Releases: meedan/check
Releases · meedan/check
v0.140.0
GraphQL API changes:
- ➕
Project
type: added new boolean fieldis_default
- now every workspace requires a default folder (a.k.a.,Project
) where items go when a destination folder is not defined... this also means that now every item belongs to a folder.
Database changes:
- ➕
projects
table: added new boolean fieldis_default
, which is also indexed - now every workspace requires a default folder (a.k.a.,Project
) where items go when a destination folder is not defined... this also means that now every item belongs to a folder.
v0.139.0
GraphQL API changes:
- ➕
ProjectMedia
type: added fieldcreator_name
, type string, with the name of the user who created the item - ➕
ProjectMedia
type: added fieldchannel
, with the integer value that represents the immutable channel from which this item came from (list of possible channel values available at https://github.com/meedan/check-api/blob/develop/lib/check_channels.rb) - ➖
Project
type: deleted fieldauto_tasks
- ➕
Team
type: added fieldsources_count
, of type integer, which returns the number of sources under a team, optionally matching akeyword
v0.138.0
- Sensitive content warning feature: the item will display a warning cover when it's detected as sensitive content
- ➕
ProjectMedia
type: add fieldshow_warning_cover
of type Boolean to return if the item is flagged to show a warning cover - ➕ Added new migration to update the Flag Dynamic Annotation schema. The migration should run very fast during
docker-compose up
- ➕
- ➕
ProjectMedia
type: add connectionsimilar_items
of type ProjectMedia to return the items related as similar to the item - ➕
TeamTask
type: add fieldstasks_count
andtasks_with_answers_count
of type Integer to return the number of tasks and number of answered tasks related to the team task respectively - ➕ New package (
ffmpeg
) and gem on CheckAPI (streamio-ffmpeg
) to fix auto-transcription and similarity matching: they should be installed duringdocker-compose up build
- ➕ New task to cache item picture and improve performance. To run the task:
bundle exec rake check:migrate:cache_item_picture
- ♻️ Update task to generate statistics about workspaces. The task returns the following information: 'Number of fact-checks published', 'Number of fact-checks sent to users', 'Number of requests' and 'Number of similar media'. To run the task:
bundle exec rake check:data:statistics[year,start_month,end_month,group_by_month (0 or 1),workspace_slugs_as_a_dot_separated_values_string]
- ➕ New audio similarity algorithm on Alegre
v0.137.0
- ➕
About
type: added new fieldcountries
of typeJsonStringType
to return the list of workspace countries - ➕
Team
type: added new fieldscountry
of typeString
andget_trends_enabled
of typeBoolean
to get the country defined for the workspace and to identify if the trends are enabled, respectively - ➕ Added the mutation
searchUpload
to upload a media to be searched. The file should be sent as form data and it returns a file handle - ➕ New task to generate some statistics about some workspaces. To run the task:
bundle exec rake check:data:statistics[year,start_month,end_month,group_by_month (0 or 1),workspace_slugs_as_a_dot_separated_values_string]
- Added new migrations. The migrations should run very fast during
docker-compose up
- ➕ Migration to add column and index
country
toteams
table - ➕ Migration to add column
platform
totipline_subscriptions
table - ➕ Migration to add timestamps (
created_at
andupdated_at
) totipline_subscriptions
table
- ➕ Migration to add column and index
v0.136.0
- Audio transcription feature: speech in audio and video files are converted into written text when added to a workspace
- ➕ Added the mutation
transcribeAudio
. The input field isid
which is the item's GraphQL ID and the mutation returns the project_media (item) and the annotation created with the transcription data returned by Alegre Bot - ➕ New migration to create the annotation type
Transcription
and the related annotation fields. The migration should run very fast duringdocker-compose up
- ➕ Added the mutation
- Tipline inbox filters feature: let users define what is visible on Tipline inbox so they can keep track of their work
- ➕
Team
type: added the fieldget_tipline_inbox_filters
of typeJsonStringType
to get the filters enabled on Tipline inbox - ➕
updateTeam
mutation: added the input fieldtipline_inbox_filters
of type String to set the filters enabled on Tipline inbox
- ➕
- Tipline newsletter feature: collect subscriptions and send weekly newsletters direct from the bot
- ➕
Team
type: added the fieldget_tipline_inbox_filters
of typeJsonStringType
andslack_notifications_json_schema
of type String to get the configuration related to Slack notifications - ➕
TeamBotInstallation
type: added the fieldsmooch_newsletter_information
of typeJsonStringType
