Bugfix/new lines rule dot qualified expression and safe access expres…

…sion (#1337)

### Whats added:
 * corrected logic fix and warn long Dot Qualified Expression and Safe Access Expression and Postfix Expression
 * commented on the choice of different logics with these expressions
 * added fix test in NewLineRuleFixTest
 * corrected and added warnings in NewLineRuleWarnTest
 
  ### Issue (#865)

diktat-gradle-plugin/src/test/kotlin/org/cqfn/diktat/plugin/gradle/DiktatGradlePluginTest.kt

@@ -28,7 +28,10 @@ class DiktatGradlePluginTest {
     @Test
     fun `check default extension properties`() {
         val diktatExtension = project.extensions.getByName("diktat") as DiktatExtension
-        val actualInputs = project.tasks.named("diktatCheck", DiktatJavaExecTaskBase::class.java).get().actualInputs
+        val actualInputs = project.tasks
+            .named("diktatCheck", DiktatJavaExecTaskBase::class.java)
+            .get()
+            .actualInputs
         Assertions.assertFalse(diktatExtension.debug)
         Assertions.assertIterableEquals(project.fileTree("src").files, actualInputs.files)
         Assertions.assertTrue(actualInputs.files.isNotEmpty())

diktat-rules/src/main/kotlin/org/cqfn/diktat/ruleset/rules/chapter2/comments/HeaderCommentRule.kt

@@ -202,7 +202,10 @@ class HeaderCommentRule(configRules: List<RulesConfig>) : DiktatRule(
 
     // Check if provided copyright node differs only in the first date from pattern
     private fun isCopyRightTextMatchesPattern(copyrightNode: ASTNode?, copyrightPattern: String): Boolean {
-        val copyrightText = copyrightNode?.text?.replace("/*", "")?.replace("*/", "")?.replace("*", "")
+        val copyrightText = copyrightNode?.text
+            ?.replace("/*", "")
+            ?.replace("*/", "")
+            ?.replace("*", "")
 
         val datesInPattern = hyphenRegex.find(copyrightPattern)?.value
         val datesInCode = copyrightText?.let { hyphenRegex.find(it)?.value }

diktat-rules/src/main/kotlin/org/cqfn/diktat/ruleset/rules/chapter2/kdoc/CommentsFormatting.kt

@@ -333,7 +333,10 @@ class CommentsFormatting(configRules: List<RulesConfig>) : DiktatRule(
     } else if (node.treeParent.elementType != FILE && node.treeParent.treePrev != null &&
         node.treeParent.treePrev.treePrev != null) {
         // When comment inside of a PROPERTY
-        node.treeParent.treePrev.treePrev.elementType == LBRACE
+        node.treeParent
+            .treePrev
+            .treePrev
+            .elementType == LBRACE
     } else {
         node.treeParent.getAllChildrenWithType(node.elementType).first() == node
     }

diktat-rules/src/main/kotlin/org/cqfn/diktat/ruleset/rules/chapter2/kdoc/KdocComments.kt

@@ -222,9 +222,15 @@ class KdocComments(configRules: List<RulesConfig>) : DiktatRule(
         propertyInLocalKdoc: ASTNode?
     ) {
         val kdocText = if (prevComment.elementType == KDOC) {
-            prevComment.text.removePrefix("/**").removeSuffix("*/").trim('\n', ' ')
+            prevComment.text
+                .removePrefix("/**")
+                .removeSuffix("*/")
+                .trim('\n', ' ')
         } else {
-            prevComment.text.removePrefix("/*").removeSuffix("*/").trim('\n', ' ')
+            prevComment.text
+                .removePrefix("/*")
+                .removeSuffix("*/")
+                .trim('\n', ' ')
         }
         // if property is documented with KDoc, which has a property tag inside, then it can contain some additional more complicated
         // structure, that will be hard to move automatically

diktat-rules/src/main/kotlin/org/cqfn/diktat/ruleset/rules/chapter3/CollapseIfStatementsRule.kt

