v1.0.152 v1.23.0
high risk
tt-a1i/archifySep 22, 2026 → Sep 22, 20260 days between releases
What's New
Major PDF text extraction overhaul with RTL, invisible text, and Form XObject support.
Breaking Changes
- Changes to text extraction order and detection logic
New Features
- Right-to-left visual order reading
- Invisible text layer detection
- Form XObject text extraction for large numerals
- ToUnicode interpolation for single-glyph gaps
- Detached spacing accent handling
Improvements
- Form XObject BBox saturation for large numerals
- Zero-advance dependent sign handling
- ToUnicode CMap coverage reporting
- Right-to-left logical order reading for invisible text
- Per-page budget for text rendering mode analysis
Bug Fixes
- Form XObject text extraction when numerals are too large
- Right-to-left text reading in visual order
- ToUnicode interpolation for unambiguous gaps
- Detached spacing accent composition
- Zero-advance dependent sign handling
Upgrade Notes
This is a major version jump (1.0 to 1.23) involving a complete overhaul of text extraction. Consumers relying on the previous text order or detection logic should review output changes, particularly regarding right-to-left text and invisible layers.
The Two Releases
v1.0.152
v1.0.152
Sep 22, 2026· 1 month ago
A minor release that corrects a query parsing bug where parent predicates were incorrectly propagated to child elements when using the `>` combinator.
Full release notesv1.23.0
1.23.0 — Text-layer repairs
Sep 22, 2026· 1 month ago
Fixes text-layer repairs for complex PDFs, including handling Form XObjects and right-to-left text.
Full release notes