Skip to content

Latest commit

 

History

History
550 lines (444 loc) · 24.6 KB

CHANGELOG.md

File metadata and controls

550 lines (444 loc) · 24.6 KB
Raw

Change Log

All notable changes to this project will be documented in this file.

The format is based on Keep a Changelog and this project adheres to Semantic Versioning.

Please view this file on the master branch, on stable branches it's out of date.

Unreleased

Added

Changed

Deprecated

Removed

Fixed

Security

0.10.0

Added

  • User can select segmentation run mode before start (@mkasztelnik)
  • Possibility to configure custom Ansys licenses for pipeline computation (@mkasztelnik)
  • pipeline_identifier case_number and token computation script helpers (@mkasztelnik)
  • Ability to parameterize UI elements according to platform type (EurValve or Process) (@patrykwojtowicz, @JanKapala)

Changed

  • Use Gitlab review procedure instead of labels (@mkasztelnik)
  • stage_in returns error code when unable to download FileStore file (@mkasztelnik)
  • Set JWT token expiration time to 24h (@mkasztelnik)

Deprecated

Removed

  • Computation script repositories removed from eurvalve.yml, step repository configuration used instead (@mkasztelnik)

Fixed

  • Fix unused pipeline title missing (@mkasztelnik)
  • Fixed GitLab integration spec (@Nuanda)

Security

0.9.1

Security

  • Upgrade Sprockets gem to avoid CVE-2018-3760 vulnerability (@mkasztelnik)

0.9.0

Added

  • Added AVD/MVD ratio to patients' statistics (@Nuanda)
  • Profile link for left avatar picture and user name (@mkasztelnik)
  • Colors for patient tile connected with last pipeline status (@mkasztelnik)
  • Instruction how to remove concrete IP from rack-attack fail2ban list (@mkasztelnik)
  • User cannot be removed or blocked when she/he owns exclusively group (@mkasztelnik)
  • Button to minimize left menu is visible always (@mkasztelnik)
  • Dedicated information about deleted pipeline owner (@mkasztelnik)

Changed

  • Reintegration of segmentation service using File Store in place of OwnCloud (@jmeizner)
  • Remove brakeman from Gemfile and use latest version while executing gitab-ci checks (@mkasztelnik)
  • Trigger segmentation start after whole input file is uploaded (@mkasztelnik)

Deprecated

Removed

  • Remove faker gem and replace it with factory_bot sequences (@mkasztelnik)

Fixed

  • Patients' statistics work correctly when turbolink-loaded (@Nuanda)
  • Fix GITLAB_HOST markup formatting in README.md (@mkasztelnik)

Security

0.8.0

Added

  • Reload pipeline step status on patient view (@mkasztelnik)
  • Reload segmentation status after it is started (@mkasztelnik)
  • Proper handling for rimrock computation start failure (@mkasztelnik)
  • Execution time updated each second for active computation (@mkasztelnik)
  • Pipeline specific input (@mkasztelnik)
  • Patient clinical details now includes patient's state (preop/postop) (@Nuanda)
  • New statistics about the current state of EurValve's prospective cohort (@Nuanda)

Changed

  • Labels for pipelines list view improved (@mkasztelnik)
  • Segmentation temp file is removed from local disc after it is transferred into segmentation Philips service (@mkasztelnik)
  • Use stages instead of types in Gitlab CI yml (@mkasztelnik)
  • Upgrade to rubocop 0.51.0 (@mkasztelnik)
  • Use addressable gem to parse URLs (@mkasztelnik)
  • Upgraded rails into 5.1.4 and other gems into latest supported versions (@mkasztelnik)
  • Change factory_girl into factory_bot (@mkasztelnik)
  • Use preconfigured docker image for builds (@mkasztelnik)
  • Extracted DataSets::Client from Patients::Details for reusability (@Nuanda)
  • Lock redis version into 3.x (@mkasztelnik)
  • New patient case widget to reflect new developments in pipelining (@Nuanda)
  • Flow model class which stores information about pipeline flow steps (@mkasztelnik)
  • Switch from PhantomJS into Chrome headless (@mkasztelnik)
  • Add escaping in the documentation pdp curl example (@mkasztelnik)
  • Unique while fetching resources in pdp (@mkasztelnik)
  • Change comparison show method to index (@mkasztelnik)
  • Pipeline steps definition refactored and generalized (@mkasztelnik)
  • Change defaults for data sets (@mkasztelnik)
  • Change the default root path to patients index (@Nuanda)
  • Renamed not_used_flow into unused_steps (@mkasztelnik)

