Re: File I/O - New Line Character - C# Discussion Boards

I'm making changes to a file, saving the edits to a StringBuilder, then writing the mods to a temporary file before deleting the original and then renaming the temporary file.

Now, I want to be able to determine what new line character(s) to use for the file being modified.

At the moment if I look at the changes made to a text file from a UNIX machine with WinMerge, it'll show that every line has been modified because of the AppendLine() method of the StringBuilder, which defaults to I'm guessing "\r\n" where UNIX uses "\n".

"That's fine, I'll just use Append(..) then Append("\n")..." I thought.
But then WinMerge will highlight the same issue for Windows based text files that require "\r\n".

Currently I use a StreamReader.ReadLine() to whip through each line of the file, but that just consumes the EOL characters so I can't determine it that way.

Do I really have to Peek() through the file until I come accross the "\n" or "\r\n",
or is there a smarter way to do it?

Thanks y'all.

Smile | :)

"If you can't explain it simply, you don't understand it well enough" ~ Albert Einstein

"If you think it's expensive to hire a professional to do the job, wait until you hire an amateur." ~ Paul Neal "Red" Adair

Now reading: 'The Third Reich', by Michael Burleigh

Hi,

here are some facts that should help you out:
- Environment.NewLine (read-only) holds the line terminator used on your system; you can't change it;
- StringBuilder.AppendLine relies on Environment.NewLine;
- TextWriter is an output text stream that has a NewLine property, which you can set;
- TextReader is an input text stream that is tolerant to all kinds of line terminators (but doesn't tell you which ones it sees).

So you probably should read your file with TextReader, write the new one with TextWriter, and not just rely on StringBuilder. AFAIK the text streams are buffered, so going through a StringBuilder isn't paying off, just do TextWriter.Write instead of StringBuilder.Append

Smile | :)

Luc Pattyn [Forum Guidelines] [Why QA sucks] [My Articles]

I only read code that is properly formatted, adding PRE tags is the easiest way to obtain that.
All Toronto weekends should be extremely wet until we get it automated in regular forums, not just QA.

Hhmmmmmmm...interesting.

Thanks Luc, I shall look at the TextReader/Writer classes and see what I can find.

The main thing I'm trying to do is just to find out what EOL terminator the file is using,
and then use the same again for the 'modified' file.

Smile | :)

"If you can't explain it simply, you don't understand it well enough" ~ Albert Einstein

"If you think it's expensive to hire a professional to do the job, wait until you hire an amateur." ~ Paul Neal "Red" Adair

Now reading: 'The Third Reich', by Michael Burleigh

OK, been there. First read (part of) the file in binary, looking for \n and \r, and make up your mind. Then start using TextReader and TextWriter.
BTW: what if the input file's NewLine is inconsistent?

Smile | :)

Luc Pattyn [Forum Guidelines] [Why QA sucks] [My Articles]

I only read code that is properly formatted, adding PRE tags is the easiest way to obtain that.
All Toronto weekends should be extremely wet until we get it automated in regular forums, not just QA.

Luc Pattyn wrote:
BTW: what if the input file's NewLine is inconsistent?

Then...I start to cry.

As I would have to go through each line, then look for the appropriate EOL terminators,
then use that EOL for the modified file.

I thought this would be easy!

Where are those tissues...

"If you can't explain it simply, you don't understand it well enough" ~ Albert Einstein

"If you think it's expensive to hire a professional to do the job, wait until you hire an amateur." ~ Paul Neal "Red" Adair

Now reading: 'The Third Reich', by Michael Burleigh

it gets easy once you have all the requirements in front of you.

are you also considering ANSI/UTF8/Unicode, and the byte-order-mask?
TextReader may happily skip over those...
(and what if the original doesn't support the kind of change you want to apply?)

Have a look at StreamWriter, it takes an Encoding...

Smile | :)

Luc Pattyn [Forum Guidelines] [Why QA sucks] [My Articles]

I only read code that is properly formatted, adding PRE tags is the easiest way to obtain that.
All Toronto weekends should be extremely wet until we get it automated in regular forums, not just QA.

Yeah, well, currently I have a StreamReader on the source file, and then a StreamWriter on the 'modified' file.

I've set the StreamWriter to have the same Encoding as the StreamReader.

As far as the 'byte-order-mask' goes, I have no idea what that is.

Smile | :)

"If you can't explain it simply, you don't understand it well enough" ~ Albert Einstein

"If you think it's expensive to hire a professional to do the job, wait until you hire an amateur." ~ Paul Neal "Red" Adair

Now reading: 'The Third Reich', by Michael Burleigh

the BOM is an optional 2- or 3-byte code prefixed to the actual text, and indicating the file is encoded using UTF8 or some kind of Unicode; have a look at some text files with a hex viewer.
You could use Notepad (watch the bottom part of the Save File Dialog) to force a particular encoding, at least since Vista.

Most classes deal with BOM automatically, without telling you. If you're not careful, you might add or remove the BOM without even realizing.

Smile | :)

Luc Pattyn [Forum Guidelines] [Why QA sucks] [My Articles]

I only read code that is properly formatted, adding PRE tags is the easiest way to obtain that.
All Toronto weekends should be extremely wet until we get it automated in regular forums, not just QA.

Woah, I never knew that.

I shall google hence forth and see what I can find.

Thanks for the information Luc.

Thumbs Up | :thumbsup:

"If you can't explain it simply, you don't understand it well enough" ~ Albert Einstein

"If you think it's expensive to hire a professional to do the job, wait until you hire an amateur." ~ Paul Neal "Red" Adair

Now reading: 'The Third Reich', by Michael Burleigh

The XP version of notepad has it too

If you knew what sort of system they were using to create the file you might be able to code around that.

