                  RFC #  822n  '      Obsoletes:  RFC #733  (NIC #41952)c                        2                         STANDARD FOR THE FORMAT OF  3                         ARPA INTERNET TEXT MESSAGESi            -                               August 13, 1982n            *                                 Revised by  -                              David H. Crockerd    5                       Dept. of Electrical Engineering=:                  University of Delaware, Newark, DE  197115                       Network:  DCrocker @ UDel-Relay                                -      Standard for ARPA Internet Text Messages*    .                              TABLE OF CONTENTS    F      PREFACE ....................................................   ii  F      1.  INTRODUCTION ...........................................    1  F          1.1.  Scope ............................................    1F          1.2.  Communication Framework ..........................    2  F      2.  NOTATIONAL CONVENTIONS .................................    3  F      3.  LEXICAL ANALYSIS OF MESSAGES ...........................    5  F          3.1.  General Description ..............................    5F          3.2.  Header Field Definitions .........................    9F          3.3.  Lexical Tokens ...................................   10F          3.4.  Clarifications ...................................   11  F      4.  MESSAGE SPECIFICATION ..................................   17  F          4.1.  Syntax ...........................................   17F          4.2.  Forwarding .......................................   19F          4.3.  Trace Fields .....................................   20F          4.4.  Originator Fields ................................   21F          4.5.  Receiver Fields ..................................   23F          4.6.  Reference Fields .................................   23F          4.7.  Other Fields .....................................   24  F      5.  DATE AND TIME SPECIFICATION ............................   26  F          5.1.  Syntax ...........................................   26F          5.2.  Semantics ........................................   26  F      6.  ADDRESS SPECIFICATION ..................................   27  F          6.1.  Syntax ...........................................   27F          6.2.  Semantics ........................................   27F          6.3.  Reserved Address .................................   33  F      7.  BIBLIOGRAPHY ...........................................   34    %                              APPENDIX   F      A.  EXAMPLES ...............................................   36F      B.  SIMPLE FIELD PARSING ...................................   40F      C.  DIFFERENCES FROM RFC #733 ..............................   41F      D.  ALPHABETICAL LISTING OF SYNTAX RULES ...................   44    F      August 13, 1982               - i -                      RFC #822 d       -      Standard for ARPA Internet Text Messagesy    )                                   PREFACE     F           By 1977, the Arpanet employed several informal standards forF      the  text  messages (mail) sent among its host computers.  It wasF      felt necessary to codify these practices and  provide  for  thoseF      features  that  seemed  imminent.   The result of that effort wasF      Request for Comments (RFC) #733, "Standard for the Format of ARPAF      Network Text Message", by Crocker, Vittal, Pogran, and Henderson.F      The specification attempted to avoid major  changes  in  existing5      software, while permitting several new features.;  F           This document revises the specifications  in  RFC  #733,  inF      order  to  serve  the  needs  of the larger and more complex ARPAF      Internet.  Some of RFC #733's features failed  to  gain  adequateF      acceptance.   In  order to simplify the standard and the softwareF      that follows it, these features have been removed.   A  differentF      addressing  scheme  is  used, to handle the case of inter-networkB      mail; and the concept of re-transmission has been introduced.  F           This specification is intended for use in the ARPA Internet.F      However, an attempt has been made to free it of any dependence onF      that environment, so that it can be applied to other network text      message systems.   F           The specification of RFC #733 took place over the course  ofF      one  year, using the ARPANET mail environment, itself, to provideF      an on-going forum for discussing the capabilities to be included.F      More  than  twenty individuals, from across the country, partici-F      pated in  the  original  discussion.   The  development  of  thisF      revised specification has, similarly, utilized network mail-basedF      group discussion.  Both specification efforts  greatly  benefited5      from the comments and ideas of the participants.   F           The syntax of the standard,  in  RFC  #733,  was  originallyF      specified  in  the  Backus-Naur Form (BNF) meta-language.  Ken L.F      Harrenstien, of SRI International, was responsible for  re-codingF      the  BNF  into  an  augmented  BNF  that makes the representation&      smaller and easier to understand.                        F      August 13, 1982              - ii -                      RFC #822 /    i-      Standard for ARPA Internet Text Messagesu          1.  INTRODUCTIONe        1.1.  SCOPE  F           This standard specifies a syntax for text messages that  areF      sent  among  computer  users, within the framework of "electronicF      mail".  The standard supersedes  the  one  specified  in  ARPANETF      Request  for Comments #733, "Standard for the Format of ARPA Net-      work Text Messages".p  F           In this context, messages are viewed as having  an  envelopeF      and  contents.   The  envelope  contains  whatever information isF      needed to accomplish transmission  and  delivery.   The  contentsF      compose  the object to be delivered to the recipient.  This stan-F      dard applies only to the format and some of the semantics of mes-F      sage  contents.   It contains no specification of the information      in the envelope.   F           However, some message systems may use information  from  theF      contents  to create the envelope.  It is intended that this stan-E      dard facilitate the acquisition of such information by programs.   F           Some message systems may  store  messages  in  formats  thatF      differ  from the one specified in this standard.  This specifica-F      tion is intended strictly as a definition of what message content*      format is to be passed BETWEEN hosts.  F      Note:  This standard is NOT intended to dictate the internal for-F             mats  used  by sites, the specific message system featuresF             that they are expected to support, or any of  the  charac-F             teristics  of  user interface programs that create or read             messages.   F           A distinction should be made between what the  specificationF      REQUIRES  and  what  it ALLOWS.  Messages can be made complex andF      rich with formally-structured components of information or can beF      kept small and simple, with a minimum of such information.  Also,F      the standard simplifies the interpretation  of  differing  visualF      formats  in  messages;  only  the  visual  aspect of a message isF      affected and not the interpretation  of  information  within  it.@      Implementors may choose to retain such visual distinctions.  F           The formal definition is divided into four levels.  The bot-F      tom level describes the meta-notation used in this document.  TheF      second level describes basic lexical analyzers that  feed  tokensF      to  higher-level  parsers.   Next is an overall specification forF      messages; it permits distinguishing individual fields.   Finally,F      there is definition of the contents of several structured fields.      F      August 13, 1982               - 1 -                      RFC #822 d     -      Standard for ARPA Internet Text MessagesC    "      1.2.  COMMUNICATION FRAMEWORK  F           Messages consist of lines of text.   No  special  provisionsF      are  made for encoding drawings, facsimile, speech, or structuredF      text.  No significant consideration has been given  to  questionsF      of  data  compression  or to transmission and storage efficiency,F      and the standard tends to be free with the number  of  bits  con-F      sumed.   For  example,  field  names  are specified as free text,%      rather than special terse codes.@  F           A general "memo" framework is used.  That is, a message con-F      sists of some information in a rigid format, followed by the mainF      part of the message, with a format that is not specified in  thisF      document.   The  syntax of several fields of the rigidly-formatedF      ("headers") section is defined in  this  specification;  some  of3      these fields must be included in all messages..  F           The syntax  that  distinguishes  between  header  fields  isF      specified  separately  from  the  internal  syntax for particularF      fields.  This separation is intended to allow simple  parsers  toF      operate on the general structure of messages, without concern forF      the detailed structure of individual header fields.   Appendix  B=      is provided to facilitate construction of these parsers..  F           In addition to the fields specified in this document, it  isF      expected  that  other fields will gain common use.  As necessary,F      the specifications for these "extension-fields" will be publishedF      through  the same mechanism used to publish this document.  UsersF      may also  wish  to  extend  the  set  of  fields  that  they  use:      privately.  Such "user-defined fields" are permitted.  F           The framework severely constrains document tone and  appear-F      ance and is primarily useful for most intra-organization communi-F      cations and  well-structured   inter-organization  communication.F      It  also  can  be used for some types of inter-process communica-F      tion, such as simple file transfer and remote job entry.  A  moreF      robust  framework might allow for multi-font, multi-color, multi-F      dimension encoding of information.  A  less  robust  one,  as  isF      present  in  most  single-machine  message  systems,  would  moreF      severely constrain the ability to add fields and the decision  toF      include specific fields.  In contrast with paper-based communica-F      tion, it is interesting to note that the RECEIVER  of  a  messageF      can   exercise  an  extraordinary  amount  of  control  over  theF      message's appearance.  The amount of actual control available  toF      message  receivers  is  contingent upon the capabilities of their       individual message systems.          F      August 13, 1982               - 2 -                      RFC #822 g    3-      Standard for ARPA Internet Text Messages           2.  NOTATIONAL CONVENTIONSo  F           This specification uses an augmented Backus-Naur Form  (BNF)F      notation.  The differences from standard BNF involve naming rules8      and indicating repetition and "local" alternatives.        2.1.  RULE NAMING  F           Angle brackets ("<", ">") are not  used,  in  general.   TheF      name  of  a rule is simply the name itself, rather than "<name>".F      Quotation-marks enclose literal text (which may be  upper  and/orF      lower  case).   Certain  basic  rules  are  in uppercase, such asF      SPACE, TAB, CRLF, DIGIT, ALPHA, etc.  Angle brackets are used  inF      rule  definitions,  and  in  the rest of this  document, wheneverE      their presence will facilitate discerning the use of rule names.e  '      2.2.  RULE1 / RULE2:  ALTERNATIVESp  F           Elements separated by slash ("/") are alternatives.   