Using/PatchIndex

Patch index is an internal optimization in darcs. darcs uses patch index to quickly identify the patches that modified a given file. This is especially useful for the commands log and annotate.

Patch index is created when a new repository is created (either by init or clone). To disable patch index at creation, pass flag –disable-patch-index. At clone the patch index will be disabled if the user:

  • passes –disable-patch-index or –lazy
  • interrupts using Ctrl-C when getting patches

When darcs is first run on an existing repository, and repository state is going to be modified, patch index will be created. The user will receive a dialog saying so, and they can stop the creation and disable patch index by using Ctrl-C. The command will continue to execute thereafter.

Patch index can be enabled or disabled on a repository by using optimize enable-patch-index or optimize disable-patch-index.

If patch index is enabled, it is automatically updated and used.

Patch index tries to be as invisible as possible to the user. To this end, creating/using/modifying patch index is almost invisible to the end user.

There are however, some commands which lets the user glean the state of patch index. They are primarily meant for debugging.

  • show patch-index-status returns information on whether patch index is not created/out of date/in sync.
  • show patch-index-test tests the patch index in the repository with some logical consistency properties.
  • show patch-index-all dumps the patch index
  • show patch-index-files displays the files tracked by patch index (This has to be same as the files tracked by darcs)