frank.conrads/pptx-image-compress
1
0
Fork 0
Code Actions Packages Activity

Bumped CaesiumCLT to 1.3.0, python to 3.14.3

Browse Source
This commit is contained in:
Frank Conrads
2026-04-08 13:32:08 +02:00
parent 85876228eb
commit 0338fd6524
4 changed files with 51 additions and 4 deletions
Download Patch File Download Diff File Expand all files Collapse all files

52
pptx_image_compress.py
View File

@@ -18,6 +18,8 @@ Highlights:
import argparse
import os
import re
import xml.etree.ElementTree as ET
import sys
import zipfile
import tempfile
@@ -30,6 +32,8 @@ from pathlib import Path
from datetime import timedelta
from concurrent.futures import ThreadPoolExecutor, as_completed
from threading import Lock
from typing import List, Optional
__version__ = "1.1.4"
@@ -117,6 +121,38 @@ def format_duration(seconds: float) -> str:
return f"{hms}.{frac[:2]}"
return base
def get_slide_numbers_for_image(rels_dir: str, image_filename: str) -> Optional[List[int]]:
"""
Durchsucht alle .rels-Dateien im angegebenen Verzeichnis und gibt die Slide-Nummern zurück,
in denen die angegebene Bilddatei referenziert wird.
:param rels_dir: Pfad zum Verzeichnis ppt/slides/_rels
:param image_filename: z. B. 'image80.png'
:return: Liste von Slide-Nummern oder None
"""
slide_numbers = []
for rels_file in os.listdir(rels_dir):
if rels_file.startswith("slide") and rels_file.endswith(".xml.rels"):
rels_path = os.path.join(rels_dir, rels_file)
try:
tree = ET.parse(rels_path)
root = tree.getroot()
for rel in root.findall(".//{http://schemas.openxmlformats.org/package/2006/relationships}Relationship"):
target = rel.attrib.get("Target", "")
if image_filename in target:
match = re.search(r"slide(\d+).xml.rels", rels_file)
if match:
slide_number = int(match.group(1))
slide_numbers.append(slide_number)
except ET.ParseError:
print(f"Fehler beim Parsen von {rels_file}")
return slide_numbers if slide_numbers else None
# -------------------- Core per-deck processing --------------------
def process_single_deck(input_pptx: Path, output_pptx: Path, threads: int, quality: int) -> dict:
start_time = time.perf_counter()
@@ -143,7 +179,7 @@ def process_single_deck(input_pptx: Path, output_pptx: Path, threads: int, quali
log_file = output_pptx.with_suffix(".log.csv")
ensure_clean_file(log_file)
log_lines = ["image_name;size_before(kb);size_after(kb);saving(kb);saving_percent(%)\n"]
log_lines = ["image_name;size_before(kb);size_after(kb);saving(kb);saving_percent(%);in_slide_number\n"]
size_before = input_pptx.stat().st_size
result["size_before"] = size_before
@@ -151,8 +187,11 @@ def process_single_deck(input_pptx: Path, output_pptx: Path, threads: int, quali
with zipfile.ZipFile(input_pptx, "r") as z:
z.extractall(work_dir)
slides_dir = work_dir / "ppt" / "slides"
media_dir = work_dir / "ppt" / "media"
images = []
if media_dir.exists():
for f in sorted(media_dir.iterdir()):
if f.is_file() and f.suffix.lower() in ALLOWED_EXT:
@@ -174,7 +213,15 @@ def process_single_deck(input_pptx: Path, output_pptx: Path, threads: int, quali
ext = img_path.suffix.lower()
orig_size = img_path.stat().st_size
chosen_size = orig_size
found_in_slide=None
slide_nr=""
try:
found_in_slide = get_slide_numbers_for_image(slides_dir.name, img_path.name)
if found_in_slide is None:
slide_nr = "NOT_USED"
else:
slide_nr = str(found_in_slide)
out_sub = scratch_dir / f"img_{idx:06d}"
caesium_out = compress_with_caesium(img_path, out_sub, caesium_threads, quality)
if caesium_out and caesium_out.exists():
@@ -189,8 +236,9 @@ def process_single_deck(input_pptx: Path, output_pptx: Path, threads: int, quali
finally:
saving = orig_size - chosen_size
saving_percent = round((saving / orig_size) * 100, 2) if orig_size > 0 else 0.0
with lock:
log_lines.append(f"{img_path.name};{human_kb(orig_size)};{human_kb(chosen_size)};{human_kb(saving)};{saving_percent}\n")
log_lines.append(f"{img_path.name};{human_kb(orig_size)};{human_kb(chosen_size)};{human_kb(saving)};{saving_percent};{slide_nr}\n")
done_count += 1
print_progress(done_count, total)