Skip to main content

Changelog

This changelog is generated automatically from GitHub Releases.

v0.2.1

2026-09-02 · GitHub

raghilda 0.2.1 adds Python 3.14 support and fixes Markdown chunking for documents that use bare carriage-return line endings.

Added

  • Added Python 3.14 support by selecting compatible Magika and ONNX Runtime versions.

Changed

  • Affected store and attribute APIs now raise TypeError instead of ValueError when callers pass values of unsupported types.

Fixed

  • Fixed MarkdownChunker failures and source offsets for Markdown containing bare carriage-return line endings.

v0.2.0

2026-06-30 · GitHub

raghilda 0.2.0 expands the package from the core RAG workflow into a more complete toolkit for building and maintaining retrieval stores. The release adds crawl and ingest APIs with caching and concurrency, a Cloudflare-backed crawler for JavaScript-rendered sites, a PostgreSQL store backend, and NVIDIA NIM embedding support.

Added

  • Added raghilda.crawl, including CrawlScope, FetchedSource, DirectoryCrawler, WebCrawler, and CloudflareCrawler, for discovering directory, web, and Cloudflare sources and converting them to markdown documents.
  • Added BaseStore.ingest() and IngestSummary for bulk document ingestion with optional document preparation, parallel writes, and inserted, replaced, and skipped counts.
  • Added crawler caching so repeated or interrupted crawls can reuse fetched and converted content.
  • Added CloudflareCrawler for crawling and converting JavaScript-rendered sites through Cloudflare’s Browser Rendering API.
  • Added PostgreSQLStore, backed by psycopg2 and pgvector, with full-text search, vector search, combined retrieval, attributes, and HNSW index support.
  • Added EmbeddingNVIDIA for NVIDIA NIM embeddings, including differentiated query and document input types and rate-limit backoff.
  • Added user guide pages for quickstart, crawling and ingestion, Cloudflare crawling, and chatlas integration.

Changed

  • CrawlScope.include_patterns and CrawlScope.exclude_patterns now use one glob-style pattern syntax across DirectoryCrawler, WebCrawler, and CloudflareCrawler.
  • Existing regex strings passed to WebCrawler or DirectoryCrawler should be rewritten as globs or passed as compiled re.Pattern objects.
  • Reorganized the user guide onboarding pages and refreshed the README.

Fixed

  • Fixed sitemap URL extraction so each <loc> entry is collected as one URL.
  • Improved DuckDB BM25 retrieval errors when the index has not been built or has become stale.

v0.1.1

2026-03-16 · GitHub

Initial raghilda release