Closes #1574 - Rename SegmentedArray to SegmentedString

arkouda/strings.py

@@ -1421,7 +1421,7 @@ def group(self) -> pdarray:
 
         Notes
         -----
-        If the arkouda server is compiled with "-sSegmentedArray.useHash=true",
+        If the arkouda server is compiled with "-sSegmentedString.useHash=true",
         then arkouda uses 128-bit hash values to group strings, rather than sorting
         the strings directly. This method is fast, but the resulting permutation
         merely groups equivalent strings and does not sort them. If the "useHash"

src/ArgSortMsg.chpl

@@ -24,7 +24,7 @@ module ArgSortMsg
     use ServerErrorStrings;
 
     use RadixSortLSD;
-    use SegmentedArray;
+    use SegmentedString;
     use Reflection;
     use ServerErrors;
     use Logging;

src/ArraySetopsMsg.chpl

@@ -14,7 +14,7 @@ module ArraySetopsMsg
 
     use MultiTypeSymbolTable;
     use MultiTypeSymEntry;
-    use SegmentedArray;
+    use SegmentedString;
     use ServerErrorStrings;
 
     use ArraySetops;

src/Cast.chpl

@@ -2,7 +2,7 @@ module Cast {
   use MultiTypeSymbolTable;
   use MultiTypeSymEntry;
   use Reflection;
-  use SegmentedArray;
+  use SegmentedString;
   use ServerErrors;
   use Logging;
   use CommAggregation;

src/CastMsg.chpl

@@ -2,7 +2,7 @@ module CastMsg {
   use MultiTypeSymbolTable;
   use MultiTypeSymEntry;
   use Reflection;
-  use SegmentedArray;
+  use SegmentedString;
   use ServerErrors;
   use Logging;
   use Message;

src/ConcatenateMsg.chpl

@@ -11,7 +11,7 @@ module ConcatenateMsg
 
     use MultiTypeSymbolTable;
     use MultiTypeSymEntry;
-    use SegmentedArray;
+    use SegmentedString;
     use ServerErrorStrings;
     use CommAggregation;
     use PrivateDist;

src/FindSegmentsMsg.chpl

@@ -12,7 +12,7 @@ module FindSegmentsMsg
     use MultiTypeSymbolTable;
     use MultiTypeSymEntry;
     use ServerErrorStrings;
-    use SegmentedArray;
+    use SegmentedString;
 
     use PrivateDist;
     use CommAggregation;

src/Flatten.chpl

@@ -2,14 +2,13 @@ module Flatten {
   use ServerConfig;
 
   use AryUtil;
-  use SegmentedArray;
+  use SegmentedString;
   use ServerErrors;
   use SymArrayDmap;
   use CommAggregation;
   use Reflection;
   use ArkoudaRegexCompat;
   use CTypes;
-  use SegmentedArray only checkCompile, _unsafeCompileRegex;
 
   config const NULL_STRINGS_VALUE = 0:uint(8);
 

src/FlattenMsg.chpl

@@ -5,7 +5,7 @@ module FlattenMsg {
   use Reflection;
   use Flatten;
   use ServerConfig;
-  use SegmentedArray;
+  use SegmentedString;
   use Logging;
   use Message;
 

src/GenSymIO.chpl

@@ -16,7 +16,7 @@ module GenSymIO {
     use Logging;
     use Message;
     use ServerConfig;
-    use SegmentedArray;
+    use SegmentedString;
 
     private config const logLevel = ServerConfig.logLevel;
     const gsLogger = new Logger(logLevel);

src/HDF5Msg.chpl

@@ -21,7 +21,7 @@ module HDF5Msg {
     use ServerConfig;
     use ServerErrors;
     use ServerErrorStrings;
-    use SegmentedArray;
+    use SegmentedString;
     use Sort;
 
     require "c_helpers/help_h5ls.h", "c_helpers/help_h5ls.c";
@@ -36,8 +36,8 @@ module HDF5Msg {
     const ARKOUDA_HDF5_FILE_VERSION_KEY = "file_version";
     const ARKOUDA_HDF5_FILE_VERSION_VAL = 1.0:real(32);
     type ARKOUDA_HDF5_FILE_VERSION_TYPE = real(32);
-    config const SEGARRAY_OFFSET_NAME = "segments";
-    config const SEGARRAY_VALUE_NAME = "values";
+    config const SEGSTRING_OFFSET_NAME = "segments";
+    config const SEGSTRING_VALUE_NAME = "values";
     config const NULL_STRINGS_VALUE = 0:uint(8);
     config const TRUNCATE: int = 0;
     config const APPEND: int = 1;
@@ -279,38 +279,38 @@ module HDF5Msg {
         var dataclass: C_HDF5.H5T_class_t;
         var bytesize: int;
         var isSigned: bool;
-        var isSegArray: bool;
+        var isSegString: bool;
 
