All notable changes to this project will be documented here.
- Haskell Tests - allow displaying of compilation errors (#554)
- Add status api for monitoring if Gunicorn is down (#555)
- Ensure all Haskell test cases still run within same file when there are failed test cases (#543)
- Ensure R packages are correctly installed (#535)
- Make PyTA version a setting (#536)
- Add
libxml2-dev
to serverDockerfile
, required by Rtidyverse
library (#539) - Display stderr contents if R packages fail to install (#539)
- Do not display
testthat
failure messages when test case passes (#539)
- Add tidyverse as a default R tester package (#512)
- For the Haskell tester, make stack resolver a test setting (#526)
- Clean up tmp directory after test runs (#528)
- Omit skipped test cases in Python tester (#522)
- Ensure _env_status is updated to "setup" earlier when a request to update test settings is made (#499)
- Fix bug that prevented copies of instructor directories from being deleted (#483)
- Add STACK_ROOT to containers as well as notes in readme (#484)
- Fix bug that prevented test results from being returned when a feedback file could not be found (#458)
- Add support for Python 3.11 and 3.12 (#467)
- Track test environment setup status and report errors when running tests if environment setup is in progress or raised an error (#468)
- Update Haskell tester to use Stack to install dependencies (#469)
- Improve default error message when a test group times out (#470)
- Updated python-ta to 2.6.2 (#454)
- Fix a bug in the Java tester, where failed/error tests were being detected as passing. (#451)
- updated python-ta to 2.6.1 (#452)
- Fix a bug that prevented test file from being copied from a zip file to another location on disk (#426)
- Remove support for python3.6 and add support for python3.10 (#399)
- Remove requirement to skip top level directory in zip archive when downloading test files (#412)
- Fix a bug in the java tester where errors were not reported (#401)
- Add explicit namespaces to R test runner script so that test code will not interfere with result reporting (#407)
- Raise error if rq or supervisord executables can't be found when running start_stop.py (#390)
- Bump python-ta version to 2.3.2 (#391)
- Support dependencies on specific package versions and non-CRAN sources for R tester (#323)
- Testers no longer generate automated content for feedback files (#375)
- Multiple feedback files can now be created per test run (#375)
- PyTA plaintext reports are sent as part of the test result, not as a feedback file (#375)
- Allow client to define environment variables to pass to individual test runs (#370)
- Add ability to clean up test scripts that haven't been used for X days (#379)
- Support dependencies on specific package versions and non-CRAN sources for R tester (#323)
- Remove the requirement for clients to send unique user name (#318)
- Add R tester (#310)
- Keep result object alive for longer than the default 500 seconds (#302)
- Update python-ta tester to be compatible with python-ta version 2 (#296)
- Improve error message when tester virtual environment fails (#297)
- Fix bug when reporting a schema that doesn't exist (#298)
- Full rewrite of autotester with a server/client architecture (#283)
- Add Jupyter tester (#284)
- Fix bug where zip archive was unpacked two levels deep instead of just one (#271)
- Pass PGHOST and PGINFO environment variables to tests (#272)
- Update to new version of markus-api that supports uploading binary files (#273)
- Fix bug where environment variables were not string types (#274)
- Add python3.9 as a tester option for the py and pyta testers, as well as the installer (#275)
- Updated java tester to support configurable classpaths and source files (#268)
- Fixed bug where relevant test data was not included in the client while enqueuing tests (#265)
- Updated development docker image to connect to the development MarkUs docker image (#238)
- Removed Tasty-Stats as a dependency for the haskell tester and added our own ingredient instead to collect stats (#259)
- Updated/improved the interface between the autotester and clients that use it (#257)
- allow tests to write to existing subdirectories but not overwrite existing test script files (#237).
- add ability to create a docker container for the autotester in development mode (#236).
- major reorganization of the structure of this package (#236).
- additional usage options for the server installation script (bin/install.sh).
- testers can/should now be installed using the server installation script instead of individually.
- configuration files now use yaml format.
- configuration file defaults are now included in the source code so the autotester can be run with or without a user specific configuration file.
- changed the default location for the workspace directory.
NOTE: This changelog starts from version 1.8.1 (changes prior to this version are not documented)
- changelog
- for all changes prior to this version see https://github.com/MarkUsProject/markus-autotesting/pulls?utf8=%E2%9C%93&q=is%3Apr+created%3A%3C2019-12-19+