A Multi-Platform Specification Language and Dataset for the Analysis of DevOps Pipelines
Bibliographic record
Abstract
To meet market demand for products that are delivered faster, while also delivering high-quality products, businesses are seeking to streamline and accelerate the design, development, and delivery process. The DevOps methodology addresses automation and faster delivery processes. Platforms such as GitHub, GitLab, Bitbucket, Azure DevOps, and Jenkins are commonly used to specify automation pipelines. With the proliferation of these platforms, it has become more difficult to analyze pipelines across individual platforms. An analysis environment that abstracts from individual platforms and can understand several pipeline dialects could address these issues. In this paper, we present a language and an Xtext-based editor for the analysis of multi-platform pipeline specifications that covers dialects from the GitHub Actions, GitLab, BitBucket, Bamboo, Circle CI, and Azure DevOps platforms. Furthermore, we present a heterogeneous dataset of automation pipelines from different platforms. We conducted a systematic analysis of existing pipeline specifications before defining the multi-platform language, and we mined and preprocessed 42,106 pipelines from open-source projects such as GitHub and Software Heritage for validation. According to our results, the proposed editor successfully parsed 40,160 pipelines after applying minor pre-processing. Based on a random sample of the remaining 1,946 pipelines, these pipelines were not parsed successfully due a malformed pipeline specification, or the files being intended for other purposes. The proposed analysis environment including language, editor, and dataset paves the way for further cross-platform analysis of automation pipelines. To demonstrate a use case for the analysis environment, we identify five distinct pipeline specification patterns from the successfully parsed pipelines to better understand common pipeline usage.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame machine prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.
Distilled classifier scores by category (both heads)
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.003 | 0.013 |
| Meta-epidemiology (narrow) | 0.002 | 0.001 |
| Meta-epidemiology (broad) | 0.001 | 0.002 |
| Bibliometrics | 0.006 | 0.006 |
| Science and technology studies | 0.001 | 0.001 |
| Scholarly communication | 0.001 | 0.002 |
| Open science | 0.002 | 0.003 |
| Research integrity | 0.001 | 0.002 |
| Insufficient payload (model declined to judge) | 0.004 | 0.006 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".