There--      fore "foo / bar" will accept foo or bar.y  -      2.3.  (RULE1 RULE2):  LOCAL ALTERNATIVES   F           Elements enclosed in parentheses are  treated  as  a  singleF      element.   Thus,  "(elem  (foo  /  bar)  elem)"  allows the token3      sequences "elem foo elem" and "elem bar elem".         2.4.  *RULE:  REPETITIONd  F           The character "*" preceding an element indicates repetition.      The full form is:  ,                               <l>*<m>element  F      indicating at least <l> and at most <m> occurrences  of  element.F      Default values are 0 and infinity so that "*(element)" allows anyF      number, including zero; "1*element" requires at  least  one;  and$      "1*2element" allows one or two.        2.5.  [RULE]:  OPTIONAL  F           Square brackets enclose optional elements; "[foo  bar]"   is!      equivalent to "*1(foo bar)".v  &      2.6.  NRULE:  SPECIFIC REPETITION  F           "<n>(element)" is equivalent to "<n>*<n>(element)"; that is,F      exactly  <n>  occurrences  of (element). Thus 2DIGIT is a 2-digitC      number, and 3ALPHA is a string of three alphabetic characters.d    F      August 13, 1982               - 3 -                      RFC #822 B    e-      Standard for ARPA Internet Text MessagesI          2.7.  #RULE:  LISTS  A           A construct "#" is defined, similar to "*", as follows:h  ,                               <l>#<m>element  F      indicating at least <l> and at most <m> elements, each  separatedF      by  one  or more commas (","). This makes the usual form of listsF      very easy; a rule such as '(element *("," element))' can be shownF      as  "1#element".   Wherever this construct is used, null elementsF      are allowed, but do not  contribute  to  the  count  of  elementsF      present.   That  is,  "(element),,(element)"  is  permitted,  butF      counts as only two elements.  Therefore, where at least one  ele-F      ment  is required, at least one non-null element must be present.F      Default values are 0 and infinity so that "#(element)" allows anyF      number,  including  zero;  "1#element" requires at least one; and$      "1#2element" allows one or two.        2.8.  ; COMMENTSr  F           A semi-colon, set off some distance to  the  right  of  ruleF      text,  starts  a comment that continues to the end of line.  ThisF      is a simple way of including useful notes in  parallel  with  the      specifications.                                                      F      August 13, 1982               - 4 -                      RFC #822 t    q-      Standard for ARPA Internet Text Messages     %      3.  LEXICAL ANALYSIS OF MESSAGESs        3.1.  GENERAL DESCRIPTION  F           A message consists of header fields and, optionally, a body.F      The  body  is simply a sequence of lines containing ASCII charac-F      ters.  It is separated from the headers by a null line  (i.e.,  a+      line with nothing preceding the CRLF).         3.1.1.  LONG HEADER FIELDSc  F         Each header field can be viewed as a single, logical  line  ofF         ASCII  characters,  comprising  a field-name and a field-body.F         For convenience, the field-body  portion  of  this  conceptualF         entity  can be split into a multiple-line representation; thisF         is called "folding".  The general rule is that wherever  thereF         may  be  linear-white-space  (NOT  simply  LWSP-chars), a CRLFF         immediately followed by AT LEAST one LWSP-char may instead  be(         inserted.  Thus, the single line  8             To:  "Joe & J. Harvey" <ddd @Org>, JJV @ BBN           can be represented as:  /             To:  "Joe & J. Harvey" <ddd @ Org>,                      JJV@BBNh           ands  "             To:  "Joe & J. Harvey"+                             <ddd@ Org>, JJVv              @BBN            andr               To:  "Joe & .              J. Harvey" <ddd @ Org>, JJV @ BBN  F              The process of moving  from  this  folded   multiple-lineF         representation  of a header field to its single line represen-F         tation is called "unfolding".  Unfolding  is  accomplished  byF         regarding   CRLF   immediately  followed  by  a  LWSP-char  as$         equivalent to the LWSP-char.  F         Note:  While the standard  permits  folding  wherever  linear-F                white-space is permitted, it is recommended that struc-F                tured fields, such as those containing addresses, limitF                folding  to higher-level syntactic breaks.  For addressF                fields, it  is  recommended  that  such  folding  occur    F      August 13, 1982               - 5 -                      RFC #822 w    h-      Standard for ARPA Internet Text Messagesa    =                between addresses, after the separating comma.s  '      3.1.2.  STRUCTURE OF HEADER FIELDS   F         Once a field has been unfolded, it may be viewed as being com-F         posed of a field-name followed by a colon (":"), followed by aF         field-body, and  terminated  by  a  carriage-return/line-feed.F         The  field-name must be composed of printable ASCII charactersF         (i.e., characters that  have  values  between  33.  and  126.,F         decimal, except colon).  The field-body may be composed of anyF         ASCII characters, except CR or LF.  (While CR and/or LF may beF         present  in the actual text, they are removed by the action of         unfolding the field.)o  F         Certain field-bodies of headers may be  interpreted  accordingF         to  an  internal  syntax  that some systems may wish to parse.F         These  fields  are  called  "structured   fields".    ExamplesF         include  fields containing dates and addresses.  Other fields,F         such as "Subject"  and  "Comments",  are  regarded  simply  as         strings of text.  F         Note:  Any field which has a field-body  that  is  defined  asF                other  than  simply <text> is to be treated as a struc-                tured field.o  F                Field-names, unstructured field bodies  and  structuredF                field bodies each are scanned by their own, independent#                "lexical" analyzers.   &      3.1.3.  UNSTRUCTURED FIELD BODIES  F         For some fields, such as "Subject" and "Comments",  no  struc-F         turing  is assumed, and they are treated simply as <text>s, asF         in the message body.  Rules of folding apply to these  fields,F         so  that  such  field  bodies  which occupy several lines mustF         therefore have the second and successive lines indented by  at         least one LWSP-char.  $      3.1.4.  STRUCTURED FIELD BODIES  F         To aid in the creation and reading of structured  fields,  theF         free  insertion   of linear-white-space (which permits foldingF         by inclusion of CRLFs)  is  allowed  between  lexical  tokens.F         Rather  than  obscuring  the  syntax  specifications for theseF         structured fields with explicit syntax for this  linear-white-F         space, the existence of another "lexical" analyzer is assumed.F         This analyzer does not apply  for  unstructured  field  bodiesF         that  are  simply  strings  of  text, as described above.  TheF         analyzer provides  an  interpretation  of  the  unfolded  text    F      August 13, 1982               - 6 -                      RFC #822 e    i-      Standard for ARPA Internet Text Messagesp    F         composing  the body of the field as a sequence of lexical sym-
         bols.t           These symbols are:  5                      -  individual special charactersr&                      -  quoted-strings'                      -  domain-literalsy                       -  comments                      -  atomse  F         The first four of these symbols  are  self-delimiting.   AtomsF         are not; they are delimited by the self-delimiting symbols andF         by  linear-white-space.   For  the  purposes  of  regeneratingF         sequences  of  atoms  and quoted-strings, exactly one SPACE isF         assumed to exist, and should be used, between them.  (Also, inF         the "Clarifications" section on "White Space", below, note theA         rules about treatment of multiple contiguous LWSP-chars.)i  <         So, for example, the folded body of an address field  .             ":sysmail"@  Some-Group. Some-Org,=             Muhammed.(I am  the greatest) Ali @(the)Vegas.WBA                                                         F      August 13, 1982               - 7 -                      RFC #822      t-      Standard for ARPA Internet Text Messagesg    A         is analyzed into the following lexical symbols and types:   7                     :sysmail              quoted stringr1                     @                     specialI.                     Some-Group            atom1                     .                     special .                     Some-Org              atom1                     ,                     specialr.                     Muhammed              atom1                     .                     speciala1                     (I am  the greatest)  commenti.                     Ali                   atom.                     @                     atom1                     (the)                 commentr.                     Vegas                 atom1                     .                     special .                     WBA                   atom  F         The canonical representations for the data in these  addresses"         are the following strings:  6                         ":sysmail"@Some-Group.Some-Org           and   2                             Muhammed.Ali@Vegas.WBA  F         Note:  For purposes of display, and when passing  such  struc-F                tured information to other systems, such as mail proto-F                col  services,  there  must  be  NO  linear-white-spaceF                between  <word>s  that are separated by period (".") orF                at-sign ("@") and exactly one SPACE between  all  otherB                <word>s.  Also, headers should be in a folded form.                                    F      August 13, 1982               - 8 -                      RFC #822 l    r-      Standard for ARPA Internet Text Messagesy    #      3.2.  HEADER FIELD DEFINITIONS"  F           These rules show a field meta-syntax, without regard for theF      particular  type  or internal syntax.  Their purpose is to permitF      detection of fields; also, they present to  higher-level  parsers3      an image of each field as fitting on one line.e  6      field       =  field-name ":" [ field-body ] CRLF  @      field-name  =  1*<any CHAR, excluding CTLs, SPACE, and ":">  '      field-body  =  field-body-contentsr/                     [CRLF LWSP-char field-body]9        field-body-contents =E                    <the ASCII characters making up the field-body, asME                     defined in the following sections, and consistingd?                     of combinations of atom, quoted-string, and A                     specials tokens, or else consisting of texts>e                                                                F      August 13, 1982               - 9 -                      RFC #822 *    e-      Standard for ARPA Internet Text Messages           3.3.  LEXICAL TOKENS,  F           The following rules are used to define an underlying lexicalF      analyzer,  which  feeds  tokens to higher level parsers.  See the*      ANSI references, in the Bibliography.  F                                                  ; (  Octal, Decimal.)F      CHAR        =  <any ASCII character>        ; (  0-177,  0.-127.)4      ALPHA       =  <any ASCII alphabetic character>F                                                  ; (101-132, 65.- 90.)F                                                  ; (141-172, 97.-122.)F      DIGIT       =  <any ASCII decimal digit>    ; ( 60- 71, 48.- 57.)F      CTL         =  <any ASCII control           ; (  0- 37,  0.