Join GitHub today
GitHub is home to over 28 million developers working together to host and review code, manage projects, and build software together.Sign up
run flunc tests (suites of twill scripts with variables) through the web http://www.coactivate.org/projects/flunc
Fetching latest commit…
Cannot retrieve the latest commit at this time.
|Type||Name||Latest commit message||Commit time|
|Failed to load latest commit information.|
fluncrunner runs flunc tests (suites of twill scripts with variables) through the web. The API ======= There is a web API for running a flunc test. Executing a test ---------------- To execute a test, POST to the test's execution URI. To execute a test with global variable overrides, provide a www-form-urlencoded set of key=value pairs in the request body. For example:: POST /basic_test/ HTTP/1.1 Content-Type: application/x-www-form-urlencoded username=lammy&password=testy&name_of_friend=bob will execute a command like:: flunc -D username=lammy,password=testy,name_of_friend=bob basic_test Inspecting a test ----------------- To discover the available variables in a test, GET the test's inspection URI. An HTML form will be returned, suitable for POSTing back to the server. If available, default values (from the test's corresponding .conf) will be detailed in the form fields' pre-filled values. Configuring the server ====================== The fluncrunner server uses Paste Deploy for configuration. A sample configuration, suitable for forking, can be found in /docs/devel_config.ini An app_factory is provided. It implements the test execution API and requires one piece of configuration: a filesystem path that points to a directory of flunc tests that may be run. This is equivalent to flunc's -p option on the command line:: [app:flunc] use = egg:fluncrunner search_path = %(here)s/ftests/ You will want to configure this app to respond to POST requests. A composite_factory `request_method` is provided which will dispatch requests to applications based on request method:: [composite:main] use = egg:fluncrunner#request_method post = flunc No automated implementation of the test inspection API is provided. If you have a directory of static HTML forms that implement the API for your flunc tests, you may want to configure a static file server to respond to GET requests:: [composite:main] use = egg:fluncrunner#request_method post = flunc get = static [app:static] use = egg:Paste#static document_root = %(here)s/ftest_forms/ The Paste static file server infers Content-type from file extensions, so if you want to serve flunc test forms that don't have *.html names you may want to use the provided `force_html` filter_factory which sets Content-type: text/html on all responses:: [app:static] use = egg:Paste#static document_root = %(here)s/ftest_forms/ filter-with = force_html [filter:force_html] use = egg:fluncrunner#force_html If you are not serving fluncrunner from Paste, you can configure this WSGI pipeline in code. You can find the needed WSGI factories in `fluncrunner`'s `main.py`, `force_html.py` and `request_method.py`. Generating flunc test forms =========================== This section wants writing. twill.js ======== There's http://bitbucket.org/ianb/twilljs/ which executes simple twill scripts (no variables, no suites) A similar API could transform flunc tests into twill scripts. Somehow that feels more friendly to me than variables in twill.js. A server would be responsible for interpolating variables provided by the requesting client and serving a "standard" twill script. Flunc suites could also be compiled down to twill scripts this way.