UITextField限制漢字數量最正確的姿式,解決iOS7下substringToIndex方法致使的崩潰

今天在寫一個限制UITextField漢字數量的需求,原覺得很是簡單的一個需求,在實際開發中遇到了不少問題.

首先,漢字的輸入時的聯想詞在輸入到TextFiled時,並不會走

- (BOOL)textField:(UITextField *)textField shouldChangeCharactersInRange:(NSRange)range replacementString:(NSString *)string;

這致使了在這裏進行長度限制的代碼都會失效.此路不通.

通過查找資料,得知能夠用註冊UITextFieldTextDidChangeNotification通知來監測TextField的文字的變化

[[NSNotificationCenter defaultCenter]addObserver:self selector:@selector(textFiledEditChanged:) 
          name:@"UITextFieldTextDidChangeNotification" object:myTextField];

　　

在textFiledEditChanged:方法中

- (void)textDidChanged:(NSNotification *)notification{
    
    NSInteger maxLength = 6;
    UITextField *textField = (UITextField *)notification.object;

    if (textField.text.length > maxLength)
    {
        textField.text = [textField.text substringToIndex:maxLength];
    }
}

　　這種方法看起來沒什麼問題,然而實測在iOS7下輸入漢字時,一旦長度到達最大長度,再進行輸入(此時輸入狀態仍是拼音)就會當即崩潰.

*** Terminating app due to uncaught exception 'NSRangeException', reason: 'NSMutableRLEArray replaceObjectsInRange:withObject:length:: Out of bounds'

該問題在iOS8以上的系統中不會出現,多是蘋果修復了該BUG.

再次查找資料,找到這篇博客http://www.jianshu.com/p/2d1c06f2dfa4

這篇博客解決了UItextFiled在截取字符串時,若是對emoji進行截取會變亂碼的問題(緣由是emoji實際上佔了3個長度).

UITextField *textField = (UITextField *)obj.object;
    NSString *toBeString = textField.text;

    //獲取高亮部分
    UITextRange *selectedRange = [textField markedTextRange];
    UITextPosition *position = [textField positionFromPosition:selectedRange.start offset:0];

    // 沒有高亮選擇的字，則對已輸入的文字進行字數統計和限制
    if (!position)
    {
        if (toBeString.length > MAX_STARWORDS_LENGTH)
        {
            NSRange rangeIndex = [toBeString rangeOfComposedCharacterSequenceAtIndex:MAX_STARWORDS_LENGTH];
            if (rangeIndex.length == 1)
            {
                textField.text = [toBeString substringToIndex:MAX_STARWORDS_LENGTH];
            }
            else
            {
                NSRange rangeRange = [toBeString rangeOfComposedCharacterSequencesForRange:NSMakeRange(0, MAX_STARWORDS_LENGTH)];
                textField.text = [toBeString substringWithRange:rangeRange];
            }
        }
    }

測試後發現,這個方法在iOS8以上的系統運行良好,可是在iOS7下竟然徹底失效了.

緣由是在iOS7下,position對象老是不爲nil,致使走不進處理長度的代碼.

研究後發如今須要增長一個判斷條件,

改進後的代碼:

UITextField *textField = (UITextField *)obj.object;
    NSString *toBeString = textField.text;

    //獲取高亮部分
    UITextRange *selectedRange = [textField markedTextRange];
    UITextPosition *position = [textField positionFromPosition:selectedRange.start offset:0];

    // 沒有高亮選擇的字，則對已輸入的文字進行字數統計和限制
    if (!position || !selectedRange)
    {
        if (toBeString.length > MAX_STARWORDS_LENGTH)
        {
            NSRange rangeIndex = [toBeString rangeOfComposedCharacterSequenceAtIndex:MAX_STARWORDS_LENGTH];
            if (rangeIndex.length == 1)
            {
                textField.text = [toBeString substringToIndex:MAX_STARWORDS_LENGTH];
            }
            else
            {
                NSRange rangeRange = [toBeString rangeOfComposedCharacterSequencesForRange:NSMakeRange(0, MAX_STARWORDS_LENGTH)];
                textField.text = [toBeString substringWithRange:rangeRange];
            }
        }
    }

再次測試後沒有發現崩潰問題了.