136 lines
4.7 KiB
Python
136 lines
4.7 KiB
Python
"""PPT online preview helpers.
|
|
|
|
Browsers cannot reliably render .pptx files inline, so the app converts a
|
|
deck to a cached PDF and embeds that PDF in the observability preview page.
|
|
"""
|
|
|
|
from __future__ import annotations
|
|
|
|
import hashlib
|
|
import os
|
|
import shutil
|
|
import subprocess
|
|
import tempfile
|
|
from dataclasses import dataclass
|
|
from pathlib import Path
|
|
|
|
|
|
@dataclass(frozen=True)
|
|
class PPTPreviewResult:
|
|
ok: bool
|
|
pdf_path: str | None = None
|
|
cache_hit: bool = False
|
|
converter: str | None = None
|
|
error: str | None = None
|
|
|
|
|
|
@dataclass(frozen=True)
|
|
class PPTPreviewCacheInfo:
|
|
pdf_path: str | None = None
|
|
cache_exists: bool = False
|
|
cache_size_kb: float | None = None
|
|
cache_mtime_ts: float | None = None
|
|
|
|
|
|
def find_libreoffice_binary() -> str | None:
|
|
return shutil.which("libreoffice") or shutil.which("soffice")
|
|
|
|
|
|
def _preview_cache_path(pptx_path: Path, cache_dir: Path) -> Path:
|
|
stat = pptx_path.stat()
|
|
cache_key = hashlib.sha256(
|
|
f"{pptx_path.resolve()}:{stat.st_size}:{int(stat.st_mtime)}".encode("utf-8")
|
|
).hexdigest()[:16]
|
|
safe_stem = "".join(ch if ch.isalnum() or ch in ("-", "_") else "_" for ch in pptx_path.stem)[:96]
|
|
return cache_dir / f"{safe_stem}_{cache_key}.pdf"
|
|
|
|
|
|
def get_ppt_preview_cache_info(
|
|
pptx_path: str | os.PathLike[str],
|
|
*,
|
|
cache_dir: str | os.PathLike[str] | None = None,
|
|
) -> PPTPreviewCacheInfo:
|
|
"""只讀取預期 PDF 快取狀態，不啟動 LibreOffice 轉檔。"""
|
|
source = Path(pptx_path)
|
|
if not source.is_file() or source.suffix.lower() != ".pptx":
|
|
return PPTPreviewCacheInfo()
|
|
|
|
target_dir = Path(cache_dir or os.getenv("PPT_PREVIEW_CACHE_DIR", "/app/data/ppt_previews"))
|
|
target_pdf = _preview_cache_path(source, target_dir)
|
|
if not target_pdf.is_file() or target_pdf.stat().st_size <= 0:
|
|
return PPTPreviewCacheInfo(pdf_path=str(target_pdf), cache_exists=False)
|
|
|
|
stat = target_pdf.stat()
|
|
return PPTPreviewCacheInfo(
|
|
pdf_path=str(target_pdf),
|
|
cache_exists=True,
|
|
cache_size_kb=round(stat.st_size / 1024, 1),
|
|
cache_mtime_ts=stat.st_mtime,
|
|
)
|
|
|
|
|
|
def build_ppt_preview(
|
|
pptx_path: str | os.PathLike[str],
|
|
*,
|
|
cache_dir: str | os.PathLike[str] | None = None,
|
|
timeout_sec: int = 90,
|
|
) -> PPTPreviewResult:
|
|
source = Path(pptx_path)
|
|
if not source.is_file():
|
|
return PPTPreviewResult(ok=False, error="pptx not found")
|
|
if source.suffix.lower() != ".pptx":
|
|
return PPTPreviewResult(ok=False, error="unsupported file type")
|
|
|
|
converter = find_libreoffice_binary()
|
|
if not converter:
|
|
return PPTPreviewResult(
|
|
ok=False,
|
|
error="LibreOffice is not installed in the app container.",
|
|
)
|
|
|
|
target_dir = Path(cache_dir or os.getenv("PPT_PREVIEW_CACHE_DIR", "/app/data/ppt_previews"))
|
|
target_dir.mkdir(parents=True, exist_ok=True)
|
|
target_pdf = _preview_cache_path(source, target_dir)
|
|
if target_pdf.is_file() and target_pdf.stat().st_size > 0:
|
|
return PPTPreviewResult(ok=True, pdf_path=str(target_pdf), cache_hit=True, converter=converter)
|
|
|
|
with tempfile.TemporaryDirectory(prefix="ppt_preview_") as tmp:
|
|
tmpdir = Path(tmp)
|
|
try:
|
|
proc = subprocess.run(
|
|
[
|
|
converter,
|
|
"--headless",
|
|
"--convert-to",
|
|
"pdf",
|
|
"--outdir",
|
|
str(tmpdir),
|
|
str(source),
|
|
],
|
|
capture_output=True,
|
|
timeout=timeout_sec,
|
|
check=False,
|
|
)
|
|
except subprocess.TimeoutExpired:
|
|
return PPTPreviewResult(ok=False, converter=converter, error="LibreOffice conversion timed out.")
|
|
except Exception as exc:
|
|
return PPTPreviewResult(ok=False, converter=converter, error=f"{type(exc).__name__}: {str(exc)[:200]}")
|
|
|
|
if proc.returncode != 0:
|
|
stderr = proc.stderr.decode("utf-8", errors="ignore").strip()
|
|
return PPTPreviewResult(
|
|
ok=False,
|
|
converter=converter,
|
|
error=f"LibreOffice conversion failed: {stderr[:240] or proc.returncode}",
|
|
)
|
|
|
|
generated = tmpdir / f"{source.stem}.pdf"
|
|
if not generated.is_file():
|
|
candidates = sorted(tmpdir.glob("*.pdf"))
|
|
generated = candidates[0] if candidates else generated
|
|
if not generated.is_file() or generated.stat().st_size <= 0:
|
|
return PPTPreviewResult(ok=False, converter=converter, error="LibreOffice did not produce a PDF.")
|
|
|
|
shutil.move(str(generated), str(target_pdf))
|
|
return PPTPreviewResult(ok=True, pdf_path=str(target_pdf), cache_hit=False, converter=converter)
|