/// <summary>
/// The main entry point for the application.
/// </summary>
public static int Main(string[] args)
{
string OutputFormat = " {0,-14} 0x{1,-14:x}";
//IO variables.
string InFilename = null;
FileStream InFileStream = null;
BinaryReader InputReader = null;
string OutFilename = null;
BinaryWriter OutputWriter = null;
//Used to pass line termination chars to FixOrConvert.
byte [] crlfArr = new byte[] { 0x0d, 0x0a };
byte [] lfArr = new byte[] { 0x0a };
//Get file expression from argument.
try
{
DirectoryInfo curdir = new DirectoryInfo(System.Environment.CurrentDirectory);
string FileExpression = null;
//Usage
if (args.Length < 1)
{
Console.WriteLine(nl + "Usage: crlf <filename | expression>" + nl);
Console.WriteLine("Requires filename or wildcard expression." + nl);
}//end if
else
{
FileExpression = args[0];
FileInfo[] files = curdir.GetFiles(FileExpression);
if (files.Length < 1)
{
Console.WriteLine("File does not exist or invalid file expression.");
return(1);
}//end if
//Loop to cover all files in expression list
foreach (FileInfo fi in files)
{
int c1 = 0;
int c2 = 0;
TermListType TermList = new TermListType();
string Reply = null;
bool NeedsFix = false;
bool AllLF = false;
bool AllCRLF = false;
bool CRLF = false;
bool LFOnly = false;
bool CROnlyFix = false;
CRFixType CRFix = CRFixType.Delete;
FileStream OutputFileStream = null;
string Prompt = null;
Console.WriteLine("Processing file: {0}", fi.Name);
//Check again that file exists. Probably redundant but doesn't hurt.
if (!File.Exists(fi.FullName))
{
Console.WriteLine("{0} does not exist, skipping file.");
continue;
}//end if
try
{
InFilename = fi.FullName;
InFileStream = new FileStream(InFilename, FileMode.Open);
InputReader = new BinaryReader(InFileStream, System.Text.Encoding.ASCII);
}//end try
catch (Exception e)
{
bool Exit = OutputException("Problem opening file: " + InFilename, e);
if (Exit)
{
return(1);
}
else //Go to next file in list.
{
continue;
}
}//end catch
InputReader.BaseStream.Seek(0, SeekOrigin.Begin);
InputReader.BaseStream.Position = 0;
if (InputReader.BaseStream.Length > 2)
{
c1 = InputReader.Read();
c2 = InputReader.Read();
} //end if
else
{
Console.WriteLine("File too small");
//Go to next file in list.
continue;
} //end else
if (c1 == 0x0a)
{
TermListItem tli = new TermListItem((long)1, TermTypeEnum.LFOnly);
TermList.Add(tli);
} //end if
long Count = 2;
//Loop through file
while (c2 > -1)
{
if (c1 == 0x0d && c2 != 0x0a)
{
TermListItem CRtli = new TermListItem((long)Count - 1, TermTypeEnum.CROnly);
TermList.Add(CRtli);
}//end if
if (c1 != 0x0d && c2 == 0x0a)
{
TermListItem LFtli = new TermListItem((long)Count, TermTypeEnum.LFOnly);
TermList.Add(LFtli);
}//end if
if (c1 == 0x0d && c2 == 0x0a)
{
TermListItem CRLFtli = new TermListItem((long)Count, TermTypeEnum.CRLF);
TermList.Add(CRLFtli);
}//end if
c1 = c2;
c2 = InputReader.Read();
Count++;
}//end while
Console.WriteLine(nl + "Number of unmatched CRs: {0, 12}", TermList.CROnlyCount);
Console.WriteLine("Number of LF-only chars {0, 12}", TermList.LFOnlyCount);
