Framework EDI Reference. Encoding Mechanism
Quoted Printable

The Quoted-Printable encoding converts 8-bit binary characters to US-ASCII printable characters.  Mostly used on text that is already human recognizeable, the purpose of this encoding is to preserve the integrity of the text as it passes through multiple transfer agents through the internet.  It does this by preserving the alphabetic, numeric and punctuation characters, and only transforming special characters that can be modified by transfer agents via the internet.

Framework EDI performs the following when encoding text to Quoted-Printable:

• All characters can be replaced with "=<Hex>" where "<Hex>" is the 2-character hexadecimal representation of the character's decimal value.  For example, the binary character having decimal value 215 is represented as "=D7".  In general,
  • All binary characters having decimal value greater than 127 are always converted this way.
  • All control characters other than horizontal tab (HT) and carriage-return line feed (CRLF) are always converted this way.
  • When encoding, the conversion from decimal to hexadecimal always results in the upper case letters (Example. "=D7").
  • When encoding, the following range of characters are not converted,
    • Range 33 (0x21) to 60 (0x3C).
    • Range 62 (0x3E) to 126 (0x7E).
  • When decoding, upper and lower case letters are treated equally and convert to the corresponding decimal value.  For example, "=D7" and "=d7" will decode to 215.  Please note that this is for robust implementation, and lower case is not normally allowed according to standard.
  • When decoding, if an "=" sign is followed by a non-hexadecimal character, a warning is generated and the original characters are returned.  For example,  the encoded string "=XZ" is not legal because "XZ" is not a hexadecimal representation of a character.
    
• Carriage-Return (CR) and Line-Feed (LF) characters that are not part of the CRLF sequence are converted like a binary character.  That is, CR is converted to "=0D", and LF is converted to "= 0A".  TheCRLF sequence is not changed because it indicates the end of a line -- except if the CRLF sequence is part of non-text data. In that case, CRLF should be convertedto"=0D= 0A".  When decoding, any occurrence of single CR and LF characters is illegal and generates a warning when encountered. 
  
• Space (SP) and tab (HT) must be converted like a binary character if it occurs at the end of a line.  That is, SP is converted to "=20" and HT is converted to "=09".  This is done because transfer agents may pad end of line with spaces, or trim the line by removing spaces/tab.

In the character table below, only the colored characters are transformed when encoding to Quoted-Printable.

Decimal  Hexadecimal  Abbr/Glyph  Encode Char  Notes 000 00 NUL =00 001 01 SOH =01 002 02 STX =02 003 03 ETX =03 004 04 EOT =04 005 05 ENQ =05 006 06 ACK =06 007 07 BEL =07 008 08 BS =08 009 09 HTAB =09 Only at the end of line 010 0A LF =0A Text document encode only if not part of CRLF 011 0B VT =0B 012 0C FF =0C 013 0D CR =0D Text document encode only if not part of CRLF 014 0E SO =0E 015 0F SI =0F 016 10 DLE =10 017 11 DC1 =11 018 12 DC2 =12 019 13 DC3 =13 020 14 DC4 =14 021 15 NAK =15 022 16 SYN =16 023 17 ETB =17 024 18 CAN =18 025 19 EM =19 026 1A SUB =1A 027 1B ESC =1B 028 1C FS =1C 029 1D GS =1D 030 1E RS =1E 031 1F US =1F 032 20 SP =20 Only at the end of line 033 21 ! ! 034 22 " " 035 23 # # 036 24 $ $ 037 25 % % 038 26 & & 039 27 ' ' 040 28 ( ( 041 29 ) ) 042 2A * * 043 2B + + 044 2C , , 045 2D - - 046 2E . . 047 2F / / 048 30 0 0 049 31 1 1 050 32 2 2 051 33 3 3 052 34 4 4 053 35 5 5 054 36 6 6 055 37 7 7 056 38 8 8 057 39 9 9 058 3A : : 059 3B ; ; 060 3C < < 061 3D = =3D 062 3E > > 063 3F ? ? 064 40 @ @ 065 41 A A 066 42 B B 067 43 C C 068 44 D D 069 45 E E

Decimal  Hexadecimal  Abbr/Glyph  Encode Char  Notes 070 46 F F 071 47 G G 072 48 H H 073 49 I I 074 4A J J 075 4B K K 076 4C L L 077 4D M M 078 4E N N 079 4F O O 080 50 P P 081 51 Q Q 082 52 R R 083 53 S S 084 54 T T 085 55 U U 086 56 V V 087 57 W W 088 58 X X 089 59 Y Y 090 5A Z Z 091 5B [ [ 092 5C \ \ 093 5D ] ] 094 5E ^ ^ 095 5F _ _ 096 60 ` ` 097 61 a a 098 62 b b 099 63 c c 100 64 d d 101 65 e e 102 66 f f 103 67 g g 104 68 h h 105 69 i i 106 6A j j 107 6B k k 108 6C l l 109 6D m m 110 6E n n 111 6F o o 112 70 p p 113 71 q q 114 72 r r 115 73 s s 116 74 t t 117 75 u u 118 76 v v 119 77 w w 120 78 x x 121 79 y y 122 7A z z 123 7B { { 124 7C | | 125 7D } } 126 7E ~ ~ 127 7F DEL =7F BINARY DATA 128 80 =80 129 81 =81 : : 254 FE =FE 255 FF =FF

For example, the following sentence "e = mc² is cool! " has a trailing space.  Also, assume that it is terminated by a carriage-return line-feed (CRLF) sequence.  Also, assume that the superscript "2" is represented by the decimal character 129.

Text

e

=

m

c

2

i

s

c

o

o

l

!

Quoted-Printable

e

SP

=3D

SP

m

c

=81

SP

i

s

SP

c

o

o

l

!

=20

CR

LF

Byte Stream

0x65

0x20

0x3Dx33x44

0x20

0x6D

0x63

0x3Dx38x31

0x20

0x69

0x73

0x20

0x63

0x6F

0x6F

0x6C

0x21

0x3Dx32x30

0x0D

0x0A

Notice that most of the characters are not transformed making text encoded in Quoted-Printable still human readable.