0.12.0.rc2

• fix: explicit fingerprinter param in custom dupefilter by @SheezZarR in #73

Full Changelog: 0.12.0.rc1...0.12.0.rc2

0.12.0.rc1

• Flatten hierarchy_radio fields during record transformation by @nascosto in #70

Full Changelog: 0.11.0...0.12.0.rc1

0.11.0

• feat: multi-arch by @darkweaver87 in #58
• Switched base OS to Debian from Ubuntu

Full Changelog: 0.10.0...0.11.0

0.11.0.rc1

• feat: multi-arch support by @darkweaver87 in #58
• Switched base OS to Debian from Ubuntu

Full Changelog: 0.10.0...0.11.0.rc1

0.10.0

New features:

• feat: carry curation rules and synonyms to new collection after scraper runs by @tharropoulos in #66
• feat: add support for http_auth_domain spider attribute by @nkls-so in #67

Dev Workflow Contributions:

• Fix start URL with JS test by @CodeSandwich in #55
• Move from circleci to GitHub actions by @nascosto in #60
• Fix link to the guide by @teners in #59

⚠️ Breaking Change:

If you're using API keys with limited permissions in the scraper, you now need to add additional permissions for synonyms and override API endpoints to the API key, to allow Typesense to copy these resources over from the old collection to the new collection.

Here's the full set of permissions you'd need in the API Key:

actions: ["collections:*", "documents:*", "aliases:*", "synonyms:*", "overrides:*"]

Full Changelog: 0.9.1...0.10.0

0.9.1

What's Changed

• Include "Bot" token in default user agent string by @Krinkle in #48
• Increase connection timeout, to account for potential embedding generation for semantic search

Full Changelog: 0.8.0...0.9.1