Contents - What's in the release?

Fixes

• Sped up Spacy document processing, required for the intelligent typeahead on item content

• Visual fixes around item detail when body is empty.

• Fixed the import of Cognitive search project template.

• Fixed the Importing a project from a Squirro instance which does not have the refactored pipeline steps does not work

• [Properies] Facets dropdown not scrollable

• [Pipeline Editor] Step facet properties not being saved after changing value

• Explore: Sometimes sub panel fails to render with entities selected

• Remove Data URLs from emails other than Newsletter

• "[Data Sources] ""Add feed source URL"" button has a weird circle animation"

• "Pipeline editor - The link under Related in the left panel ""Add new Relate stepts in the AI Studio"" links a user to the 'dashboards' space"

• [Pipeline] Pipeline edit middle draggable section frontend implementation

• Pipeline testing

• Unable to rename step in pipeline workflow

• Extend translations support for Studio plugins

• Extend translations support for Guides

• Translate options for Frontend Connectors

• Alignment/Overlap issues with some languages

• "[Data Sources] ""Run"" on CSV and FEED data source returns 204 - no content"

• [backend] Serialize the default_value of this kind of config option to a JSON formatted string

• [backend] Enable modification of pipeline step names

• "All existing KEE steps should be migrated to ""Relate"" automatically"

• Content based typeahead - Caching key needs to add project identifier

• Ground Truth list view: list of documents overflows

• Item detail: x button is misaligned.

• Project selector (name) is missing if there's only one dashboard

• Ground Truth: labeling bigger documents triggers 414 error

• Settings pages are missing space at the bottom

• Setup experimentation project (index some 100k to million documents)

• "[Explore Dashboard] Creating ""Favourite"" with alert returns 503"

• Unable to create alerts

• [Sprint goal] Finish full implementation for AI Studio feedback processing

• Redesign frontend coding challenges

• Design concept for selection of sentence splitting rule set

• Implementation of sentence splitter picking concept

• Use the new sentence splitting step in the DSS templates

• Create rule set to split MRN data

• Move sentence splitting logic out of integration sub module

• Create advanced sentence splitting step in lib/nlp with addable rule option

• "Gather other document styles, which cause issues"

• Extend Ground Truth to store sentence splitting info

• View for configuring a Ground Truth is not rendered properly

• Data loader frontend config - Defaults don't show up

• Data loader frontend config - booleans break frontend when mappings.json provided

• Language switching in dashboards / frontend

• Ability to rerun whole pipeline for already processed data

• Pre-populate pipeline with default steps

• Weird mixed state leads to inability to logout

• Pipeline design review

• Pipeline UX design

• Salesforce SDK: Downloading fonts from Salesforce instead of Squirro

• [Sprint goal] Migrate pipeline API for new frontend

• [Sprint goal] Pipeline frontend phase 2

• Salesforce SDK: Styles bleed from Squirro to Salesforce

• Salesforce SDK: unable to open items

• Cannot change avatar picture - save button missing

• Search bar overlaps tabs on communities screen

• Cognitive Search: PDF Highlighting of Query Matches not working

• Fresh installation of filbeat prints out errors

• [Sprint goal] Finalise Sales Insights UI

• Update newsletter template to look like the original designs

• Ensure that out of the box connectors are available on new installations

• Upgrade Cognitive Search Template to support additional typeahead strategies

• Brush up connector page

• Adapt the SQ Dataloader behavior to not kill jobs automatically every 30 mins.

• Newsletters: Item links should redirect the user to a squirro dashboard instead of original source

• Type-ahead refresh for cognitive search

• [Sprint goal] Better understand performance limitations of current query processing

• Load activity_tracking.json logs from frontend service into elastic

• Provide image uploading capability for GrapeJS email editing

• Investigate performance of initial fuzzy auto-completer implementation

• Triggering a sample email for testing purpose does not get populated with data.

• Add support for PDF documents

• Tests for newsletter sending

• Community item count not displayed inside the card

• Configure upgrades with Ansible Tower

• Community list: sorted list should not be case sensitive

• Have a dedicated endpoint to display a community

• On getting the link to the detail page the overlay with the link is not shown

• funny overlay in cognitive search

• Search option on communities list page to find a community

• Change default sort order for cognitive search to: “By Relevance”

• Cognitive Search: Communities - Horizontal Scrolling and Search Bar overflowing community tabs

• Recent Query Suggestor Service: Add Log extractor

• Add infinity scroll loading for communities

• Cognitive Search: Highlights from global feed -> only first item of the cards has community display -> for starred items

• Write smoke-tests to preserve functionality after upgrades

• Cognitive Search: Global Search Bar not taking right colours

• [Sprint goal] Finish newsletter implementation

• Migration scripts fail in cluster context

• Item detail view: pixel perfection

• "If you enable compact mode for the navbar and search using global search, the dashboard tabs move down"

• Tab widget pixel perfection

• Community query displayed inside the breadcrumb instead of community name

• Cognitive search: Global search not showing typeahead suggestions

• Favorite filters search is broken.

