Right now, the tool is fairly straightforward to extend if one goes through the source code, but there's currently no documentation on how to do it. Developers looking to leverage scraper for their source control platforms that are not yet supported (e.g., Subversion) would benefit from some kind of reference in order to figure out the integration points.
I'm curious what approach might work best:
1) Add a lightweight plugin system, where contributors can write their own Python packages, install them via pip, and add their config to config.json. Scraper simply invokes each plugin listed in the config and integrates the returned metadata into the final code.json file.
or
2) Don't modify scraper, just add comprehensive documentation on how contributors can add a new module. Contributors can then open a PR to merge the new feature.
Right now, the tool is fairly straightforward to extend if one goes through the source code, but there's currently no documentation on how to do it. Developers looking to leverage scraper for their source control platforms that are not yet supported (e.g., Subversion) would benefit from some kind of reference in order to figure out the integration points.
I'm curious what approach might work best:
1) Add a lightweight plugin system, where contributors can write their own Python packages, install them via pip, and add their config to config.json. Scraper simply invokes each plugin listed in the config and integrates the returned metadata into the final code.json file.
or
2) Don't modify scraper, just add comprehensive documentation on how contributors can add a new module. Contributors can then open a PR to merge the new feature.