- 31.)F                      character and DEL>          ; (    177,     127.)F      CR          =  <ASCII CR, carriage return>  ; (     15,      13.)F      LF          =  <ASCII LF, linefeed>         ; (     12,      10.)F      SPACE       =  <ASCII SP, space>            ; (     40,      32.)F      HTAB        =  <ASCII HT, horizontal-tab>   ; (     11,       9.)F      <">         =  <ASCII quote mark>           ; (     42,      34.)      CRLF        =  CR LFt  D      LWSP-char   =  SPACE / HTAB                 ; semantics = SPACE  D      linear-white-space =  1*([CRLF] LWSP-char)  ; semantics = SPACEB                                                  ; CRLF => folding  E      specials    =  "(" / ")" / "<" / ">" / "@"  ; Must be in quoted-iB                  /  "," / ";" / ":" / "\" / <">  ;  string, to useB                  /  "." / "[" / "]"              ;  within a word.  ;      delimiters  =  specials / linear-white-space / commentf  F      text        =  <any CHAR, including bare    ; => atoms, specials,@                      CR & bare LF, but NOT       ;  comments andF                      including CRLF>             ;  quoted-strings areC                                                  ;  NOT recognized.O  @      atom        =  1*<any CHAR except specials, SPACE and CTLs>  C      quoted-string = <"> *(qtext/quoted-pair) <">; Regular qtext or B                                                  ;   quoted chars.  C      qtext       =  <any CHAR excepting <">,     ; => may be folded ,                      "\" & CR, and including(                      linear-white-space>  5      domain-literal =  "[" *(dtext / quoted-pair) "]"         F      August 13, 1982              - 10 -                      RFC #822 a    "-      Standard for ARPA Internet Text Messages     C      dtext       =  <any CHAR excluding "[",     ; => may be folded /                      "]", "\" & CR, & including (                      linear-white-space>  <      comment     =  "(" *(ctext / quoted-pair / comment) ")"  C      ctext       =  <any CHAR excluding "(",     ; => may be foldede/                      ")", "\" & CR, & includingo(                      linear-white-space>  E      quoted-pair =  "\" CHAR                     ; may quote any chara  D      phrase      =  1*word                       ; Sequence of words  (      word        =  atom / quoted-string          3.4.  CLARIFICATIONS"        3.4.1.  QUOTING  F         Some characters are reserved for special interpretation,  suchF         as  delimiting lexical tokens.  To permit use of these charac-F         ters as uninterpreted data, a quoting mechanism  is  provided.@         To quote a character, precede it with a backslash ("\").  F         This mechanism is not fully general.  Characters may be quotedF         only  within  a subset of the lexical constructs.  In particu-.         lar, quoting is limited to use within:  -                              -  quoted-stringy.                              -  domain-literal'                              -  commentr  F         Within these constructs, quoting is REQUIRED for  CR  and  "\"F         and for the character(s) that delimit the token (e.g., "(" andF         ")" for a comment).  However, quoting  is  PERMITTED  for  any         character.  F         Note:  In particular, quoting is NOT permitted  within  atoms.F                For  example  when  the local-part of an addr-spec mustF                contain a special character, a quoted  string  must  be9                used.  Therefore, a specification such as:t  -                             Full\ Name@Domainb  5                is not legal and must be specified as:o  .                             "Full Name"@Domain    F      August 13, 1982              - 11 -                      RFC #822 n     -      Standard for ARPA Internet Text Messages           3.4.2.  WHITE SPACE  F         Note:  In structured field bodies, multiple linear space ASCIIF                characters  (namely  HTABs  and  SPACEs) are treated asF                single spaces and may freely surround any  symbol.   InF                all header fields, the only place in which at least oneF                LWSP-char is REQUIRED is at the beginning of  continua-,                tion lines in a folded field.  F         When passing text to processes  that  do  not  interpret  textF         according to this standard (e.g., mail protocol servers), thenF         NO linear-white-space characters should occur between a periodF         (".") or at-sign ("@") and a <word>.  Exactly ONE SPACE shouldF         be used in place of arbitrary linear-white-space  and  comment         sequences.  F         Note:  Within systems conforming to this standard, wherever  aF                member of the list of delimiters is allowed, LWSP-chars5                may also occur before and/or after it.s  F         Writers of  mail-sending  (i.e.,  header-generating)  programsF         should realize that there is no network-wide definition of theF         effect of ASCII HT (horizontal-tab) characters on the  appear-F         ance  of  text  at another network host; therefore, the use ofB         tabs in message headers, though permitted, is discouraged.        3.4.3.  COMMENTSa  F         A comment is a set of ASCII characters, which is  enclosed  inF         matching  parentheses  and which is not within a quoted-stringF         The comment construct permits message originators to add  textF         which  will  be  useful  for  human readers, but which will beF         ignored by the formal semantics.  Comments should be  retainedF         while  the  message  is subject to interpretation according toF         this standard.  However, comments  must  NOT  be  included  inF         other  cases,  such  as  during  protocol  exchanges with mail         servers.  F         Comments nest, so that if an unquoted left parenthesis  occursF         in  a  comment  string,  there  must  also be a matching rightF         parenthesis.  When a comment acts as the delimiter  between  aF         sequence of two lexical symbols, such as two atoms, it is lex-F         ically equivalent with a single SPACE,  for  the  purposes  ofF         regenerating  the  sequence, such as when passing the sequenceF         onto a mail protocol server.  Comments are  detected  as  such6         only within field-bodies of structured fields.  F         If a comment is to be "folded" onto multiple lines,  then  theF         syntax  for  folding  must  be  adhered to.  (See the "Lexical    F      August 13, 1982              - 12 -                      RFC #822 i     -      Standard for ARPA Internet Text Messagesr    F         Analysis of Messages" section on "Folding Long Header  Fields"F         above,  and  the  section on "Case Independence" below.)  NoteF         that  the  official  semantics  therefore  do  not  "see"  anyF         unquoted CRLFs that are in comments, although particular pars-F         ing programs may wish to note their presence.  For these  pro-F         grams,  it would be reasonable to interpret a "CRLF LWSP-char"F         as being a CRLF that is part of the comment; i.e., the CRLF isF         kept  and  the  LWSP-char is discarded.  Quoted CRLFs (i.e., aF         backslash followed by a CR followed by a  LF)  still  must  be+         followed by at least one LWSP-char.   .      3.4.4.  DELIMITING AND QUOTING CHARACTERS  F         The quote character (backslash) and  characters  that  delimitF         syntactic  units  are not, generally, to be taken as data thatF         are part of the delimited or quoted unit(s).   In  particular,F         the   quotation-marks   that   define   a  quoted-string,  theF         parentheses that define  a  comment  and  the  backslash  thatF         quotes  a  following  character  are  NOT  part of the quoted-F         string, comment or quoted character.  A quotation-mark that isF         to  be  part  of  a quoted-string, a parenthesis that is to beF         part of a comment and a backslash that is to be part of eitherF         must  each be preceded by the quote-character backslash ("\").F         Note that the syntax allows any character to be quoted  withinF         a  quoted-string  or  comment; however only certain charactersF         MUST be quoted to be included as data.  These  characters  areF         the  ones that are not part of the alternate text group (i.e.,         ctext or qtext).  F         The one exception to this rule  is  that  a  single  SPACE  isF         assumed  to  exist  between  contiguous words in a phrase, andF         this interpretation is independent of  the  actual  number  ofF         LWSP-chars  that  the  creator  places  between the words.  ToF         include more than one SPACE, the creator must make  the  LWSP-)         chars be part of a quoted-string.   F         Quotation marks that delimit a quoted string  and  backslashesF         that  quote  the  following character should NOT accompany theF         quoted-string when the string is passed to processes  that  doF         not interpret data according to this specification (e.g., mail         protocol servers).        3.4.5.  QUOTED-STRINGSa  F         Where permitted (i.e., in words in structured fields)  quoted-F         strings  are  treated  as a single symbol.  That is, a quoted-F         string is equivalent to an atom, syntactically.  If a  quoted-F         string  is to be "folded" onto multiple lines, then the syntaxF         for folding must be adhered to.  (See the "Lexical Analysis of    F      August 13, 1982              - 13 -                      RFC #822 =     -      Standard for ARPA Internet Text Messagesi    F         Messages"  section  on "Folding Long Header Fields" above, andF         the section on "Case  Independence"  below.)   Therefore,  theF         official  semantics  do  not  "see" any bare CRLFs that are inF         quoted-strings; however particular parsing programs  may  wishF         to  note  their presence.  For such programs, it would be rea-F         sonable to interpret a "CRLF LWSP-char" as being a CRLF  whichF         is  part  of the quoted-string; i.e., the CRLF is kept and theF         LWSP-char is discarded.  Quoted CRLFs (i.e., a backslash  fol-F         lowed  by  a CR followed by a LF) are also subject to rules ofF         folding, but the presence of the quoting character (backslash)F         explicitly  indicates  that  the  CRLF  is  data to the quotedF         string.  Stripping off the first following LWSP-char  is  also.         appropriate when parsing quoted CRLFs.  "      3.4.6.  BRACKETING CHARACTERS  F         There is one type of bracket which must occur in matched pairs4         and may have pairs nested within each other:  F             o   Parentheses ("(" and ")") are used  to  indicate  com-                 ments.  F         There are three types of brackets which must occur in  matched+         pairs, and which may NOT be nested:c  F             o   Colon/semi-colon (":" and ";") are   used  in  addressF                 specifications  to  indicate that the included list of7                 addresses are to be treated as a group.r  F             o   Angle brackets ("<" and ">")  are  generally  used  toF                 indicate  the  presence of a one machine-usable refer-F                 ence (e.g., delimiting mailboxes), possibly  including.                 source-routing to the machine.  F             o   Square brackets ("[" and "]") are used to indicate theF                 presence  of  a  domain-literal, which the appropriateF                 name-domain  is  to  use  directly,  bypassing  normal+                 name-resolution mechanisms.e        3.4.7.  CASE INDEPENDENCE  F         Except as noted, alphabetic strings may be represented in  anyF         combination of upper and lower case.  