pdfer_forms 0.3.2

Fast pure-Rust PDF form filling, AcroForm inspection, and document operations (merge, split, rotate, encrypt) — a pypdf / PyPDF2 compatibility layer.
Documentation 
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29

# Changelog

## 0.3.1

### Added
- `update_page_form_field_values` now also matches a requested field by its leaf
  `/T` (the final `.`-segment) when the fully-qualified name and partial name
  don't match. Callers (LLM agents) reliably copy the leaf field name but often
  drop/mangle intermediate AcroForm group segments (e.g. request
  `…Page1[0].f1_07[0]` for a field actually at `…Page1[0].Address_ReadOrder[0].f1_07[0]`);
  the leaf `/T` identifies the widget, so these now resolve instead of silently
  not matching.


## 0.3.0

### Fixed
- `update_page_form_field_values` now fills *merged field+widget* annotations
  (the annotation itself carries `/FT` and `/T`, as in most IRS / USCIS / DMV
  forms) by their **fully-qualified name** — the same name `get_fields()`
  returns. Previously the qualified name was computed from the widget's
  `/Parent`, dropping the leaf `/T` segment, so filling by the name reported by
  inspection silently matched nothing.

### Changed (breaking)
- `update_page_form_field_values` now returns `Result<FillReport>` instead of
  `Result<()>`. `FillReport` reports `matched` / `unmatched` requested field
  names and `widgets_updated`, so callers can detect (and surface) a fill that
  matched no widget instead of treating a no-op as success.