Neue Sturktur

This commit is contained in:
2025-09-10 11:50:23 +02:00
parent 97ae257e44
commit 00f4337d59
5 changed files with 316 additions and 363 deletions

View File

@@ -1,37 +1,47 @@
# PPTX Image Compressor (CaesiumCLT only) # PPTX Image Compressor (CaesiumCLT only)
**Version 1.0.0** **Version 1.0.0**
Ein CLITool, das **PPTX-Dateien** entpackt, alle Bilder in `ppt/media` mit **CaesiumCLT** komprimiert und anschließend wieder zu einer PPTX packt. Dieses Paket enthält:
Es ersetzt nur dann Bilder, **wenn die komprimierte Version kleiner ist**, schreibt ein **CSV-Log** und zeigt eine **Fortschrittsanzeige** sowie eine **Summary inkl. Laufzeit** an.
--- ```
PPTX-Image-Compressor-1.0.0/
├─ README.md
├─ install_and_run.bat
├─ pptx_image_compress.py
├─ bin/
│ └─ PUT_caesiumclt_here.txt
└─ samples/
└─ README.txt
```
## Features ## Schnellstart (ohne Admin-Rechte)
- **Nur CaesiumCLT** als Bildkompressor (keine Pillow-Abhängigkeit). 1) Lege `caesiumclt.exe` in den Ordner `bin/` **oder** sorge dafür, dass es im `PATH` liegt.
- Unterstützte Bildtypen: **JPG/JPEG, PNG, WebP** (GIF wird übersprungen). 2) Doppelklicke `install_and_run.bat` **oder** rufe es in CMD/PowerShell auf, z.B.:
- **Overwrite-Policy:** `-O bigger` (nur überschreiben, wenn Ziel größer ist).
- **Multi-Threading**: Parallele Bildverarbeitung.
- **Log** (`.log` neben der Output-Datei) mit:
`image_name,size_before,size_after,saving,saving_percent`.
- **Sauberes Cleanup**: Keine Caesium-Tempdateien in der finalen PPTX.
--- ```bat
install_and_run.bat -i "C:\Slides\Deck.pptx" -t 8
```
## Systemvoraussetzungen Die Batch lädt bei Bedarf automatisch das **Windows Embeddable Python Package**, entpackt es lokal und führt das Tool aus.
- **Windows 10/11 (64-bit)**
- **Keine Admin-Rechte erforderlich**
- **CaesiumCLT** (portabler Download)
- **Python** (portabel, ohne Installation) **oder** fertige EXE
--- ## Was das Tool macht
- Entpackt die PPTX in einen TempOrdner
- Komprimiert **JPG/JPEG, PNG, WebP** mit **CaesiumCLT** (`-q 90`, `-O bigger`)
- Ersetzt Bilder nur, wenn die komprimierte Datei kleiner ist
- Schreibt ein CSVLog (`.log` neben der OutputPPTX)
- Baut eine neue PPTX und zeigt eine Summary (Name, Größe vorher/nachher, Ersparnis %, Zeit)
- Räumt alle temporären Dateien auf (keine CaesiumTempfiles in der finalen PPTX)
## Installation & Setup ## Hinweise
- **GIF** wird übersprungen (keine Rekodierung).
- `-t` steuert die Parallelität (PythonThreads); intern wird `caesiumclt --threads 1` gesetzt, sobald `-t > 1`, um Oversubscription zu vermeiden.
### 1) CaesiumCLT installieren ## Manuelle Nutzung des .py (falls Python vorhanden)
1. Lade **CaesiumCLT** von GitHub herunter: ```bat
[https://github.com/Lymphatus/caesium-clt/releases](https://github.com/Lymphatus/caesium-clt) python pptx_image_compress.py -i "C:\Pfad\input.pptx" -t 8
2. Entpacke `caesiumclt.exe` in einen Ordner, z.B. `C:\Tools\caesiumclt`. ```
3. Füge den Ordner temporär zum PATH hinzu (optional):
```bat ## Quellen & Tools
set PATH=C:\Tools\caesiumclt;%PATH% - CaesiumCLT Projekt/Downloads: https://github.com/Lymphatus/caesium-clt
``` - Windows Embeddable Python Package Doku/Downloads: https://docs.python.org/3/using/windows.html

View File

@@ -0,0 +1 @@
Place caesiumclt.exe in this folder or ensure it is available on PATH.

