Documentation for IBM extended characters

Hi,

I was wondering if there is documentation related to the IBM extended characters (cp437, https://en.wikipedia.org/wiki/Code_page_437) and what (escape sequences?) I need to send to MUSHclient from my server to translate to these characters?

That is, I was wanting to implement cp437 and I am unsure what is the best way to get MUSHclient to recognize the data I send. I assume there is some standard for this since a few MU*s already implement unicode or cp437 or parts of them.


Thanks in advance!

EDIT: it occurred to me it might be enough just to send the characters themselves. Is this the usual method?

Convert to UTF8.

I am using PennMUSH as the MU*base, and allowing for all 255 characters to show yields that a few are not working. Namely:
2,7,9,10,13,27.

How would one go about converting to UTF-8 in the case of PennMUSH? Using safe_chr(<cp437 int>) as my function output, I would assume that the changes to UTF-8 would be to whatever processes the buff character.

Characters less than 32 are control characters. For example 10 is a linefeed and 13 is a carriage return.

I have a feeling that someone recently got CP437 to work using an appropriate font in the output window. I might be wrong, and can't find the post right now.

Here it is:

http://www.gammon.com.au/forum/bbshowpost.php?id=13797

Different font, same general idea.

10 being linefeed makes sense as it goes to a new line after being used. The rest show up blank. I am using terminal right now which has full support for cp437, indeed these are all the characters I can already see:
http://i.imgur.com/tCcNkIJ.png

As you can probably tell, 9 is a tab, 10 is linefeed. The other problem-characters simply do not appear. I think the problem comes before the client (and thus the typeface), probably how PennMUSH interprets the buffer character of the integers in question. I will check M*U*S*H to see if anyone has a solution.

They aren't really problem characters. You shouldn't be trying to "see" control characters. If you do attempt it, the server may discard them.

Sorry, maybe I wasn't clear with my goal. My goal is to see the entirety of cp437. For example, 3 is a heart character. I want to see the heart character, not actually what PennMUSH is currently trying to interpret the third character as (presumably a control character).

If you declare a specific decidedly-non-standard codepage, then everyone who connects must initiate MUSHclient with that codepage in mind and modify mushclient on their end. If you're controlling the server, convert to UTF8 which MUSHclient already supports.

I don't see how your goal will be accomplished. For example, character 10 is a linefeed. You at least need that to get through as a linefeed or you will never have any new lines.

Therefore the inverted "O" symbol from the page you referenced ( ◙ ) will never be able to be shown (thus your goal "My goal is to see the entirety of cp437." will fail).

From the Wikipedia page you mentioned:

MUSHclient itself (in the input processing state machine) either ignores \r (carriage return) or uses it to clear the current line.

Similarly the \a character (bell/alert) is used to play a bell sound.

Tab (\t), ESC (0x1B) and IAC (0xFF) also have special treatment.

Thus those characters won't make it to the output rendering system.

Perhaps I am still being unclear. I am not asking "how can I replace the control sequences with cp437 characters", but instead I am asking "what can I do to show the cp437 characters I am missing so I can hack this into my local functions for PennMUSH".

But now I realize this is not something MUSHclient handles, it is to do with PennMUSH. Thanks for your help in any case.

Well, I suppose you could use libiconv. http://pubs.opengroup.org/onlinepubs/9699919799/functions/iconv.html

Or, you basically* find&replace according to https://en.wikipedia.org/w/index.php?title=Code_page_437&oldid=565442465#Characters

* - slightly more complicated than that, since you're often replacing 1 byte with 2 bytes.

In Windows you could also call MultiByteToWideChar to convert from CP437 to UTF16, and then WideCharToMultiByte to convert that to UTF8.