formalpdf 0.1.5

An Apache-licensed package for extracting, creating, filling, and flattening PDF Forms.

formalpdf

formalpdf is an Apache-licensed python library for PDF forms. It's a unified API for extracting, creating, filling, and flattening forms. It has a similar, but not drop-in, high-level API to PyMuPDF. All of this is possible by wrapping pdfium, thanks to the low-level bindings made available through pypdfium2.

Installation

Using uv

uv pip install formalpdf

Using pip

pip install formalpdf

CLI Usage

🚧 Soon you can use formalpdf to fill a PDF from an FDF document using:

formalpdf fill <input.pdf> <input.fdf> <output.pdf>

Programmatic Usage

Get All Widget Objects in a PDF

import formalpdf

doc = formalpdf.open("path/to/document.pdf") 

for page in doc:
    widgets = page.widgets()

A Widget object has information about the location, type, and contents of form fields/wdigets in the PDF. For isntance, a widget might look like:

Widget(
    # name of the widget
    field_name='Text110',
    # label/alternate name, if provided
    field_label='Date (MM/DD/YYYY)',
    # current value (always a string, even if checkbox or combobox)
    field_value='',
    # widget type enum value
    field_type=6,
    # widget type string value 
    field_type_string='Text',
    # widget location Rect
    rect=Rect(
        top=36.95610046386719,
        left=473.5320129394531,
        bottom=24.171100616455078,
        right=587.177978515625
    )
)

Filling out Forms: Updating Widget Values

Let's say we had some textbox widget:

w = doc[0].widgets()[0]

We can update it with:

w.update("New Value")

doc.save("new_doc.pdf")

And when we open new_doc.pdf we'll find a the value filled out!

Creating Forms

🚧 Work in Progress

Adding Annotations (Highlights, Links, etc.)

🚧 Work in Progress

Rendering, Extracting Text, Extracting Images

🚧 Work in Progress

Navigating Unsupported Operations

You can access the raw PdfDocument from pypdfium2 using by calling:

from formalpdf import Document

doc = Document("/path/to/your.pdf")

pdfium_doc = doc.document

You can use this to do lower-level operations that aren't yet supported by formalpdf. For instance, if you want to render the first page of the document (currently an unsupported option):

import formalpdf

doc = formalpdf.open("path/to/document.pdf")

pdfium_doc = doc.document
bmp = pdfium_doc[0].render(scale=scale)
pil = bmp.to_pil()

Testing

uv run pytest

There are a large number of test PDFs found in tests/data.

Roadmap

• create PyPI package
• finish widget extraction
• widget updating
  • textbox
  • checkbox
  • combobox
• widget creation
• tests