The only syntactic units                F      August 13, 1982              - 14 -                      RFC #822 R    s-      Standard for ARPA Internet Text Messagesg    <         which requires preservation of case information are:                       -  text                      -  qtext                     -  dtext                     -  ctext"                     -  quoted-pair6                     -  local-part, except "Postmaster"  F         When matching any other syntactic unit, case is to be ignored.F         For  example, the field-names "From", "FROM", "from", and evenF         "FroM" are semantically equal and should all be treated ident-         ically.:  F         When generating these units, any mix of upper and  lower  caseF         alphabetic  characters  may  be  used.  The case shown in thisB         specification is suggested for message-creating processes.  F         Note:  The reserved local-part address unit, "Postmaster",  isF                an  exception.   When  the  value "Postmaster" is beingF                interpreted, it must be  accepted  in  any  mixture  of>                case, including "POSTMASTER", and "postmaster".  '      3.4.8.  FOLDING LONG HEADER FIELDSd  F         Each header field may be represented on exactly one line  con-F         sisting  of the name of the field and its body, and terminatedF         by a CRLF; this is what the parser sees.  For readability, theF         field-body  portion of long header fields may be "folded" ontoF         multiple lines of the actual field.  "Long" is commonly inter-F         preted  to  mean greater than 65 or 72 characters.  The formerF         length serves as a limit, when the message is to be viewed  onF         most  simple terminals which use simple display software; how-8         ever, the limit is not imposed by this standard.  F         Note:  Some display software often can selectively fold lines,F                to  suit  the display terminal.  In such cases, sender-F                provided  folding  can  interfere  with   the   display                software.  !      3.4.9.  BACKSPACE CHARACTERSt  F         ASCII BS characters (Backspace, decimal 8) may be included  inF         texts and quoted-strings to effect overstriking.  However, anyF         use of backspaces which effects an overstrike to the  left  ofA         the beginning of the text or quoted-string is prohibited.e          F      August 13, 1982              - 15 -                      RFC #822 f    g-      Standard for ARPA Internet Text Messagesm    .      3.4.10.  NETWORK-SPECIFIC TRANSFORMATIONS  F         During transmission through heterogeneous networks, it may  beF         necessary  to  force data to conform to a network's local con-F         ventions.  For example, it may be required that a CR  be  fol-F         lowed  either by LF, making a CRLF, or by <null>, if the CR isF         to stand alone).  Such transformations are reversed, when  the#         message exits that network.,  F         When  crossing  network  boundaries,  the  message  should  beF         treated  as  passing  through  two modules.  It will enter theF         first module containing whatever network-specific  transforma-F         tions  that  were  necessary  to  permit migration through the?         "current" network.  It then passes through the modules:e  '             o   Transformation Reversala  F                 The "current" network's idiosyncracies are removed andF                 the  message  is returned to the canonical form speci-&                 fied in this standard.               o   Transformation  F                 The "next" network's local idiosyncracies are  imposed                 on the message.   2                                 ------------------2                     From   ==>  | Remove Net-A   |2                     Net-A       | idiosyncracies |2                                 ------------------)                                        || )                                        \/h-                                   Conformance /                                   with standarde)                                        ||i)                                        \/ 2                                 ------------------;                                 | Impose Net-B   |  ==>  Too>                                 | idiosyncracies |       Net-B2                                 ------------------                      F      August 13, 1982              - 16 -                      RFC #822 g    e-      Standard for ARPA Internet Text Messagesn          4.  MESSAGE SPECIFICATION        4.1.  SYNTAXf  F      Note:  Due to an artifact of the notational conventions, the syn-F             tax  indicates that, when present, some fields, must be inF             a particular order.  Header fields  are  NOT  required  toF             occur  in  any  particular  order, except that the messageF             body must occur AFTER  the  headers.   It  is  recommendedF             that,  if  present,  headers be sent in the order "Return-F             Path", "Received", "Date",  "From",  "Subject",  "Sender",             "To", "cc", etc.  F             This specification permits multiple  occurrences  of  mostF             fields.   Except  as  noted,  their  interpretation is not9             specified here, and their use is discouraged.o  F           The following syntax for the bodies of various fields shouldF      be  thought  of  as  describing  each field body as a single longF      string (or line).  The "Lexical Analysis of Message"  section  onF      "Long  Header Fields", above, indicates how such long strings canF      be represented on more than one line in  the  actual  transmitted
      message.s  C      message     =  fields *( CRLF *text )       ; Everything afterkC                                                  ;  first null line C                                                  ;  is message bodyl  A      fields      =    dates                      ; Creation time, C                       source                     ;  author id & oneeD                     1*destination                ;  address requiredC                      *optional-field             ;  others optional.  A      source      = [  trace ]                    ; net traversalsE@                       originator                 ; original mail<                    [  resent ]                   ; forwarded  A      trace       =    return                     ; path to sender ?                     1*received                   ; receipt tagsc  A      return      =  "Return-path" ":" route-addr ; return addressg  @      received    =  "Received"    ":"            ; one per relay?                        ["from" domain]           ; sending hostTA                        ["by"   domain]           ; receiving hostp@                        ["via"  atom]             ; physical pathE                       *("with" atom)             ; link/mail protocolOB                        ["id"   msg-id]           ; receiver msg id?                        ["for"  addr-spec]        ; initial formm    F      August 13, 1982              - 17 -                      RFC #822 y     -      Standard for ARPA Internet Text Messagesn    @                         ";"    date-time         ; time received  E      originator  =   authentic                   ; authenticated addr 2                    [ "Reply-To"   ":" 1#address] )  @      authentic   =   "From"       ":"   mailbox  ; Single authorC                  / ( "Sender"     ":"   mailbox  ; Actual submittortC                      "From"       ":" 1#mailbox) ; Multiple authorsiA                                                  ;  or not senderu  %      resent      =   resent-authenticp8                    [ "Resent-Reply-To"  ":" 1#address] )        resent-authentic = 5                  =   "Resent-From"      ":"   mailboxt5                  / ( "Resent-Sender"    ":"   mailboxa8                      "Resent-From"      ":" 1#mailbox  )  ;      dates       =   orig-date                   ; Originale<                    [ resent-date ]               ; Forwarded  1      orig-date   =  "Date"        ":"   date-time3  1      resent-date =  "Resent-Date" ":"   date-time8  :      destination =  "To"          ":" 1#address  ; Primary/                  /  "Resent-To"   ":" 1#addressn<                  /  "cc"          ":" 1#address  ; Secondary/                  /  "Resent-cc"   ":" 1#address ?                  /  "bcc"         ":"  #address  ; Blind carbon /                  /  "Resent-bcc"  ":"  #address"        optional-field = 4                  /  "Message-ID"        ":"   msg-id4                  /  "Resent-Message-ID" ":"   msg-id?                  /  "In-Reply-To"       ":"  *(phrase / msg-id)l?                  /  "References"        ":"  *(phrase / msg-id)h4                  /  "Keywords"          ":"  #phrase2                  /  "Subject"           ":"  *text2                  /  "Comments"          ":"  *text3                  /  "Encrypted"         ":" 1#2wordr@                  /  extension-field              ; To be definedD                  /  user-defined-field           ; May be pre-empted  D      msg-id      =  "<" addr-spec ">"            ; Unique message id            F      August 13, 1982              - 18 -                      RFC #822 D     -      Standard for ARPA Internet Text Messages           extension-field =<                    <Any field which is defined in a document;                     published as a formal extension to thiseA                     specification; none will have names beginninga)                     with the string "X-">o        user-defined-field =s8                    <Any field which has not been defined<                     in this specification or published as an>                     extension to this specification; names for9                     such fields must be unique and may bev7                     pre-empted by published extensions>         4.2.  FORWARDINGa  F           Some systems permit mail recipients to  forward  a  message,F      retaining  the original headers, by adding some new fields.  ThisF      standard supports such a service, through the "Resent-" prefix to      field names.A  F           Whenever the string "Resent-" begins a field name, the fieldF      has  the  same  semantics as a field whose name does not have theF      prefix.  However, the message is assumed to have  been  forwardedF      by  an original recipient who attached the "Resent-" field.  ThisF      new field is treated as being more recent  than  the  equivalent,F      original  field.   For  example, the "Resent-From", indicates theF      person that forwarded the message, whereas the "From" field indi-      cates the original author.e  F           Use of such precedence  information  depends  upon  partici-F      pants'  communication needs.  For example, this standard does notF      dictate when a "Resent-From:" address should receive replies,  in1      lieu of sending them to the "From:" address.a  F      Note:  In general, the "Resent-" fields should be treated as con-F             taining  a  set  of information that is independent of theF             set of original fields.  Information for  one  set  shouldF             not  automatically be taken from the other.  The interpre-F             tation of multiple "Resent-" fields, of the same type,  is             undefined.  F           In the remainder of this specification, occurrence of  legalF      "Resent-"  fields  are treated identically with the occurrence of                F      August 13, 1982              - 19 -                      RFC #822       -      Standard for ARPA Internet Text Messages     3      fields whose names do not contain this prefix.l        4.3.  TRACE FIELDSp  F           Trace information is used to provide an audit trail of  mes-F      sage  handling.   In  addition,  it indicates a route back to the      sender of the message.   F           The list of known "via" and  "with"  values  are  registeredF      with  the  Network  Information  Center, SRI International, Menlo      Park, California.        