What is PeekChar() ???

  • Thread starter Thread starter Guest
  • Start date Start date
G

Guest

I was trying to use BinaryReader.PeekChar to "peek" the next byte, that works
fine unless the byte value's above 127 .... is there anything I can do abouit
it?

thnx,
mikolas
 
Don't you have PeekByte? PeekChar will read a unicode char, this can be
more than 1 byte.

Kind regards

Alexander
 
Hi Mikolas,

The default constructor for BinaryReader is set to use UTF8 encoding which
will be ascii for the first 127 byte characters but will mean unicode for
characters above 127 (ie reading two bytes etc).

Use the encoding used in the text. This will create a reader with the
default windows page on your machine.
BinaryReader br = new BinaryReader(stream, Encoding.Default);
 
Morten Wennevik said:
Hi Mikolas,

The default constructor for BinaryReader is set to use UTF8 encoding which
will be ascii for the first 127 byte characters but will mean unicode for
characters above 127 (ie reading two bytes etc).

Use the encoding used in the text. This will create a reader with the
default windows page on your machine.
BinaryReader br = new BinaryReader(stream, Encoding.Default);
Click to expand...

I admit I don't know much about all those different encodings, but .Default
sounds a bit dangerous since I'm writing code that should be running on
different devices. So should I use UnicodeEncoding and then mask the desired
byte?

thanks,
mikolas
 
Morten Wennevik said:
Hi Mikolas,

The default constructor for BinaryReader is set to use UTF8 encoding
which
will be ascii for the first 127 byte characters but will mean unicode
for
characters above 127 (ie reading two bytes etc).

Use the encoding used in the text. This will create a reader with the
default windows page on your machine.
BinaryReader br = new BinaryReader(stream, Encoding.Default);
Click to expand...

I admit I don't know much about all those different encodings, but
.Default
sounds a bit dangerous since I'm writing code that should be running on
different devices. So should I use UnicodeEncoding and then mask the
desired
byte?

thanks,
mikolas
Click to expand...

Hard to say. If the text could be in different encodings you would
probably benefit from using Unicode or UTF8. If you only need the
standard default extended English ascii you could get that specific
encoding with
Encoding e = Encoding.GetEncoding("Windows-1252");
Encoding e = Encoding.GetEncoding("ISO-8859-1");
I believe the code table are the same for these two code pages. There is
also "US ASCII" listed under western code pages.

You can read more about Unicode and UTF8 on this page
http://www.pobox.com/~skeet/csharp/unicode.html
 
Morten Wennevik said:
Hard to say. If the text could be in different encodings you would
probably benefit from using Unicode or UTF8. If you only need the
standard default extended English ascii you could get that specific
encoding with
Encoding e = Encoding.GetEncoding("Windows-1252");
Encoding e = Encoding.GetEncoding("ISO-8859-1");
I believe the code table are the same for these two code pages.
Click to expand...

No they're not - they differ between 128 and 140.
 
No they're not - they differ between 128 and 140.
Click to expand...

I see. Which is considered the standard western codepage? Or is the
standard something that differs between europe and usa?
 
Morten Wennevik said:
I see. Which is considered the standard western codepage? Or is the
standard something that differs between europe and usa?
Click to expand...

Encoding.Default is Windows 1252, I believe.
 
Encoding.Default is Windows 1252, I believe.
Click to expand...

Well, on this system, Norwegian Windows 98, Encoding.Default is
iso-8859-1. I believe Encoding.Default varies with whatever codepage the
OS uses.
 
Morten Wennevik said:
Well, on this system, Norwegian Windows 98, Encoding.Default is
iso-8859-1.
Click to expand...

Interesting. I thought 1252 was the default throughout Western
Europe... Which property did you use to display iso-8859-1 though? The
BodyName property for 1252 returns iso-8859-1, but the others return
Windows-1252.
I believe Encoding.Default varies with whatever codepage the
OS uses.
Click to expand...

It certainly does.
 
Interesting. I thought 1252 was the default throughout Western
Europe... Which property did you use to display iso-8859-1 though? The
BodyName property for 1252 returns iso-8859-1, but the others return
Windows-1252.


It certainly does.
Click to expand...

