Darcs.Repository.State

Documentation

restrictSubpaths :: (RepoPatch p, ApplyState p ~ Tree) => Repository rt p wR wU wT -> [SubPath] -> IO (TreeFilter m)

From a repository and a list of SubPath's, construct a filter that can be used on a Tree (recorded or unrecorded state) of this repository. This constructed filter will take pending into account, so the subpaths will be translated correctly relative to pending move patches.

restrictBoring :: Tree m -> IO (TreeFilter m)

Construct a TreeFilter that removes any boring files that are not also contained in the argument Tree.

The standard use case is for the argument to be the recorded state, possibly with further patches applied, so as not to discard any files already known to darcs. The result is usually applied to the full working state.

newtype TreeFilter m

Constructors

TreeFilter

Fields

• applyTreeFilter :: forall tr. FilterTree tr m => tr m -> tr m

restrictDarcsdir :: TreeFilter m

Construct a Tree filter that removes any darcs metadata files the Tree might have contained.

maybeRestrictSubpaths :: (RepoPatch p, ApplyState p ~ Tree) => FL (PrimOf p) wT wP -> Repository rt p wR wU wT -> Maybe [SubPath] -> IO (TreeFilter m)

Diffs

unrecordedChanges :: (RepoPatch p, ApplyState p ~ Tree) => (UseIndex, ScanKnown, DiffAlgorithm) -> LookForMoves -> LookForReplaces -> Repository rt p wR wU wT -> Maybe [SubPath] -> IO (FL (PrimOf p) wT wU)

For a repository and an optional list of paths (when Nothing, take everything) compute a (forward) list of prims (i.e. a patch) going from the recorded state of the repository (pristine) to the unrecorded state of the repository (the working copy + pending). When a list of paths is given, at least the files that live under any of these paths in either recorded or unrecorded will be included in the resulting patch. NB. More patches may be included in this list, eg. the full contents of the pending patch. This is usually not a problem, since selectChanges will properly filter the results anyway.

This also depends on the options given:

• -look-for-moves: Detect pending file moves using the index. The resulting patches are added to pending and taken into consideration, when filtering the tree according to the given path list.
• -look-for-adds: Include files in the working state that do not exist in the recorded + pending state.
• -include-boring: Include even boring files.
• -look-for-replaces: Detect pending replace patches. Like detected moves, these are added to the pending patch. Note that, like detected moves, these are mere proposals for the user to consider or reject.
• -ignore-times: Disables index usage completely -- for each file, we read both the unrecorded and the recorded copy and run a diff on them. This is very inefficient, although in extremely rare cases, the index could go out of sync (file is modified, index is updated and file is modified again within a single second).

Note that use of the index is also disabled when we detect moves or replaces, since this implies that the index is out of date.

readPending :: (RepoPatch p, ApplyState p ~ Tree) => Repository rt p wR wU wT -> IO (Tree IO, Sealed (FL (PrimOf p) wT))

Obtains the recorded Tree with the pending patch applied, plus the pending patch itself. The pending patch should start at the recorded state (we even verify that it applies, and degrade to renaming pending and starting afresh if it doesn't), but we've set to say it starts at the tentative state.

Question (Eric Kow) Is this a bug? Darcs.Repository.Pending.readPending says it is

Trees

readRecorded :: Repository rt p wR wU wT -> IO (Tree IO)

Obtains a Tree corresponding to the "recorded" state of the repository: this is the same as the pristine cache, which is the same as the result of applying all the repository's patches to an empty directory.

readUnrecorded :: (RepoPatch p, ApplyState p ~ Tree) => Repository rt p wR wU wT -> Maybe [SubPath] -> IO (Tree IO)

Obtains a Tree corresponding to the "unrecorded" state of the repository: the modified files of the working tree plus the "pending" patch. The optional list of paths allows to restrict the query to a subtree.

