You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Running syft pointed at a directory that contains the source code of a Ruby gem should find that gem and its dependencies.
Why is this needed:
Syft's current Ruby catalogers, which focus on Gemfile.lock (directory cataloger) and **/specifications/**/*.gemspec in an image context. However, when running Syft on a directory that contains a checkout of the source of a gem, neither of these catalogers finds anything. (For example, running git clone git@github.com:CanCanCommunity/cancancan.git && syft packages --catalogers all dir:cancancan prints No packages discovered; it should instead print cancancan itself and its dependencies.)
The new code added in 0.95.0 will show the gem name as a package but none of the dependencies present in the gemspec file, unless you have a Gemfile.lock
if you don't have a Gemfile.lock you won't see anything added in Gemfile or *.gemspec. That is unfortunate.
What would you like to be added:
Running
syft
pointed at a directory that contains the source code of a Ruby gem should find that gem and its dependencies.Why is this needed:
Syft's current Ruby catalogers, which focus on
Gemfile.lock
(directory cataloger) and**/specifications/**/*.gemspec
in an image context. However, when running Syft on a directory that contains a checkout of the source of a gem, neither of these catalogers finds anything. (For example, runninggit clone git@github.com:CanCanCommunity/cancancan.git && syft packages --catalogers all dir:cancancan
printsNo packages discovered
; it should instead printcancancan
itself and its dependencies.)Additional context:
https://yehudakatz.com/2010/12/16/clarifying-the-roles-of-the-gemspec-and-gemfile/ is a helpful post on understanding how working on a gem is different from having an installed gem and different from working on an application.
Thanks @mscottford for pointing out this issue!
The text was updated successfully, but these errors were encountered: