pamutil

type FileInfo struct {
	// Path is the value passed to ParsePath.
	Path string

	// FileType is the type of the file. For "dir/0:0,46:1653469.mapq", the type
	// is FileTypeFieldData. For "dir/0:0,46:1653469.mapq", the type is
	// FileTypeFieldIndex.
	Type FileType

	// Field stores the field part of the filename. Field=="mapq" if the pathname
	// is "dir/0:0,46:1653469.mapq". It is meaningful iff Type ==
	// FileTypeFieldData.
	Field string

	// Dir is the directory under which the file is stored. Dir="dir" if the
	// pathname is "dir/0:0,46:1653469.mapq".
	Dir string
	// Range is the record range that the file stores. Range={Start:{0,0},
	// Limit:{46,1653469}} if the pathname is "dir/0:0,46:1653469.mapq".
	Range biopb.CoordRange
}

type FileType int

type GenerateReadShardsOpts struct {
	// Range defines an optional row shard range. Only records in this range will
	// be returned by Scan() and Read(). If Range is unset, the universal range is
	// assumed. See also ReadOpts.Range.
	Range biopb.CoordRange

	// SplitMappedCoords allows GenerateReadShards to split mapped reads of
	// the same <refid, alignment position> into multiple shards. Setting
	// this flag true will cause shard size to be more even, but the caller
	// must be able to handle split reads.
	SplitMappedCoords bool
	// SplitUnmappedCoords allows GenerateReadShards to split unmapped
	// reads into multiple shards. Setting this flag true will cause shard
	// size to be more even, but the caller must be able to handle split
	// unmapped reads.
	SplitUnmappedCoords bool
	// CombineMappedAndUnmappedCoords allows creating a shard that contains both
	// mapped and unmapped reads. If this flag is false, shards are always split
	// at the start of unmapped reads.
	AlwaysSplitMappedAndUnmappedCoords bool

	// BytesPerShard is the target shard size, in bytes across all fields.  If
	// this field is set, NumShards is ignored.
	BytesPerShard int64
	// NumShards specifies the number of shards to create. This field is ignored
	// if BytePerShard>0. If neither BytesPerShard nor NumShards is set,
	// runtime.NumCPU()*4 shards will be created.
	NumShards int
}

type ShardIndex struct {
	// Range is the coordinate range that this object represents. Records and indexes from the
	// source PAM that don't intersect this range were ignored.
	Range biopb.CoordRange
	// ApproxFileBytes is an estimate of the total file size of records in Range (in the
	// underlying PAM)
	ApproxFileBytes int64
	// Blocks is a sequence of index entries from one PAM field that span Range.
	Blocks []biopb.PAMBlockIndexEntry
}