Merge pull request #8 from yaslab/feature/trim-fields

Trim each field

CSV.swift.podspec

@@ -1,6 +1,6 @@
 Pod::Spec.new do |s|
   s.name = 'CSV.swift'
-  s.version = '1.0.0'
+  s.version = '1.1.0'
   s.license = 'MIT'
   s.summary = 'CSV reading library written in Swift.'
   s.homepage = 'https://github.com/yaslab/CSV.swift'

README.md

@@ -58,7 +58,7 @@ var csv = try! CSV(
     string: "id,name\n1,foo",
     hasHeaderRow: true) // It must be true.
 
-while csv.next() != nil {
+while let _ = csv.next() {
     print("\(csv["id"]!)")   // => "1"
     print("\(csv["name"]!)") // => "foo"
 }
@@ -83,13 +83,13 @@ let csv = try! CSV(
 ### CocoaPods
 
 ```ruby
-pod 'CSV.swift', '~> 1.0'
+pod 'CSV.swift', '~> 1.1'
 ```
 
 ### Carthage
 
 ```
-github "yaslab/CSV.swift" ~> 1.0
+github "yaslab/CSV.swift" ~> 1.1
 ```
 
 ### Swift Package Manager
@@ -100,7 +100,7 @@ import PackageDescription
 let package = Package(
     name: "PackageName",
     dependencies: [
-        .Package(url: "https://github.com/yaslab/CSV.swift", majorVersion: 1, minor: 0)
+        .Package(url: "https://github.com/yaslab/CSV.swift", majorVersion: 1, minor: 1)
     ]
 )
 ```

Sources/CSV+init.swift

@@ -13,10 +13,11 @@ extension CSV {
     public init(
         stream: InputStream,
         hasHeaderRow: Bool = defaultHasHeaderRow,
+        trimFields: Bool = defaultTrimFields,
         delimiter: UnicodeScalar = defaultDelimiter)
         throws
     {
-        try self.init(stream: stream, codecType: UTF8.self, hasHeaderRow: hasHeaderRow, delimiter: delimiter)
+        try self.init(stream: stream, codecType: UTF8.self, hasHeaderRow: hasHeaderRow, trimFields: trimFields, delimiter: delimiter)
     }
 
 }
@@ -26,11 +27,12 @@ extension CSV {
     public init(
         string: String,
         hasHeaderRow: Bool = defaultHasHeaderRow,
+        trimFields: Bool = defaultTrimFields,
         delimiter: UnicodeScalar = defaultDelimiter)
         throws
     {
         let iterator = string.unicodeScalars.makeIterator()
-        try self.init(iterator: iterator, hasHeaderRow: hasHeaderRow, delimiter: delimiter)
+        try self.init(iterator: iterator, hasHeaderRow: hasHeaderRow, trimFields: trimFields, delimiter: delimiter)
     }
     
 }

Sources/CSV.swift

@@ -13,11 +13,13 @@ private let CR = UnicodeScalar("\r")!
 private let DQUOTE = UnicodeScalar("\"")!
 
 internal let defaultHasHeaderRow = false
+internal let defaultTrimFields = false
 internal let defaultDelimiter = UnicodeScalar(",")!
 
 public struct CSV: IteratorProtocol, Sequence {
 
     private var iterator: AnyIterator<UnicodeScalar>
+    private let trimFields: Bool
     private let delimiter: UnicodeScalar
 
     private var back: UnicodeScalar? = nil
@@ -28,15 +30,23 @@ public struct CSV: IteratorProtocol, Sequence {
     public var headerRow: [String]? { return _headerRow }
     private var _headerRow: [String]? = nil
     
+    private let whitespaces: CharacterSet
+
     internal init<T: IteratorProtocol>(
         iterator: T,
         hasHeaderRow: Bool,
+        trimFields: Bool,
         delimiter: UnicodeScalar)
         throws where T.Element == UnicodeScalar
     {
         self.iterator = AnyIterator(base: iterator)
+        self.trimFields = trimFields
         self.delimiter = delimiter
 
+        var whitespaces = CharacterSet.whitespaces
+        whitespaces.remove(delimiter)
+        self.whitespaces = whitespaces
+        
         if hasHeaderRow {
             guard let headerRow = next() else {
                 throw CSVError.cannotReadHeaderRow
@@ -55,13 +65,14 @@ public struct CSV: IteratorProtocol, Sequence {
         stream: InputStream,
         codecType: T.Type,
         hasHeaderRow: Bool = defaultHasHeaderRow,
+        trimFields: Bool = defaultTrimFields,
         delimiter: UnicodeScalar = defaultDelimiter)
         throws
         where T.CodeUnit == UInt8
     {
         let reader = try BinaryReader(stream: stream, endian: .unknown, closeOnDeinit: true)
         let iterator = UnicodeIterator(input: reader.makeUInt8Iterator(), inputEncodingType: codecType)
-        try self.init(iterator: iterator, hasHeaderRow: hasHeaderRow, delimiter: delimiter)
+        try self.init(iterator: iterator, hasHeaderRow: hasHeaderRow, trimFields: trimFields, delimiter: delimiter)