View File

@@ -1,3 +1,4 @@
@echo off @echo off
setlocal EnableExtensions EnableDelayedExpansion setlocal EnableExtensions EnableDelayedExpansion
@@ -29,7 +30,7 @@ if exist "%SELF_DIR%%CAE_EXE%" (
where /q %CAE_EXE% where /q %CAE_EXE%
if errorlevel 1 ( if errorlevel 1 (
echo [ERROR] ^> 'caesiumclt.exe' nicht gefunden. echo [ERROR] ^> 'caesiumclt.exe' nicht gefunden.
echo Lege 'caesiumclt.exe' neben diese BAT ^(empfohlen^) echo Lege 'caesiumclt.exe' neben diese BAT (empfohlen)
echo oder sorge dafuer, dass es im PATH liegt. echo oder sorge dafuer, dass es im PATH liegt.
exit /b 2 exit /b 2
) )
@@ -89,7 +90,7 @@ if not defined PY_CMD (
exit /b 5 exit /b 5
) )
rem ---- Optional: 'import site' im Embeddable aktivieren (freundlich fuer spaetere Erweiterungen) ---- rem ---- Optional: 'import site' im Embeddable aktivieren ----
if exist "%PY_DIR%" ( if exist "%PY_DIR%" (
for /f "delims=" %%F in ('dir /b "%PY_DIR%\python3*.pth" 2^>nul') do ( for /f "delims=" %%F in ('dir /b "%PY_DIR%\python3*.pth" 2^>nul') do (
set "PTH_FILE=%PY_DIR%\%%F" set "PTH_FILE=%PY_DIR%\%%F"
@@ -111,7 +112,6 @@ echo.
echo [%APP_NAME%] Starte ... echo [%APP_NAME%] Starte ...
echo Command: "%PY_CMD%" "%SCRIPT%" %* echo Command: "%PY_CMD%" "%SCRIPT%" %*
echo. echo.
"%PY_CMD%" "%SCRIPT%" %* "%PY_CMD%" "%SCRIPT%" %*
set "RC=%ERRORLEVEL%" set "RC=%ERRORLEVEL%"

