7Z_Python_Script/old_01/compress_folder.py

#!/usr/bin/env python3
"""
Folder Compressor using 7-Zip
Lists available folders, lets you select one, and compresses it
using 7-Zip with no compression (store) at fastest speed.
Real-time progress bar shown during compression.
"""

import os
import re
import sys
import shutil
import subprocess
import time
from pathlib import Path


# ── Terminal color helpers ────────────────────────────────────────────────────

SUPPORTS_COLOR = (
    sys.stdout.isatty()
    and (sys.platform != "win32" or os.environ.get("WT_SESSION") or os.environ.get("TERM"))
)

def _c(code: str, text: str) -> str:
    return f"\033[{code}m{text}\033[0m" if SUPPORTS_COLOR else text

def green(t):  return _c("32", t)
def yellow(t): return _c("33", t)
def cyan(t):   return _c("36", t)
def bold(t):   return _c("1",  t)
def dim(t):    return _c("2",  t)
def red(t):    return _c("31", t)


# ── Size / time formatters ────────────────────────────────────────────────────

def fmt_size(b: float) -> str:
    for unit in ("B", "KB", "MB", "GB", "TB"):
        if b < 1024:
            return f"{b:.1f} {unit}"
        b /= 1024
    return f"{b:.1f} PB"


def fmt_time(seconds: float) -> str:
    s = int(seconds)
    if s < 60:
        return f"{s}s"
    m, s = divmod(s, 60)
    if m < 60:
        return f"{m}m {s:02d}s"
    h, m = divmod(m, 60)
    return f"{h}h {m:02d}m {s:02d}s"


# ── Folder size calculator ────────────────────────────────────────────────────

def folder_size(path: Path) -> int:
    total = 0
    for root, _, files in os.walk(path):
        for f in files:
            try:
                total += os.path.getsize(os.path.join(root, f))
            except OSError:
                pass
    return total


# ── 7-Zip locator ─────────────────────────────────────────────────────────────

def find_7zip() -> str | None:
    for name in ("7z", "7za", "7zz", "7zip"):
        path = shutil.which(name)
        if path:
            return path
    if sys.platform == "win32":
        for p in (
            r"C:\Program Files\7-Zip\7z.exe",
            r"C:\Program Files (x86)\7-Zip\7z.exe",
        ):
            if os.path.isfile(p):
                return p
    return None


# ── Folder listing / selection ────────────────────────────────────────────────

def list_folders(base_dir: str) -> list[Path]:
    base = Path(base_dir)
    return sorted(
        (e for e in base.iterdir() if e.is_dir()),
        key=lambda p: p.name.lower(),
    )


def prompt_folder_selection(folders: list[Path]) -> Path:
    print("\nAvailable folders:\n")
    for i, folder in enumerate(folders, start=1):
        try:
            sz = folder_size(folder)
            size_label = dim(f"  ({fmt_size(sz)})")
        except PermissionError:
            size_label = dim("  (no access)")
        print(f"  {dim(f'[{i:>3}]')}  {folder.name}{size_label}")

    print()
    while True:
        try:
            raw = input("Enter the number of the folder to compress: ").strip()
            idx = int(raw)
            if 1 <= idx <= len(folders):
                return folders[idx - 1]
            print(f"  Please enter a number between 1 and {len(folders)}.")
        except ValueError:
            print("  Invalid input — please enter a number.")
        except (EOFError, KeyboardInterrupt):
            print("\nAborted.")
            sys.exit(0)


# ── Progress bar renderer ─────────────────────────────────────────────────────

BAR_WIDTH = 36

# Matches 7-Zip progress segments like "47% 3 + path/to/file.bin"
# or just "47%" when no file is on the same segment
_PROGRESS_RE = re.compile(r"(\d{1,3})%(?:\s+(\d+)\s*\+?\s*(.*))?")

# Split on ANY run of backspaces (\x08) or carriage returns (\r).
# - Linux/macOS 7-Zip uses \x08 to erase and rewrite the progress line
# - Windows 7-Zip uses \r to overwrite the progress line
# Both need to be handled so the script works cross-platform.
_SPLIT_RE = re.compile(rb"[\x08\r]+")


def _move_up(n: int):
    if SUPPORTS_COLOR:
        sys.stdout.write(f"\033[{n}A")


def _clear_line():
    if SUPPORTS_COLOR:
        sys.stdout.write("\033[2K\r")
    else:
        sys.stdout.write("\r" + " " * 78 + "\r")


def _redraw(pct: int, elapsed: float, files_done: int,
            current_file: str, speed_bps: float):
    """Overwrite the two reserved progress lines in-place."""
    filled  = int(BAR_WIDTH * pct / 100)
    bar     = "█" * filled + "░" * (BAR_WIDTH - filled)
    color   = green if pct == 100 else cyan
    speed_s = f"{fmt_size(speed_bps)}/s" if speed_bps > 0 else "─"

    _move_up(2)

    # Line 1 — bar + stats
    _clear_line()
    sys.stdout.write(
        f"  {color(bar)}  {bold(f'{pct:>3}%')}"
        f"  {dim('elapsed')} {dim(fmt_time(elapsed))}"
        f"  {yellow(speed_s)}"
        f"  {dim(f'files: {files_done}')}\n"
    )