     }
 
     /// Create an instance with `InputStream`.
@@ -76,13 +87,14 @@ public struct CSV: IteratorProtocol, Sequence {
         codecType: T.Type,
         endian: Endian = .big,
         hasHeaderRow: Bool = defaultHasHeaderRow,
+        trimFields: Bool = defaultTrimFields,
         delimiter: UnicodeScalar = defaultDelimiter)
         throws
         where T.CodeUnit == UInt16
     {
         let reader = try BinaryReader(stream: stream, endian: endian, closeOnDeinit: true)
         let iterator = UnicodeIterator(input: reader.makeUInt16Iterator(), inputEncodingType: codecType)
-        try self.init(iterator: iterator, hasHeaderRow: hasHeaderRow, delimiter: delimiter)
+        try self.init(iterator: iterator, hasHeaderRow: hasHeaderRow, trimFields: trimFields, delimiter: delimiter)
     }
 
     /// Create an instance with `InputStream`.
@@ -97,13 +109,14 @@ public struct CSV: IteratorProtocol, Sequence {
         codecType: T.Type,
         endian: Endian = .big,
         hasHeaderRow: Bool = defaultHasHeaderRow,
+        trimFields: Bool = defaultTrimFields,
         delimiter: UnicodeScalar = defaultDelimiter)
         throws
         where T.CodeUnit == UInt32
     {
         let reader = try BinaryReader(stream: stream, endian: endian, closeOnDeinit: true)
         let iterator = UnicodeIterator(input: reader.makeUInt32Iterator(), inputEncodingType: codecType)
-        try self.init(iterator: iterator, hasHeaderRow: hasHeaderRow, delimiter: delimiter)
+        try self.init(iterator: iterator, hasHeaderRow: hasHeaderRow, trimFields: trimFields, delimiter: delimiter)
     }
     
     // MARK: IteratorProtocol
@@ -160,6 +173,13 @@ public struct CSV: IteratorProtocol, Sequence {
         var field: String
         var end: Bool
         while true {
+            if trimFields {
+                // Trim the leading spaces
+                while next != nil && whitespaces.contains(next!) {
+                    next = moveNext()
+                }
+            }
+            
             if next == nil {
                 (field, end) = ("", true)
             }
@@ -169,6 +189,11 @@ public struct CSV: IteratorProtocol, Sequence {
             else {
                 back = next
                 (field, end) = readField(quoted: false)
+                
+                if trimFields {
+                    // Trim the trailing spaces
+                    field = field.trimmingCharacters(in: whitespaces)
+                }
             }
             row.append(field)
             if end {
@@ -188,7 +213,15 @@ public struct CSV: IteratorProtocol, Sequence {
         while let c = next {
             if quoted {
                 if c == DQUOTE {
-                    let cNext = moveNext()
+                    var cNext = moveNext()
+                    
+                    if trimFields {
+                        // Trim the trailing spaces
+                        while cNext != nil && whitespaces.contains(cNext!) {
+                            cNext = moveNext()
+                        }
+                    }
+                    
                     if cNext == nil || cNext == CR || cNext == LF {
                         if cNext == CR {
                             let cNextNext = moveNext()
@@ -208,7 +241,7 @@ public struct CSV: IteratorProtocol, Sequence {
                         field.append(String(DQUOTE))
                     }
                     else {
-                        // ERROR??
+                        // ERROR?
                         field.append(String(c))
                     }
                 }
@@ -239,6 +272,7 @@ public struct CSV: IteratorProtocol, Sequence {
             next = moveNext()
         }
         
+        // END FILE
         return (field, true)
     }
     

Sources/Info.plist

@@ -15,7 +15,7 @@
 	<key>CFBundlePackageType</key>
 	<string>FMWK</string>
 	<key>CFBundleShortVersionString</key>
-	<string>1.0.0</string>
+	<string>1.1.0</string>
 	<key>CFBundleSignature</key>
 	<string>????</string>
 	<key>CFBundleVersion</key>

Tests/CSV/CSVTests.swift

@@ -163,7 +163,7 @@ class CSVTests: XCTestCase {
     
     func testCSVState1() {