4.3.1.  RETURN-PATH  F         This field  is  added  by  the  final  transport  system  thatF         delivers  the message to its recipient.  The field is intendedF         to contain definitive information about the address and  route)         back to the message's originator.o  F         Note:  The "Reply-To" field is added  by  the  originator  andF                serves  to  direct  replies,  whereas the "Return-Path"F                field is used to identify a path back to  the  origina-                tor.   F         While the syntax  indicates  that  a  route  specification  isF         optional,  every attempt should be made to provide that infor-         mation in this field.         4.3.2.  RECEIVEDs  F         A copy of this field is added by each transport  service  thatF         relays the message.  The information in the field can be quite.         useful for tracing transport problems.  F         The names of the sending  and  receiving  hosts  and  time-of-F         receipt may be specified.  The "via" parameter may be used, toF         indicate what physical mechanism the message  was  sent  over,F         such  as  Arpanet or Phonenet, and the "with" parameter may beF         used to indicate the mail-,  or  connection-,  level  protocolF         that  was  used, such as the SMTP mail protocol, or X.25 tran-         sport protocol.y  F         Note:  Several "with" parameters may  be  included,  to  fully;                specify the set of protocols that were used.i  F         Some transport services queue mail; the internal message iden-F         tifier that is assigned to the message may be noted, using theF         "id" parameter.  When the  sending  host  uses  a  destinationF         address specification that the receiving host reinterprets, by    F      August 13, 1982              - 20 -                      RFC #822       -      Standard for ARPA Internet Text Messagesl    F         expansion or transformation, the receiving host  may  wish  toF         record  the original specification, using the "for" parameter.F         For example, when a copy of mail is sent to the  member  of  aF         distribution  list,  this  parameter may be used to record the;         original address that was used to specify the list.         4.4.  ORIGINATOR FIELDS  F           The standard allows only a subset of the combinations possi-F      ble  with the From, Sender, Reply-To, Resent-From, Resent-Sender,@      and Resent-Reply-To fields.  The limitation is intentional.        4.4.1.  FROM / RESENT-FROM   F         This field contains the identity of the person(s)  who  wishedF         this  message to be sent.  The message-creation process shouldF         default this field  to  be  a  single,  authenticated  machineF         address,  indicating  the  AGENT  (person,  system or process)F         entering the message.  If this is not done, the "Sender" fieldF         MUST  be  present.  If the "From" field IS defaulted this way,F         the "Sender" field is  optional  and  is  redundant  with  theF         "From"  field.   In  all  cases, addresses in the "From" fieldF         must be machine-usable (addr-specs) and may not contain  named         lists (groups).   #      4.4.2.  SENDER / RESENT-SENDERt  F         This field contains the authenticated identity  of  the  AGENTF         (person,  system  or  process)  that sends the message.  It isF         intended for use when the sender is not the author of the mes-F         sage,  or  to  indicate  who among a group of authors actuallyF         sent the message.  If the contents of the "Sender" field wouldF         be  completely  redundant  with  the  "From"  field,  then theF         "Sender" field need not be present and its use is  discouragedF         (though  still legal).  In particular, the "Sender" field MUST=         be present if it is NOT the same as the "From" Field.:  F         The Sender mailbox  specification  includes  a  word  sequenceF         which  must correspond to a specific agent (i.e., a human userF         or a computer program) rather than a standard  address.   ThisF         indicates  the  expectation  that  the field will identify theF         single AGENT (person,  system,  or  process)  responsible  forF         sending  the mail and not simply include the name of a mailboxF         from which the mail was sent.  For example in the  case  of  aF         shared login name, the name, by itself, would not be adequate.F         The local-part address unit, which refers to  this  agent,  isF         expected to be a computer system term, and not (for example) aF         generalized person reference which can  be  used  outside  the%         network text message context.     F      August 13, 1982              - 21 -                      RFC #822       -      Standard for ARPA Internet Text Messages     F         Since the critical function served by the  "Sender"  field  isF         identification  of  the agent responsible for sending mail andF         since computer programs cannot be held accountable  for  theirF         behavior, it is strongly recommended that when a computer pro-F         gram generates a message, the HUMAN  who  is  responsible  forF         that program be referenced as part of the "Sender" field mail-         box specification.  '      4.4.3.  REPLY-TO / RESENT-REPLY-TO   F         This field provides a general  mechanism  for  indicating  anyF         mailbox(es)  to which responses are to be sent.  Three typicalF         uses for this feature can  be  distinguished.   In  the  firstF         case,  the  author(s) may not have regular machine-based mail-F         boxes and therefore wish(es) to indicate an alternate  machineF         address.   In  the  second case, an author may wish additionalF         persons to be made aware of, or responsible for,  replies.   AF         somewhat  different  use  may be of some help to "text messageF         teleconferencing" groups equipped with automatic  distributionF         services:   include the address of that service in the "Reply-F         To" field of all messages  submitted  to  the  teleconference;F         then  participants  can  "reply"  to conference submissions toF         guarantee the correct distribution of any submission of  their         own.  F         Note:  The "Return-Path" field is added by the mail  transportF                service,  at the time of final deliver.  It is intendedF                to identify a path back to the orginator  of  the  mes-F                sage.   The  "Reply-To"  field  is added by the message<                originator and is intended to direct replies.  6      4.4.4.  AUTOMATIC USE OF FROM / SENDER / REPLY-TO  F         For systems which automatically  generate  address  lists  forD         replies to messages, the following recommendations are made:  F             o   The "Sender" field mailbox should be sent  notices  ofF                 any  problems in transport or delivery of the originalF                 messages.  If there is no  "Sender"  field,  then  the4                 "From" field mailbox should be used.  F             o   The  "Sender"  field  mailbox  should  NEVER  be  used>                 automatically, in a recipient's reply message.  F             o   If the "Reply-To" field exists, then the reply  shouldF                 go to the addresses indicated in that field and not to>                 the address(es) indicated in the "From" field.        F      August 13, 1982              - 22 -                      RFC #822 o     -      Standard for ARPA Internet Text Messagese    F             o   If there is a "From" field, but no  "Reply-To"  field,F                 the  reply should be sent to the address(es) indicated$                 in the "From" field.  F         Sometimes, a recipient may actually wish to  communicate  withF         the  person  that  initiated  the  message  transfer.  In such<         cases, it is reasonable to use the "Sender" address.  F         This recommendation is intended  only  for  automated  use  ofF         originator-fields  and is not intended to suggest that repliesF         may not also be sent to other recipients of messages.   It  isF         up  to  the  respective  mail-handling programs to decide what/         additional facilities will be provided.   ,         Examples are provided in Appendix A.        4.5.  RECEIVER FIELDS        4.5.1.  TO / RESENT-TO-  F         This field contains the identity of the primary recipients  of         the message.        4.5.2.  CC / RESENT-CCR  F         This field contains the identity of  the  secondary  (informa-*         tional) recipients of the message.        4.5.3.  BCC / RESENT-BCCI  F         This field contains the identity of additional  recipients  ofF         the  message.   The contents of this field are not included inF         copies of the message sent to the primary and secondary  reci-F         pients.   Some  systems  may choose to include the text of theF         "Bcc" field only in the author(s)'s  copy,  while  others  mayF         also include it in the text sent to all those indicated in the         "Bcc" list.d        4.6.  REFERENCE FIELDSa  +      4.6.1.  MESSAGE-ID / RESENT-MESSAGE-IDn  F              This field contains a unique identifier  (the  local-partF         address  unit)  which  refers to THIS version of THIS message.F         The uniqueness of the message identifier is guaranteed by  theF         host  which  generates  it.  This identifier is intended to beF         machine readable and not necessarily meaningful to humans.   AF         message  identifier pertains to exactly one instantiation of aF         particular message; subsequent revisions to the message should    F      August 13, 1982              - 23 -                      RFC #822 f    v-      Standard for ARPA Internet Text Messages     -         each receive new message identifiers.o        4.6.2.  IN-REPLY-TO  F              The contents of this field identify  previous  correspon-F         dence  which this message answers.  Note that if message iden-F         tifiers are used in this  field,  they  must  use  the  msg-id         specification format.o        4.6.3.  REFERENCES   F              The contents of this field identify other  correspondenceF         which  this message references.  Note that if message identif-E         iers are used, they must use the msg-id specification format.e        4.6.4.  KEYWORDSn  F              This field contains keywords  or  phrases,  separated  by         commas.a        4.7.  OTHER FIELDSc        4.7.1.  SUBJECT  F              This is intended to provide a summary,  or  indicate  the         nature, of the message.l        4.7.2.  COMMENTSa  F              Permits adding text comments  onto  the  message  without6         disturbing the contents of the message's body.        4.7.3.  ENCRYPTED  F              Sometimes,  data  encryption  is  used  to  increase  theF         privacy  of  message  contents.   If the body of a message hasF         been encrypted, to keep its contents private, the  "Encrypted"F         field  can be used to note the fact and to indicate the natureF         of the encryption.  The first <word> parameter  indicates  theF         software  used  to  encrypt the body, and the second, optionalF         <word> is intended to  aid  the  recipient  in  selecting  theF         proper  decryption  key.   This  code word may be viewed as an7         index to a table of keys held by the recipient.A  F         Note:  Unfortunately, headers must contain envelope,  as  wellF                as  contents,  information.  Consequently, it is neces-F                sary that they remain unencrypted, so that  mail  tran-F                sport   services   may   access   them.   