Console.WriteLine("Number of CRLF sets: {0, 12}" + nl, TermList.CRLFCount);
Console.WriteLine("Size of input file: {0}", InputReader.BaseStream.Length);
if (TermList.LFOnlyCount > 0 && TermList.CRLFCount == 0)
{
Console.WriteLine("File is LF-only terminated" + nl);
Console.WriteLine("Total number of LF terminated lines: {0}", TermList.LFOnlyCount);
AllLF = true;
}//end if
else
{
if (TermList.CRLFCount > 0 && TermList.LFOnlyCount == 0)
{
Console.WriteLine("File is CRLF terminated");
Console.WriteLine("Total number of CRLF terminated lines: {0}", TermList.CRLFCount);
AllCRLF = true;
}
else
{
if (TermList.LFOnlyCount > TermList.CRLFCount && TermList.LFOnlyCount > 0)
{
Console.WriteLine("File is mostly LF-only terminated");
NeedsFix = true;
}//end if
else
{
if (TermList.CRLFCount > TermList.LFOnlyCount && TermList.CRLFCount > 0)
{
Console.WriteLine("File is mostly CRLF terminated" + nl);
NeedsFix = true;
} //end if
} //end else
Console.WriteLine("Number of unexpected CR chars if file is LF-only: {0}" + nl,
TermList.CRLFCount);
if (TermList.CRLFCount > 0)
{
OutputLocations(TermList, TermList.FirstCRLF, OutputFormat, false, TermList.CRLFCount);
}//end if
Console.WriteLine("Number of unexpected LF-only chars if file is CRLF: {0}" + nl,
TermList.LFOnlyCount);
if (TermList.LFOnlyCount > 0)
{
OutputLocations(TermList, TermList.FirstLFOnly, OutputFormat, false, TermList.LFOnlyCount);
}//end if
else
{
Console.WriteLine("Insufficient line termination chars to " +
"determine termination type." + nl);
//Jump to next file in list.
continue;
} //end else
} //end else
} //end else
if (TermList.CROnlyCount > 0)
{
CROnlyFix = true;
Console.WriteLine("Number of unmatched CR-only chars: {0}" + nl, TermList.CROnlyCount);
OutputLocations(TermList, TermList.FirstCROnly, OutputFormat, true, TermList.CROnlyCount);
}//end if
if (NeedsFix || CROnlyFix)
{
Prompt = nl + "Do you wish to fix this file? ";
}//end if
else
{
if (AllLF)
{
Prompt = "Do you wish to convert this file to all CRLF? ";
}//end if
else
{
if (AllCRLF)
{
Prompt = "Do you wish to convert this file to all LF-only? ";
} //end if
} //end else
} //end else
Console.Write(Prompt);
Reply = Console.ReadLine();
if (Reply.ToLower() == "y" || Reply.ToLower() == "yes")
{
try
{
OutFilename = Path.GetFileNameWithoutExtension(InFilename) +
".fixed" +
Path.GetExtension(InFilename);
OutputFileStream = new FileStream(OutFilename, FileMode.Create);
OutputWriter = new BinaryWriter(OutputFileStream, System.Text.Encoding.ASCII);
}//end try
catch (Exception e)
{
bool Exit = OutputException("Problem opening file: " + OutFilename, e);
if (Exit)
{
return(1);
}
else
{
continue;
}
}//end catch
//Reposition input file seek for second read from start.
InputReader.BaseStream.Seek(0, SeekOrigin.Begin);
InputReader.BaseStream.Position = 0;
//If we got here with 'yes' then we will convert to opposite if no fixes required.
if (AllLF)
{
CRLF = true;
}//end if
else
{
if (AllCRLF)
{
LFOnly = true;
} //end if
} //end else
//Since we need fixes, ask which way to convert.
