All repositories in this organization, where our UCCA data is kept, currently use Travis CI for continuous integration: on every push, validations are run on the xml files, and then they are split to sentences, visualized and converted to various formats. The results are pushed automatically to GitHub branches in each repository (see e.g. the visualizations for EWT). This is all defined in .travis.yml for each repository (e.g. for EWT).
All repositories in this organization, where our UCCA data is kept, currently use Travis CI for continuous integration: on every push, validations are run on the xml files, and then they are split to sentences, visualized and converted to various formats. The results are pushed automatically to GitHub branches in each repository (see e.g. the visualizations for EWT). This is all defined in
.travis.ymlfor each repository (e.g. for EWT).As recently posted, at the end of 2020, Travis CI will stop providing free jobs by default. Open source projects can apply to keep using it for free, but it is unclear how and to what extend this is possible. Therefore, the best course of actions seems to be switching to another service. GitHub Actions seems to be the most straightforward choice, and they have a guide for migrating from Travis CI. I have already done this for https://github.com/huji-nlp/ucca, https://github.com/huji-nlp/semstr and https://github.com/huji-nlp/tupa (see
.githubdirectory in each one), but not yet for the repositories under https://github.com/UniversalConceptualCognitiveAnnotation.