@@ -109,10 +109,11 @@ class CollapseIfStatementsRule(configRules: List<RulesConfig>) : DiktatRule(
 
     private fun takeCommentsBeforeNestedIf(node: ASTNode): List<ASTNode> {
         val thenNode = (node.psi as KtIfExpression).then?.node
-        return thenNode?.children()?.takeWhile { it.elementType != IF }?.filter {
-            it.elementType == EOL_COMMENT ||
-                it.elementType == BLOCK_COMMENT
-        }
+        return thenNode?.children()
+            ?.takeWhile { it.elementType != IF }
+            ?.filter {
+                it.elementType == EOL_COMMENT || it.elementType == BLOCK_COMMENT
+            }
             ?.toList() ?: emptyList()
     }
 

diktat-rules/src/main/kotlin/org/cqfn/diktat/ruleset/rules/chapter3/LineLength.kt

@@ -219,7 +219,10 @@ class LineLength(configRules: List<RulesConfig>) : DiktatRule(
         }
         // check, that space to split is a part of text - not code
         // If the space split is part of the code, then there is a chance of breaking the code when fixing, that why we should ignore it
-        val isSpaceIsWhiteSpace = node.psi.findElementAt(delimiterIndex)!!.node.isWhiteSpace()
+        val isSpaceIsWhiteSpace = node.psi
+            .findElementAt(delimiterIndex)!!
+            .node
+            .isWhiteSpace()
         if (isSpaceIsWhiteSpace) {
             return None()
         }

diktat-rules/src/main/kotlin/org/cqfn/diktat/ruleset/rules/chapter3/StringTemplateFormatRule.kt

@@ -96,7 +96,10 @@ class StringTemplateFormatRule(configRules: List<RulesConfig>) : DiktatRule(
             ?.elementType == ARRAY_ACCESS_EXPRESSION
 
         return if (onlyOneRefExpr && !isArrayAccessExpression) {
-            (!(node.treeNext.text.first().isLetterOrDigit()  // checking if first letter is valid
+            (!(node.treeNext
+                .text
+                .first()
+                .isLetterOrDigit()  // checking if first letter is valid
                 || node.treeNext.text.startsWith("_")) || node.treeNext.elementType == CLOSING_QUOTE)
         } else if (!isArrayAccessExpression) {
             node.hasAnyChildOfTypes(FLOAT_CONSTANT, INTEGER_CONSTANT)  // it also fixes "${1.0}asd" cases

diktat-rules/src/main/kotlin/org/cqfn/diktat/ruleset/rules/chapter3/files/FileStructureRule.kt

@@ -228,7 +228,9 @@ class FileStructureRule(configRules: List<RulesConfig>) : DiktatRule(
     ) {
         findAllReferences(node)
         packageName = (node.findChildByType(PACKAGE_DIRECTIVE)?.psi as KtPackageDirective).qualifiedName
-        node.findChildByType(IMPORT_LIST)?.getChildren(TokenSet.create(IMPORT_DIRECTIVE))?.toList()
+        node.findChildByType(IMPORT_LIST)
+            ?.getChildren(TokenSet.create(IMPORT_DIRECTIVE))
+            ?.toList()
             ?.forEach { import ->
                 val ktImportDirective = import.psi as KtImportDirective
                 val importName = ktImportDirective.importPath?.importedName?.asString()

diktat-rules/src/main/kotlin/org/cqfn/diktat/ruleset/rules/chapter3/files/IndentationRule.kt

