Subject: | |
From: | |
Reply To: | St. John's University Cerebral Palsy List |
Date: | Sat, 10 Apr 1999 01:40:46 +0100 |
Content-Type: | text/plain |
Parts/Attachments: |
|
|
> SGVsbG8gZXZlcnlvbmUsDQoNCk9oIHdlbGwsIHNvcnJ5IGFib3V0IG15IHByZXZpb3VzIG1lc3N5
> IGVtYWlsLiANClNpbmNlIEkgdXNlIGtvcmVhbiB3aW5kb3dzLCBteSBlbWFpbCB3YXMgZW5jb2Rl
> ZCANCmludG8ga29yZWFuIGxhbmd1YWdlLiBOb3cgSSBjaGFuZ2VkIGNoYXJhY3RlciBzZXQuDQpI
> b3BlZnVsbHkgdGhpcyBlbWFpbCB3b3Jrcy4gTmV2ZXJ0aGVsZXNzLCBtYW55IHBlb3BsZSANCnNl
> bnQgbWUgcmVwbGllcyB3aXRoIGdvb2QgaW5mb3JtYXRpb24uIFRoYW5rIGV2ZXJ5b25lIA0Kb2Yg
> dGhlbS4gDQoNCkhhcHB5IGRheXMuLi4uDQoNClRhZS1TdW5nIFNoaW4NCg==
>
>
This decodes to:-
Hello everyone,
Oh well, sorry about my previous messy email.
Since I use korean windows, my email was encoded
into korean language. Now I changed character set.
Hopefully this email works. Nevertheless, many people
sent me replies with good information. Thank everyone
of them.
Happy days....
Tae-Sung Shin
I think I know what's happening:-
Shin's original posting was a 2 part MIME encoded e-mail, both
parts were also encoded as Base64 (more on this later), to
further confuse the issue the character set used was "euc-kr". So
to understand the message your mailer needs to understand:-
MIME format
Base64 decoding
euc-cr character set
To further muddy the waters the Maelstrom listserver has a very
nasty habit of stripping out some of the mail headers from
peoples posts.
The 2nd part of his post was the same information in HTML format
(I hate Outlook Express for doing this!!).
For this second post Shin turned off Mime encoding and reset the
charset to USAscii, the base64 encoding remains, and I suspect
the listserver chewed the header lines which would have specified
this to your mailers.
If Shin can now turn off the Base64 encoding as well, we can all
read the posts.
Base64
======
This encoding system is designed for Internet Mail transfer
(since there are some fundamental flaws in UUEncoding - an older
method).
Take the encoded stuff in groups of 4 characters and turn each
character into a code 0 to 63 thus:
* A-Z map to 0 to 25
* a-z map to 26 to 51
* 0-9 map to 52 to 61
* + maps to 62
* / maps to 63
Express the four numbers thus found (all 0 to 63) in binary:
00aaaaaa 00bbbbbb 00cccccc 00dddddd
This then maps to _three_ real bytes formed thus:
aaaaaabb bbbbcccc ccdddddd
Equals signs (one or two) are used at the end of the encoded
block to indicate that the text was not an integer multiple of
three bytes long.
So the first 4 chars in the message are:-
SGVs map from table above
18 6 21 44 convert to 6-bit binary
010010 000110 010101 101100 regrouping as 8 bit chars yields
01001000 01100101 01101100 convert to Ascii codes
72 101 108 and back to letters again (Hel)lo
Hel
The reason for all this shenanigans is because the original
Internet network only allowed for 7-bit transfers where many
foreign characters require the 8th bit (and of course graphics),
so UUencoding was developed. If you think this is complex wait
until UniCode becomes popular - 65536 characters in each
character set!!
Fascinating stuff huh!!??!! Cheers Tae-Sung Shin.
--
Deri James
10 REPEAT PRINT "Hello world!":UNTIL FALSE
|
|
|