Deprecated

Removed

Fixed

  • Remove n+1 query when updating computation status (@mkasztelnik)
  • Sidebar's hamburger button is operational properly toggling the left menu (@dharezlak)
  • Fix patient web socket unsubscribe (@mkasztelnik)
  • Patient and pipeline creation silent failures (@jmeizner)

Security

0.7.0

Added

  • Show started rimrock computation source link (@mkasztelnik)
  • Parameter extraction pipeline step (@amber7b)
  • Automatic and manual pipeline execution mode (@mkasztelnik)
  • Pipeline flows (pipeline with different steps) (@jmeizner)
  • Configure redis based cache (@mkasztelnik)
  • Add cache for repositories tags and versions (@mkasztelnik)
  • Manual gitlab-ci push master to production step (@mkasztelnik)
  • Patient details are fetched from the external data set service and shown in the patient page (@dharezlak)
  • Possibility to configure automatic pipeline steps during pipeline creation (@mkasztelnik)
  • Wrench automatic pipeline step icon when configuration is needed (@mkasztelnik)
  • Add extra pipeline steps for CFD, ROM, 0D, PV visualization and uncertainty analysis (@amber7b)
  • Computation update interval can be configured using ENV variable (@mkasztelnik)
  • Add loading indicator when reloading current pipeline step (@mkasztelnik)
  • Show manual pipeline steps during pipeline creation (@mkasztelnik)
  • External data set service is called to fetch patient's inferred details (@dharezlak)
  • Additional mocked pipelines and pipelines steps (@mkasztelnik)
  • Show pipeline details (flow and owner) (@Nuanda)
  • Design new pipeline disabled button (@mkasztelnik)
  • Links to 4 EurValve datasets query interfaces (@mkasztelnik)
  • Add possibility to configure data sets site path though ENV variable (@mkasztelnik)
  • Add link to segmentation status output directory (@mkasztelnik)
  • Show computation error message (@mkasztelnik)
  • Gitlab endpoint can be configured using ENV variable (@mkasztelnik)
  • Gitlab clone url can be configured though ENV variable (@mkasztelnik)
  • Add clone_repo(repo) script generator helper (@mkasztelnik)

Changed

  • Segmentation run mode can be configured using yaml or ENV variable (@mkasztelnik)
  • Default segmentation run mode changed into 3 (@mkasztelnik)
  • Patient.case_number now used as patient ID in HTTP requests (@amber7b)
  • Computation update interval changed into 30 seconds (@mkasztelnik)
  • Load patient details only when needed (@mkasztelnik)
  • Patient's case number is now easily available in the script generator (@dharezlak)
  • Fetch patient details timeout set to 2 seconds (@mkasztelnik)
  • Patient details loaded using ajax (@mkasztelnik)
  • Don't show run segmentation button when segmentation is active (@mkasztelnik)
  • Don't show blank option for pipeline mode (@mkasztelnik)
  • Patient age inferred value expressions changed to conform with values coming from real data sets (@dharezlak)
  • File browsers embedded in the pipeline views sort contents by date descendingly (@dharezlak)

Deprecated

Removed

Fixed

  • Output from one pipeline is not taken as different pipeline input (@mkasztelnik)
  • Change step status after segmentation is submitted (@mkasztelnik)
  • Error when deleting a pipeline which has data files (@jmeizner)
  • Set segmentation status to failure when zip cannot be unzipped (@jmeizner)
  • Don't create webdav structure when pipeline cannot be created (@mkasztelnik)
  • Show error details when pipeline cannot be created (@mkasztelnik)
  • Fix proxy expired email title (@mkasztelnik)
  • JSON query requesting inferred patient details updated to comply with data set service's new API (@dharezlak)
  • Don't show run button when manual pipeline and proxy is not valid (@mkasztelnik)
  • Update computation status after computation is started (@mkasztelnik)
  • Extract computations_helper GitLab host name to an env variable (@Nuanda)

Security

0.6.1

Fixed

  • Fix segmentation status update after WebDav computation finished (@mkasztelnik)