//Ask to convert towards most common one first.
if (NeedsFix)
{
if (TermList.CRLFCount >= TermList.LFOnlyCount) //Mostly CRLF
{
Console.Write("Do you wish to make file CRLF? ");
Reply = Console.ReadLine();
if (Reply.ToLower() == "y" || Reply.ToLower() == "yes")
{
CRLF = true;
}//end if
else
{
Console.WriteLine("Modified file will be LF-only...");
LFOnly = true;
} //end else
} //end if
else //Mostly LF-only
{
Console.Write("Do you wish to make file LF-only? ");
Reply = Console.ReadLine();
if (Reply.ToLower() == "y" || Reply.ToLower() == "yes")
{
LFOnly = true;
}//end if
else
{
Console.WriteLine("Modified file will be CRLF...");
CRLF = true;
} //end else
}
} //end if
if (TermList.CROnlyCount > 0)
{
Console.Write("Do you wish to delete unmatched CR chars? ");
Reply = Console.ReadLine();
if (Reply.ToLower() == "y" || Reply.ToLower() == "yes")
{
CRFix = CRFixType.Delete;
}
else
{
Console.WriteLine("Unmatched CR chars will be replaced with normal line termination.");
CRFix = CRFixType.Replace;
}
}//end if
if (NeedsFix || CROnlyFix)
{
if (CROnlyFix && !NeedsFix) //CROnlyFix case
{
if (CRLF)
{
FixOrConvert(TermList,
InputReader,
OutputWriter,
crlfArr,
TermList.FirstCROnly,
CRFix);
}//end if
else
{
if (LFOnly)
{
FixOrConvert(TermList,
InputReader,
OutputWriter,
lfArr,
TermList.FirstCROnly,
CRFix);
} //end if
} //end else
} //end if
else
{
if (CRLF) //Change LF-onlys to CRLF during fix
{
FixOrConvert(TermList,
InputReader,
OutputWriter,
crlfArr,
TermList.FirstLFOnly,
CRFix);
}//end if
else
{
if (LFOnly) //Change CRLFs to LF-only during fix
{
FixOrConvert(TermList,
InputReader,
OutputWriter,
lfArr,
TermList.FirstCRLF,
CRFix);
} //end if
} //end else
} //end if
} //end else
else //File conversion only
{
if (CRLF) //Convert to CRLF
{
FixOrConvert(TermList, InputReader, OutputWriter, crlfArr, -1, CRFix);
}//end if
else
{
if (LFOnly) //Convert to LF-only
{
FixOrConvert(TermList, InputReader, OutputWriter, lfArr, -1, CRFix);
} //end if
} //end else
} //end else
OutputWriter.BaseStream.Flush();
OutputWriter.Close();
OutputWriter = null;
Console.WriteLine(nl + "Modified file saved to: {0}", OutFilename);
} //end if
else //Reply was 'no'
{
Console.WriteLine("Exiting without creating fixed or modified file.");
}//end else
InputReader.Close();
InputReader = null;
Console.WriteLine();
} //end foreach
} //end else
return(0);
} //end try
catch (Exception e)
{
OutputException("Unexpected exception", e);
return(1);
}//end catch
finally
{
if (InputReader != null)
{
InputReader.Close();
}//end if
if (OutputWriter != null)
{
OutputWriter.Close();
} //end if
} //end finally
} //end Main
/// <summary>
/// This function does the actual file fixing or conversion.
/// It contains the main logic of this application.
/// </summary>
/// <param name="TermList"></param>
/// <param name="InputReader"></param>
/// <param name="OutputWriter"></param>
/// <param name="LineEnd"></param>
/// <param name="StartIndex"></param>
/// <param name="CRFix"></param>
internal static void FixOrConvert(TermListType TermList,
BinaryReader InputReader,
BinaryWriter OutputWriter,
byte[] LineEnd,
int StartIndex,
CRFixType CRFix)
{
try
{
long finalposition = 0;
long chunk = 0;
long position = 0;
int jumpsize = 0;
int jsize = 0;
if (LineEnd.Length == 2) //Replace LF-only with CRLF so jumpsize is only 1.
{
jumpsize = 1;
}
else
{
if (LineEnd.Length == 1) //Replace CRLF with LF-only so jumpsize is 2.