• Studio plugins in AI studio space fail

• Pipeline Editor: Cannot delete workflow which was recently created and saved without reentering the editor

• Feedback AI Studio: Rejected feedback doesn't get marked as processed

• Communities headline widget: Arrow shape broken

• build Model: There is a miss configured link

• Incorrect textcolor

• Feedback AI Studio: Recommend bubble is not completely displayed and thumb-down button isn't alligned

• NLPTagger: Spacy NER cannot handle large text inputs

• Scheduled newsletters not sent on testing server

• Update scheduling frequency on project settings does not reflect in already scheduled newsletter tasks.

• [Projects Sidebar] Projects names are not visible

• "Pipeline editor: On using scroll arrows in the top left of the editor, the pipelines should scroll instead of jump"

• Refactor squirro-ansible upgrades into AWX

• "Feedback widget: incorrect classification, there is no option to select the correct class"

• Pipeline Editor: Displaying issue with the config options of the Near-Duplicate Detection

• Add an option to disable number of items for communities

• Pipelet implementing getArguments() is not rendered as 1st class widget -- throws error instead

• Some steps which are already part of a workflow are missing values for certain properties

• Provide a config key for every step returned by the GET /v0/projects/{project_id}/pipeline_sections endpoint

• Add support for multiple language files to the translator tool

• IB App UI / Dashboard development: The UI of the IB App requires a refinement to reflect business and usage requirements.

• Error when configuring source with pipeline steps that come from a DL plugin

• [Pipeline] Pipeline edit left sidebar frontend implementation

• Add mechanism that automatically removes the processed raw data after a certain time period or disk space threshold

• Enable triggering the rerunning of a pipeline workflow on the already processed raw data

• Modify ingester/lib.stream to move the raw data to a processed directory after executing the pipeline when on dev mode

• Add configuration option that denotes whether we are running in development or production mode

• Add new built-in pipeline step Transform Input that does the work of mapping

• Remove validation logic regarding the mappings from the api.provider

• Remove the transformation logic of mapping source fields to item fields and facets from the Dataloader

• Add config_options to built-in steps

• Newsletter sending fails if there is no full name

• Newsletter test sending fails when no communities subscribed

• Newsletter test sending silently fails with additional_data error

• Cognitive Search: Communities - Add Tooltip to see full name

Fresh Installation Instructions

Please follow the regular installation steps

Upgrade Instructions

yum update java-1.8.0-openjdk
yum update squirro-storage-node-users

# update storage node
yum update squirro-storage-node

yum update squirro-cluster-node-users
yum update squirro-*

# Resolve any `rpmnew` files. We anticipate `/etc/squirro/storage.ini`, `/etc/squirro/machinelearning.ini` and `/etc/nginx/conf.d/frontend.conf` to at least be resolved.
systemctl reload nginx
squirro_restart

# Remove all orphaned python 2 packages
yum erase squirro-python27*

yum update java-1.8.0-openjdk
yum update squirro-storage-node-users

# update storage node
yum update squirro-storage-node

yum update java-1.8.0-openjdk

yum update squirro-cluster-node-users
yum update squirro-*

# Resolve any `rpmnew` files. We anticipate `/etc/squirro/storage.ini`, `/etc/squirro/machinelearning.ini` and `/etc/nginx/conf.d/frontend.conf` to at least be resolved.
systemctl reload nginx
squirro_restart

# Remove all orphaned python 2 packages
yum erase squirro-python27*

yum update java-1.8.0-openjdk
yum update squirro-storage-node-users

# update storage node
yum update squirro-storage-node

yum update java-1.8.0-openjdk

yum update squirro-cluster-node-users
yum update squirro-*

# Resolve any `rpmnew` files. We anticipate `/etc/squirro/storage.ini`, `/etc/squirro/machinelearning.ini` and `/etc/nginx/conf.d/frontend.conf` to at least be resolved.
systemctl reload nginx
squirro_restart

# Remove all orphaned python 2 packages
yum erase squirro-python27*