         try {
             defer { // Close the file on exit
                 C_HDF5.H5Fclose(file_id);
             }
             if isStringsDataset(file_id, dsetName) {
                 if ( !skipSegStringOffsets ) {
-                    var offsetDset = dsetName + "/" + SEGARRAY_OFFSET_NAME;
+                    var offsetDset = dsetName + "/" + SEGSTRING_OFFSET_NAME;
                     var (offsetClass, offsetByteSize, offsetSign) = 
                                             try get_dataset_info(file_id, offsetDset);
                     if (offsetClass != C_HDF5.H5T_INTEGER) {
                         throw getErrorWithContext(
                         msg="dataset %s has incorrect one or more sub-datasets" +
-                        " %s %s".format(dsetName,SEGARRAY_OFFSET_NAME,SEGARRAY_VALUE_NAME), 
+                        " %s %s".format(dsetName,SEGSTRING_OFFSET_NAME,SEGSTRING_VALUE_NAME), 
                         lineNumber=getLineNumber(),
                         routineName=getRoutineName(),
                         moduleName=getModuleName(),
-                        errorClass='SegArrayError');                    
+                        errorClass='SegStringError');                    
                     }
                 }
-                var valueDset = dsetName + "/" + SEGARRAY_VALUE_NAME;
+                var valueDset = dsetName + "/" + SEGSTRING_VALUE_NAME;
                 try (dataclass, bytesize, isSigned) = 
                                            try get_dataset_info(file_id, valueDset);
-                isSegArray = true;
+                isSegString = true;
             } else if isBooleanDataset(file_id, dsetName) {
                 var booleanDset = dsetName + "/" + "booleans";
                 (dataclass, bytesize, isSigned) = get_dataset_info(file_id, booleanDset);
-                isSegArray = false;            
+                isSegString = false;            
             } else {
                 (dataclass, bytesize, isSigned) = get_dataset_info(file_id, dsetName);
-                isSegArray = false;
+                isSegString = false;
             }
         } catch e : Error {
             //:TODO: recommend revisiting this catch block 
@@ -321,7 +321,7 @@ module HDF5Msg {
                 moduleName=getModuleName(),
                 errorClass='Error');
         }
-        return (isSegArray, dataclass, bytesize, isSigned);
+        return (isSegString, dataclass, bytesize, isSigned);
     }
 
