Skip to content

Perq 0.2.0

Choose a tag to compare

View all tags
@shaneaevans shaneaevans released this 06 Sep 10:48
· 1 commit to master since this release
v0.2.0
7fd437a

Perq matches incoming documents against large sets of saved queries inside a single Python process. This release modernizes PSearch for saved-search alerts, scraping pipelines, and document routing.

  • A dependency-free Python API and JSONL CLI, with MemoryStore, SQLite, and optional LMDB.
  • Atomic index replacement, consistent read snapshots, stable string/integer IDs, AND/OR token rules, and numeric range filters.
  • Runnable scraper examples, JSON schemas, agent integration guidance, and reproducible benchmarks. The short uniform synthetic workloads measured 5,801 documents/second against 100,000 queries and 599 documents/second against one million queries in MemoryStore.

Migration: import and CLI names are now perq. The query API and persistent index format have changed; rebuild old indexes from their original query definitions. Read the migration guide before upgrading.

Install the wheel attached to this release:

python -m pip install https://github.com/shaneaevans/perq/releases/download/v0.2.0/perq-0.2.0-py3-none-any.whl

Perq is also available on PyPI. Install this version with python -m pip install perq==0.2.0. See the README, benchmark methodology and results, and changelog.

Assets 5
Loading