Squirro 3.3.5 - Release Notes
Features
NLP-Tagger improvements:
Extend the VaderSentimentAnalyzer to extract phrases associated with negative or positive sentiment. For more information please refer to this page: NLP Tagger#Sentiment-Analysis.
Key-Phrase cleaning (strip determiners).
Add a simple Topic Tagger based on key-phrase deduplication and more aggressive cleaning.
Add a term/phrase filter step to exclude unwanted Named-Entities or Sentiment-Terms.
Extend text-cleaning to remove Emails & HTTP-Links.
Update the SVC and the Naive Bayes ML templates from using Bag of Words embeddings to TF-IDF embeddings.
Add the ability to apply sentence-level models on PDF documents and potentially get entities (but without highlighting available on the PDF viewer yet).
When using the scrolling API, do not scroll on scores but rely on document-index order as is.
Show which rerun types are available for a pipeline workflow.
Add an endpoint to create a timeline chart based on a facet's count with aggregation on another facet.
Bug Fixes
Removed leftovers from prior query-template handling for typeahead - typeahead failed after initial ACL-refactoring.
Fix full screen item detail for mobile.
Improve the performance of processing PDF documents in the machinelearning service.
Fixed issue with downloading attachments without content.
Close scroll context after scrolling is done.
Fix duplicate test name; Change Percy logging level.
Ensure tokens are not included in log files as part of authentication flow.
Fix word selection on PDF documents.
Fix updates of user profiles.
Document level ground truths will not offer word selection anymore.
Results per page select should be working now users and AI studio listings.
Known Issues
Editing an existing data source may result in some cases to a security-releated error (e.g. It is forbidden to use server files). A fix will be available in the next release, but if you encounter this error using this release, you can try one of the following workarounds:
Set the configuration option
allowed_server_path_directories
in the sectiondata_load
of the/etc/squirro/common.ini
file to/var/lib/squirro/temp
and then restart thedatasource
service (systemctl restart sqdatasourced
).Delete the data source and recreate it.
Installation and Upgrade
You will have to resolve at least the following config files when upgrading from Squirro 3.3.0
/etc/nginx/conf.d/ssl.inc
/etc/squirro/common.ini
For new installations, please follow the Setup on Linux instructions.
To upgrade an existing installation, please consult the Upgrades for Squirro 3.2.0 and later guide.