Having cloned the brooklyn repo a number of times over the last days I was curious why it's so big. As expected there are a number of binary files, probably unmodified since their first commit, having been deleted a long time ago polluting the history. Here's a summary of the largest offenders:
All sizes are in kB's. The pack column is the size of the object, compressed, inside the pack file. size pack SHA location 57722 57718 01760133fa9dedc811762019a2aaa691bbe61988 monterey-example/src/main/resources/booking-mvc.war 23917 23905 d0b52371d4ba447ee9b0f882854b4b5d4d39147f monterey-example/src/main/resources/jboss-booking.war 23915 23903 4f6c7f01e0f49c5e3574d648bcab991f7cab5236 monterey-example/src/main/resources/monterey-booking-as7.war 21737 20984 ab9740889e11e06df060595684c2bd803c1baac1 examples/simple-nosql-cluster/src/main/resources/cumulusrdf-0.6.1-pre.jar 20988 20993 04a0a6fd94873543cfec769b5a272bbe36f0a914 examples/simple-nosql-cluster/src/main/resources/cumulusrdf.war 12124 12119 7ccd0ef45879c78941575d47a8521eef49b0b704 sandbox/examples/src/main/resources/swf-booking-mvc.war 7642 7063 0249fd11430b461f918e2b95adb52425caf81230 gemfire/lib/gemfire-6.5.1.4.jar 7217 7219 9d384383e5542b04ca2b6332aedba92bc92bc5bb examples/simple-nosql-cluster/src/main/resources/cumulusrdf.war 2431 2428 5d7d1216a4276d82c3fabdfa25852e527e1bf78e monterey-example/src/main/resources/booking-mvc.war 2430 2427 9fc9f74223ed9704e91810d5dd9fe79d41ff4b84 com.cloudsoftcorp.monterey.brooklyn/src/main/resources/booking-mvc.war The only sane way to delete them without losing the history is to rewrite it with the big-file-commits excluded. This has the downside that any PRs or cloned repositories will need manual intervention after the changes. I think it's worth the effort of removing them, having in mind the benefit of shrinking the repo to a few MBs in long run. There's never a "right" moment to do such a change, but the more it is delayed, the worse it becomes. So what are the thoughts of the community? Svet.