{
jumpsize = 2;
}
}//end else
if (StartIndex == -1) //Convert only.
{
//Presumed to get here only with pure conversion and no fixes.
//Flag to indicate this is StartIndex == -1.
//Entire list will be walked and converted.
foreach (TermListItem tli in TermList.List)
{
position = tli.Position;
chunk = position - finalposition - jumpsize;
OutputWriter.Write(InputReader.ReadBytes((int)chunk));
OutputWriter.Write(LineEnd);
byte[] jump = InputReader.ReadBytes(jumpsize);
finalposition = position;
} //end foreach
} //end if
else
{
//Fix the file: jump through the list starting with CROnly if at beginning.
//The fix is complicated since it fixes both unmatched CRs by and mismatched line termination.
//It works by jumping back and forth through the CROnly locations via the CROnlyNext field.
int savedNext = -1;
bool switched = false;
TermListItem tli = null;
//Start with CROnly if it's first in the list. We'll switch back in the loop.
if (TermList.FirstCROnly > -1 && TermList.FirstCROnly < StartIndex)
{
switched = true;
savedNext = StartIndex;
tli = TermList[TermList.FirstCROnly];
}//end if
else
{
tli = TermList[StartIndex];
savedNext = -1;
switched = false;
}//end else
//************ Main jump loop ***********
//Can come into this loop already switched onto the CROnly track
//by way of code above.
while (true)
{
position = tli.Position;
if (tli.TermType == TermTypeEnum.CROnly)
{
jsize = 1;
}
else
{
jsize = jumpsize;
}
chunk = position - finalposition - jsize;
if (chunk > -1) //protect against case if position < finalposition + jumpsize
{
OutputWriter.Write(InputReader.ReadBytes((int)chunk));
}
//Handle CROnly case to delete char
if (tli.TermType != TermTypeEnum.CROnly)
{
OutputWriter.Write(LineEnd);
}//end if
//Still write if CR fix is to replace, otherwise delete by not writing anything.
else
{
if (CRFix == CRFixType.Replace)
{
OutputWriter.Write(LineEnd);
} //end if
} //end else
byte[] jump = InputReader.ReadBytes(jsize);
//finalposition carries critical info around to the start of the loop
finalposition = position;
//Loop exit point. When there are no more jumps anywhere then exit.
if (tli.Next < 0 && tli.CROnlyNext < 0 && savedNext < 0)
{
break;
}//end if
//If we are on an entry where there there is a CROnly next then switch to that track.
//The CROnly next should always be "nearer" than the normal next due to the way that
//it is updated in the list.
if (tli.CROnlyNext > -1)
{
if (!switched)
{
savedNext = tli.Next;
switched = true;
} //end if
tli = TermList[tli.CROnlyNext];
} //end if
else
{
//We are still on the CROnly track.
if (switched)
{
//Check to see if we need to switch back.
if (tli.Next > savedNext && savedNext > -1 || tli.Next < 0)
{
tli = TermList[savedNext];
switched = false;
savedNext = -1;
} //end if
else //Get another CROnly before switching
{
tli = TermList[tli.Next]; //We're on the CROnly track here.
} //end else
} //end if
else
{
tli = TermList[tli.Next];
} //end else
} //end else
} //end while
} //end else
//Get remainder of file.
OutputWriter.Write(InputReader.ReadBytes((int)(InputReader.BaseStream.Length - finalposition)));
}//end try
catch (Exception e)
{
OutputException("Exception in FixOrConvert", e);
} //end catch
} //end FixOrConvert()
/// <summary>
/// This function does the actual file fixing or conversion.
/// It contains the main logic of this application.
