analyze_files.py

import os

# ----------------------------
# Configuration
# ----------------------------


IGNORE_DIRS = [
    '.git/', 'obj/', 'Debug/', 'Release/', 'bin/', 'ise_work/',
    'x64/', 'x86/', 'tmp/', '_xmsgs/', '_ngo/'
]

IGNORE_SUFFIXES = [
    '.log', '.pdb', '.obj', '.tmp', '.cache', '.xwbt', '.mrp',
    '.vcd', '.jou', '.xml', '.html', '.wlf', '.sdf', '.hex', '.ngc'
]

# explicit filenames to ignore
IGNORE_EXACT = {
    'wdf_from_fpga.bin', 'transcript', 'wdf_from_fpga.hex'
}

# explicit filenames to keep
KEEP_EXACT = {
    'tw_model1_outbuf_golden.hex'
}

KEEP_SUFFIX = {
    '.exe', '.dll', '.bit'
    }

# ----------------------------
# Helper functions
# ----------------------------

def human(size):
    for unit in ['B','K','M','G','T']:
        if size < 1024:
            return f"{size:.1f}{unit}"
        size /= 1024
    return f"{size:.1f}P"

def should_drop(path, suffix):



    # This removes the filename, leaving only the folder path
    folder_path = os.path.dirname(path)
    folder_path = folder_path.replace("\\", "/") + "/"

    # Extract file name (leaf)
    leaf_file = os.path.basename(path)     # → "abc.xyz"

    # explicit suffix keep
    if suffix in KEEP_SUFFIX :
        return 'KEEP_SUFFIX', False
        

    # explicit ignore
    if os.path.basename(path) in IGNORE_EXACT:
        return 'IGNORE_EXACT', True

    # explicit keep
    if os.path.basename(path) in KEEP_EXACT:
        return 'KEEP_EXACT', False

    
    # print(folder_path)
    # directory-based ignore
    for d in IGNORE_DIRS:
        if d in folder_path :
            return 'IGNORE_DIRS', True

    # suffix-based ignore
    if suffix in IGNORE_SUFFIXES:
        return 'IGNORE_SUFFIXES', True

    return 'KEEP', False


casue, drop = should_drop('./test_cases/target_model/logs_artsim_perl/tsim_mem_image.bin', 'bin')

# print(casue, drop)
# exit()

# ----------------------------
# Main logic
# ----------------------------

tracked_count = 0
tracked_size = 0

with open("all_files.txt") as f:
    for line in f:
        path = line.strip()
        if not path or not os.path.isfile(path):
            continue
        
        # compute suffix
        base = os.path.basename(path)
        if "." in base:
            suffix = "." + base.split(".")[-1]
        else:
            suffix = "NULL"

        

        # get size
        try:
            size = os.path.getsize(path)
        except:
            size = -1

        # determine TRACK / DROP
        cause, drop = should_drop(path, suffix)
        decision = "DROP" if drop else "TRACK"

        # accumulate statistics
        if not drop and size >= 0:
            tracked_count += 1
            tracked_size += size

        # print formatted row
        print(f"{path:<150}{suffix:<20}{size:>10}  {decision:<10}  {cause}")

        if decision == 'TRACK' :
            print(f"git add {path:<150}; ## {suffix:<20}{size:>10}  {decision:<10}  {cause}")

# final summary
print("\n---------------------------")
print(f"Tracked files: {tracked_count}")
print(f"Total tracked size: {human(tracked_size)}")
print("---------------------------")