         let it = "あ,い1,\"う\",えお\n,,x,".unicodeScalars.makeIterator()
-        var csv = try! CSV(iterator: it, hasHeaderRow: defaultHasHeaderRow, delimiter: defaultDelimiter)
+        var csv = try! CSV(iterator: it, hasHeaderRow: defaultHasHeaderRow, trimFields: defaultTrimFields, delimiter: defaultDelimiter)
         
         var rows = [[String]]()
         
@@ -175,4 +175,112 @@ class CSVTests: XCTestCase {
         XCTAssertEqual(rows[1], ["", "", "x", ""])
     }
     
+    func testTrimFields1() {
+        let csvString = "abc,def,ghi"
+        let csv = try! CSV(string: csvString, trimFields: true)
+        for row in csv {
+            XCTAssertEqual(row, ["abc", "def", "ghi"])
+        }
+    }
+    
+    func testTrimFields2() {
+        let csvString = " abc,  def,   ghi"
+        let csv = try! CSV(string: csvString, trimFields: true)
+        for row in csv {
+            XCTAssertEqual(row, ["abc", "def", "ghi"])
+        }
+    }
+    
+    func testTrimFields3() {
+        let csvString = "abc ,def  ,ghi   "
+        let csv = try! CSV(string: csvString, trimFields: true)
+        for row in csv {
+            XCTAssertEqual(row, ["abc", "def", "ghi"])
+        }
+    }
+    
+    func testTrimFields4() {
+        let csvString = " abc ,  def  ,   ghi   "
+        let csv = try! CSV(string: csvString, trimFields: true)
+        for row in csv {
+            XCTAssertEqual(row, ["abc", "def", "ghi"])
+        }
+    }
+    
+    func testTrimFields5() {
+        let csvString = "\"abc\",\"def\",\"ghi\""
+        let csv = try! CSV(string: csvString, trimFields: true)
+        for row in csv {
+            XCTAssertEqual(row, ["abc", "def", "ghi"])
+        }
+    }
+
+    func testTrimFields6() {
+        let csvString = " \"abc\",  \"def\",   \"ghi\""
+        let csv = try! CSV(string: csvString, trimFields: true)
+        for row in csv {
+            XCTAssertEqual(row, ["abc", "def", "ghi"])
+        }
+    }
+    
+    func testTrimFields7() {
+        let csvString = "\"abc\" ,\"def\"  ,\"ghi\"   "
+        let csv = try! CSV(string: csvString, trimFields: true)
+        for row in csv {
+            XCTAssertEqual(row, ["abc", "def", "ghi"])
+        }
+    }
+    
+    func testTrimFields8() {
+        let csvString = " \"abc\" ,  \"def\"  ,   \"ghi\"   "
+        let csv = try! CSV(string: csvString, trimFields: true)
+        for row in csv {
+            XCTAssertEqual(row, ["abc", "def", "ghi"])
+        }
+    }
+    
+    func testTrimFields9() {
+        let csvString = "\" abc \",\" def \",\" ghi \""
+        let csv = try! CSV(string: csvString, trimFields: true)
+        for row in csv {
+            XCTAssertEqual(row, [" abc ", " def ", " ghi "])
+        }
+    }
+
+    func testTrimFields10() {
+        let csvString = "\tabc,\t\tdef\t,ghi\t"
+        let csv = try! CSV(string: csvString, trimFields: true)
+        for row in csv {
+            XCTAssertEqual(row, ["abc", "def", "ghi"])
+        }
+    }
+    
+    func testTrimFields11() {
+        let csvString = " abc \n def "
+        var csv = try! CSV(string: csvString, trimFields: true)
+        
+        let row1 = csv.next()!
+        XCTAssertEqual(row1, ["abc"])
+        let row2 = csv.next()!
+        XCTAssertEqual(row2, ["def"])
+    }
+    
+    func testTrimFields12() {
+        let csvString = " \"abc \" \n \" def\" "
+        var csv = try! CSV(string: csvString, trimFields: true)
+        
+        let row1 = csv.next()!
+        XCTAssertEqual(row1, ["abc "])
+        let row2 = csv.next()!
+        XCTAssertEqual(row2, [" def"])
+    }
+    
+    func testTrimFields13() {
+        let csvString = " abc \t\tdef\t ghi "
+        let csv = try! CSV(string: csvString, trimFields: true, delimiter: UnicodeScalar("\t")!)
+        for row in csv {
+            XCTAssertEqual(row, ["abc", "", "def", "ghi"])
+        }
+    }
+    
 }

Tests/CSV/ReadmeTests.swift

@@ -46,7 +46,7 @@ class ReadmeTests: XCTestCase {
             string: "id,name\n1,foo",
             hasHeaderRow: true) // It must be true.
         
-        while csv.next() != nil {
+        while let _ = csv.next() {
             print("\(csv["id"]!)")   // => "1"
             print("\(csv["name"]!)") // => "foo"
         }