View File

@@ -28,21 +28,16 @@ from datetime import timedelta
from concurrent.futures import ThreadPoolExecutor, as_completed from concurrent.futures import ThreadPoolExecutor, as_completed
from threading import Lock from threading import Lock
# -------------------- Version --------------------
__version__ = "1.0.0" __version__ = "1.0.0"
# -------------------- Konfiguration -------------------- ALLOWED_EXT = {".jpg", ".jpeg", ".png", ".webp", ".gif"}
ALLOWED_EXT = {".jpg", ".jpeg", ".png", ".webp", ".gif"} # GIF wird übersprungen CAESIUM_QUALITY = 90
CAESIUM_QUALITY = 90 # -q 90
PROGRESS_BAR_LEN = 40 PROGRESS_BAR_LEN = 40
TEMP_PREFIX = "pptx_compress_" TEMP_PREFIX = "pptx_compress_"
# -------------------- Utilities --------------------
def human_mb(nbytes: int) -> float: def human_mb(nbytes: int) -> float:
return round(nbytes / (1024 * 1024), 2) return round(nbytes / (1024 * 1024), 2)
def ensure_clean_file(path: Path): def ensure_clean_file(path: Path):
if path.exists(): if path.exists():
try: try:
@@ -53,7 +48,6 @@ def ensure_clean_file(path: Path):
except Exception: except Exception:
pass pass
def cleanup_old_temps(): def cleanup_old_temps():
tmp_root = Path(tempfile.gettempdir()) tmp_root = Path(tempfile.gettempdir())
for p in tmp_root.glob(f"{TEMP_PREFIX}*"): for p in tmp_root.glob(f"{TEMP_PREFIX}*"):
@@ -65,15 +59,13 @@ def cleanup_old_temps():
except Exception: except Exception:
pass pass
def print_progress(i: int, total: int): def print_progress(i: int, total: int):
if total <= 0: if total <= 0:
return return
done = int(PROGRESS_BAR_LEN * i / total) done = int(PROGRESS_BAR_LEN * i / total)
bar = "" * done + "-" * (PROGRESS_BAR_LEN - done) bar = "" * done + "-" * (PROGRESS_BAR_LEN - done)
pct = int(i * 100 / total) pct = int(i * 100 / total)
print(f"\rBilder: |{bar}| {i}/{total} ({pct}%)", end="", flush=True) print(f"
Bilder: |{bar}| {i}/{total} ({pct}%)", end="", flush=True) Bilder: |{bar}| {i}/{total} ({pct}%)", end="", flush=True)
def zip_dir_to_pptx(src_dir: Path, out_pptx: Path): def zip_dir_to_pptx(src_dir: Path, out_pptx: Path):
@@ -83,53 +75,37 @@ def zip_dir_to_pptx(src_dir: Path, out_pptx: Path):
full = Path(root) / f full = Path(root) / f
rel = full.relative_to(src_dir) rel = full.relative_to(src_dir)
z.write(full, arcname=str(rel)) z.write(full, arcname=str(rel))
def which(cmd: str) -> str | None:
def which(cmd: str): def which(cmd: str):
return shutil.which(cmd) return shutil.which(cmd)
"""
Ruft caesiumclt auf, um eine komprimierte Version zu erzeugen.
Output wird ins out_dir geschrieben (gleicher Filename).
Gibt Pfad zur erzeugten Datei zurück oder None bei Fehler.
"""
def compress_with_caesium(original: Path, out_dir: Path, caesium_threads: int | None) -> Path | None: def compress_with_caesium(original: Path, out_dir: Path, caesium_threads: int | None) -> Path | None:
exe = which("caesiumclt") exe = which("caesiumclt")
raise RuntimeError( if not exe:
"'caesiumclt' wurde nicht gefunden. Bitte CaesiumCLT installieren und in PATH verfügbar machen."
)
raise RuntimeError("'caesiumclt' wurde nicht gefunden. Bitte CaesiumCLT installieren und in PATH verfügbar machen.") raise RuntimeError("'caesiumclt' wurde nicht gefunden. Bitte CaesiumCLT installieren und in PATH verfügbar machen.")
# Nur Formate an Caesium geben, die es unterstützt: JPG/JPEG, PNG, WEBP
out_dir.mkdir(parents=True, exist_ok=True) out_dir.mkdir(parents=True, exist_ok=True)
ext = original.suffix.lower() ext = original.suffix.lower()
return None # GIF & andere werden übersprungen if ext not in {".jpg", ".jpeg", ".png", ".webp"}:
