This repository aims to contain the input/output data comprising the HTTP compliance test suite. It also contains a simplistic script to extract the rules from the XML.
This project is based on the two conversations, one took place on the HTTP Working Group mailing list and the other on the HTTP Workshop. The blog post Towards Validated HTTP Implementation summarizes both discussions and gives a high-level overview of the project.
Test suite execution engines (TBD: To Be Developed) should map the requirements violations to log levels and results as follows:
| Requirement | Log Level |
|---|---|
| MUST | ERROR |
| MUST NOT | ERROR |
| REQUIRED | ERROR |
| SHALL | ERROR |
| SHALL NOT | ERROR |
| SHOULD | WARNING |
| SHOULD NOT | WARNING |
| RECOMMENDED | WARNING |
| NOT RECOMMENDED | WARNING |
| MAY | INFO |
| OPTIONAL | INFO |
The mapping table is built according to the RFC 2119.
The proecss is yet to be refined. Currenty the XML files are placed in the rfcs directory. For each RFC, a subdirectory is created under rfcs. The subtree has immediate descendants text and data. The text content is generated by the program in rfc-requirements-extractor. The data sub-directory is meant to contain the codified data corresponding to the requirements as defined in the RFC.
- Refine the process
- Define the data format & structure as suitable for the respective RFC section
- Define the test cases
To find rules in the XML, search using the regular expression:
<bcp14>\w+(\s\w+)?</bcp14>