Since  names,F                addresses, and "Subject"  field  contents  may  contain    F      August 13, 1982              - 24 -                      RFC #822 w    y-      Standard for ARPA Internet Text Messages     F                sensitive  information,  this  requirement limits total                message privacy.n  F              Names of encryption software are registered with the Net-F         work  Information Center, SRI International, Menlo Park, Cali-         fornia.T        4.7.4.  EXTENSION-FIELD  F              A limited number of common fields have  been  defined  inF         this  document.   As  network mail requirements dictate, addi-F         tional fields may be standardized.   To  provide  user-definedF         fields  with  a  measure  of  safety,  in name selection, suchF         extension-fields will never have names  that  begin  with  the         string "X-".  F              Names of Extension-fields are registered with the NetworkF         Information Center, SRI International, Menlo Park, California.        4.7.5.  USER-DEFINED-FIELDS  F              Individual users of network mail are free to  define  andF         use  additional  header  fields.   Such fields must have namesF         which are not already used in the current specification or  inF         any definitions of extension-fields, and the overall syntax ofF         these user-defined-fields must conform to this specification'sF         rules   for   delimiting  and  folding  fields.   Due  to  theF         extension-field  publishing  process,  the  name  of  a  user-'         defined-field may be pre-emptedn  F         Note:  The prefatory string "X-" will never  be  used  in  theF                names  of Extension-fields.  This provides user-defined4                fields with a protected set of names.                                    F      August 13, 1982              - 25 -                      RFC #822 m    e-      Standard for ARPA Internet Text Messagese    $      5.  DATE AND TIME SPECIFICATION        5.1.  SYNTAXe  ;      date-time   =  [ day "," ] date time        ; dd mm yyo@                                                  ;  hh:mm:ss zzz  4      day         =  "Mon"  / "Tue" /  "Wed"  / "Thu"+                  /  "Fri"  / "Sat" /  "Sun"   A      date        =  1*2DIGIT month 2DIGIT        ; day month year B                                                  ;  e.g. 20 Jun 82  6      month       =  "Jan"  /  "Feb" /  "Mar"  /  "Apr"6                  /  "May"  /  "Jun" /  "Jul"  /  "Aug"6                  /  "Sep"  /  "Oct" /  "Nov"  /  "Dec"  D      time        =  hour zone                    ; ANSI and Military  2      hour        =  2DIGIT ":" 2DIGIT [":" 2DIGIT]F                                                  ; 00:00:00 - 23:59:59  A      zone        =  "UT"  / "GMT"                ; Universal TimeeF                                                  ; North American : UTF                  /  "EST" / "EDT"                ;  Eastern:  - 5/ - 4F                  /  "CST" / "CDT"                ;  Central:  - 6/ - 5F                  /  "MST" / "MDT"                ;  Mountain: - 7/ - 6F                  /  "PST" / "PDT"                ;  Pacific:  - 8/ - 7D                  /  1ALPHA                       ; Military: Z = UT;F                                                  ;  A:-1; (J not used)F                                                  ;  M:-12; N:+1; Y:+12E                  / ( ("+" / "-") 4DIGIT )        ; Local differential E                                                  ;  hours+min. (HHMM).        5.2.  SEMANTICS  F           If included, day-of-week must be the day implied by the date      specification.e  F           Time zone may be indicated in several ways.  "UT" is Univer-F      sal  Time  (formerly called "Greenwich Mean Time"); "GMT" is per-F      mitted as a reference to Universal Time.  The  military  standardF      uses  a  single  character for each zone.  "Z" is Universal Time.F      "A" indicates one hour earlier, and "M" indicates 12  hours  ear-F      lier;  "N"  is  one  hour  later, and "Y" is 12 hours later.  TheF      letter "J" is not used.  The other remaining two forms are  takenF      from ANSI standard X3.51-1975.  One allows explicit indication ofF      the amount of offset from UT; the other uses  common  3-character8      strings for indicating time zones in North America.    F      August 13, 1982              - 26 -                      RFC #822 l     -      Standard for ARPA Internet Text Messages           6.  ADDRESS SPECIFICATION        6.1.  SYNTAXo  @      address     =  mailbox                      ; one addressee=                  /  group                        ; named listE  -      group       =  phrase ":" [#mailbox] ";"   A      mailbox     =  addr-spec                    ; simple address C                  /  phrase route-addr            ; name & addr-specd  -      route-addr  =  "<" [route] addr-spec ">"m  @      route       =  1#("@" domain) ":"           ; path-relative  A      addr-spec   =  local-part "@" domain        ; global addressl  @      local-part  =  word *("." word)             ; uninterpretedA                                                  ; case-preservedm  0      domain      =  sub-domain *("." sub-domain)  /      sub-domain  =  domain-ref / domain-literal   E      domain-ref  =  atom                         ; symbolic reference         6.2.  SEMANTICS  F           A mailbox receives mail.  It is a  conceptual  entity  whichF      does  not necessarily pertain to file storage.  For example, someF      sites may choose to print mail on their line printer and  deliver(      the output to the addressee's desk.  F           A mailbox specification comprises a person, system  or  pro-F      cess name reference, a domain-dependent string, and a name-domainF      reference.  The name reference is optional and is usually used toF      indicate  the  human name of a recipient.  The name-domain refer-F      ence specifies a sequence of sub-domains.   The  domain-dependentF      string is uninterpreted, except by the final sub-domain; the restA      of the mail service merely transmits it as a literal string.r        6.2.1.  DOMAINS  F         A name-domain is a set of registered (mail)  names.   A  name-F         domain  specification  resolves  to  a subordinate name-domainF         specification  or  to  a  terminal  domain-dependent   string.F         Hence,  domain  specification  is  extensible,  permitting any&         number of registration levels.    F      August 13, 1982              - 27 -                      RFC #822 r    r-      Standard for ARPA Internet Text Messagesc    F         Name-domains model a global, logical, hierarchical  addressingF         scheme.   The  model is logical, in that an address specifica-F         tion is related to name registration and  is  not  necessarilyF         tied  to  transmission  path.   The  model's  hierarchy  is  aF         directed graph, called an in-tree, such that there is a singleF         path  from  the root of the tree to any node in the hierarchy.F         If more than one path actually exists, they are considered  to         be different addresses.   F         The root node is common to all addresses; consequently, it  isF         not  referenced.   Its  children  constitute "top-level" name-F         domains.  Usually, a service has access to its own full domainE         specification and to the names of all top-level name-domains.n  F         The "top" of the domain addressing hierarchy -- a child of theF         root  --  is  indicated  by  the right-most field, in a domainF         specification.  Its child is specified to the left, its  child         to the left, and so on.r  F         Some groups provide formal registration services;  these  con-F         stitute   name-domains   that  are  independent  logically  ofF         specific machines.  In addition, networks and machines  impli-F         citly  compose name-domains, since their membership usually is"         registered in name tables.  F         In the case of formal registration, an organization implementsF         a  (distributed)  data base which provides an address-to-route2         mapping service for addresses of the form:  5                          person@registry.organization   F         Note that "organization" is a logical  entity,  separate  from-         any particular communication network.   F         A mechanism for accessing "organization" is universally avail-F         able.   That mechanism, in turn, seeks an instantiation of theF         registry; its location is not indicated in the address specif-F         ication.   It  is assumed that the system which operates underF         the name "organization" knows how to find a subordinate regis-F         try.  The registry will then use the "person" string to deter-2         mine where to send the mail specification.  F         The latter,  network-oriented  case  permits  simple,  direct,:         attachment-related address specification, such as:  /                               user@host.networko  F         Once the network is accessed, it is expected  that  a  messageF         will  go  directly  to the host and that the host will resolve    F      August 13, 1982              - 28 -                      RFC #822 .    c-      Standard for ARPA Internet Text Messages     A         the user name, placing the message in the user's mailbox.n  -      6.2.2.  ABBREVIATED DOMAIN SPECIFICATION   F         Since any number of  levels  is  possible  within  the  domainF         hierarchy,  specification  of  a  fully  qualified address canF         become inconvenient.  This standard permits abbreviated domain)         specification, in a special case:   B             For the address of  the  sender,  call  the  left-mostB             sub-domain  Level  N.   In a header address, if all ofB             the sub-domains above (i.e., to the right of) Level  NB             are  the same as those of the sender, then they do notB             have to appear in the specification.   Otherwise,  the,             address must be fully qualified.  B             This feature is subject  to  approval  by  local  sub-B             domains.   Individual  sub-domains  may  require theirB             member systems, which originate mail, to provide  fullB             domain  specification only.  When permitted, abbrevia-B             tions may be present  only  while  the  message  stays0             within the sub-domain of the sender.  B             Use of this mechanism requires the sender's sub-domainB             to reserve the names of all top-level domains, so thatB             full specifications can be distinguished from abbrevi-              ated specifications.  .         For example, if a sender's address is:  <                  sender@registry-A.registry-1.organization-X  '         and one recipient's address is:   >                 recipient@registry-B.registry-1.organization-X           and another's is:   >                 recipient@registry-C.registry-2.organization-X  F         then ".registry-1.organization-X" need not be specified in theF         the  message,  but  "registry-C.registry-2"  DOES  have  to beF         specified.  That is, the first two addresses may  be  abbrevi-<         ated, but the third address must be fully specified.  F         When a message crosses a domain boundary, all  addresses  mustF         be  specified  in  the  full format, ending with the top-levelF         name-domain in the right-most field.  It is the responsibilityF         of  mail  forwarding services to ensure that addresses conform    F      August 13, 1982              - 29 -                      RFC #822       -      Standard for ARPA Internet Text Messages     F         with this requirement.  In the case of abbreviated  addresses,F         the  relaying  service must make the necessary expansions.  ItF         should be noted that it often is difficult for such a  serviceF         to locate all occurrences of address abbreviations.  For exam-F         ple, it will not be possible to find such abbreviations withinF         the  body  of  the  message.   