return None
cmd = [
exe,
"-q", str(CAESIUM_QUALITY),
"-O", "bigger", # <<< nur überschreiben, wenn Ziel größer ist
"-o", str(out_dir),
]
cmd = [exe, "-q", str(CAESIUM_QUALITY), "-O", "bigger", "-o", str(out_dir)] cmd = [exe, "-q", str(CAESIUM_QUALITY), "-O", "bigger", "-o", str(out_dir)]
if caesium_threads is not None: if caesium_threads is not None:
cmd += ["--threads", str(caesium_threads)] cmd += ["--threads", str(caesium_threads)]
cmd += [str(original)] cmd += [str(original)]
try: try:
r = subprocess.run(cmd, capture_output=True, text=True) r = subprocess.run(cmd, capture_output=True, text=True)
sys.stderr.write(f"\n[caesiumclt] Fehler bei {original.name}:\n{r.stderr}\n") if r.returncode != 0:
sys.stderr.write(f"
[caesiumclt] Fehler bei {original.name}:
{r.stderr}
") ")
return None return None
out_file = out_dir / original.name out_file = out_dir / original.name
return out_file if out_file.exists() else None return out_file if out_file.exists() else None
sys.stderr.write(f"\n[caesiumclt] Ausnahme bei {original.name}: {ex}\n") except Exception as ex:
sys.stderr.write(f"
[caesiumclt] Ausnahme bei {original.name}: {ex}
") ")
return None return None
def format_duration(seconds: float) -> str: def format_duration(seconds: float) -> str:
total_ms = int(round(seconds * 1000)) total_ms = int(round(seconds * 1000))
@@ -139,27 +115,14 @@ def format_duration(seconds: float) -> str:
hms, frac = base.split(".", 1) hms, frac = base.split(".", 1)
return f"{hms}.{frac[:2]}" return f"{hms}.{frac[:2]}"
return base return base
def main(): def main():
start_time = time.perf_counter() start_time = time.perf_counter()
parser = argparse.ArgumentParser(
description="PPTX Grafik-Komprimier-Tool (nur CaesiumCLT, Multi-Thread, sauberes Cleanup)",
formatter_class=argparse.ArgumentDefaultsHelpFormatter,
)
parser = argparse.ArgumentParser(description="PPTX Grafik-Komprimier-Tool (nur CaesiumCLT, Multi-Thread, sauberes Cleanup)", formatter_class=argparse.ArgumentDefaultsHelpFormatter) parser = argparse.ArgumentParser(description="PPTX Grafik-Komprimier-Tool (nur CaesiumCLT, Multi-Thread, sauberes Cleanup)", formatter_class=argparse.ArgumentDefaultsHelpFormatter)
parser.add_argument("-i", "--input", help="Input-PPTX", required=False) parser.add_argument("-i", "--input", help="Input-PPTX", required=False)
parser.add_argument( parser.add_argument("-o", "--output", help="Output-PPTX", required=False)
"-t", "--threads", parser.add_argument("-t", "--threads", type=int, default=min(32, os.cpu_count() or 4), help="Anzahl paralleler Threads für die Bildverarbeitung")
type=int,
default=min(32, os.cpu_count() or 4),
help="Anzahl paralleler Threads für die Bildverarbeitung"
)
parser.add_argument(
"--version",
action="version",
version=f"%(prog)s {__version__}"
)
parser.add_argument("--version", action="version", version=f"%(prog)s {__version__}") parser.add_argument("--version", action="version", version=f"%(prog)s {__version__}")
args = parser.parse_args() args = parser.parse_args()
@@ -172,28 +135,22 @@ def main():
if not input_pptx.exists() or input_pptx.suffix.lower() != ".pptx": if not input_pptx.exists() or input_pptx.suffix.lower() != ".pptx":
print("❌ Eingabedatei existiert nicht oder ist keine .pptx") print("❌ Eingabedatei existiert nicht oder ist keine .pptx")
sys.exit(2) sys.exit(2)
if args.output:
output_pptx = Path(args.output).resolve()
else:
output_pptx = input_pptx.with_name(f"{input_pptx.stem}_compressed.pptx")
