exclude page parts from indexing using a regex

Version upped

translation update

Fix PT-BR Translation

As stated in https://github.com/dokuwiki/dokuwiki/pull/4353, I believe I created the translation on the wrong language (PT instead of PT-BR), so as this is specific to PT-BR I'

Fix PT-BR Translation

As stated in https://github.com/dokuwiki/dokuwiki/pull/4353, I believe I created the translation on the wrong language (PT instead of PT-BR), so as this is specific to PT-BR I've renamed it.

show more ...

Translate error message

Version upped

updated Ollama models

I only added a couple of popular ones. It's hard to find the context and
dimension limits.

support the Google Gemini API

updated model definitions

translation update

Update VoyageAI multilingual model

Add missing field for "Ollama" model configuration

Version upped

added configurable custom prompt

This makes it easier to add to the existing internal prompt instead of
completely overwriting it using the localization mechanism.

Version upped

store embedding meta data on start

This way errors during the a new embedding run will not trigger a clear
index on retry

retry failed QDrant queries

hosted qdrant seems to be a bit unreliable. retrying might help.

show used query when doing similarity queries

updated model lists

This adds a CLI tool to fetch some model data from @BerriAI/litellm

Version upped

monkey patching vanderlee/php-sentence#/27

Version upped

added claude 3.5 sonnet model

auto codestyle cleanup

Use custom renderer when creating embeddings

Rendering makes plugin output available and and handles includes. It
might also help with #15.
The renderer uses markdown like output since all LLMs seem

Use custom renderer when creating embeddings

Rendering makes plugin output available and and handles includes. It
might also help with #15.
The renderer uses markdown like output since all LLMs seem to be very
familiar with it's syntax. This might help them to understand the
document structure better.
This also adds a breadcrumb trail at the top of each chunk which might
help with contextulization as well.

show more ...