 
@@ -1865,7 +1865,7 @@ module HDF5Msg {
         } else {
             filenames = filelist;
         }
-        var segArrayFlags: [filedom] bool;
+        var segStringFlags: [filedom] bool;
         var dclasses: [filedom] C_HDF5.hid_t;
         var bytesizes: [filedom] int;
         var signFlags: [filedom] bool;
@@ -1881,7 +1881,7 @@ module HDF5Msg {
             for (i, fname) in zip(filedom, filenames) {
                 var hadError = false;
                 try {
-                    (segArrayFlags[i], dclasses[i], bytesizes[i], signFlags[i]) = get_dtype(fname, dsetName, calcStringOffsets);
+                    (segStringFlags[i], dclasses[i], bytesizes[i], signFlags[i]) = get_dtype(fname, dsetName, calcStringOffsets);
                 } catch e: FileNotFoundError {
                     fileErrorMsg = "File %s not found".format(fname);
                     h5Logger.error(getModuleName(),getRoutineName(),getLineNumber(),fileErrorMsg);
@@ -1902,8 +1902,8 @@ module HDF5Msg {
                     h5Logger.error(getModuleName(),getRoutineName(),getLineNumber(),fileErrorMsg);
                     hadError = true;
                     if !allowErrors { return new MsgTuple(fileErrorMsg, MsgType.ERROR); }
-                } catch e: SegArrayError {
-                    fileErrorMsg = "SegmentedArray error: %s".format(e.message());
+                } catch e: SegStringError {
+                    fileErrorMsg = "SegmentedString error: %s".format(e.message());
                     h5Logger.error(getModuleName(),getRoutineName(),getLineNumber(),fileErrorMsg);
                     hadError = true;
                     if !allowErrors { return new MsgTuple(fileErrorMsg, MsgType.ERROR); }
@@ -1922,12 +1922,12 @@ module HDF5Msg {
                     fileErrorCount += 1;
                 }
             }
-            const isSegArray = segArrayFlags[filedom.first];
+            const isSegString = segStringFlags[filedom.first];
             const dataclass = dclasses[filedom.first];
             const bytesize = bytesizes[filedom.first];
             const isSigned = signFlags[filedom.first];
-            for (name, sa, dc, bs, sf) in zip(filenames, segArrayFlags, dclasses, bytesizes, signFlags) {
-              if ((sa != isSegArray) || (dc != dataclass)) {
+            for (name, sa, dc, bs, sf) in zip(filenames, segStringFlags, dclasses, bytesizes, signFlags) {
+              if ((sa != isSegString) || (dc != dataclass)) {
                   var errorMsg = "Inconsistent dtype in dataset %s of file %s".format(dsetName, name);
                   h5Logger.error(getModuleName(),getRoutineName(),getLineNumber(),errorMsg);
                   return new MsgTuple(errorMsg, MsgType.ERROR);
@@ -1946,11 +1946,11 @@ module HDF5Msg {
             var len: int;
             var nSeg: int;
             try {
-                if isSegArray {
+                if isSegString {
                     if (!calcStringOffsets) {
-                        (segSubdoms, nSeg, skips) = get_subdoms(filenames, dsetName + "/" + SEGARRAY_OFFSET_NAME);
+                        (segSubdoms, nSeg, skips) = get_subdoms(filenames, dsetName + "/" + SEGSTRING_OFFSET_NAME);
                     }
-                    (subdoms, len, skips) = get_subdoms(filenames, dsetName + "/" + SEGARRAY_VALUE_NAME);
+                    (subdoms, len, skips) = get_subdoms(filenames, dsetName + "/" + SEGSTRING_VALUE_NAME);
                 } else {
                     (subdoms, len, skips) = get_subdoms(filenames, dsetName);
                 }
@@ -1967,18 +1967,18 @@ module HDF5Msg {
             h5Logger.debug(getModuleName(),getRoutineName(),getLineNumber(),
                                            "Got subdomains and total length for dataset %s".format(dsetName));
 
-            select (isSegArray, dataclass) {
+            select (isSegString, dataclass) {
                 when (true, C_HDF5.H5T_INTEGER) {
                     if (bytesize != 1) || isSigned {
                         var errorMsg = "Error: detected unhandled datatype: segmented? %t, class %i, size %i, signed? %t".format(
-                                                isSegArray, dataclass, bytesize, isSigned);
+                                                isSegString, dataclass, bytesize, isSigned);
                         h5Logger.error(getModuleName(),getRoutineName(),getLineNumber(),errorMsg);
                         return new MsgTuple(errorMsg, MsgType.ERROR);
                     }
 
                     // Load the strings bytes/values first
                     var entryVal = new shared SymEntry(len, uint(8));
-                    read_files_into_distributed_array(entryVal.a, subdoms, filenames, dsetName + "/" + SEGARRAY_VALUE_NAME, skips);
+                    read_files_into_distributed_array(entryVal.a, subdoms, filenames, dsetName + "/" + SEGSTRING_VALUE_NAME, skips);
 
                     proc _buildEntryCalcOffsets(): shared SymEntry throws {
                         var offsetsArray = segmentedCalcOffsets(entryVal.a, entryVal.aD);
@@ -1987,7 +1987,7 @@ module HDF5Msg {
 
                     proc _buildEntryLoadOffsets() throws {
                         var offsetsEntry = new shared SymEntry(nSeg, int);
-                        read_files_into_distributed_array(offsetsEntry.a, segSubdoms, filenames, dsetName + "/" + SEGARRAY_OFFSET_NAME, skips);
+                        read_files_into_distributed_array(offsetsEntry.a, segSubdoms, filenames, dsetName + "/" + SEGSTRING_OFFSET_NAME, skips);
                         fixupSegBoundaries(offsetsEntry.a, segSubdoms, subdoms);
                         return offsetsEntry;
                     }
@@ -2059,7 +2059,7 @@ module HDF5Msg {
                 }
                 otherwise {
                     var errorMsg = "detected unhandled datatype: segmented? %t, class %i, size %i, " +
-                                   "signed? %t".format(isSegArray, dataclass, bytesize, isSigned);
+                                   "signed? %t".format(isSegString, dataclass, bytesize, isSigned);
                     h5Logger.error(getModuleName(),getRoutineName(),getLineNumber(),errorMsg);
                     return new MsgTuple(errorMsg, MsgType.ERROR);
                 }

