯 Purpose & Impact

• 🛠 Fixes potential bugs by streamlining the suffix handling process, which could prevent errors when working with file paths.
• 📁 Improves the clarity of the code, making it easier to maintain and understand, which is beneficial for both developers working on the project and users running into path incrementation issues.
• ✅ Users should experience more reliable file handling when the system manages similar file or directory names, particularly during automated tasks such as logging model training runs.