0.6.0

Added

  • rack-attack bans are logged to log/rack-attack.log (@mkasztelnik)
  • Show user email in admin users index view (@mkasztelnik)
  • Pipeline step script generation uses ERB templates (@mkasztelnik)
  • Blood flow computation is self-contained - Ansys files are downloaded from git repository (@mkasztelnik)
  • Gitlab integration services (list branches/tags and download specific file) (@amber7b)
  • Comparison of image files in pipeline comparison view (@amber7b)
  • Fetch computation slurm start script from Gitlab (@mkasztelnik)
  • Pipeline comparison view uses OFF viewers to show 3D mesh differences (@dharezlak)
  • Enabled selection of rimrock computation version (@jmeizner, @mkasztelnik)
  • Brakeman security errors check is executed by Gitlab CI (@mkasztelnik)
  • Data sets API reference added to the Help section as a separate entry (@dharezlak)
  • Use action cable (web sockets) to refresh computation (@mkasztelnik)
  • Computation stores information about selected tag/branch and revision (@mkasztelnik)
  • Pipelines comparison shows sources diff link to GitLab (@Nuanda)

Changed

  • Upgrade to ruby 2.4.1 (@mkasztelnik)
  • Upgrade dependencies (@mkasztelnik)
  • Upgrade to rails 5.1.2 (@mkasztelnik)
  • Update rubocop to 0.49.1 (@mkasztelnik)
  • Move Webdav::Client into app/models to avoid auto loading problems (@mkasztelnik)
  • OFF viewer's height in pipeline's compare mode takes half of the available width (@dharezlak)
  • Patient case_number is checked for unsafe characters (@Nuanda)

Deprecated

Removed

Fixed

  • Patient left menu focus when showing patient pipeline computation (@mkasztelnik)
  • Avoid n+1 queries in patient view (@mkasztelnik)
  • Disable turbolinks in links to cloud view (turbolinks does not work well with GWT) (@mkasztelnik)
  • Fix random failing test connected with html escaping user name in generated emails (@mkasztelnik)
  • Fix content type mismatch for file stage out (@mkasztelnik)
  • Turn on files integration tests (@mkasztelnik)
  • Fix missing tooltip messages for queued computation state (@mkasztelnik)
  • Preventing WebDAV folder creation for patients with incorrect case_number (@Nuanda)

Security

0.5.0

Added

  • Pipelines have own DataFiles which represent produced results (@Nuanda)
  • Pipelines comparison view which present differences between files of two Pipelines (@Nuanda)
  • Segmentation output is unzipped into pipeline directory (@amber7b, @mkasztelnik)
  • Store in FileStore heart model computation PNG results (@mkasztelnik).
  • Integration tests (tag "files") are run by GitlabCI (@mkasztelnik)

Changed

  • File browser template for multiple embed mode created (@dharezlak)
  • WebDAV synchronisation takes care for pipeline-related and patient input files (@Nuanda)
  • Computations are executed in the pipeline scope (@mkasztelnik)
  • Move FileStore configuration into one place (@mkasztelnik)
  • Make patient case_number factory field unique (@mkasztelnik)
  • Redirect user back to previous page after proxy is regenerated successfully (@mkasztelnik)
  • Prefix is removed from segmentation outputs (@mkasztelnik)
  • Show compare two pipelines button only when there is more than on pipeline (@mkasztelnik)

Deprecated

Removed

  • DataFile#handle removed since it is not going to be used anymore (@Nuanda)

Fixed

  • Policy management API handles paths with escaped characters well and limits path scope to a single service (@dharezlak)
  • Move and copy policy API calls handle well resources with escaped characters (@dharezlak)
  • Copy and move destinations for policy management API treated as pretty paths for proper processing (@dharezlak)
  • JS resources required for the OFF viewer are loaded in the file browser direct embed mode (@dharezlak)
  • Resource pretty path decodes and encodes values according to URI restrictions (@dharezlak)
  • Local path exclusion validation query fixed (@ddharezlak)
  • Proxy warning is shown only for active Rimrock computations (@mkasztelnik)
  • Correct computation tooltip text on pipelines list (@mkasztelnik)
  • When removing policies via API access method existence check is scoped to a given service (@dharezlak)
  • Set Rimrock computation job id to nil while restarting computation (@mkasztelnik)
  • Show missing patient outputs on pipeline diff view (@mkasztelnik)
  • Path processing fixed for policy move/copy operations invoked via API (@dharezlak)
  • Set correct order for pipelines computations (@mkasztelnik)