The "Return-Path" field can aid3         recipients in recovering from these errors.   F         Note:  When passing any portion of an addr-spec onto a processF                which  does  not interpret data according to this stan-F                dard (e.g., mail protocol servers).  There must  be  NOF                LWSP-chars  preceding  or  following the at-sign or anyF                delimiting period ("."), such as  shown  in  the  aboveF                examples,   and   only  ONE  SPACE  between  contiguous                <word>s.e        6.2.3.  DOMAIN TERMS   F         A domain-ref must be THE official name of a registry, network,F         or  host.   It  is  a  symbolic  reference, within a name sub-F         domain.  At times, it is necessary to bypass standard  mechan-F         isms  for  resolving  such  references,  using  more primitiveF         information, such as a network host address  rather  than  its         associated host name.   F         To permit such references, this standard provides the  domain-F         literal  construct.   Its contents must conform with the needs5         of the sub-domain in which it is interpreted.   F         Domain-literals which refer to domains within the ARPA  Inter-F         net  specify  32-bit  Internet addresses, in four 8-bit fieldsF         noted in decimal, as described in Request for  Comments  #820,)         "Assigned Numbers."  For example:U  ,                                  [10.0.3.19]  F         Note:  THE USE OF DOMAIN-LITERALS IS STRONGLY DISCOURAGED.  ItF                is  permitted  only  as  a means of bypassing temporaryF                system limitations, such as name tables which  are  not                complete.  F         The names of "top-level" domains, and  the  names  of  domainsF         under  in  the  ARPA Internet, are registered with the NetworkF         Information Center, SRI International, Menlo Park, California.  *      6.2.4.  DOMAIN-DEPENDENT LOCAL STRING  F         The local-part of an  addr-spec  in  a  mailbox  specificationF         (i.e.,  the  host's  name for the mailbox) is understood to be    F      August 13, 1982              - 30 -                      RFC #822 h    "-      Standard for ARPA Internet Text Messagess    F         whatever the receiving mail protocol server allows.  For exam-F         ple,  some systems do not understand mailbox references of the,         form "P. D. Q. Bach", but others do.  F         This specification treats periods (".") as lexical separators.F         Hence,  their  presence  in  local-parts which are not quoted-F         strings, is detected.   However,  such  occurrences  carry  NOF         semantics.  That is, if a local-part has periods within it, anF         address parser will divide the local-part into several tokens,F         but  the  sequence  of  tokens will be treated as one uninter-F         preted unit.  The sequence  will  be  re-assembled,  when  theF         address is passed outside of the system such as to a mail pro-         tocol service.  !         For example, the address:   2                            First.Last@Registry.Org  F         is legal and does not require the local-part to be  surroundedF         with  quotation-marks.   (However,  "First  Last" DOES requireF         quoting.)  The local-part of the address, when passed  outsideF         of  the  mail  system,  within  the  Registry.Org  domain,  is4         "First.Last", again without quotation marks.  ,      6.2.5.  BALANCING LOCAL-PART AND DOMAIN  F         In some cases, the boundary between local-part and domain  canF         be  flexible.  The local-part may be a simple string, which isF         used for the final determination of the  recipient's  mailbox.F         All  other  levels  of  reference  are, therefore, part of the         domain.t  F         For some systems, in the case of abbreviated reference to  theF         local  and  subordinate  sub-domains,  it  may  be possible toF         specify only one reference within the domain  part  and  placeF         the  other,  subordinate  name-domain  references  within  the*         local-part.  This would appear as:  5                         mailbox.sub1.sub2@this-domain   F         Such a specification would be acceptable  to  address  parsersF         which  conform  to  RFC  #733,  but  do not support this newerF         Internet standard.  While contrary to the intent of this stan-          dard, the form is legal.  F         Also, some sub-domains have a specification syntax which  does3         not conform to this standard.  For example:l  7                       sub-net.mailbox@sub-domain.domaind    F      August 13, 1982              - 31 -                      RFC #822       -      Standard for ARPA Internet Text Messagesu    F         uses a different parsing  sequence  for  local-part  than  for         domain.   F         Note:  As a rule,  the  domain  specification  should  containF                fields  which  are  encoded  according to the syntax ofF                this standard and which contain  generally-standardizedF                information.   The local-part specification should con-F                tain only that portion of the  address  which  deviates>                from the form or intention of the domain field.        6.2.6.  MULTIPLE MAILBOXESl  F         An individual may have several mailboxes and wish  to  receiveF         mail  at  whatever  mailbox  is  convenient  for the sender toF         access.  This standard does not provide a means of  specifying,         "any member of" a list of mailboxes.  F         A set of individuals may wish to receive mail as a single unitF         (i.e.,  a  distribution  list).  The <group> construct permitsF         specification of such a list.  Recipient mailboxes are  speci-F         fied  within  the  bracketed  part (":" - ";").  A copy of theF         transmitted message is to be  sent  to  each  mailbox  listed.F         This  standard  does  not  permit  recursive  specification of         groups within groups.   F         While a list must be named, it is not required that  the  con-F         tents  of  the  list be included.  In this case, the <address>F         serves only as an indication of group distribution  and  would         appear in the form:s  *                                     name:;  F         Some mail  services  may  provide  a  group-list  distributionF         facility,  accepting  a single mailbox reference, expanding itF         to the full distribution list, and relaying the  mail  to  theF         list's  members.   This standard provides no additional syntaxF         for indicating such a  service.   Using  the  <group>  addressF         alternative,  while listing one mailbox in it, can mean eitherF         that the mailbox reference will be expanded to a list or  that)         there is a group with one member.s  (      6.2.7.  EXPLICIT PATH SPECIFICATION  F         At times, a  message  originator  may  wish  to  indicate  theF         transmission  path  that  a  message  should  follow.  This isF         called source routing.  The normal addressing scheme, used  inF         an  addr-spec,  is  carefully separated from such information;F         the <route> portion of a route-addr is provided for such occa-F         sions.  It specifies the sequence of hosts and/or transmission    F      August 13, 1982              - 32 -                      RFC #822       -      Standard for ARPA Internet Text Messagesh    F         services that are  to  be  traversed.   Both  domain-refs  and$         domain-literals may be used.  F         Note:  The use of source routing is discouraged.   Unless  theF                sender has special need of path restriction, the choiceF                of transmission route should be left to the mail  tran-                sport service.n        6.3.  RESERVED ADDRESS   F           It often is necessary to send mail to a site, without  know-F      ing  any  of its valid addresses.  For example, there may be mailF      system dysfunctions, or a user may wish to find  out  a  person's#      correct address, at that site.c  F           This standard specifies a single, reserved  mailbox  addressF      (local-part)  which  is  to  be valid at each site.  Mail sent toF      that address is to be routed to  a  person  responsible  for  theF      site's mail system or to a person with responsibility for generalE      site operation.  The name of the reserved local-part address is:   *                                 Postmaster  9      so that "Postmaster@domain" is required to be valid.P  F      Note:  This reserved local-part must be  matched  without  sensi-F             tivity to alphabetic case, so that "POSTMASTER", "postmas-:             ter", and even "poStmASteR" is to be accepted.                                              F      August 13, 1982              - 33 -                      RFC #822 g    h-      Standard for ARPA Internet Text Messagesn          7.  BIBLIOGRAPHYk    F      ANSI.  "USA Standard Code  for  Information  Interchange,"  X3.4.F         American  National Standards Institute: New York (1968).  AlsoF         in:  Feinler, E.  and J. Postel, eds., "ARPANET Protocol Hand-         book", NIC 7104.  F      ANSI.  "Representations of Universal Time, Local  Time  Differen-F         tials,  and United States Time Zone References for InformationF         Interchange," X3.51-1975.  American National Standards  Insti-         tute:  New York (1975).s  F      Bemer, R.W., "Time and the Computer."  In:  Interface  Age  (Feb.         1979).  F      Bennett, C.J.  "JNT Mail Protocol".  Joint Network Team,  Ruther-7         ford and Appleton Laboratory:  Didcot, England.a  F      Bhushan, A.K., Pogran, K.T., Tomlinson,  R.S.,  and  White,  J.E.F         "Standardizing  Network  Mail  Headers,"   ARPANET Request forF         Comments No. 561, Network Information Center  No.  18516;  SRI4         International:  Menlo Park (September 1973).  F      Birrell, A.D., Levin, R.,  Needham,  R.M.,  and  Schroeder,  M.D.F         "Grapevine:  An Exercise in Distributed Computing," Communica-5         tions of the ACM 25, 4 (April 1982), 260-274.   F      Crocker,  D.H.,  Vittal,  J.J.,  Pogran,  K.T.,  Henderson,  D.A.F         "Standard  for  the  Format  of  ARPA  Network  Text Message,"F         ARPANET Request for  Comments  No.  733,  Network  InformationF         Center  No.  41952.   SRI International:  Menlo Park (November         1977).  F      Feinler, E.J. and Postel, J.B.  ARPANET Protocol  Handbook,  Net-F         work  Information  Center  No.  7104   (NTIS AD A003890).  SRI0         International:  Menlo Park (April 1976).  F      Harary, F.   "Graph  Theory".   Addison-Wesley:   Reading,  Mass.         (1969).e  F      Levin, R. and Schroeder, M.  "Transport  of  Electronic  MessagesF         through  a  Network,"   TeleInformatics  79, pp. 29-33.  NorthF         Holland (1979).  Also  as  Xerox  Palo  Alto  Research  Center"         Technical Report CSL-79-4.  F      Myer, T.H. and Henderson, D.A.  "Message Transmission  Protocol,"F         ARPANET  Request  for  Comments,  No. 680, Network InformationA         Center No. 32116.  SRI International:  Menlo Park (1975).     F      August 13, 1982              - 34 -                      RFC #822       -      Standard for ARPA Internet Text MessagesS    F      NBS.  "Specification of Message Format for Computer Based MessageF         Systems, Recommended Federal Information Processing Standard."F         National  Bureau   of   Standards:    Gaithersburg,   Maryland         (October 1981).N  F      NIC.  Internet Protocol Transition Workbook.  Network InformationF         Center,   SRI-International,  Menlo  Park,  California  (March         1982).  F      Oppen, D.C. and Dalal, Y.K.  "The Clearinghouse:  A DecentralizedF         Agent  for  Locating  Named  Objects in a Distributed Environ-F         ment," OPD-T8103.  