# 3.10.1 - Release Notes Squirro 3.10.1 was released April 4, 2024. Learn more about the [Squirro Release Process](../../squirro-release-process.md#getting-squirro-process). ## What’s New - Add a new special chip for the favorite filter in GlobalSearch that allows editing the chip and expanding the saved query. - Pipeline steps executed in `batch` execution mode now feature a `timeout` configuration option. It enforces a limit (in seconds) after which a timeout error is raised if the step takes more than that to process a received batch of items. This helps in unblocking the ingester processor that picked up this batch. ## Improvements - Create RPM packages for nlp services required by semantic search projects. This allows to run nlp services on the Squirro node server for fast semantic search PoCs. - The configuration options for the `rerun_from_index` queue were split from the `queues_local` section into their own section called `queues_rerun_from_index`. - Add support for all tab tooltip in the `tabs` widget. - Added `query_context` parameter to the `/starter-questions` endpoint of the GenAI Studio plugin, which allows filtering/limiting the set of Squirro Items used to generate the questions. - Search: Improve editable chip targetability. - Search on flat items: Answer is initially highlighted and scrolled to when user opens item detail. - Search on flat items: Answer match in item detail sidebar is marked as answer in title. - The `squirro_lxml_html_extractor` option in the `common.ini` has no effect anymore and it has been removed from the packaged `common.ini` file. The lxml parser is now used by default. - Added context length to `sqgpt` configuration. Users can now manually set the context length that will be sent to an LLM. This will overwrite the context length estimation and may lead to issues if set wrong. - Re-enable project translations for widget title in React widgets. - The `ingester` service can now be configured without any processors and solely serve its web API. Previously, it would exit immediately with a vague message when no processors were configured. - In addition to the `timeout` option, the `Text Chunking` step now exposes an option to specify the max item body size. This prevents the step from processing items with a very large body. - Remove the `ingester.priorities.enabled` server configuration option and make it the only way of operation for `ingester`. The number of processors for each priority can still be adjusted by the `ingester.priorities.pool-*-processors` server configuration options. - The configuration option `processors` in the `ingester.ini` file is now deprecated and it will be removed in an upcoming release. You can safely remove it from your `/etc/squirro/ingester.ini` file. - Added automatic context length determination to summarization endpoint. - Enable query auto submit after chips deletion for global search and search widget. - Multiple improvements to the `Tabs` widget. - An active dashboard store will not be cleared when the global search is cleared. - GenAI Squirro Retriever will now only use `query_context` to run queries. This deprecates the usage of the `query` param. This improvement will result in better copilot results. - Improved Context length guessing by adding the newest gpt 4 models and updating legacy models. - Add item title typeahead suggestions which will open ItemDetail with the item directly from global search. ## Bug Fixes - Fix dashboard not reacting to in-page search change if parse request fails. - Handling of empty `query_context` values: genai will now handle empty values for query context the same as None, this is a workaround for an issue where specifying empty values will return unexpected results. - Fixed the issue where an item had an answer to the query, and inside the item detail of HTML the item highlight matches have been shifted by 1. - [Search] Scoring profile `prefix_match` was failing on queries that contained only numeric tokens. - Fixed issue where rerun from index tasks, exceeding the configured `redis_visibility_timeout_secs` value, became available for execution again, resulting in duplicated rerun batches. - Fixed `communityWizard` value not being properly updated when modifying/creating a dashboard. - Fix issue that was preventing the instantiation of a content streamer by an ingester processor. - Set default `top_p` to 1.0 in SquirroGPT config schema to support Mistral on Anyscale. - [GenAI Studio Plugin] Fix `max_num_words` not being a cast to int. Fix `context_length == -1` not getting resolved to default value. - Fix favorite filter’s query that couldn’t be edited. ## Installation and Upgrade For new installations, find step-by-step instructions in [Install and Manage Squirro with Ansible](../../../install/ansible/index.md#ansible) (recommended) or [Installing Squirro on Linux](../../../install/linux.md#install-linux). To upgrade an existing installation, see [Upgrading Squirro](../../../install/upgrade.md#install-upgrade).