Security

0.4.4

Changed

  • JWT token parameter name changed to access_token for the data sets subpage request URL (@dharezlak)

0.4.3

Fixed

  • JWT token passed as a URL query parameter for the data sets subpage (@dharezlak)

0.4.2

Changed

  • JWT token structure description updated (information about sub record) in documentation (@mkasztelnik)

0.4.1

Changed

  • sub changed from integer into string (which is required by specification) (@mkasztelnik)

0.4.0

Added

  • Basic rack-attack configuration (@mkasztelnik)
  • Patient pipelines (@mkasztelnik)
  • Patient/Pipeline file store structure is created while creating/destroying patient/pipeline (@mkasztelnik)
  • Segmentation patient case pipeline step (@tbartynski)
  • User id was added into JWT token using sub key (@mkasztelnik)

Changed

  • Update rubocop into 0.47.1, fix new discovered offenses (@mkasztelnik)
  • JWT token expiration time extended to 6 hours (@mkasztelnik)
  • Change zeroclipboard into clipboard.js because support for flash is dropped (@mkasztelnik)
  • Factories fields which should be unique use unique Faker methods (@mkasztelnik)
  • Simplify Webdav operations for FileStore and OwnCloud (@mkasztelnik)

Deprecated

Removed

  • PLGrid left menu removed since we are not showing any item there (@mkasztelnik)
  • PLGrid patient data synchronizer removed. FileStore is the only supported backend (@mkasztelnik)

Fixed

  • Left menu can be scrolled when high is small (@mkasztelnik)
  • Policy API filters policies according to the service identified by the passed id (@dharezlak)
  • A 404 error code is returned instead of the 500 code when copying/moving policy for a non-existent source policy (@dharezlak)
  • All user groups assignments are removed while user is deleted from the system (@mkasztelnik)

Security

0.3.4

Changed

  • Default Patient Case File Synchronizer changed into WebDav (@mkasztelnik)

0.3.3

Fixed

  • Data sets page url can be overwritten with an environment property (@dharezlak)

0.3.2

Fixed

  • Fixed pushing tag into production from different branch than master (@mkasztelnik)

0.3.1

Fixed

  • Policy API filters policies according to the service identified by the passed id (@dharezlak)

0.3.0

Added

  • PLGrid proxy details shown in the profile view, warning shown when proxy is outdated while active computations are present (@mkasztelnik)
  • Support for running specs with JS support (@mkasztelnik)
  • When a computation finishes the patient view is automatically reloaded in order to show the user the latest files produced by the computation (@Nuanda)
  • OD Heart model is now incorporated in the patient case pipeline (@Nuanda)
  • Show alert when unable to update patient files or run computations because PLGrid proxy is outdated (@mkasztelnik)
  • User with active computations is notified (via email) while proxy expired (@mkasztelnik)
  • Users management view redesigned, added possibility to remove user by admin (@mkasztelnik)
  • Set plgrid login in OpenId request while regenerating proxy certificate (@mkasztelnik)
  • Long living JWT tokens can be generated using internal API (@mkasztelnik)
  • Resource paths support wildcard characters through UI and API (@dharezlak)
  • User can download patient case computation stdout and stderr (@mkasztelnik)
  • Data sets view from an external server was integrated as an iframe (@dharezlak)

Changed

  • Make jwt pem path configurable though system variable (@mkasztelnik)
  • Externalize JWT token generation from user into separate class (@mkasztelnik)
  • Make PDP URI and access method case insensitive (@mkasztelnik)
  • Improve service URL form placeholder (@mkasztelnik)
  • Externalize PLGrid grant id into configuration file (@mkasztelnik)
  • Rename User.with_active_computations into User.with_submitted_computations (@mkasztelnik)
  • Corrected SMTP config to include Auth support (@jmeizner)
  • Updated PDP and Services documentation (@jmeizner)
  • Update rails into 5.0.2 and other dependencies (@mkasztelnik)
  • Make Computation a base class for more specialized classes (@tomek.bartynski)
  • Introduced WebdavComputation and RimrockComputation that inherit from Computation class (@tomek.bartynski)
  • Patient Case pipeline is now hardcoded in Patient model (@tomek.bartynski)
  • Implemented classes that run Blood Flow Simulation and Heart Model Computation pipeline steps (@tomek.bartynski)