output_pptx = Path(args.output).resolve() if args.output else input_pptx.with_name(f"{input_pptx.stem}_compressed.pptx") output_pptx = Path(args.output).resolve() if args.output else input_pptx.with_name(f"{input_pptx.stem}_compressed.pptx")
# Vorherige Temp-Files & existierendes Output löschen
cleanup_old_temps() cleanup_old_temps()
ensure_clean_file(output_pptx) ensure_clean_file(output_pptx)
# --- Zwei getrennte Temp-Verzeichnisse ---
work_dir = Path(tempfile.mkdtemp(prefix=TEMP_PREFIX + "work_")) # entpackte PPTX work_dir = Path(tempfile.mkdtemp(prefix=TEMP_PREFIX + "work_"))
scratch_dir = Path(tempfile.mkdtemp(prefix=TEMP_PREFIX + "scratch_")) # Caesium-Ausgaben (außerhalb!)
scratch_dir = Path(tempfile.mkdtemp(prefix=TEMP_PREFIX + "scratch_")) scratch_dir = Path(tempfile.mkdtemp(prefix=TEMP_PREFIX + "scratch_"))
# Logdatei neben Output
log_file = output_pptx.with_suffix(".log") log_file = output_pptx.with_suffix(".log")
log_lines = ["image_name,size_before,size_after,saving,saving_percent\n"] ensure_clean_file(log_file)
log_lines = ["image_name,size_before,size_after,saving,saving_percent
"] "]
size_before = input_pptx.stat().st_size size_before = input_pptx.stat().st_size
# Entpacken
try: try:
with zipfile.ZipFile(input_pptx, "r") as z: with zipfile.ZipFile(input_pptx, "r") as z:
z.extractall(work_dir) z.extractall(work_dir)
@@ -208,15 +165,13 @@ def main():
total = len(images) total = len(images)
print(f"🔧 Finde Bilder in {media_dir} ... {total} Kandidaten") print(f"🔧 Finde Bilder in {media_dir} ... {total} Kandidaten")
print_progress(0, total) print_progress(0, total)
# Vorab prüfen, ob caesiumclt verfügbar ist
print("\n'caesiumclt' nicht gefunden. Bitte installieren und in PATH verfügbar machen.") if not which("caesiumclt"):
print("
'caesiumclt' nicht gefunden. Bitte installieren und in PATH verfügbar machen.") 'caesiumclt' nicht gefunden. Bitte installieren und in PATH verfügbar machen.")
sys.exit(3) sys.exit(3)
# Oversubscription vermeiden: viele Python-Threads -> caesium intern 1 Thread
caesium_threads = 1 if args.threads and args.threads > 1 else None caesium_threads = 1 if args.threads and args.threads > 1 else None
# Thread-sichere Fortschritts- & Log-Verwaltung
lock = Lock() lock = Lock()
done_count = 0 done_count = 0
@@ -224,52 +179,43 @@ def main():
def worker(idx: int, img_path: Path): def worker(idx: int, img_path: Path):
nonlocal done_count nonlocal done_count
ext = img_path.suffix.lower() ext = img_path.suffix.lower()
# GIF überspringen
orig_size = img_path.stat().st_size orig_size = img_path.stat().st_size
if ext == ".gif": if ext == ".gif":
with lock: with lock:
log_lines.append(f"{img_path.name},{orig_size},{orig_size},0,0.0\n") done_count += 1
log_lines.append(f"{img_path.name},{orig_size},{orig_size},0,0.0
") ")
print_progress(done_count, total) print_progress(done_count, total)
return return
chosen_size = orig_size chosen_size = orig_size
# Eigener Output-Unterordner pro Bild, um Kollisionen zu vermeiden
try: try:
out_sub = scratch_dir / f"img_{idx:06d}" out_sub = scratch_dir / f"img_{idx:06d}"
caesium_out = compress_with_caesium(img_path, out_sub, caesium_threads) caesium_out = compress_with_caesium(img_path, out_sub, caesium_threads)
if caesium_out and caesium_out.exists(): if caesium_out and caesium_out.exists():
s = caesium_out.stat().st_size s = caesium_out.stat().st_size
# kleineren ersetzen (atomar)
if s < orig_size: if s < orig_size:
tmp_target = img_path.with_suffix(img_path.suffix + ".tmp") tmp_target = img_path.with_suffix(img_path.suffix + ".tmp")
shutil.copy2(caesium_out, tmp_target) shutil.copy2(caesium_out, tmp_target)
