If you are using some other scripting or programming environment, you can directly write out the DAX format using the provided schema using any language. For instance, LIGO, the Laser Interferometer Gravitational Wave Observatory, generate their DAX files as XML using their own Python code, not using our provided API.
If you write your own XML, you must ensure that the generated XML is well formed and valid with respect to the DAX schema. You can use the pegasus-dax-validator to verify the validity of your generated file. Typically, you generate a smallish test file to, validate that your generator creates valid XML using the validator, and then ramp it up to produce the full workflow(s) you want to run. At this point the pegasus-dax-validator is a very simple program that will only take exactly one argument, the name of the file to check.The following snippet checks a black-diamond file that uses an improper osversion attribute in its executable element:
blackdiamond.daxERROR: cvc-pattern-valid: Value '2.6.18-194.26.1.el5' is not facet-valid with respect to pattern '[0-9]+(\.[0-9]+(\.[0-9]+)?)?' for type 'VersionPattern'. ERROR: cvc-attribute.3: The value '2.6.18-194.26.1.el5' of attribute 'osversion' on element 'executable' is not valid with respect to its type, 'VersionPattern'. 0 warnings, 2 errors, and 0 fatal errors detected.
We are working on improving this program, e.g. provide output with regards to the line number where the issue occurred. However, it will return with a non-zero exit code whenever errors were detected.