Deprecated

Removed

Fixed

  • Correct default pundit permission denied message is returned when no custom message is defined (@mkasztelnik)
  • Missing toastr Javascript map file added (@mkasztelnik)
  • Show add group member and group management help only to group group owner (@mkasztelnik)
  • Unify cloud resources view with other views (surrounding white box added) (@mkasztelnik)
  • Fix path parsing when alias service name is used in PDP request (@mkasztelnik)
  • PLGrid profile view visible only for connected accounts (@mkasztelnik)
  • PDP - wildcard at the beginning and end of resource path should not be default (@mkasztelnik)

Security

0.2.0

Added

  • Application version and revision in layout footer (@mkasztelnik)
  • Pundit authorized error messages for groups and services (@mkasztelnik)
  • Notifications are using the JS toastr library for fancier popups (@dharezlak)
  • The file store component uses portal's notification system to report errors (@dharezlak)
  • Service owner can manage local policies through UI (@mkasztelnik)
  • Unique user name composed with full name and email (@mkasztelnik)
  • Embed atmosphere UI into cloud resources section (@nowakowski)

Changed

  • Redirect status set to 302 instead of 404 (when record not found), 401 (when user does not have permission to perform action) to avoid ugly "You are being redirected" page (@mkasztelnik)
  • PDP denies everything when user is not approved (@mkasztelnik)
  • Add/remove group members redesigned (@mkasztelnik)
  • Update rubocop and remove new offenses (@mkasztelnik)
  • Update project dependencies (@mkasztelnik)
  • Update rubocop to 0.46.0, remove new discovered offences (@mkasztelnik)
  • Update to rails 5.0.1 (@mkasztelnik)

Deprecated

Removed

Fixed

  • Service factory that used to randomly produce invalid objects (@tomek.bartynski)
  • Edit/destroy group buttons visible only for group owners (@mkasztelnik)
  • Administration side menu item displayed only if it is not empty (@tomek.bartynski)
  • Corresponding resource entities are removed when policy API called with only a path param (@dharezlak)

Security

0.1.0

Added

  • Basic project structure (@mkasztelnik, @dharezlak)
  • Use gentelalla theme (@dharezlak, @mkasztelnik)
  • Integration with PLGrid openId (@mkasztelnik)
  • Accepting new users by supervisor (@dharezlak)
  • Policy Decision Point (PDP) REST API (@mkasztelnik)
  • JWT Devise login strategy (@mkasztelnik)
  • JWT configuration (@tomek.bartynski)
  • Resource and permission management (@dharezlak)
  • Patient case POC (@Nuanda)
  • Sentry integration (@mkasztelnik)
  • Get user gravatar (@mkasztelnik)
  • Store PLGrid proxy in DB while logging in using PLGrid openId (@mkasztelnik)
  • JWT token expiry set to 1 hour (@tomek.bartynski)
  • Help pages for REST API (@mkasztelnik)
  • Custom error pages (@mkasztelnik)
  • Group hierarchies performance tests (@tomek.bartynski)
  • Push patient case files into PLGrid using PLGData (@Nuanda)
  • Use Rimrock for PLGrid job submission (@mkasztelnik)
  • Approve user account after PLGrid login (@mkasztelnik)
  • Asynchronous email sending (@mkasztelnik)
  • Notify supervisors after new user registered (@mkasztelnik)
  • Find resources using regular expressions (@dharezlak)
  • Introduce service (@mkasztelnik)
  • WebDav file browser (@dharezlak)
  • REST interface for local resource management (@dharezlak)
  • User profile page (@mkasztelnik)
  • Project description (@amber7b, @mkasztelnik, @dharezlak)
  • Sent confirmation email after user account is approved (@mkasztelnik)
  • Setup continuous delivery (@tomek.bartynski)
  • Setup rubocop (@tomek.bartynski)
  • Service ownership (@Nuanda)
  • Delegate user credentials in service resource management REST API (@mkasztelnik)
  • Service management UI (@Nuanda, @mkasztelnik)
  • Group management UI (@mkasztelnik)
  • Show service token on show view (@mkasztelnik)
  • Deploy to production when tag is pushed into remote git repository (@tomek.bartynski)
  • Group hierarchies management UI (@mkasztelnik)
  • Add new registered user into default groups (@mkasztelnik)
  • Resource management REST API documentation (@dharezlak)
  • Service aliases (@jmeizner)
  • Issue and merge request templates (@mkasztelnik)
  • Rake task for generating sample data for development purposes (@tomek.bartynski)
  • Users can set Access Methods for new and existing Services (@Nuanda)
  • There are global Access Methods which work for all Services (@Nuanda)
  • Bullet gem warns about database queries performance problems (@tomek.bartynski)
  • Omnipotent admin (@Nuanda)
  • Global resource access policies management UI (@mkasztelnik)
  • Additional attribute with policy proxy URL is passed to the file store browser (@dharezlak)
  • Provide help panels for resource actions (@nowakowski)
  • Hint regarding resource paths in global policies tab (@tomek.bartynski)
  • Support path component in service uri (@tomek.bartynski)

