Packages

All packages were built from source checkpoint 06a9bab.

• Rust crate pdf-inspector 1.15.0 — cargo add pdf-inspector@1.15.0
• Python pdf-inspector 1.15.0 — pip install pdf-inspector==1.15.0
• Node @firecrawl/pdf-inspector 1.15.0 — npm install @firecrawl/pdf-inspector@1.15.0
• WebAssembly @firecrawl/pdf-inspector-wasm 1.15.0 — npm install @firecrawl/pdf-inspector-wasm@1.15.0

Highlights

• Adds selective OCR for scanned pages and mixed PDFs using PDFium rendering and PP-OCRv6 Small through ONNX Runtime (#348, #357, #358).
• Recovers credible PDFium text layers before OCR, routes only affected pages, and adaptively fuses OCR with usable native text (#359, #393, #394).
• Reuses OCR sessions and preserves row boundaries for faster, cleaner extraction (#390, #391).
• Exposes the OCR pipeline through Rust, pdf2md, Node.js, and Python, including page-level provenance, confidence, timings, warnings, and hosted-parser recommendations (#360, #403, #405).
• Adds pinned runtime and model caching, offline operation, and cross-language smoke coverage (#404, #409).

Other improvements

• Rejoins words hyphenated across line breaks (#388).
• Improves running-header, running-footer, and numbered-reference-note handling (#374, #392).

Compare changes since 1.14.2

Packages

All packages were built from source checkpoint 4bee4f9.

• Rust crate pdf-inspector 1.14.2 — cargo add pdf-inspector@1.14.2
• Python pdf-inspector 1.14.2 — pip install pdf-inspector==1.14.2
• Node @firecrawl/pdf-inspector 1.14.2 — npm install @firecrawl/pdf-inspector@1.14.2
• WebAssembly @firecrawl/pdf-inspector-wasm 1.14.2 — npm install @firecrawl/pdf-inspector-wasm@1.14.2

Highlights

• Caps Form XObject expansion, CID /W ranges, Encoding/ToUnicode CMaps, content-stream decode, detector Tj/TJ lookback, and disjoint-rect clustering so crafted PDFs cannot force unbounded CPU or memory (#370, #372, #373, #375, #379, #380, #381).
• Form XObjects now track the text line matrix and honor T* / TL / ' / " / Tc / Tw, so nested form text keeps the right spacing (#369).
• Small-caps runs merge instead of being read as extra table columns (#371).

Changes since 1.14.1

Source: f4aab3b...4bee4f9

Resource bounds

• Bound Form XObject expansion per page (#370)
• Bound CID /W range expansion (#372)
• Cap content-stream decode before allocating operators (#373)
• Bound Encoding CMap cidrange expansion (#375)
• Bound ToUnicode bfrange expansion during subset remap (#379)
• Bound detector Tj/TJ operand lookback to the previous operator (#380)
• Bound disjoint-rect table clustering so overlap tests stay subquadratic (#381)

Extraction quality

• Track the text line matrix and handle T* / TL / ' / " / Tc / Tw in Form XObjects (#369)
• Merge small-caps runs so they stop reading as table columns (#371)

Packages

All packages were built from source checkpoint 965dc65.

• Rust crate pdf-inspector 0.1.8 — cargo add pdf-inspector@0.1.8
• Python pdf-inspector 0.2.7 — pip install pdf-inspector==0.2.7
• Node @firecrawl/pdf-inspector 1.13.0 — npm install @firecrawl/pdf-inspector@1.13.0
• WebAssembly @firecrawl/pdf-inspector-wasm 0.1.4 — npm install @firecrawl/pdf-inspector-wasm@0.1.4

Highlights

• Hardens parsing against nested-object, AcroForm /Kids, structure-tree, malformed Unicode/glyph-name, and extreme-coordinate denial-of-service or panic cases (#198, #314, #320, #321, #322, #328).
• Improves extraction correctness for contextual digit runs, redline/strikeout handling, escaped comments, Base-14 fonts, CFF CID fonts, page-prefixed Markdown, corrupt startxref pointers, OCR-routing parity, and password-protected item JSON (#201, #209, #230, #231, #240, #241, #245, #253, #259, #284).
• Improves ruled-table preservation, chart reading order, and rejection of script attachments or tiny numeric table fragments (#262, #264).
• Adds non-blocking async variants to the Node API (#337) and synchronizes Python type stubs with the shipped bindings (#250).
• Refreshes packaged API and benchmark documentation (#192).

Changes since each package's previous release

The package versions were not previously consolidated, so each changelog starts at that package's actual published commit rather than an old repository-wide v* tag.

This is the first package-umbrella GitHub release. Earlier v* tags tracked individual package history and should not be treated as a unified release line.