我有一个要求，应该使用给定的字符拆分文件。默认拆分选项是 CRLF、LF 和 CR。
在这些情况下，我分别用 \r\n 和 \n 和 \r 分割线。
我还要求处理任何大小的文件。（处理基本上是将给定的字符串插入到给定位置的文件中）。
为此，我正在读取 1024 字节的文件。
然后我正在应用 string.Split() 方法。Split() 方法提供了忽略空格和无空格的选项。
我必须将这些换行符添加回该行。为此，我正在使用二进制写入器，并且正在将字节数组写入新文件。问题：-
1) 当换行符为 CRLF 且拆分选项为 NONE 时，拆分后的数组中也会添加空格。给出了第二个选项（忽略空格）CRLF 正常工作。
2）位忽略空白选项会产生其他问题，因为我正在逐字节读取文件，所以我不能忽略空白。
3)当换行符不是默认值时（例如'|'，一个空值被添加到结果行。
任何人都可以解决我的问题吗？
这是我写的方法
private static void ProcessOriginalFile(string filePath, string destinationFilePath, int fromPosition, int toPosition, string lineBreakCharacter, string replaceString, long fileSize, bool ignoreHeader, bool ignoreFooter)
{
int chunkSize = fileSize < 1024 ? (int)fileSize : 1024;//bytes
byte[] chunkData = new byte[chunkSize];
char[] charactersSeparator=new char[1];
charactersSeparator = CommonOperations.ResolveTheRecordBreak(lineBreakCharacter);
int totalLineBreakCharactersLength = 0;
for (int i = 0; i < charactersSeparator.Length; i++)
{
if (charactersSeparator[i] == 0)
break;
totalLineBreakCharactersLength = totalLineBreakCharactersLength + BitConverter.GetBytes(charactersSeparator[i]).Length;
}
using (BinaryReader fileReader = new BinaryReader(new FileStream(filePath, FileMode.Open)))
{
string lastChunk = string.Empty;
string chunkcontents = string.Empty;
IList<byte[]> dataToBeWritten = new List<byte[]>();
while (fileSize > 0)
{
chunkData = fileReader.ReadBytes(chunkSize);
byte[] chunkToBeWritten = new byte[chunkSize];
chunkcontents = chunkcontents + System.Text.ASCIIEncoding.UTF8.GetString(chunkData);
string[] splittedArray = new string[1];
splittedArray = chunkcontents.Split(charactersSeparator, StringSplitOptions.RemoveEmptyEntries);
if (ignoreHeader)
{
splittedArray = splittedArray.Skip(1).ToArray();
ignoreHeader = false;
if (splittedArray.Count() == 0)
continue;
}
int count = 0;
if (splittedArray != null || splittedArray.Count() > 0)
{
foreach (string str in splittedArray)
{
++count;
if (count != splittedArray.Length)
{
string stringToBeEdited = string.Empty;
stringToBeEdited = str;
if (stringToBeEdited.Length < (toPosition + 1))
throw new Exception("Position exceeds the string length. Line contents are : " + stringToBeEdited + " String length is : " + stringToBeEdited.Length + " To Position is : " + toPosition);
//replace the text between from and two positions with the replaced string
stringToBeEdited = stringToBeEdited.Remove(fromPosition, toPosition + 1 - fromPosition).Insert(fromPosition, replaceString);
//clear the array
dataToBeWritten.Clear();
AddLineBreakCharacter(lineBreakCharacter, charactersSeparator, totalLineBreakCharactersLength, dataToBeWritten, ref chunkToBeWritten, ref stringToBeEdited);
//write data using binary writer
WriteBinaryDataToFile(destinationFilePath, dataToBeWritten);
totalLinesProcessed++;
}
else
{
lastChunk = str;
chunkcontents = str;
}
}
}
fileSize = fileSize - chunkSize;
/*if file size is less than the chunksize*2, then chunk size should be the file size
* and chunkdata array length should be that of the file size*/
if (fileSize < (chunkSize * 2))
{
chunkSize = (int)fileSize;
chunkData = new byte[chunkSize];
}
}
if (!string.IsNullOrEmpty(lastChunk))
{
if (ignoreFooter == false)
{
if (lastChunk.Length >= toPosition + 1)
lastChunk = lastChunk.Remove(fromPosition, toPosition + 1 - fromPosition).Insert(fromPosition, replaceString);
else
throw new Exception("Position exceeds in the last line of the file. Line contents are : " + lastChunk + " String length is : " + lastChunk.Length + " To Position is : " + toPosition);
dataToBeWritten.Clear();
byte[] chunkToBeWritten = new byte[chunkSize];
AddLineBreakCharacter(lineBreakCharacter, charactersSeparator, totalLineBreakCharactersLength, dataToBeWritten, ref chunkToBeWritten, ref lastChunk);
WriteBinaryDataToFile(destinationFilePath, dataToBeWritten);
totalLinesProcessed++;
}
}
}
}
private static void WriteBinaryDataToFile(string destinationFilePath, IList<byte[]> chunkToBeWritten)
{
using (FileStream fileSream = new FileStream(destinationFilePath, FileMode.Append))
{
using (BinaryWriter outfile = new BinaryWriter(fileSream, Encoding.ASCII))
{
foreach (byte[] item in chunkToBeWritten)
{
outfile.Write(item);
}
}
}
}
private static void AddLineBreakCharacter(string lineBreakCharacter, char[] charactersSeparator, int totalLineBreakCharactersLength, IList<byte[]> dataToBeWritten, ref byte[] chunkToBeWritten, ref string stringToBeEdited)
{
switch (lineBreakCharacter)
{
case CommonConstants.NEW_LINE:
stringToBeEdited = stringToBeEdited + CommonConstants.CARRIAGE_RETURN_CHARACTER + CommonConstants.NEW_LINE_CHARACTER;
chunkToBeWritten = System.Text.Encoding.UTF8.GetBytes(stringToBeEdited);
//add modified line
dataToBeWritten.Add(chunkToBeWritten);
break;
case CommonConstants.LINE_FEED:
stringToBeEdited = stringToBeEdited + CommonConstants.NEW_LINE_CHARACTER;
chunkToBeWritten = System.Text.Encoding.UTF8.GetBytes(stringToBeEdited);
//add modified line
dataToBeWritten.Add(chunkToBeWritten);
break;
case CommonConstants.CARRIAGE_RETURN:
stringToBeEdited = stringToBeEdited + CommonConstants.CARRIAGE_RETURN_CHARACTER;
chunkToBeWritten = System.Text.Encoding.UTF8.GetBytes(stringToBeEdited);
//add modified line
dataToBeWritten.Add(chunkToBeWritten);
break;
default:
chunkToBeWritten = System.Text.Encoding.UTF8.GetBytes(stringToBeEdited);
//add modified line
dataToBeWritten.Add(chunkToBeWritten);
byte[] seperatorCharArray = new byte[totalLineBreakCharactersLength];
int destinationIndex = 0;
foreach (char ch in charactersSeparator)
{
if (ch == 0)
break;
//convert the character to byte array
byte[] charArr = BitConverter.GetBytes(ch);
//copy character bytes to seperator character array
Array.Copy(charArr, 0, seperatorCharArray, destinationIndex, charArr.Length);
}
//add line break characters array to list
dataToBeWritten.Add(seperatorCharArray);
break;
}
}