officework

xlsx and docx engines built not to destroy your forms — and that print. Read a spreadsheet or a document, change it, write back only what you changed, and turn it into a PDF. No office suite, no headless browser, no print driver.

Written in Rust (about 109,000 lines in the engine crates, 1,600+ tests), exposed to Python through PyO3.

日本語の説明は GitHub にあります (Japanese documentation on GitHub): Python の手引き

Install

$ pip install officework

0.5.0 is in beta. It is published as a pre-release, so the line above still gives you 0.4.0. To try the beta:

$ pip install --pre officework

Wheels are abi3 (CPython 3.10+), so one wheel per platform covers every version; Linux, macOS and Windows are published. The wheel is just the engines — a few MB, no GUI, nothing to install alongside. pandas is imported only if you ask for it (pip install officework[pandas]).

Spreadsheets

from officework import sheet
b = sheet.Book.open("form7.xlsx")
s = b["quote"]
s.insert_row(30)                         # make room; the formulas below follow the move
s["A30"] = "Sample Trading Co., Ltd."    # only the value is rewritten
s["C30"] = "=B30*100"                    # a formula; recalculated on the spot
print(s["C30"].value)                    # the computed value, as in openpyxl
b.save("out.xlsx")                       # the xlsx to send on
b.save("quote.pdf")                      # the same sheet, straight to paper

Indexing follows openpyxl exactly: ws["A1"] is a cell, ws["A1:C3"] a tuple of rows, ws[1] a row, ws["A"] a column. Reading an untouched address gives you a cell whose .value is None, not None itself — so writing into a merged region's top-left works the way the articles show.

Documents

from officework import doc
d = doc.Doc.open("report.docx")
d.replace("Old Name Ltd.", "New Name Ltd.")   # replaces the text, run by run
d.fill("customer", "Sample Trading K.K.")     # fill a named form field
d.save("out.docx")                       # the docx to send on
d.save("report.pdf")                     # the same document, typeset for paper

PDF and PNG — the part the others cannot do

save() looks at the extension. The same book or document you just edited becomes a PDF, laid out by the same typesetting engine that drives the desktop app, so the paper matches the screen:

b.save("quote.pdf")                      # the sheet, paginated, repeating header rows
d.save("report.pdf")                     # the document, typeset

.png gives you the same page as an image — for a thumbnail, a preview in a web page, or a picture to drop into a chat message:

b.save("quote.png")                      # 150 dpi by default; A4 comes out 1240x1754
d.save("report.png", dpi=300)            # print resolution

One file per page. The first page keeps the name you gave it, and later pages get -2, -3 appended, so a three-page document writes report.png, report-2.png and report-3.png. Both formats come off the same laid-out page, so the image and the PDF agree.

This runs on a server with nothing else installed. There is no LibreOffice to launch, no Chromium to drive, no wkhtmltopdf, no temporary HTML. It is one library call, and it is fast enough to sit inside a request handler.

Fonts are subsetted, so only the glyphs you used are embedded. A Japanese page comes out under 30 KB — measured here at 8 KB for plain text and 25 KB for a page with a table, colour and shading — where embedding a whole CJK font costs 20 MB. Line breaking follows JIS X 4051, so Japanese text does not break before a closing bracket or after an opening one.

Neither openpyxl nor python-docx can produce a PDF at all. The commercial libraries that can are priced accordingly.

Shapes in a document

python-docx can only place pictures. This engine can place real shapes — rectangles, rounded rectangles, ellipses, arrows, diamonds and lines — pinned to the page, with fill, outline, text inside, rotation, opacity and a drop shadow:

d.add_shape("roundRect", 25, 80, 40, 25, fill="DDE7F0", line="2E5A87",
            text="Approved", shadow=True)

Coordinates and sizes are millimetres from the top-left of the page. They are written as DrawingML, so Word and LibreOffice open them as shapes you can select and edit — not as a flattened picture. They also come out in the PDF, they are read back when you reopen the file, and page= puts a shape on any page, not just the first.

Charts

Charts are drawn by this library, as shapes, rather than written as an instruction for Excel to render later. So they appear in the PDF, not only after someone opens the file in Excel:

ws.add_chart("bar", data="B3:C8", categories="A4:A8", at="A10",
             title="Target and actual")

All eleven chart types openpyxl can write are built in: bar, line, area, pie, doughnut, projected pie, radar, scatter, bubble, stock (high-low-close) and surface. For finer control there is a small chart layer whose shape is borrowed from d3 — build a scale, then place marks through it:

from officework import chart
c = chart.Chart(340, 180, title="Attainment")
x = c.band(branches)
y = c.linear([0, 150])
c.axis_left(y, fmt=lambda v: f"{int(v)}%")
c.bars(x, y, rates, color="70AD47", labels=True)
c.place(ws, "A20")

What you give up is a live Excel chart: ours is fixed at the data it was drawn from. Redraw it to update it.

Equations

officework.tex takes LaTeX and returns SVG or PNG. With TeX installed it typesets there (matrix columns align); without it, matplotlib's mathtext does the job; with neither, it refuses with the reason — never a silent empty picture.

Your old vocabulary still works

Code written for openpyxl or python-docx largely runs as-is:

ws = wb.active                          # openpyxl: cell(), append, iter_rows,
ws.cell(2, 3).value                     #   dimensions, create_sheet,
ws.append(["Aug", "pens", 5000])        #   copy_worksheet, freeze_panes …
d.tables[0].cell(0, 1).text             # python-docx: row_cells, columns,
d[3].runs[0].font.name                  #   runs, clear …

The inventory — every core member of the reference libraries, judged one by one — is in the repo: docs/pysheet-gokan.ja.adoc. Interop is proven with the originals' own eyes: openpyxl reads what this engine writes, including the computed values it cannot produce itself. See the Python manual for the details and the deliberate differences.

Why the engines exist

openpyxl and python-docx rewrite the parts of the file they do not understand. For a document used as a printed form — the way most Japanese offices use one — that means the borders, merged cells, column widths, shapes, styles and headers you spent an afternoon on come back wrong.

When these engines open an xlsx or docx and write it back, they rewrite only what changed. b.unsupported / d.unsupported list what they recognised as unreadable, as far as they can tell.

The docx side is checked against an independent reader (genoffice's TypeScript docx engine) over 51 real documents, 43 of which this project did not write: 46 survive an open-and-save untouched, and no document loses a single part of its zip. The rest — footnote marks, second and later section breaks, equations — are listed in d.unsupported rather than dropped quietly.

Measured on one machine, 1096 rows × 20 columns (21,920 cells):

License

AGPL-3.0-or-later.

Using it inside your company — building forms, running ledgers, writing scripts — carries no obligations at all. Obligations appear only if you ship something built on it to third parties, or offer a modified version as a network service.