@@ -232,13 +232,17 @@ class IndentationRule(configRules: List<RulesConfig>) : DiktatRule(
     /**
      * If it is triple-quoted string template we need to indent all its parts
      */
+    @Suppress("LOCAL_VARIABLE_EARLY_DECLARATION")
     private fun fixStringLiteral(
         whiteSpace: PsiWhiteSpace,
         expectedIndent: Int,
         actualIndent: Int
     ) {
         val textIndent = " ".repeat(expectedIndent + INDENT_SIZE)
-        val templateEntries = whiteSpace.node.treeNext.firstChildNode.getAllChildrenWithType(LITERAL_STRING_TEMPLATE_ENTRY)
+        val templateEntries = whiteSpace.node
+            .treeNext
+            .firstChildNode
+            .getAllChildrenWithType(LITERAL_STRING_TEMPLATE_ENTRY)
         templateEntries.forEach { node ->
             if (!node.text.contains("\n")) {
                 fixFirstTemplateEntries(node, textIndent, actualIndent)
@@ -263,8 +267,14 @@ class IndentationRule(configRules: List<RulesConfig>) : DiktatRule(
     ) {
         val correctedText = StringBuilder()
         // shift of the node depending on its initial string template indent
-        val nodeStartIndent = if (node.firstChildNode.text.takeWhile { it == ' ' }.count() - actualIndent - INDENT_SIZE > 0) {
-            node.firstChildNode.text.takeWhile { it == ' ' }.count() - actualIndent - INDENT_SIZE
+        val nodeStartIndent = if (node.firstChildNode
+            .text
+            .takeWhile { it == ' ' }
+            .count() - actualIndent - INDENT_SIZE > 0) {
+            node.firstChildNode
+                .text
+                .takeWhile { it == ' ' }
+                .count() - actualIndent - INDENT_SIZE
         } else {
             0
         }

diktat-rules/src/main/kotlin/org/cqfn/diktat/ruleset/rules/chapter3/files/NewlinesRule.kt

@@ -12,8 +12,10 @@ import org.cqfn.diktat.ruleset.utils.appendNewlineMergingWhiteSpace
 import org.cqfn.diktat.ruleset.utils.emptyBlockList
 import org.cqfn.diktat.ruleset.utils.extractLineOfText
 import org.cqfn.diktat.ruleset.utils.findAllDescendantsWithSpecificType
+import org.cqfn.diktat.ruleset.utils.findAllNodesWithCondition
 import org.cqfn.diktat.ruleset.utils.findParentNodeWithSpecificType
 import org.cqfn.diktat.ruleset.utils.getFilePath
+import org.cqfn.diktat.ruleset.utils.getFirstChildWithType
 import org.cqfn.diktat.ruleset.utils.getIdentifierName
 import org.cqfn.diktat.ruleset.utils.getRootNode
 import org.cqfn.diktat.ruleset.utils.hasParent
@@ -124,22 +126,82 @@ class NewlinesRule(configRules: List<RulesConfig>) : DiktatRule(
     private val configuration by lazy {
         NewlinesRuleConfiguration(configRules.getRuleConfig(WRONG_NEWLINES)?.configuration ?: emptyMap())
     }
+
     override fun logic(node: ASTNode) {
         when (node.elementType) {
             SEMICOLON -> handleSemicolon(node)
             OPERATION_REFERENCE, EQ -> handleOperatorWithLineBreakAfter(node)
+            // this logic regulates indentation with elements - so that the symbol and the subsequent expression are on the same line
             in lineBreakBeforeOperators -> handleOperatorWithLineBreakBefore(node)
             LPAR -> handleOpeningParentheses(node)
             COMMA -> handleComma(node)
             COLON -> handleColon(node)
             BLOCK -> handleLambdaBody(node)
             RETURN -> handleReturnStatement(node)
             SUPER_TYPE_LIST, VALUE_PARAMETER_LIST, VALUE_ARGUMENT_LIST -> handleList(node)
+            // this logic splits long expressions into multiple lines
+            DOT_QUALIFIED_EXPRESSION, SAFE_ACCESS_EXPRESSION, POSTFIX_EXPRESSION -> handDotQualifiedAndSafeAccessExpression(node)
             else -> {
             }
         }
     }
 
