new .buffer_position()

Author: zhengxwen

DESCRIPTION

@@ -1,8 +1,8 @@
 Package: SeqArray
 Type: Package
 Title: Data management of large-scale whole-genome sequence variant calls using GDS files
-Version: 1.47.6
-Date: 2025-03-24
+Version: 1.47.7
+Date: 2025-03-31
 Depends: R (>= 3.5.0), gdsfmt (>= 1.31.1)
 Imports: methods, parallel, digest, IRanges, GenomicRanges, GenomeInfoDb, Biostrings,
         S4Vectors

NAMESPACE

@@ -16,7 +16,7 @@ useDynLib(SeqArray,
     SEQ_Apply_Variant, SEQ_Apply_Sample, SEQ_BApply_Variant,
     SEQ_ConvBED2GDS,
     SEQ_SelectFlag, SEQ_ResetChrom,
-    SEQ_IntAssign, SEQ_AppendFill, SEQ_ClearVarMap,
+    SEQ_IntAssign, SEQ_AppendFill, SEQ_ClearVarMap, SEQ_BufferPosition,
     SEQ_Pkg_Init,
     SEQ_bgzip_create, SEQ_Progress, SEQ_ProgressAdd,
     SEQ_Unit_SlidingWindows,

R/Internal.R

@@ -245,6 +245,16 @@ process_count <- 1L
 
 
 
+#######################################################################
+# Get or clear the memory buffer storing variant positions (return pointer or NULL)
+#
+.buffer_position <- function(gdsfile, clear=FALSE)
+{
+    .Call(SEQ_BufferPosition, gdsfile, clear)
+}
+
+
+
 #######################################################################
 # Open and close a connection,
 # Please always call '.close_conn' after '.open_bin' and '.open_text'

src/Index.cpp

@@ -2,7 +2,7 @@
 //
 // Index.cpp: Indexing Objects
 //
-// Copyright (C) 2016-2024    Xiuwen Zheng
+// Copyright (C) 2016-2025    Xiuwen Zheng
 //
 // This file is part of SeqArray.
 //
@@ -933,6 +933,14 @@ vector<C_Int32> &CFileInfo::Position()
 	return _Position;
 }
 
+void CFileInfo::ClearPosition()
+{
+	if (!_Root)
+		throw ErrSeqArray(ERR_FILE_ROOT);
+	_Position.clear();
+	std::vector<C_Int32>().swap(_Position);
+}
+
 CGenoIndex &CFileInfo::GenoIndex()
 {
 	if (_GenoIndex.Empty())

src/Index.h

@@ -2,7 +2,7 @@
 //
 // Index.h: Indexing Objects
 //
-// Copyright (C) 2016-2022    Xiuwen Zheng
+// Copyright (C) 2016-2025    Xiuwen Zheng
 //
 // This file is part of SeqArray.
 //
@@ -414,6 +414,8 @@ class COREARRAY_DLL_LOCAL CFileInfo
 	void ResetChromosome();
 	/// return _Position which has been initialized
 	vector<C_Int32> &Position();
+	/// clear the buffer for variant positions
+	void ClearPosition();
 
 	/// return _GenoIndex which has been initialized
 	CGenoIndex &GenoIndex();

src/SeqArray.cpp

@@ -2,7 +2,7 @@
 //
 // SeqArray.cpp: the C++ codes for the SeqArray package
 //
-// Copyright (C) 2013-2024    Xiuwen Zheng
+// Copyright (C) 2013-2025    Xiuwen Zheng
 //
 // This file is part of SeqArray.
 //
@@ -1362,6 +1362,29 @@ COREARRAY_DLL_EXPORT SEXP SEQ_ClearVarMap(SEXP gdsfile)
 
 
 
+// ===========================================================
+// Clear VarMap in a GDS file
+// ===========================================================
+
+COREARRAY_DLL_EXPORT SEXP SEQ_BufferPosition(SEXP gdsfile, SEXP clear)
+{
+	int clear_flag = Rf_asLogical(clear);
+	COREARRAY_TRY
+		CFileInfo &File = GetFileInfo(gdsfile);
+		if (clear_flag == 1)  // TRUE
+		{
+			File.ClearPosition();
+			rv_ans = R_NilValue;
+		} else {
+			vector<C_Int32> &pos = File.Position();
+			SEXP n = Rf_ScalarInteger(pos.size());  // # of positions
+			rv_ans = R_MakeExternalPtr(&pos[0], R_NilValue, n);
+		}
+	COREARRAY_CATCH
+}
+
+
+
 // ===========================================================
 // Get system configuration
 // ===========================================================
@@ -1653,7 +1676,7 @@ COREARRAY_DLL_EXPORT void R_init_SeqArray(DllInfo *info)
 		CALL(SEQ_SelectFlag, 2),            CALL(SEQ_ResetChrom, 1),
 
 		CALL(SEQ_IntAssign, 2),             CALL(SEQ_AppendFill, 3),
-		CALL(SEQ_ClearVarMap, 1),
+		CALL(SEQ_ClearVarMap, 1),           CALL(SEQ_BufferPosition, 2),
 
 		CALL(SEQ_bgzip_create, 1),
 		CALL(SEQ_ToVCF_Init, 6),            CALL(SEQ_ToVCF_Done, 0),