Interesting, I only checked the BodyName and indeed the other values are
1252/windows 1252, but if BodyName reports the codename for mail agents
one would think the two are identical. Odd.
 
Well, this is is strange. Obviously the encoding thing's pretty complex so do
I really have to deal with all this? All I want is to peek one byte ahead ...
I remember like 10 years ago when I started learning programming things like
that were possible(even here in th Czech rep.) - what is it that happened in
the meantime?

Many Thanks,
mikolas
 
Alexander Muylaert said:
Don't you have PeekByte?
Click to expand...

this would certainly be great and I wouldn't have to spend 1hour debugging
but I haven't seen it anywhere...

mikolas
 
Well, this is is strange. Obviously the encoding thing's pretty complex
so do
I really have to deal with all this? All I want is to peek one byte
ahead ...
I remember like 10 years ago when I started learning programming things
like
that were possible(even here in th Czech rep.) - what is it that
happened in
the meantime?

Many Thanks,
mikolas
Click to expand...

Well, if you aren't going to read the bytes as characters just use the
overloaded
constructor for the BinaryReader.

new BinaryReader(stream, Encoding.Default);

This should force a character to be considered 8 bits in size so PeekChar
should return a single byte.
 
Morten Wennevik said:
Well, if you aren't going to read the bytes as characters just use the
overloaded
constructor for the BinaryReader.

new BinaryReader(stream, Encoding.Default);

This should force a character to be considered 8 bits in size so PeekChar
should return a single byte.
Click to expand...

I'm sorry but I still don't get it. How do you know that? My SW should also
be running on WinCE and as far as I know there's only unicode.
This is just so weird - why there's PeekChar and not PeekByte, one would
expect to use binary reader for binary reading....

thanks,
mikolas
 
Mikolas said:
I'm sorry but I still don't get it. How do you know that? My SW should also
be running on WinCE and as far as I know there's only unicode.
This is just so weird - why there's PeekChar and not PeekByte, one would
expect to use binary reader for binary reading....
Click to expand...

I know, it's very strange that there isn't a PeekByte :(

Rather than trust to Encoding.Default, you could use Encoding.ASCII or
Encoding.GetEncoding(28591), the latter of which is ISO-8859-1.
 
Mikolas,

Hm, I mistakingly though Encoding.Default always would use an 8-bit
encoding, but a simple test proved me wrong. I would use
Encoding.GetEncoding("ISO-8859-1") or any of the encodings you know are
8-bit. You should not however use Encoding.ASCII as that is only 7-bit
and will truncate any values above 127.
 
Morten Wennevik said:
Hm, I mistakingly though Encoding.Default always would use an 8-bit
encoding, but a simple test proved me wrong. I would use
Encoding.GetEncoding("ISO-8859-1") or any of the encodings you know are
8-bit. You should not however use Encoding.ASCII as that is only 7-bit
and will truncate any values above 127.
Click to expand...

I don't see why that causes an issue if the encoding is only being used
for PeekChar, to detect the end of the file. If it's *not* only being
used for PeekChar, then the encoding of the actual text in the file is
what should be used, whether that's ASCII, ISO-8859-1 or whatever.
Detecting the end of the file might then become significantly more
difficult if the desired encoding is multi-byte.
 
I don't see why that causes an issue if the encoding is only being used
for PeekChar, to detect the end of the file. If it's *not* only being
used for PeekChar, then the encoding of the actual text in the file is
what should be used, whether that's ASCII, ISO-8859-1 or whatever.
Detecting the end of the file might then become significantly more
difficult if the desired encoding is multi-byte.
Click to expand...

Well, the original message didn't say anything about only reading the end
of file, nor using it for reading text. PeekChar with Encoding.ASCII will
read 7 bits so a byte value of say 130 would be returned as 63, ie not
what you would want.
 
Morten Wennevik said:
Well, the original message didn't say anything about only reading the end
of file, nor using it for reading text. PeekChar with Encoding.ASCII will
read 7 bits so a byte value of say 130 would be returned as 63, ie not
what you would want.
Click to expand...

Oops - I was getting confused with a different thread asking why there
isn't a PeekByte method...
 
Back
Top