/// </summary>
/// <param name="TermList"></param>
/// <param name="InputReader"></param>
/// <param name="OutputWriter"></param>
/// <param name="LineEnd"></param>
/// <param name="StartIndex"></param>
/// <param name="CRFix"></param>
internal static void FixOrConvert(TermListType TermList,
BinaryReader InputReader,
BinaryWriter OutputWriter,
byte[] LineEnd,
int StartIndex,
CRFixType CRFix)
{
try
{
long finalposition = 0;
long chunk = 0;
long position = 0;
int jumpsize = 0;
int jsize = 0;
if (LineEnd.Length == 2) //Replace LF-only with CRLF so jumpsize is only 1.
jumpsize = 1;
else
{
if (LineEnd.Length == 1) //Replace CRLF with LF-only so jumpsize is 2.
jumpsize = 2;
}//end else
if (StartIndex == -1) //Convert only.
{
//Presumed to get here only with pure conversion and no fixes.
//Flag to indicate this is StartIndex == -1.
//Entire list will be walked and converted.
foreach (TermListItem tli in TermList.List)
{
position = tli.Position;
chunk = position - finalposition - jumpsize;
OutputWriter.Write(InputReader.ReadBytes((int)chunk));
OutputWriter.Write(LineEnd);
byte[] jump = InputReader.ReadBytes(jumpsize);
finalposition = position;
}//end foreach
}//end if
else
{
//Fix the file: jump through the list starting with CROnly if at beginning.
//The fix is complicated since it fixes both unmatched CRs by and mismatched line termination.
//It works by jumping back and forth through the CROnly locations via the CROnlyNext field.
int savedNext = -1;
bool switched = false;
TermListItem tli = null;
//Start with CROnly if it's first in the list. We'll switch back in the loop.
if (TermList.FirstCROnly > -1 && TermList.FirstCROnly < StartIndex )
{
switched = true;
savedNext = StartIndex;
tli = TermList[TermList.FirstCROnly];
}//end if
else
{
tli = TermList[StartIndex];
savedNext = -1;
switched = false;
}//end else
//************ Main jump loop ***********
//Can come into this loop already switched onto the CROnly track
//by way of code above.
while (true)
{
position = tli.Position;
if (tli.TermType == TermTypeEnum.CROnly)
jsize = 1;
else
jsize = jumpsize;
chunk = position - finalposition - jsize;
if (chunk > -1) //protect against case if position < finalposition + jumpsize
{
OutputWriter.Write(InputReader.ReadBytes((int)chunk));
}
//Handle CROnly case to delete char
if (tli.TermType != TermTypeEnum.CROnly)
{
OutputWriter.Write(LineEnd);
}//end if
//Still write if CR fix is to replace, otherwise delete by not writing anything.
else
{
if (CRFix == CRFixType.Replace)
{
OutputWriter.Write(LineEnd);
}//end if
}//end else
byte[] jump = InputReader.ReadBytes(jsize);
//finalposition carries critical info around to the start of the loop
finalposition = position;
//Loop exit point. When there are no more jumps anywhere then exit.
if (tli.Next < 0 && tli.CROnlyNext < 0 && savedNext < 0)
{
break;
}//end if
//If we are on an entry where there there is a CROnly next then switch to that track.
//The CROnly next should always be "nearer" than the normal next due to the way that
//it is updated in the list.
if (tli.CROnlyNext > -1)
{
if (!switched)
{
savedNext = tli.Next;
switched = true;
}//end if
tli = TermList[tli.CROnlyNext];
}//end if
else
{
//We are still on the CROnly track.
if (switched)
{
//Check to see if we need to switch back.
if (tli.Next > savedNext && savedNext > -1 || tli.Next < 0)
{
tli = TermList[savedNext];
switched = false;
savedNext = -1;
}//end if
else //Get another CROnly before switching
{
tli = TermList[tli.Next]; //We're on the CROnly track here.
}//end else
}//end if
else
{
tli = TermList[tli.Next];
}//end else
}//end else
}//end while
}//end else
//Get remainder of file.
OutputWriter.Write(InputReader.ReadBytes((int)(InputReader.BaseStream.Length - finalposition)));
}//end try
catch (Exception e)
{
OutputException("Exception in FixOrConvert", e);
}//end catch
}//end FixOrConvert()