+    @Suppress("GENERIC_VARIABLE_WRONG_DECLARATION", "MagicNumber")
+    private fun handDotQualifiedAndSafeAccessExpression(node: ASTNode) {
+        val listParentTypesNoFix = listOf(PACKAGE_DIRECTIVE, IMPORT_DIRECTIVE, VALUE_PARAMETER_LIST,
+            VALUE_ARGUMENT_LIST, DOT_QUALIFIED_EXPRESSION, SAFE_ACCESS_EXPRESSION, POSTFIX_EXPRESSION)
+        if (isNotFindParentNodeWithSpecificManyType(node, listParentTypesNoFix)) {
+            val listDot = node.findAllNodesWithCondition(
+                withSelf = true,
+                excludeChildrenCondition = { !isDotQuaOrSafeAccessOrPostfixExpression(it) }
+            ) {
+                isDotQuaOrSafeAccessOrPostfixExpression(it) && it.elementType != POSTFIX_EXPRESSION
+            }.reversed()
+            if (listDot.size > 3) {
+                val without = listDot.filterNot {
+                    val whiteSpaceBeforeDotOrSafeAccess = it.findChildByType(DOT)?.treePrev ?: it.findChildByType(SAFE_ACCESS)?.treePrev
+                    whiteSpaceBeforeDotOrSafeAccess?.elementType == WHITE_SPACE && whiteSpaceBeforeDotOrSafeAccess.text.lines().size > 1
+                }
+                if (without.size > 1 || (without.size == 1 && without[0] != listDot[0])) {
+                    WRONG_NEWLINES.warnAndFix(configRules, emitWarn, isFixMode, "should be split before second and other dot/safe access", node.startOffset, node) {
+                        fixDotQualifiedExpression(listDot)
+                    }
+                }
+            }
+        }
+    }
+
+    /**
+     * Return false, if you find parent with types in list else return true
+     */
+    private fun isNotFindParentNodeWithSpecificManyType(node: ASTNode, list: List<IElementType>): Boolean {
+        list.forEach { elem ->
+            node.findParentNodeWithSpecificType(elem)?.let {
+                return false
+            }
+        }
+        return true
+    }
+
+    /**
+     * Fix Dot Qualified Expression and Safe Access Expression -
+     * 1) Append new White Space node before second and subsequent node Dot or Safe Access
+     * in Dot Qualified Expression? Safe Access Expression and Postfix Expression
+     * 2) If before first Dot or Safe Access node stay White Space node with \n - remove this node
+     */
+    private fun fixDotQualifiedExpression(list: List<ASTNode>) {
+        list.forEachIndexed { index, astNode ->
+            val dotNode = astNode.getFirstChildWithType(DOT) ?: astNode.getFirstChildWithType(SAFE_ACCESS)
+            val nodeBeforeDot = dotNode?.treePrev
+            if (index > 0) {
+                astNode.appendNewlineMergingWhiteSpace(nodeBeforeDot, dotNode)
+            }
+        }
+    }
+
+    private fun isDotQuaOrSafeAccessOrPostfixExpression(node: ASTNode) =
+        node.elementType == DOT_QUALIFIED_EXPRESSION || node.elementType == SAFE_ACCESS_EXPRESSION || node.elementType == POSTFIX_EXPRESSION
+
     /**
      * Check that EOL semicolon is used only in enums
      */
@@ -294,9 +356,13 @@ class NewlinesRule(configRules: List<RulesConfig>) : DiktatRule(
         }
     }
 
+    @Suppress("TOO_LONG_FUNCTION")
     private fun handleLambdaBody(node: ASTNode) {
         if (node.treeParent.elementType == FUNCTION_LITERAL) {
-            val isSingleLineLambda = node.treeParent.text.lines().size == 1
+            val isSingleLineLambda = node.treeParent
+                .text
+                .lines()
+                .size == 1
             val arrowNode = node.siblings(false).find { it.elementType == ARROW }
             if (!isSingleLineLambda && arrowNode != null) {
                 // lambda with explicit arguments

diktat-rules/src/main/kotlin/org/cqfn/diktat/ruleset/rules/chapter3/files/WhiteSpaceRule.kt