tmp_target.replace(img_path) tmp_target.replace(img_path)
chosen_size = s chosen_size = s
chosen_size = orig_size # Original beibehalten except Exception:
chosen_size = orig_size chosen_size = orig_size
finally: finally:
saving = orig_size - chosen_size saving = orig_size - chosen_size
saving_percent = round((saving / orig_size) * 100, 2) if orig_size > 0 else 0.0 saving_percent = round((saving / orig_size) * 100, 2) if orig_size > 0 else 0.0
log_lines.append(f"{img_path.name},{orig_size},{chosen_size},{saving},{saving_percent}\n") with lock:
log_lines.append(f"{img_path.name},{orig_size},{chosen_size},{saving},{saving_percent}
") ")
done_count += 1 done_count += 1
print_progress(done_count, total) print_progress(done_count, total)
# Parallel ausführen
if total > 0: if total > 0:
with ThreadPoolExecutor(max_workers=max(1, args.threads)) as ex: with ThreadPoolExecutor(max_workers=max(1, args.threads)) as ex:
futures = [ex.submit(worker, i, p) for i, p in enumerate(images, start=1)] futures = [ex.submit(worker, i, p) for i, p in enumerate(images, start=1)]
pass # Fortschritt wird im Worker gezeichnet for _ in as_completed(futures):
pass pass
print() # newline nach Progressbar
print() print()
# --- Safety-Cleanup innerhalb des Arbeitsverzeichnisses ---
# 1) Entferne evtl. vorhandene caesium*-Ordner (aus alten Runs)
for p in work_dir.rglob("*"): for p in work_dir.rglob("*"):
try: try:
@@ -277,7 +223,6 @@ def main():
shutil.rmtree(p, ignore_errors=True) shutil.rmtree(p, ignore_errors=True)
except Exception: except Exception:
pass pass
# 2) Lösche eventuelle .tmp-Dateien in ppt/media
media_dir = work_dir / "ppt" / "media" media_dir = work_dir / "ppt" / "media"
if media_dir.exists(): if media_dir.exists():
@@ -287,24 +232,23 @@ def main():
f.unlink(missing_ok=True) f.unlink(missing_ok=True)
except Exception: except Exception:
pass pass
# Neue PPTX bauen (nur work_dir -> scratch_dir liegt außerhalb und ist damit sicher ausgeschlossen)
zip_dir_to_pptx(work_dir, output_pptx) zip_dir_to_pptx(work_dir, output_pptx)
size_after = output_pptx.stat().st_size size_after = output_pptx.stat().st_size
# Log schreiben
try: try:
with open(log_file, "w", encoding="utf-8") as f: with open(log_file, "w", encoding="utf-8") as f:
f.writelines(log_lines) f.writelines(log_lines)
except Exception: except Exception:
# Summary pass
savings_pct = 0.0 savings_pct = 0.0
if size_before > 0: if size_before > 0:
savings_pct = round(100.0 * (size_before - size_after) / size_before, 2) savings_pct = round(100.0 * (size_before - size_after) / size_before, 2)
elapsed = time.perf_counter() - start_time elapsed = time.perf_counter() - start_time
print("\n✅ Fertig!")
print("
Fertig!") Fertig!")
print("Summary") print("Summary")
print("-------") print("-------")
@@ -316,7 +260,6 @@ def main():
print(f"Zeit benötigt: {format_duration(elapsed)}") print(f"Zeit benötigt: {format_duration(elapsed)}")
print(f"Log-Datei: {log_file}") print(f"Log-Datei: {log_file}")
# Aufräumen ALLER temporären Dateien/Ordner
finally: finally:
try: try:
shutil.rmtree(work_dir, ignore_errors=True) shutil.rmtree(work_dir, ignore_errors=True)
@@ -325,9 +268,7 @@ def main():
try: try:
shutil.rmtree(scratch_dir, ignore_errors=True) shutil.rmtree(scratch_dir, ignore_errors=True)
except Exception: except Exception:
# Zusätzlich: ältere Reste entfernen
pass pass
cleanup_old_temps() cleanup_old_temps()
if __name__ == "__main__": if __name__ == "__main__":

1
samples/README.txt Normal file
View File

@@ -0,0 +1 @@
Place your PPTX files here for testing, or use -i with a full path.