src/KExtremeMsg.chpl

@@ -15,7 +15,7 @@ module KExtremeMsg
 
     use MultiTypeSymbolTable;
     use MultiTypeSymEntry;
-    use SegmentedArray;
+    use SegmentedString;
     use ServerErrorStrings;
 
     use KReduce;

src/Merge.chpl

@@ -1,6 +1,6 @@
 module Merge {
   use IO;
-  use SegmentedArray;
+  use SegmentedString;
   use RadixSortLSD only numTasks, calcBlock;
   use Reflection;
   use ServerConfig;

src/ParquetMsg.chpl

@@ -13,7 +13,7 @@ module ParquetMsg {
   use Sort;
   use CommAggregation;
 
-  use SegmentedArray;
+  use SegmentedString;
 
 
   // Use reflection for error information

src/RandArray.chpl

@@ -3,7 +3,7 @@ module RandArray {
   use ServerErrors;
   use Logging;
   use Random;
-  use SegmentedArray;
+  use SegmentedString;
   use ServerErrorStrings;
   use MultiTypeSymEntry;
   use Map;

src/RegistrationMsg.chpl

@@ -13,7 +13,7 @@ module RegistrationMsg
     use MultiTypeSymbolTable;
     use MultiTypeSymEntry;
     use ServerErrorStrings;
-    use SegmentedArray;
+    use SegmentedString;
 
     private config const logLevel = ServerConfig.logLevel;
     const regLogger = new Logger(logLevel);
@@ -174,7 +174,7 @@ module RegistrationMsg
     }
 
     /* 
-    Compile the component parts of a SegArray attach message 
+    Compile the component parts of a SegString attach message 
 
     :arg cmd: calling command 
     :type cmd: string 
@@ -189,7 +189,7 @@ module RegistrationMsg
     */
     proc attachSegArrayMsg(cmd: string, name: string, st: borrowed SymTab): MsgTuple throws {
         regLogger.debug(getModuleName(),getRoutineName(),getLineNumber(), 
-                            "%s: Collecting SegArray components for '%s'".format(cmd, name));
+                            "%s: Collecting SegString components for '%s'".format(cmd, name));
 
         var repMsg: string;
 

src/SegStringSort.chpl

@@ -1,5 +1,5 @@
 module SegStringSort {
-  use SegmentedArray;
+  use SegmentedString;
   use Sort;
   use Time;
   use IO;

src/SegmentedMsg.chpl

@@ -4,7 +4,7 @@ module SegmentedMsg {
   use ServerErrors;
   use Logging;
   use Message;
-  use SegmentedArray;
+  use SegmentedString;
   use ServerErrorStrings;
   use ServerConfig;
   use MultiTypeSymbolTable;

src/SegmentedArray.chpl → src/SegmentedString.chpl

@@ -1,4 +1,4 @@
-module SegmentedArray {
+module SegmentedString {
   use AryUtil;
   use CTypes;
   use MultiTypeSymbolTable;
@@ -22,7 +22,7 @@ module SegmentedArray {
   const saLogger = new Logger(logLevel);
 
   private config param useHash = true;
-  param SegmentedArrayUseHash = useHash;
+  param SegmentedStringUseHash = useHash;
 
   private config param regexMaxCaptures = ServerConfig.regexMaxCaptures;
 

src/ServerErrors.chpl

@@ -117,14 +117,14 @@ module ServerErrors {
     }
 
     /*
-     * The SegArrayError is thrown if the file corresponding to the SegArray lacks either the
-     * SEGARRAY_OFFSET_NAME or SEGARRAY_VALUE_NAME dataset.
+     * The SegStringError is thrown if the file corresponding to the SegString lacks either the
+     * SEGSTRING_OFFSET_NAME or SEGSTRING_VALUE_NAME dataset.
      */
-    class SegArrayError: ErrorWithContext { 
+    class SegStringError: ErrorWithContext { 
 
         proc init(msg : string, lineNumber: int, routineName: string, 
                                                            moduleName: string) { 
-           super.init(msg,lineNumber,routineName,moduleName,errorClass='SegArrayError'); 
+           super.init(msg,lineNumber,routineName,moduleName,errorClass='SegStringError'); 
         } 
 
         proc init(){ super.init(); }