Changed

  • Upgrade to Rails 5 (@mkasztelnik, @Nuanda)
  • Left menu refactoring - not it is server side rendered (@mkasztelnik)
  • Simplify login form pages using simple_form wrapper (@mkasztelnik)
  • Resource separation into global and local (@dharezlak)
  • Unify UIs for all CRUD views (@mkasztelnik)
  • WebDav browser JS files obtained from a dedicated CDN (@dharezlak)
  • Delete service ownership when user or service is destroyed (@mkasztelnik)
  • Service views refactoring - separation for global/local policies management (@mkasztelnik, @Nuanda)
  • Rewrite migrations to avoid using ActiveRecord (@amber7b)
  • Increase fade out time for flash messages into 10 seconds (@mkasztelnik)
  • Go to new user session path after logout (@mkasztelnik)
  • Got to new user session path after logout (@mkasztelnik)
  • Added a spec checking for correct removal of only policies specified by the API request parameter (@dharezlak)
  • File store view uses a common layout (@dharezlak)
  • Service uri must not end with a slash (@tomek.bartynski)
  • Resource path must start with a slash (@tomek.bartynski)
  • Removed path unification in Resource model (@tomek.bartynski)
  • Update to sidekiq 4.2.2 and remove unused anymore sinatra dependency (@mkasztelnik)
  • Administrator is able to see all registered services (@mkasztelnik)
  • Fixed path field in a form for Resource (@tomek.bartynski)

Deprecated

Removed

  • Remove webdav specific access methods from seeds (@mkasztelnik)

Fixed

  • Improve service URI validation (@Nuanda)
  • Fix extra spaces in markdown code blocks (@mkasztelnik)
  • Disable turbolinks on WebDav browser view to make GWT work (@mkasztelnik)
  • Fix group validation - at last one group owner is needed (@mkasztelnik)
  • Policies do not forbid access for access methods which they do not define (@Nuanda)
  • Make icheck work with turbolinks (@mkasztelnik)
  • Make zero clipboard work with turbolinks (@mkasztelnik)
  • Don't allow to create group without group owner (@mkasztelnik)
  • Show notice information to the user after new account created that account needs to be approved by supervisor (@mkasztelnik)
  • Show new view instead of edit while creating new group and validation failed (@mkasztelnik)
  • Remove n+1 query for service ownership in Services#index view (@Nuanda)
  • Made it possible to navigate to Service#show with a click if a service had no name (@Nuanda)
  • Content passed to the file store JS browser sanitized to prevent XSS attacks (@dharezlak)
  • Turbolinks disabled directly on the Files link (@dharezlak)
  • PDP returns 403 when uri or access_method query params are missing (@mkasztelnik)
  • Fix missing translations on service list and service show views (@mkasztelnik, @Nuanda)
  • Fix n+1 queries problems in service and group sections (@mkasztelnik)
  • Check service policy only once on global policies view (@mkasztelnik)
  • More advance validation for service uri overridden (@jmeizner)
  • Policy management API documentation corrected (@dharezlak)
  • Fix duplicates and missing part of Service#show (@jmeizner)
  • Policy API uses a service reference when querying for access methods (@dharezlak)

Security