Meta Issue: File classification and categorization

[pombredanne]

We have several issues related to categorization and classification. This is a meta issue to track the latest and refactor these in roadmap entries. In general, the actual classification and categorization should be best handled in ScanCode Toolkit. Applying these to automate scan processing would be best in ScanCode.io.

• Proposal: Scan deduction and summarization #377
• Proposal: high level file classification #426
• Proposal: Introduce configurable Rules for scan classification, summarization, refinements and inference  #1012
• Define patterns of file names and types for classification #1036
• Improve Programming language detection and classification #1445
• Create plugin to determine file categories #1745
• New summary/primary Content Type prototype #1754
• RFC: categorize codebase files to prioritize license-related risk analysis #2945
• Excluding unnecessary files/docs #3282

And one PR:

• [WIP] Add new file categorization plugin #2945 #2958

[armijnhemel]

What for me would be very useful is something that could be run after scancode has finished and where I could, based on configurable settings, choose to ignore the files from reporting, or include them in reporting but tag them. Example: build related files (Makefile, GNU autoconf m4 macros, etc.).

I would still want them to be scanned so the results are available if needed, but filter and ignore if possible.