Merge pull request #30 from RDPerera/main

Upgrade to lang.regex

ballerina/Dependencies.toml

@@ -14,8 +14,8 @@ version = "1.1.1"
 dependencies = [
 	{org = "ballerina", name = "file"},
 	{org = "ballerina", name = "io"},
+	{org = "ballerina", name = "lang.regexp"},
 	{org = "ballerina", name = "log"},
-	{org = "ballerina", name = "regex"},
 	{org = "ballerina", name = "test"}
 ]
 modules = [
@@ -70,14 +70,8 @@ version = "0.0.0"
 dependencies = [
 	{org = "ballerina", name = "jballerina.java"}
 ]
-
-[[package]]
-org = "ballerina"
-name = "lang.string"
-version = "0.0.0"
-dependencies = [
-	{org = "ballerina", name = "jballerina.java"},
-	{org = "ballerina", name = "lang.regexp"}
+modules = [
+	{org = "ballerina", packageName = "lang.regexp", moduleName = "lang.regexp"}
 ]
 
 [[package]]
@@ -120,18 +114,6 @@ dependencies = [
 	{org = "ballerina", name = "jballerina.java"}
 ]
 
-[[package]]
-org = "ballerina"
-name = "regex"
-version = "1.4.3"
-dependencies = [
-	{org = "ballerina", name = "jballerina.java"},
-	{org = "ballerina", name = "lang.string"}
-]
-modules = [
-	{org = "ballerina", packageName = "regex", moduleName = "regex"}
-]
-
 [[package]]
 org = "ballerina"
 name = "test"

ballerina/component_group_reader.bal

@@ -20,7 +20,7 @@ isolated function readComponentGroup(string compositeText, EdiSchema ediSchema,
         return ();
     }
 
-    string[] components = split(compositeText, ediSchema.delimiters.component);
+    string[] components = check split(compositeText, ediSchema.delimiters.component);
     if fieldSchema.truncatable {
         int minFields = getMinimumCompositeFields(fieldSchema);
         if components.length() < minFields {

ballerina/edi_translator.bal

@@ -30,7 +30,7 @@ type EdiContext record {|
 public isolated function fromEdiString(string ediText, EdiSchema schema) returns json|Error {
     EdiContext context = {schema};
     EdiUnitSchema[] currentMapping = context.schema.segments;
-    context.ediText = splitSegments(ediText, context.schema.delimiters.segment);
+    context.ediText = check splitSegments(ediText, context.schema.delimiters.segment);
     EdiSegmentGroup rootGroup = check readSegmentGroup(currentMapping, context, true);
     return rootGroup;
 }

ballerina/repetition_reader.bal

@@ -18,7 +18,7 @@ import ballerina/log;
 
 isolated function readRepetition(string repeatText, string repeatDelimiter, EdiSchema mapping, EdiFieldSchema fieldMapping)
         returns SimpleArray|EdiComponentGroup[]|Error {
-    string[] fields = split(repeatText, repeatDelimiter);
+    string[] fields = check split(repeatText, repeatDelimiter);
     SimpleArray|EdiComponentGroup[] repeatValues = getArray(fieldMapping.dataType);
     if fields.length() == 0 {
         // None of the repeating values are provided. Return an empty array.

ballerina/segment_group_reader.bal

@@ -15,7 +15,6 @@
 // under the License.
 
 import ballerina/log;
-import ballerina/regex;
 
 type SegmentGroupContext record {|
     int schemaIndex = 0;
@@ -31,9 +30,7 @@ isolated function readSegmentGroup(EdiUnitSchema[] currentUnitSchema, EdiContext
         if segSchema is () {
             return error Error("Segment schema cannot be empty.");
         }
-        string sDesc = context.ediText[context.rawIndex];
-        string segmentDesc = regex:replaceAll(sDesc, "\n", "");
-
+        string segmentDesc = removeLineBreaks(context.ediText[context.rawIndex]);
         // There can be segments that do not follow standard EDI format (e.g. EDIFACT UNA segment).
         // Therefore, it is necessary to check and skip ignore segments before spliting into fields.
         boolean ignoreCurrentSegment = false;
@@ -81,11 +78,11 @@ isolated function readSegmentGroup(EdiUnitSchema[] currentUnitSchema, EdiContext
         }
     }
     if rootGroup && ediSchema.delimiters.'field != "FL" {
-        foreach int i in context.rawIndex...(context.ediText.length() - 1) {
+        foreach int i in context.rawIndex ... (context.ediText.length() - 1) {
             string unmatchedRaw = context.ediText[context.rawIndex];
-            string[] unmatchedSegFields = split(unmatchedRaw, ediSchema.delimiters.'field);
+            string[] unmatchedSegFields = check split(unmatchedRaw, ediSchema.delimiters.'field);
             if ediSchema.ignoreSegments.indexOf(unmatchedSegFields[0], 0) == () {
-               return error Error(string `Segment text does not match with the schema. 
+                return error Error(string `Segment text does not match with the schema. 
                     Segment: ${context.ediText[context.rawIndex]}, Curren row: ${context.rawIndex}`);
             }
         }

ballerina/subcomponent_group_reader.bal

@@ -19,7 +19,7 @@ isolated function readSubcomponentGroup(string scGroupText, EdiSchema schema, Ed
         return ();
     }
 