@@ -337,12 +337,18 @@ class WhiteSpaceRule(configRules: List<RulesConfig>) : DiktatRule(
             else -> {
             }
         }
-        val isDeclaration = node.treeParent.elementType == VALUE_PARAMETER_LIST && node.treeParent.treeParent.elementType.let {
-            it == PRIMARY_CONSTRUCTOR || it == FUN || it == CALL_EXPRESSION
-        }
-        val isCall = node.treeParent.elementType == VALUE_ARGUMENT_LIST && node.treeParent.treeParent.elementType.let {
-            it == CONSTRUCTOR_DELEGATION_CALL || it == CALL_EXPRESSION
-        }
+        val isDeclaration = node.treeParent.elementType == VALUE_PARAMETER_LIST && node.treeParent
+            .treeParent
+            .elementType
+            .let {
+                it == PRIMARY_CONSTRUCTOR || it == FUN || it == CALL_EXPRESSION
+            }
+        val isCall = node.treeParent.elementType == VALUE_ARGUMENT_LIST && node.treeParent
+            .treeParent
+            .elementType
+            .let {
+                it == CONSTRUCTOR_DELEGATION_CALL || it == CALL_EXPRESSION
+            }
         if (isDeclaration || isCall) {
             handleToken(node, 0, 0)
         } else {

diktat-rules/src/main/kotlin/org/cqfn/diktat/ruleset/rules/chapter4/NullChecksRule.kt

@@ -62,7 +62,10 @@ class NullChecksRule(configRules: List<RulesConfig>) : DiktatRule(
     private fun isComplexIfStatement(parentIf: ASTNode): Boolean {
         val parentIfPsi = parentIf.psi
         require(parentIfPsi is KtIfExpression)
-        return (parentIfPsi.`else`?.node?.firstChildNode?.elementType == IF_KEYWORD)
+        return (parentIfPsi.`else`
+            ?.node
+            ?.firstChildNode
+            ?.elementType == IF_KEYWORD)
     }
 
     private fun conditionInIfStatement(node: ASTNode) {
@@ -218,7 +221,8 @@ class NullChecksRule(configRules: List<RulesConfig>) : DiktatRule(
         .treeParent
         .findChildByType(type)
         ?.let { it.findChildByType(BLOCK) ?: it }
-        ?.findAllNodesWithCondition { it.elementType == BREAK }?.isNotEmpty()
+        ?.findAllNodesWithCondition { it.elementType == BREAK }
+        ?.isNotEmpty()
         ?: false
 
     private fun ASTNode.extractLinesFromBlock(type: IElementType): List<String>? =

diktat-rules/src/main/kotlin/org/cqfn/diktat/ruleset/rules/chapter6/RunInScript.kt

@@ -59,7 +59,10 @@ class RunInScript(private val configRules: List<RulesConfig>) : Rule(NAME_ID) {
 
     private fun checkScript(node: ASTNode) {
         val isLambdaArgument = node.firstChildNode.hasChildOfType(LAMBDA_ARGUMENT)
-        val isLambdaInsideValueArgument = node.firstChildNode.findChildByType(VALUE_ARGUMENT_LIST)?.findChildByType(VALUE_ARGUMENT)?.findChildByType(LAMBDA_EXPRESSION) != null
+        val isLambdaInsideValueArgument = node.firstChildNode
+            .findChildByType(VALUE_ARGUMENT_LIST)
+            ?.findChildByType(VALUE_ARGUMENT)
+            ?.findChildByType(LAMBDA_EXPRESSION) != null
         if (!isLambdaArgument && !isLambdaInsideValueArgument) {
             warnRunInScript(node)
         }

diktat-rules/src/main/kotlin/org/cqfn/diktat/ruleset/rules/chapter6/classes/InlineClassesRule.kt