Limiting the query may be more efficient, since hashes on the uninteresting parts of the index do not need to go through an up-to-date check (which involves a relatively expensive lstat(2) per file.

readRecordedAndPending :: (RepoPatch p, ApplyState p ~ Tree) => Repository rt p wR wU wT -> IO (Tree IO)

Obtains the recorded Tree with the pending patch applied.

readWorking :: IO (Tree IO)

Obtains a Tree corresponding to the complete working copy of the repository (modified and non-modified files).

readPendingAndWorking :: forall rt p wR wU wT. (RepoPatch p, ApplyState p ~ Tree) => (UseIndex, ScanKnown, DiffAlgorithm) -> LookForMoves -> LookForReplaces -> Repository rt p wR wU wT -> Maybe [SubPath] -> IO ((FL (PrimOf p) :> FL (PrimOf p)) wT wU)

readUnrecordedFiltered :: (RepoPatch p, ApplyState p ~ Tree) => Repository rt p wR wU wT -> UseIndex -> ScanKnown -> LookForMoves -> Maybe [SubPath] -> IO (Tree IO)

A variant of readUnrecorded that takes the UseIndex and ScanKnown options into account, similar to readPendingAndWorking. We are only interested in the resulting tree, not the patch, so the DiffAlgorithm option is irrelevant.

Index

readIndex :: (RepoPatch p, ApplyState p ~ Tree) => Repository rt p wR wU wT -> IO Index

updateIndex :: (RepoPatch p, ApplyState p ~ Tree) => Repository rt p wR wU wT -> IO ()

invalidateIndex :: t -> IO ()

Mark the existing index as invalid. This has to be called whenever the listing of pristine changes and will cause darcs to update the index next time it tries to read it. (NB. This is about files added and removed from pristine: changes to file content in either pristine or working are handled transparently by the index reading code.)

data UseIndex

Constructors

UseIndex

IgnoreIndex

Instances

Eq UseIndex

Methods

(==) :: UseIndex -> UseIndex -> Bool

(/=) :: UseIndex -> UseIndex -> Bool

Show UseIndex

Methods

showsPrec :: Int -> UseIndex -> ShowS

show :: UseIndex -> String

showList :: [UseIndex] -> ShowS

YesNo UseIndex

Methods

yes :: UseIndex -> Bool

no :: UseIndex -> Bool

data ScanKnown

Constructors

ScanKnown

Just files already known to darcs

ScanAll

All files, i.e. look for new ones

ScanBoring

All files, even boring ones

Instances

Eq ScanKnown

Methods

(==) :: ScanKnown -> ScanKnown -> Bool

(/=) :: ScanKnown -> ScanKnown -> Bool

Show ScanKnown

Methods

showsPrec :: Int -> ScanKnown -> ShowS

show :: ScanKnown -> String

showList :: [ScanKnown] -> ShowS

Utilities

filterOutConflicts

Arguments

:: (RepoPatch p, ApplyState p ~ Tree)

=> RL (PatchInfoAnd rt p) wX wT

Recorded patches from repository, starting from same context as the patches to filter

-> Repository rt p wR wU wT

Repository itself, used for grabbing unrecorded changes

-> FL (PatchInfoAnd rt p) wX wZ

Patches to filter

-> IO (Bool, Sealed (FL (PatchInfoAnd rt p) wX))

True iff any patches were removed, possibly filtered patches

Remove any patches (+dependencies) from a sequence that conflict with the recorded or unrecorded changes in a repo

Pending-related functions that depend on repo state

addPendingDiffToPending :: (RepoPatch p, ApplyState p ~ Tree) => Repository rt p wR wU wT -> UpdateWorking -> FreeLeft (FL (PrimOf p)) -> IO ()

Add an FL of patches started from the pending state to the pending patch. TODO: add witnesses for pending so we can make the types precise: currently the passed patch can be applied in any context, not just after pending.

addToPending :: (RepoPatch p, ApplyState p ~ Tree) => Repository rt p wR wU wT -> UpdateWorking -> FL (PrimOf p) wU wY -> IO ()

Add an FL of patches starting from the working state to the pending patch, including as much extra context as is necessary (context meaning dependencies), by commuting the patches to be added past as much of the changes between pending and working as is possible, and including anything that doesn't commute, and the patch itself in the new pending patch.