fix(desktop): bundle document parsing libs

This commit is contained in:
zhayujie
2026-06-29 11:07:08 +08:00
parent 538281da51
commit e536232963
2 changed files with 22 additions and 1 deletions

View File

@@ -16,7 +16,7 @@ class Read(BaseTool):
"""Tool for reading file contents"""
name: str = "read"
description: str = f"Read or inspect file contents. For text/PDF files, returns content (truncated to {DEFAULT_MAX_LINES} lines or {DEFAULT_MAX_BYTES // 1024}KB). For images/videos/audio, returns metadata only (file info, size, type). Use offset/limit for large text files."
description: str = f"Read or inspect file contents. For text/PDF/Word/Excel/PPT files, returns content (truncated to {DEFAULT_MAX_LINES} lines or {DEFAULT_MAX_BYTES // 1024}KB). For images/videos/audio, returns metadata only (file info, size, type). Use offset/limit for large text files."
params: dict = {
"type": "object",

View File

@@ -73,6 +73,21 @@ hiddenimports += [
'tiktoken_ext.openai_public',
]
# Document parsing libs. The read / web_fetch tools import these lazily inside
# functions (e.g. `from pypdf import PdfReader`), so PyInstaller's static
# analysis misses them and they'd be dropped from the bundle — leaving the
# desktop client unable to read PDF/Word/Excel/PPT. List them explicitly.
hiddenimports += [
'pypdf',
'docx', # python-docx
'pptx', # python-pptx
'openpyxl',
]
hiddenimports += collect_submodules('pypdf')
hiddenimports += collect_submodules('docx')
hiddenimports += collect_submodules('pptx')
hiddenimports += collect_submodules('openpyxl')
# --- Data files -----------------------------------------------------------
# Runtime-read files/dirs that must travel with the executable. Paths are
# (source, dest_dir_in_bundle).
@@ -92,6 +107,12 @@ datas = [
# Some libraries (tiktoken encodings, etc.) ship data files.
datas += collect_data_files('tiktoken_ext', include_py_files=False)
# python-docx / python-pptx bundle template files (default.docx / default.pptx,
# content-type XML) inside their packages; they're loaded at import/parse time,
# so ship them or document parsing fails in the frozen build.
datas += collect_data_files('docx')
datas += collect_data_files('pptx')
# --- Excludes -------------------------------------------------------------
# Keep the bundle lean: drop Feishu's heavy SDK, plugins (disabled in desktop
# mode), tests/docs, and dev-only packages.