    # Line 2 — current file
    _clear_line()
    label = current_file.strip() or "starting…"
    if len(label) > 60:
        label = "…" + label[-59:]
    sys.stdout.write(f"  {dim('↳ ' + label)}\n")

    sys.stdout.flush()


# ── Compression with real-time progress ───────────────────────────────────────

def compress_folder(seven_zip: str, folder: Path, output_dir: Path) -> Path:
    archive_path = output_dir / (folder.name + ".7z")

    print(f"\n  {dim('Calculating source size…')} ", end="", flush=True)
    src_size = folder_size(folder)
    print(fmt_size(src_size))

    cmd = [
        seven_zip,
        "a",
        "-t7z",
        "-mx=0",   # store – no compression
        "-ms=off", # no solid archive
        "-bsp1",   # stream progress to stdout
        str(archive_path),
        str(folder),
    ]

    print(f"\n  {dim('Command: ' + ' '.join(cmd))}")
    print()
    print("  " + "─" * 64)

    # Reserve two blank lines for the live display
    print()
    print()

    start_time   = time.monotonic()
    last_pct     = 0
    files_done   = 0
    current_file = ""
    speed_bps    = 0.0
    last_bytes   = 0.0
    speed_ts     = start_time

    proc = subprocess.Popen(
        cmd,
        stdout=subprocess.PIPE,
        stderr=subprocess.STDOUT,
        # Raw binary — we parse bytes ourselves to handle \x08 and \r
    )

    buf = b""

    while True:
        chunk = proc.stdout.read(256)
        if not chunk:
            break

        buf   += chunk
        elapsed = time.monotonic() - start_time

        # Split on runs of \x08 (Linux/macOS) or \r (Windows).
        # Each segment between separators is one "visible write" from 7-Zip.
        segments = _SPLIT_RE.split(buf)

        # Keep the last (potentially incomplete) segment for the next iteration
        buf = segments[-1]

        for seg in segments[:-1]:
            text = seg.decode("utf-8", errors="replace").strip()
            if not text:
                continue

            m = _PROGRESS_RE.search(text)
            if m:
                pct        = min(int(m.group(1)), 100)
                files_done = int(m.group(2)) if m.group(2) else files_done
                fname      = (m.group(3) or "").strip()
                if fname:
                    current_file = fname

                # Rolling speed estimate (recalculate every 0.4 s)
                if src_size > 0 and pct > 0:
                    bytes_done = src_size * pct / 100
                    now        = time.monotonic()
                    dt         = now - speed_ts
                    if dt >= 0.4:
                        speed_bps  = (bytes_done - last_bytes) / dt
                        last_bytes = bytes_done
                        speed_ts   = now

                last_pct = pct
                _redraw(pct, elapsed, files_done, current_file, speed_bps)

    proc.wait()
    elapsed_total = time.monotonic() - start_time

    # Lock bar at 100% on success
    if proc.returncode == 0:
        _redraw(100, elapsed_total, files_done, current_file, speed_bps)

    print()
    print("  " + "─" * 64)

    if proc.returncode == 0:
        arc_size  = archive_path.stat().st_size
        avg_speed = src_size / elapsed_total if elapsed_total > 0 else 0

        print(f"\n  {green('✅  Archive created successfully!')}")
        print(f"     {bold('Path')}     : {archive_path}")
        print(f"     {bold('Source')}   : {fmt_size(src_size)}")
        print(f"     {bold('Archive')}  : {fmt_size(arc_size)}")
        print(f"     {bold('Time')}     : {fmt_time(elapsed_total)}")
        print(f"     {bold('Avg speed')}: {fmt_size(avg_speed)}/s")
    else:
        print(f"\n  {red(f'❌  7-Zip exited with code {proc.returncode}.')}")
        sys.exit(proc.returncode)

    return archive_path


# ── Entry point ───────────────────────────────────────────────────────────────

def main():
    scan_dir = os.path.abspath(sys.argv[1] if len(sys.argv) > 1 else os.getcwd())

    if not os.path.isdir(scan_dir):
        print(f"Error: '{scan_dir}' is not a valid directory.")
        sys.exit(1)

    print(f"\n{bold('📦  Folder Compressor')}")
    print(f"    Scanning: {cyan(scan_dir)}\n")

    seven_zip = find_7zip()
    if not seven_zip:
        print(
            f"\n{red('❌  7-Zip not found.')}\n"
            "    Install it and make sure '7z' (or '7za') is on your PATH.\n"
            "    • Linux  : sudo apt install p7zip-full\n"
            "    • macOS  : brew install p7zip\n"
            "    • Windows: https://www.7-zip.org/download.html"
        )
        sys.exit(1)

    print(f"  {green('✔')}  7-Zip: {dim(seven_zip)}")

    folders = list_folders(scan_dir)
    if not folders:
        print(f"\nNo sub-folders found in '{scan_dir}'.")
        sys.exit(0)

    chosen = prompt_folder_selection(folders)
    print(f"\n  {green('✔')}  Selected: {bold(chosen.name)}")

    compress_folder(seven_zip, chosen, chosen.parent)
    print()


if __name__ == "__main__":
    main()