@@ -45,7 +45,10 @@ class InlineClassesRule(configRules: List<RulesConfig>) : DiktatRule(
         // Fixme: In Kotlin 1.4.30 inline classes may be used with internal constructors. When it will be released need to check it
         if (hasValidProperties(classPsi) &&
             !isExtendingClass(classPsi.node) &&
-            classPsi.node.getFirstChildWithType(MODIFIER_LIST)?.getChildren(null)?.all { it.elementType in goodModifiers } != false) {
+            classPsi.node
+                .getFirstChildWithType(MODIFIER_LIST)
+                ?.getChildren(null)
+                ?.all { it.elementType in goodModifiers } != false) {
             // Fixme: since it's an experimental feature we shouldn't do fixer
             INLINE_CLASS_CAN_BE_USED.warn(configRules, emitWarn, isFixMode, "class ${classPsi.name}", classPsi.node.startOffset, classPsi.node)
         }
@@ -56,8 +59,14 @@ class InlineClassesRule(configRules: List<RulesConfig>) : DiktatRule(
             return !classPsi.getProperties().single().isVar
         } else if (classPsi.getProperties().isEmpty() && classPsi.hasExplicitPrimaryConstructor()) {
             return classPsi.primaryConstructorParameters.size == 1 &&
-                !classPsi.primaryConstructorParameters.first().node.hasChildOfType(VAR_KEYWORD) &&
-                classPsi.primaryConstructor?.visibilityModifierType()?.value?.let { it == "public" } ?: true
+                !classPsi.primaryConstructorParameters
+                    .first()
+                    .node
+                    .hasChildOfType(VAR_KEYWORD) &&
+                classPsi.primaryConstructor
+                    ?.visibilityModifierType()
+                    ?.value
+                    ?.let { it == "public" } ?: true
         }
         return false
     }

diktat-rules/src/main/kotlin/org/cqfn/diktat/ruleset/utils/AstNodeUtils.kt

@@ -549,15 +549,17 @@ fun ASTNode.leaveExactlyNumNewLines(num: Int) {
 }
 
 /**
- * If [whiteSpaceNode] is not null and has type [WHITE_SPACE], prepend a line break to it's text.
- * Otherwise, insert a new node with a line break before [beforeNode]
+ * If [whiteSpaceNode] is not null and has type [WHITE_SPACE] and this [WHITE_SPACE] contains 1 line, prepend a line break to it's text.
+ * If [whiteSpaceNode] is null or has`t type [WHITE_SPACE], insert a new node with a line break before [beforeNode]
  *
  * @param whiteSpaceNode a node that can possibly be modified
  * @param beforeNode a node before which a new WHITE_SPACE node will be inserted
  */
 fun ASTNode.appendNewlineMergingWhiteSpace(whiteSpaceNode: ASTNode?, beforeNode: ASTNode?) {
     if (whiteSpaceNode != null && whiteSpaceNode.elementType == WHITE_SPACE) {
-        (whiteSpaceNode as LeafPsiElement).rawReplaceWithText("\n${whiteSpaceNode.text}")
+        if (whiteSpaceNode.text.lines().size == 1) {
+            (whiteSpaceNode as LeafPsiElement).rawReplaceWithText("\n${whiteSpaceNode.text}")
+        }
     } else {
         addChild(PsiWhiteSpaceImpl("\n"), beforeNode)
     }

diktat-rules/src/main/kotlin/org/cqfn/diktat/ruleset/utils/KotlinParser.kt

@@ -151,7 +151,8 @@ class KotlinParser {
                     .createBlockCodeFragment(text, null)
                     .node
                     .findChildByType(BLOCK)!!
-                    .findChildByType(ERROR_ELEMENT)!!.findChildByType(IMPORT_LIST)!!
+                    .findChildByType(ERROR_ELEMENT)!!
+                    .findChildByType(IMPORT_LIST)!!
             }
         } else {
             ktPsiFactory

diktat-rules/src/main/kotlin/org/cqfn/diktat/ruleset/utils/StringUtils.kt

@@ -18,7 +18,9 @@ val JAVA = arrayOf("abstract", "assert", "boolean",
     "try", "void", "volatile", "while")
 
 @Suppress("VARIABLE_NAME_INCORRECT_FORMAT")
-val KOTLIN = KtTokens.KEYWORDS.types.map { line -> line.toString() }
+val KOTLIN = KtTokens.KEYWORDS
+    .types
+    .map { line -> line.toString() }
     .plus(KtTokens.SOFT_KEYWORDS.types.map { line -> line.toString() })
 
 val loggerPropertyRegex = "(log|LOG|logger)".toRegex()