-    string[] subcomponents = split(scGroupText, schema.delimiters.subcomponent);
+    string[] subcomponents = check split(scGroupText, schema.delimiters.subcomponent);
     if compSchema.truncatable {
         int minFields = getMinimumSubcomponentFields(compSchema);
         if subcomponents.length() < minFields {

ballerina/tests/edi_modification_tests.bal

@@ -12,8 +12,8 @@ function testSegmentModification(string testName) returns error? {
     string ediOut = check toEdiString(message, schema);
     string ediExpected = check getOutputEDI(testName);
 
-    ediOut = prepareEDI(ediOut, schema);
-    ediIn = prepareEDI(ediExpected, schema);
+    ediOut = check prepareEDI(ediOut, schema);
+    ediIn = check prepareEDI(ediExpected, schema);
 
     test:assertEquals(ediOut, ediIn);
 }

ballerina/tests/segment_reading_tests.bal

@@ -13,8 +13,8 @@ function testSegments(string testName) returns error? {
     string ediOut = check toEdiString(message, schema);
     check saveEDIMessage(testName, ediOut);
 
-    ediOut = prepareEDI(ediOut, schema);
-    ediIn = prepareEDI(ediIn, schema);
+    ediOut = check prepareEDI(ediOut, schema);
+    ediIn = check prepareEDI(ediIn, schema);
 
     test:assertEquals(ediOut, ediIn);
 }
@@ -32,8 +32,8 @@ function testFixedLengthEDIs(string testName) returns error? {
     string ediOut = check toEdiString(message, schema);
     check saveEDIMessage(testName, ediOut);
 
-    ediOut = prepareEDI(ediOut, schema);
-    ediIn = prepareEDI(ediIn, schema);
+    ediOut = check prepareEDI(ediOut, schema);
+    ediIn = check prepareEDI(ediIn, schema);
 
     test:assertEquals(ediOut, ediIn);
 }
@@ -51,8 +51,8 @@ function testDynamicLengthEDIs(string testName) returns error? {
     string ediOut = check toEdiString(message, schema);
     check saveEDIMessage(testName, ediOut);
 
-    ediOut = prepareEDI(ediOut, schema);
-    ediIn = prepareEDI(ediIn, schema);
+    ediOut = check prepareEDI(ediOut, schema);
+    ediIn = check prepareEDI(ediIn, schema);
 
     test:assertEquals(ediOut, ediIn);
 }

ballerina/tests/test_utils.bal

@@ -1,6 +1,6 @@
 import ballerina/io;
-import ballerina/regex;
 import ballerina/file;
+import ballerina/lang.regexp;
 
 function getTestSchema(string testName) returns EdiSchema|error {
     string schemaPath = check file:joinPath("tests", "resources", testName, "schema.json");
@@ -37,11 +37,11 @@ function saveJsonMessage(string testName, json message) returns error? {
     check io:fileWriteJson(path, message);
 }
 
-function prepareEDI(string edi, EdiSchema schema) returns string {
-    string e1 = regex:replaceAll(edi, " ", "");
-    e1 = regex:replaceAll(e1, "\n", "");
-    e1 = regex:replaceAll(e1, validateDelimiter((schema.delimiters.decimalSeparator ?: ".")) + "0", "");
-    e1 = regex:replaceAll(e1, "0", "");
-    return e1;
+function prepareEDI(string edi, EdiSchema schema) returns string|error {
+    string:RegExp decSeprator = check regexp:fromString(schema.delimiters.decimalSeparator ?: ".");
+    string:RegExp nulSpaces = re ` |\n|0`;
+    string preparedEDI = nulSpaces.replaceAll(edi, "");
+    preparedEDI = decSeprator.replaceAll(preparedEDI, "");
+    return preparedEDI;
 }
 

ballerina/utils.bal

@@ -14,19 +14,27 @@
 // specific language governing permissions and limitations
 // under the License.
 
-import ballerina/regex;
+import ballerina/lang.regexp;
 
