scitex-msword 0.3.2

MS Word (.docx) reader/writer with journal-style profiles (MDPI, IEEE, Springer, Elsevier, RESNA, IOP) — standalone module from the SciTeX ecosystem

scitex-msword

MS Word (.docx) reader/writer with journal-style profiles.

Full Documentation · uv pip install scitex-msword[all]

---

Installation

pip install scitex-msword

Quick Start

import scitex_msword as sxm

# Word -> intermediate JSON-like document
doc = sxm.load_docx("input.docx", profile="generic")

# JSON-like document -> Word (apply a journal style)
sxm.save_docx(doc, "output.docx", profile="mdpi-ijerph")

# DOCX -> LaTeX (requires the umbrella `scitex` package for the .tex export step)
sxm.convert_docx_to_tex(
    "manuscript.docx", "manuscript.tex",
    profile="resna-2025", image_dir="figures",
)

1 Interfaces

Python API

import scitex_msword as sxm

# Round-trip
doc = sxm.load_docx("paper.docx", profile="generic")
sxm.save_docx(doc, "paper-styled.docx", profile="ieee")

# Helpers
sxm.link_captions_to_images(doc)
sxm.link_captions_to_images_by_proximity(doc)
sxm.normalize_section_headings(doc)
sxm.validate_document(doc)
sxm.create_post_import_hook(doc)

# Register custom profile
sxm.register_profile("my-style", {...})

Review / dogfooding helpers (unreleased)

import docx
import scitex_msword as sxm

# 1. Diff two .docx versions by paragraph (paragraphs in/out + run-level
#    bold / italic / font / highlight deltas).
ops = sxm.diff_docx("v15.docx", "v16.docx")
sxm.summarize_diff(ops)            # {'equal': 38, 'insert': 4, 'delete': 1, 'modify': 3}

# 2. Visualize edits with highlights (BOOST review convention).
doc = docx.Document("v16.docx")
sxm.mark_additions(doc, runs=[(3, 0), (5, 2)])     # default turquoise
sxm.mark_modifications(doc, runs=[(7, 1)])         # default magenta -> Word PINK

# 3. Read highlights back, bucketed by color name.
sxm.extract_highlights(doc)        # {'turquoise': [...], 'pink': [...]}

# 4. Bold-preserve keyword tokens (Japanese tokens get MS Gothic).
sxm.preserve_bold_tokens(doc, tokens=["JST", "BOOST", "Sovereign Tech"])

# 5. Pull Word comments + their anchor ranges.
comments = sxm.extract_comments("v16.docx")
# Optionally apply REPLACE:-grammar comments as edits.
summary = sxm.apply_comments_as_edits(doc)         # {'applied': 2, 'skipped': 4, ...}

Track Changes (revision) helpers

import docx
import scitex_msword as sxm

# 1. Turn Word's "Track Changes" switch on, so subsequent operator edits
#    are recorded as revisions (writes <w:trackChanges/> to settings.xml).
doc = docx.Document("draft.docx")
sxm.enable_track_changes(doc, enabled=True)
sxm.is_track_changes_enabled(doc)        # True

# 2. Wrap agent edits as <w:ins> / <w:del> so Word renders them as
#    accept/reject-able revisions.
p = doc.paragraphs[10]
sxm.wrap_as_tracked_insertion(p, runs=[2, 3], author="agent")
sxm.wrap_as_tracked_deletion(p, runs=[5],    author="agent")

# 3. Inspect all tracked changes (structured).
for c in sxm.extract_tracked_changes(doc):
    print(c["type"], c["author"], c["text"])

# 4. Bulk accept / reject (Word's "Accept All" / "Reject All").
sxm.accept_all_tracked_changes(doc)        # or reject_all_tracked_changes
doc.save("draft_v27.docx")

MCP server (optional)

pip install scitex-msword[mcp]
python -m scitex_msword.mcp_server          # stdio transport

Tools exposed: diff_docx_tool, mark_additions_tool, mark_modifications_tool, preserve_bold_tokens_tool, extract_highlights_tool, extract_comments_tool, list_profiles_tool.

Built-in profiles

generic, mdpi-ijerph, resna-2025, iop-double-anonymous, ieee, springer, elsevier, boost-2026.

Status

Standalone fork of scitex.msword. Only runtime dep is python-docx. The umbrella scitex.msword import path is preserved via a sys.modules-alias bridge. convert_docx_to_tex lazily imports scitex.tex, so it works only when the umbrella package is also installed.

Architecture

scitex_msword/
├── _load.py              ← `load_docx` — DOCX → JSON-like document
├── _save.py              ← `save_docx` — apply profile, write DOCX
├── _convert.py           ← `convert_docx_to_tex` (lazy scitex.tex import)
├── profiles/             ← built-in journal styles
│   ├── generic.py        ← default
│   ├── ieee.py
│   ├── mdpi_ijerph.py
│   ├── resna_2025.py
│   ├── springer.py
│   └── elsevier.py
├── helpers/              ← caption-image linking, heading normalization
└── _registry.py          ← `register_profile` for user styles

Demo

flowchart LR
    A[draft.docx] -->|load_docx| B[JSON-like doc]
    B -->|save_docx<br/>profile=ieee| C[submission.docx]
    B -->|convert_docx_to_tex| D[manuscript.tex]

import scitex_msword as sxm

doc = sxm.load_docx("draft.docx", profile="generic")
sxm.save_docx(doc, "submission.docx", profile="ieee")

Round-trips DOCX through a JSON-like intermediate, then re-renders with IEEE column widths, fonts, and heading numbering applied automatically.

Part of SciTeX

scitex-msword is part of SciTeX. Install via the umbrella with pip install scitex[msword] to use as scitex.msword (Python) or scitex msword ... (CLI).

Four Freedoms for Research

0. The freedom to run your research anywhere — your machine, your terms.
1. The freedom to study how every step works — from raw data to final manuscript.
2. The freedom to redistribute your workflows, not just your papers.
3. The freedom to modify any module and share improvements with the community.

AGPL-3.0 — because we believe research infrastructure deserves the same freedoms as the software it runs on.

License

AGPL-3.0-only (see LICENSE).

---