Do not report copyright that are part of an official license text

[pombredanne]

For instance the L/GPL licenses always contain an FSF copyright. Reporting these as copyright is just noise as this is the copyright for the license text and not for the code itself.

[agneet42]

@pombredanne sir could you tell me what all files and the changes to be made so that I could do the same ?

[pombredanne]

@agneet42 Thanks for looking into this: for a start you could run the copyright detection of scancode proper to locate and collect the list of all scancode licenses and rules that do contain a copyright statement.
something such as ./scancode --copyright --format json-pp src/licensed/data <someoutput.json> .... and or do a grep on src/licensed/data ffor -ie "copyri" then these text with a copyright need to be reviewed one by one to decide for each if this is OK and should not be changed (e.g. when the copyright is REALLY part of the license and typically applies to the license text itself) vs the cases where the copyright is NOT part of the license and should be considered for removal.

[pombredanne]

@DennisClark commented in #819

Copyright detection reports the copyrights on license texts from codebases. These are a distraction, and not helpful. Typical examples are the FSF and the Apache Software Foundation.

[adityaviki]

@pombredanne
How should we go about this? From what I can think of, we can't simple maintain a list of organizations to ignore, since it will ignore their code related copyrights. We probably have to check for the name of the file ( i.e the file name should contain the word "license" and it should have no extension or ".txt" extension)