 isolated function convertToType(string value, EdiDataType dataType, string? decimalSeparator) returns SimpleType|error {
     string v = value.trim();
     match dataType {
         STRING => {
             return v;
         }
-        INT => {
-            return int:fromString(decimalSeparator != () ? regex:replace(v, decimalSeparator, ".") : v);
-        }
-        FLOAT => {
-            return float:fromString(decimalSeparator != () ? regex:replace(v, decimalSeparator, ".") : v);
+        INT|FLOAT => {
+            if decimalSeparator != () {
+                string:RegExp decimalSep = check regexp:fromString(decimalSeparator);
+                v = decimalSep.replace(v, ".");
+            }
+            match dataType {
+                INT => {
+                    return int:fromString(v);
+                }
+                FLOAT => {
+                    return float:fromString(v);
+                }
+            }
         }
     }
     return error("Undefined type for value:" + value);
@@ -96,31 +104,40 @@ isolated function splitFields(string segmentText, string fieldDelimiter, EdiUnit
                 return error Error(string `Start index and field length is not provided for fixed length schema field. Segment: ${segSchema.code}, Field: ${fieldSchema.tag}`);
             }
             int startIndex = fieldSchema.startIndex - 1;
-            int endIndex = startIndex +  feildLength;
+            int endIndex = startIndex + feildLength;
             if startIndex >= segmentText.length() {
                 break;
             }
-            endIndex = segmentText.length() < endIndex? segmentText.length() : endIndex;
+            endIndex = segmentText.length() < endIndex ? segmentText.length() : endIndex;
             string fieldText = segmentText.substring(startIndex, endIndex);
-            fields.push(fieldText);    
+            fields.push(fieldText);
         }
         return fields;
     } else {
         return split(segmentText, fieldDelimiter);
     }
 }
 
-isolated function split(string text, string delimiter) returns string[] {
+isolated function split(string text, string delimiter) returns string[]|Error {
     string preparedText = prepareToSplit(text, delimiter);
-    string validatedDelimiter = validateDelimiter(delimiter);
-    return regex:split(preparedText, validatedDelimiter);
+    string:RegExp|error validatedDelimiter = regexp:fromString(validateDelimiter(delimiter));
+    if validatedDelimiter is error {
+        return error Error("Invalid delimiter: " + delimiter);
+    }
+    return validatedDelimiter.split(preparedText);
 }
 
-isolated function splitSegments(string text, string delimiter) returns string[] {
-    string validatedDelimiter = validateDelimiter(delimiter);
-    string[] segmentLines = regex:split(text, validatedDelimiter);
+isolated function splitSegments(string text, string delimiter) returns string[]|Error {
+    string:RegExp|error validatedDelimiter = regexp:fromString(validateDelimiter(delimiter));
+    if validatedDelimiter is error {
+        return error Error("Invalid delimiter: " + delimiter);
+    }
+    string[] segmentLines = validatedDelimiter.split(text);
+    if segmentLines[segmentLines.length() - 1] == "" {
+        string _ = segmentLines.remove(segmentLines.length() - 1);
+    }
     foreach int i in 0 ... (segmentLines.length() - 1) {
-        segmentLines[i] = regex:replaceAll(segmentLines[i], "\n", "");
+        segmentLines[i] = removeLineBreaks(segmentLines[i]);
     }
     return segmentLines;
 }
@@ -227,7 +244,8 @@ isolated function serializeSimpleType(SimpleType v, EdiSchema schema, int fixedL
         if sv.endsWith(".0") {
             sv = sv.substring(0, sv.length() - 2);
         } else if schema.delimiters.decimalSeparator != "." {
-            sv = regex:replace(sv, "\\.", schema.delimiters.decimalSeparator ?: ".");
+            string:RegExp separator = re `\\.`;
+            sv = separator.replace(sv, schema.delimiters.decimalSeparator ?: ".");
         }
     }
     return fixedLength > 0 ? addPadding(sv, fixedLength) : sv;
@@ -236,9 +254,13 @@ isolated function serializeSimpleType(SimpleType v, EdiSchema schema, int fixedL
 isolated function addPadding(string value, int requiredLength) returns string {
     string paddedValue = value;
     int lengthDiff = requiredLength - value.length();
-    foreach int i in 1...lengthDiff {
+    foreach int i in 1 ... lengthDiff {
         paddedValue += " ";
     }
     return paddedValue;
 }
 
+isolated function removeLineBreaks(string value) returns string {
+    string:RegExp newline = re `\n`;
+    return newline.replaceAll(value, "");
+}

build.gradle

@@ -71,7 +71,6 @@ subprojects {
         ballerinaStdLibs "io.ballerina.stdlib:io-ballerina:${project.stdlibIoVersion}"
         ballerinaStdLibs "io.ballerina.stdlib:log-ballerina:${project.stdlibLogVersion}"
         ballerinaStdLibs "io.ballerina.stdlib:file-ballerina:${project.stdlibFileVersion}"
-        ballerinaStdLibs "io.ballerina.stdlib:regex-ballerina:${project.stdlibRegexVersion}"
 
         ballerinaStdLibs "io.ballerina.stdlib:os-ballerina:${project.stdlibOsVersion}"
         ballerinaStdLibs "io.ballerina.stdlib:time-ballerina:${project.stdlibTimeVersion}"

changelog.md

@@ -7,4 +7,5 @@ and this project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0
 ## [Unreleased]
 
 ### Added
-- [Add support for field length constraints (min/max)](https://github.com/ballerina-platform/ballerina-library/issues/5896).
+- [Add support for field length constraints (min/max)](https://github.com/ballerina-platform/ballerina-library/issues/5896).
+- [Updated dependencies to use lang.regex instead of ballerina/regex](https://github.com/ballerina-platform/ballerina-library/issues/5941)