to get the information about the newsletter. Currently, the information includes day, time, timezone, body, feed_url and number of articles - Added new migrations to support the feature. The migrations should run very fast during
docker-compose up
- ➕ Migration to add newsletter settings to Smooch Bot
- ➕ Migration to create the table
tipline_subscriptions
- ➕
- Slack integration improvements: simplify slack notifications and allow teams to configure which notifications will be sent to Slack. Currently the event types can be
any_activity
,status_changed
,item_added
- ➕
Team
type: added the fieldsget_slack_notifications
of typeJsonStringType
andslack_notifications_json_schema
of type String to get the configuration related to Slack notifications - ➕
updateTeam
mutation: added the input fieldslack_notifications
of type String to set the configuration related to Slack notifications. This field should be filled in by the JSON version of an array of hashes representing the types of notifications sent to to Check. Each notification will have a hash structure containing the keyslabel
,event_type
,values
andslack_channel
- ➖
Team
type: removed fieldget_slack_channel
. Please get it fromslack_notifications
fromTeam
type - ➖
Project
type: removed fieldsget_slack_channel
andget_slack_events
. Please get it fromslack_notifications
fromTeam
type - ➖
updateTeam
mutation: removed input fieldslack_channel
- ➖
updateProject
mutation: removed input fieldsset_slack_channel
andslack_events
. Please set it onslack_notifications
fromTeam
type - ➕ New task to migrate the deprecated
slack_channel
andslack_events
settings from teams and projects to the newslack_notifications
team settings. To run the task:bundle exec rake check:migrate:migrate_slack_notifications
- ➕
TeamTask
type updates:- ♻️ replace field
is_child
of type Boolean to fieldconditional_info
of type String to get the information about when a team task is displayed to users based on other metadata's response - ➕ add field
required
of type Boolean to return if the team task is required
- ♻️ replace field
- ➕
ProjectMedia
type: add fieldfull_url
to return the item URL - ➕
Task
type: add fieldteam_task
of typeTeamTask
to return the related team task - ➕
Team
type: add fieldurl
of type String to return the url of the team - ➕
Annotation
type: add fieldteam
of typeTeam
to return the related team - Added new migrations. The migrations should run very fast during
docker-compose up
- ➕ Migration to add index
privacy
toprojects
table - ➕ Migration to index the extracted text annotation content in a new field in ElasticSearch for each item
- ➕ Migration to remove column
is_child
and createconditional_info on
team_tasks table - ➕ Migration to remove
smooch_project_id
from Smooch Bot settings. The tipline now uses the concept ofchannels
and don't belong to projects anymore
- ➕ Migration to add index
- Added new task to change existent data
- ➕ Task to update smooch_user annotations and relate the annotation to team instead of projects. To run the task:
bundle exec rake check:migrate:relate_smooch_user_to_team
- ➕ Task to remove duplicated relationships on database. To run the task:
bundle exec rake check:migrate:remove_duplicate_relationships
- ➕ Task for indexing the field
extracted_text
on Elasticsearch. To run the task:bundle exec rake check:migrate:index_extracted_text
- ➕ Task to update smooch_user annotations and relate the annotation to team instead of projects. To run the task:
- ⬆️ Upgraded Ruby (
2.4
=>2.6
), Rails (4.2.11.1
=>5.2.6
) and the gems. Some of the changes related to the upgrade are posted on CheckAPI wiki
v0.135.0
- Channels feature: the items created store the information about how the item was submitted to Check. Currently the channels can be
MANUAL
(via app),FETCH
(via Fetch bot),BROWSER_EXTENSION
(via Check Mark extension),API
(via CheckAPI),ZAPIER
(via Zapier) orTIPLINE
(via WhatsApp, Messenger, Twitter, Telegram, Viber or Line)- ➕
About
type: added new fieldchannels
of typeJsonStringType
to return the list of channels available - ➕
createProjectMedia
mutation: added the input fieldchannel
of type Integer to set the channel the item was submitted by. The mapping between the channel and its code is defined onChannelCodes
- ➕ New migration to add
channel
toproject_medias
table and add elasticsearch mapping. The migration should run very fast duringdocker-compose up
- ➕ New task to set the channel of existing items and index the field
channel
on ElasticSearch. To run the task:bundle exec rake check:migrate:add_channel_to_project_medias
- ➕
- Privacy settings feature: folders and content inside them can be configured to be visible by (I) only admins (II) admins and editors (III) all users with access to the team
- ➕
updateProject
mutation: added the input fieldprivacy
of type Integer to set the code related to the permitted users. The mapping between the group with access and its code is defined onPrivacySettings
