pdq

Rust-native PDF split and merge MVP.

Runtime constraints:

• does not invoke the qpdf binary;
• does not use a subprocess wrapper;
• does not link libqpdf through FFI.

The first implementation uses lopdf as a pure-Rust PDF object model and writer. It focuses on valid split/merge outputs for ordinary, unencrypted PDFs. Advanced qpdf behavior such as repair, encryption, linearization, forms, outlines, and full compatibility with unusual historical PDFs is intentionally out of scope for the current MVP.

Commands

cargo test
cargo run --bin pdq -- split input.pdf --out 1-3 out-1.pdf --out 4-z out-2.pdf
cargo run --bin pdq -- split-pages --output 'page-%d.pdf' input.pdf
cargo run --bin pdq -- merge --output merged.pdf a.pdf b.pdf

Tests may use qpdf as a development validator when it is available on PATH. The runtime implementation must remain qpdf-free.

Benchmark Snapshot

Measured on 2026-07-03 with local PDFs identified only by page count. Wall time is hyperfine --warmup 1 --runs 5 mean plus standard deviation. RSS is maximum resident set size from a separate single /usr/bin/time -l run, so use it as a memory-order signal rather than a statistically averaged value.

Completed outputs were validated by page count and qpdf --warning-exit-0 --check. Split scenarios validated the first and last output file. qpdf used --remove-unreferenced-resources=no for copy-like paths where applicable.

To reproduce the timing matrix:

PDQ_BIG_PDF=/path/to/12732-pages.pdf \
PDQ_SMALL_PDF=/path/to/2642-pages.pdf \
scripts/benchmark.sh