New Line info[^]

Possibly the best explanation I have ever seen for what the heck happened with new lines.

Thanks, yeah I've had a peruse through that link as well - very interesting.

The trouble is the files could come from any system, I've got some files from a UNIX system, but have run the modification functionality on windows file as well.

I, stupidly, thought that it would be a simple thing to do...

Smile | :)

"If you can't explain it simply, you don't understand it well enough" ~ Albert Einstein

"If you think it's expensive to hire a professional to do the job, wait until you hire an amateur." ~ Paul Neal "Red" Adair

Now reading: 'The Third Reich', by Michael Burleigh

1.21 Gigawatts wrote:
The trouble is the files could come from any system, I've got some files from a UNIX system, but have run the modification functionality on windows file as well.

Dear lord I am so sorry.

Good luck.

Ignoring the issue of byte-order marks, you could try this:

1. Open the file with a StreamReader
2. Read the first line
3. Compare StreamReader.BaseStream.Position with line length

This may be able to distinguish between \n and \r\n but of course doesn't really help if you could have \r.

As an (simpler) alternative, is there some way to make WinMerge ignore line-ending changes?

Gideon Engelberth wrote:
As an (simpler) alternative, is there some way to make WinMerge ignore line-ending changes?

Yeah there is, but it's hiding the problem, rather than finding the solution to it.

I don't want to use the wrong EOL terminator for a UNIX file and then that system not being able to read the file correctly.

But as far as the BaseStream.StreamPosition goes, I hadn't thought of that.

Thanks for the info.

Smile | :)

"If you can't explain it simply, you don't understand it well enough" ~ Albert Einstein

"If you think it's expensive to hire a professional to do the job, wait until you hire an amateur." ~ Paul Neal "Red" Adair

Now reading: 'The Third Reich', by Michael Burleigh

generally speaking, if you are writing an install program, how would you ensure that a computer has a required installation prerequisite (i.e. MySql)?

in development, i add a reference to a DLL to get the MySQL libraries. in an installation environment, would the install program look for that DLL (through COM for example), or would I need to include the required DLL with the install?

HI All!

i have a panel on which i am drwaing a image control on the center of panel, now when i zoom the image i am resizing the control's left, top , height and width, now when i get negative left and top values the image gets cropped from top Image gets cropped from the top but the bottom scroll bars appears.. is there any way so that when i set the values of my image control to negative(i.e left and top) still the scroll bars appear for that control!!

Thanks in Advance
Samir

Hello,

i am very new to C# so please be patient.

i am printing orders in my app.
so i have a button which creates the printjob.
the paper will be written in the Printpage-Event.

now i want to have a function that gets th order-object as a parm
and print it on the paper.

the reason is, that an order must be printed on severall programm modules.

private void btnDruck_Click(object sender, EventArgs e)
{
	PrintDocument Druckjob = new PrintDocument();
	PrintDialog Druckermenue = new PrintDialog();
	PrintPreviewDialog Druckpreview = new PrintPreviewDialog();
	Druckjob.DocumentName = "Rechnung";

	Druckjob.PrintPage += new PrintPageEventHandler(Druckjob_PrintPage);
	Druckpreview.Document = Druckjob;
	Druckermenue.Document = Druckjob;
	if (Druckermenue.ShowDialog() == DialogResult.OK)
	{
		Druckjob.Print();
	}
	Druckjob.PrintPage -= new PrintPageEventHandler(Druckjob_PrintPage);
	 
}
void Druckjob_PrintPage(object sender, PrintPageEventArgs e)
{
	Graphics g = e.Graphics;
	StringFormat lineFmtRight = new StringFormat();
	lineFmtRight.Alignment = StringAlignment.Far;
	g.PageUnit = GraphicsUnit.Millimeter;

	// Druckausgabe ...
	// Drucken ..

	g.Dispose();
}

bye jo

Can any body tell me with an example, how loose coupling is implemented in c# in a such a way that if a class is implementing some interface and if we change something in that class then consumer class doesn't gets effected and without using interface how all the classes consumer class is changed if we didnot implemen using interface.

Only if you tell me, with an example, of how high cohesion is implemented in c# so that the module implementing the interface is written so it encapsulates only its own duties an does not effect any of the parameters passed to it.

Smile | :)

Dalek Dave: There are many words that some find offensive, Homosexuality, Alcoholism, Religion, Visual Basic, Manchester United, Butter.

Following article is much help...

Dependency Injection for Loose Coupling[^]

Thanks,
Padmanabh

Can any body tell me with an example, how loose coupling is implemented in c# in a such a way that if a class is implementing some interface and if we change something in that class then consumer class doesn't gets effected and without using interface how all the classes consumer class is changed if we didnot implemen using innterface.

Hi,

I would like to achieve the following. I want to use c# code in an if statement where if the last 2 digits are equal to "xx". How would I approach this.

I appreciate the feedback, thank you in advance.
suprsnipes

Hi,

your question isn't very clear. If the variable is an integer, all it takes is if ((var % 100) == x) .... If the value could be negative, you might need additional code.

Smile | :)

Luc Pattyn [Forum Guidelines] [Why QA sucks] [My Articles]

I only read code that is properly formatted, adding PRE tags is the easiest way to obtain that.
All Toronto weekends should be extremely wet until we get it automated in regular forums, not just QA.

You could try

if(Number.ToString().EndsWith("xx"))
{
//do something
}

Although it wont be as efficient as using a mod operator