- ➕ New migration to add
privacy
toprojects
table. The migration should run very fast duringdocker-compose up
- ➕
- Conditional annotation feature: display a metadata based on other metadata answers
- ➕
TeamTask
type: added new fieldis_child
of type Boolean to identify when a team task is displayed to users based on other metadata's response - ➕
createTeamTask
andupdateTeamTask
mutations: added the input fieldis_child
of type Boolean to identify if the metadata e dependent of other metadata - ➕ New migration to add
is_child
toteam_tasks
table. The migration should run very fast duringdocker-compose up
- ➕
- Identical media items identification feature: improve filter for identical media items (image, video and audio) submitted to Check
- ➕ New task to identify and merge identical uploaded files (image, video and audio). To run the task:
bundle exec rake check:migrate:fix_identical_uploaded_file
- ➕ New task to delete medias not associated with any project media. To run the task:
bundle exec rake check:migrate:delete_orphaned_medias
- ➕ New task to identify and merge identical uploaded files (image, video and audio). To run the task:
v0.134.0
- ➕
Team
type: add fieldget_tasks_enabled
of typeBoolean
. When it'strue
the tasks will be displayed on item pages - ➕ New migration to configure the existent teams to display the tasks on item pages. The migrations should run very fast during
docker-compose up
. - ➕ New task added to fix folder name cache. To run the task:
rake check:migrate:fix_folder_cached_value
v0.133.0
- ➕ Added new mutations:
ExtractText
: get text from images using Alegre Bot. The input field isid
which is the item's GraphQL ID and the mutation returns the project_media (item)
- ➕ New mutations to add a single file to a metadata field and to delete a single file from it:
AddFilesToTask
: add one or more files as response of a task or metadata. The input field isid
which is the task or metadata GraphQL ID and the files should be sent as form data. The return field is the taskRemoveFilesFromTask
: remove files from task or metadata's response. The input_field isid
, the task of metadata GrapqhQL ID andfilenames
the name of the files that should be removed. The return field is the task
- ➕
TeamBotInstallation
type: add fieldalegre_settings
to set/get settings from Alegre Bot - Smooch Bot integration with Turn.io
v0.132.0
- ➕ Support extraction of texts from uploded images (OCR): New migration added to create annotation type "Extracted text". The migration should run very fast during
docker-compose up
- ➕
Search
type improvements: addedoperator
as option to send onSearch
query. Tha value sent can beAND
orOR
. By default it's value isAND
- ➕
About
type: added new fieldsupload_max_size_in_bytes
of typeInteger
. Returns the maximum upload size, in bytesfile_max_size_in_bytes
of typeInteger
. Returns the maximum file upload size, in bytesvideo_max_size_in_bytes
of typeInteger
. Returns the maximum video upload size, in bytesaudio_max_size_in_bytes
of typeInteger
. Returns the maximum audio upload size, in bytes
- ➕
TeamBotIntallation
type: added optional argumentforce
of typeBoolean
on fieldsmooch_enabled_integrations
. When it'strue
, refreshed the cache and get the most updated enabled integrations - ➕
Team
type: added fieldteam_bot_installation
of typeTeamBotInstallation
with required argumentbot_identifier
. Return the team bot installation related to the team with the identifier sent on argument - ➕ New migrations added. The migrations should run very fast during
docker-compose up
:- migration to drop table
project_medias_projects
because it's not used anymore. The migration should run very fast duringdocker-compose up
: - migration to adjust indexes for dynamic annotation fields. The migration should run very fast during
docker-compose up
: - migration to hide deprecated tipline template
- migration to drop table
- ♻️ Updated task to create ElasticSearch data from database to accept a team slug as argument and create the data for the specified team only. To run the task:
rake check:create_es_data_from_pg[team_slug1, team_slug2, ...]
- ➕ New task added to update report status cached value for existing secondary items. To run the task:
rake check:migrate:update_secondary_report_status
v0.131.0
- ➕ Upgrade of Gems: the gems should be upgraded during
docker-compose build
- ➕ New event
publish_report
: when a report is published the external bots subscribing the event will be notified. - Sort items by item title:
- ➕ New migration to add mapping for sorting by title. The migration should run very fast during
docker-compose up
: - ➕ New task for indexing the field
sort_title
with the itemanalysis_title
if present ortitle
. To run the task:check:migrate:fill_es_sort_title
- ➕ New migration to add mapping for sorting by title. The migration should run very fast during