Xerox Office Products Division:  Palo Alto,         CA. (October 1981).d  F      Postel, J.B.  "Assigned Numbers,"  ARPANET Request for  Comments,?         No. 820.  SRI International:  Menlo Park (August 1982).o  F      Postel, J.B.  "Simple Mail Transfer  Protocol,"  ARPANET  RequestF         for Comments, No. 821.  SRI International:  Menlo Park (August         1982).  F      Shoch, J.F.  "Internetwork naming, addressing  and  routing,"  inF         Proc. 17th IEEE Computer Society International Conference, pp.7         72-79, Sept. 1978, IEEE Cat. No. 78 CH 1388-8C.   F      Su, Z. and Postel, J.  "The Domain Naming Convention for InternetF         User  Applications,"  ARPANET  Request  for Comments, No. 819.5         SRI International:  Menlo Park (August 1982).-                                              F      August 13, 1982              - 35 -                      RFC #822      d-      Standard for ARPA Internet Text Messages     )                                  APPENDIXl          A.  EXAMPLES         A.1.  ADDRESSES  .      A.1.1.  Alfred Neuman <Neuman@BBN-TENEXA>        A.1.2.  Neuman@BBN-TENEXA  F              These two "Alfred Neuman" examples have identical  seman-F         tics, as far as the operation of the local host's mail sendingF         (distribution) program (also sometimes  called  its  "mailer")F         and  the remote host's mail protocol server are concerned.  InF         the first example, the  "Alfred  Neuman"  is  ignored  by  theF         mailer,  as "Neuman@BBN-TENEXA" completely specifies the reci-F         pient.  The second example contains  no  superfluous  informa-F         tion,  and,  again,  "Neuman@BBN-TENEXA" is the intended reci-         pient.  F         Note:  When the message crosses name-domain  boundaries,  thenF                these specifications must be changed, so as to indicateF                the remainder of the hierarchy, starting with  the  top                level.a  1      A.1.3.  "George, Ted" <Shared@Group.Arpanet>t  F              This form might be used to indicate that a single mailboxF         is  shared  by several users.  The quoted string is ignored byF         the originating host's mailer, because  "Shared@Group.Arpanet"5         completely specifies the destination mailbox.   3      A.1.4.  Wilt . (the  Stilt) Chamberlain@NBA.USs  F              The "(the  Stilt)" is a comment, which is NOT included inF         the  destination  mailbox  address  handed  to the originatingF         system's mailer.  The local-part of the address is the  stringF         "Wilt.Chamberlain", with NO space between the first and second         words.        A.1.5.  Address Lists  :      Gourmets:  Pompous Person <WhoZiWhatZit@Cordon-Bleu>,6                 Childs@WGBH.Boston, Galloping Gourmet@@                 ANT.Down-Under (Australian National Television),*                 Cheapie@Discount-Liquors;,,        Cruisers:  Port@Portugal, Jones@SEA;,"          Another@Somewhere.SomeOrg    F      August 13, 1982              - 36 -                      RFC #822      7-      Standard for ARPA Internet Text Messagesm    F         This group list example points out the use of comments and the'         mixing of addresses and groups.w        A.2.  ORIGINATOR ITEMSc        A.2.1.  Author-sent  F              George Jones logs into his host  as  "Jones".   He  sends         mail himself.   "             From:  Jones@Group.Org  
         or  1             From:  George Jones <Jones@Group.Org>q        A.2.2.  Secretary-senti  F              George Jones logs in as Jones on his  host.   His  secre-F         tary,  who logs in as Secy sends mail for him.  Replies to the!         mail should go to George.e  /             From:    George Jones <Jones@Group>r%             Sender:  Secy@Other-Groupu  9      A.2.3.  Secretary-sent, for user of shared directory   F              George Jones' secretary sends mail  for  George.  Replies         should go to George.  4             From:     George Jones<Shared@Group.Org>&             Sender:   Secy@Other-Group  F         Note that there need not be a space between  "Jones"  and  theF         "<",  but  adding a space enhances readability (as is the case         in other examples.  0      A.2.4.  Committee activity, with one author  F              George is a member of a committee.  He wishes to have any;         replies to his message go to all committee members.   3             From:     George Jones <Jones@Host.Net>               Sender:   Jones@Host4             Reply-To: The Committee: Jones@Host.Net,5                                      Smith@Other.Org,o8                                      Doe@Somewhere-Else;  F         Note  that  if  George  had  not  included  himself   in   the    F      August 13, 1982              - 37 -                      RFC #822 a     -      Standard for ARPA Internet Text Messagest    F         enumeration  of  The  Committee,  he  would not have gotten anF         implicit reply; the presence of the  "Reply-to"  field  SUPER-F         SEDES the sending of a reply to the person named in the "From"         field.  5      A.2.5.  Secretary acting as full agent of author   F              George Jones asks his secretary  (Secy@Host)  to  send  aF         message for him in his capacity as Group.  He wants his secre-#         tary to handle all replies.,  /             From:     George Jones <Group@Host>g             Sender:   Secy@Hostr             Reply-To: Secy@Hostd  2      A.2.6.  Agent for user without online mailbox  F              A friend  of  George's,  Sarah,  is  visiting.   George'sF         secretary  sends  some  mail to a friend of Sarah in computer-F         land.  Replies should go to George, whose mailbox is Jones  at         Registry.   4             From:     Sarah Friendly <Secy@Registry>/             Sender:   Secy-Name <Secy@Registry>s%             Reply-To: Jones@Registry.s  ,      A.2.7.  Agent for member of a committee  F              George's secretary sends out a message which was authoredF         jointly by all the members of a committee.  Note that the nameF         of the committee cannot be specified, since <group> names  are(         not permitted in the From field.               From:   Jones@Host,d%                     Smith@Other-Host, &                     Doe@Somewhere-Else             Sender: Secy@SHost                            F      August 13, 1982              - 38 -                      RFC #822      v-      Standard for ARPA Internet Text Messages           A.3.  COMPLETE HEADERS         A.3.1.  Minimum required   E      Date:     26 Aug 76 1429 EDT        Date:     26 Aug 76 1429 EDTtE      From:     Jones@Registry.Org   or   From:     Jones@Registry.OrgSE      Bcc:                                To:       Smith@Registry.Org   F         Note that the "Bcc" field may be empty, while the  "To"  field1         is required to have at least one address.   0      A.3.2.  Using some of the additional fields  !      Date:     26 Aug 76 1430 EDTD'      From:     George Jones<Group@Host>e      Sender:   Secy@SHOSTI$      To:       "Al Neuman"@Mad-Host,$                Sam.Irving@Other-Host%      Message-ID:  <some.string@SHOST>s  4      A.3.3.  About as complex as you're going to get  #      Date     :  27 Aug 76 0932 PDT 6      From     :  Ken Davis <KDavis@This-Host.This-net>*      Subject  :  Re: The Syntax in the RFC!      Sender   :  KSecy@Other-HostB,      Reply-To :  Sam.Irving@Reg.Organization6      To       :  George Jones <Group@Some-Reg.An-Org>,(                  Al.Neuman@MAD.Publisher       cc       :  Important folk:2                    Tom Softwood <Balsa@Tree.Root>,,                    "Sam Irving"@Other-Host;,'                  Standard Distribution:d:                    /main/davis/people/standard@Other-Host,:                    "<Jones>standard.dist.3"@Tops-20-Host>;?      Comment  :  Sam is away on business. He asked me to handle,?                  his mail for him.  He'll be able to provide  aF?                  more  accurate  explanation  when  he  returns                   next week.w;      In-Reply-To: <some.string@DBM.Group>, George's messageo?      X-Special-action:  This is a sample of user-defined field-a9                  names.  There could also be a field-name >                  "Special-action", but its name might later be                  preempted0      Message-ID: <4231.629.XYzi-What@Other-Host>            F      August 13, 1982              - 39 -                      RFC #822 h    r-      Standard for ARPA Internet Text Messages           B.  SIMPLE FIELD PARSINGo  F           Some mail-reading software systems may wish to perform  onlyF      minimal  processing,  ignoring  the internal syntax of structuredF      field-bodies and treating them the  same  as  unstructured-field-:      bodies.  Such software will need only to distinguish:  1          o   Header fields from the message body,   C          o   Beginnings of fields from lines which continue fields,   -          o   Field-names from field-contents.   F           The abbreviated set of syntactic rules  which  follows  willF      suffice  for  this  purpose.  It describes a limited view of mes-F      sages and is a subset of the syntactic rules provided in the mainF      part of this specification.  One small exception is that the con-0      tents of field-bodies consist only of text:        B.1.  SYNTAX     -      message         =   *field *(CRLF *text)n  :      field           =    field-name ":" [field-body] CRLF  D      field-name      =  1*<any CHAR, excluding CTLs, SPACE, and ":">  :      field-body      =   *text [CRLF LWSP-char field-body]          B.2.  SEMANTICS  F           Headers occur before the message body and are terminated  by.      a null line (i.e., two contiguous CRLFs).  F           A line which continues a header field begins with a SPACE orF      HTAB  character,  while  a  line  beginning a field starts with a.      printable character which is not a colon.  F           A field-name consists of one or  more  printable  charactersF      (excluding  colon,  space, and control-characters).  A field-nameF      MUST be contained on one line.  Upper and lower case are not dis-+      tinguished when comparing field-names.o              F      August 13, 1982              - 40 -                      RFC #822 E    q-      Standard for ARPA Internet Text Messagesn    "      C.  DIFFERENCES FROM RFC #733  F           The following summarizes the differences between this  stan-F      dard  and the one specified in Arpanet Request for Comments #733,F      "Standard for the Format of ARPA  Network  Text  Messages".   TheF      differences  are  listed  in the order of their occurrence in the      current specification.         C.1.  FIELD DEFINITIONS        C.1.1.  FIELD NAMES  F         These now must be a sequence of  printable  characters.   They'         may not contain any LWSP-chars.i        C.2.  LEXICAL TOKENSo        C.2.1.  SPECIALSl  F         The characters period ("."), left-square  bracket  ("["),  andF         right-square  bracket ("]") have been added.  For presentationF         purposes, and when passing a specification to  a  system  thatF         does  not conform to this standard, periods are to be contigu-F         ous with their surrounding lexical tokens.   No  linear-white-F         space  is  permitted  between them.  The presence of one LWSP-4         char between other tokens is still directed.        C.2.2.  ATOMt  $         Atoms may not contain SPACE.        C.2.3.  SPECIAL TEXTi  F         ctext and qtext have had backslash ("\") added to the list  of         prohibited characters.        C.2.4.  DOMAINS  F         The lexical tokens  <domain-literal>  and  <dtext>  have  been         added.         C.3.  MESSAGE SPECIFICATION        C.3.1.  TRACE  F         The "Return-path:" and "Received:" fields have been specified.          F      August 13, 1982              - 41 -                      RFC #822 U     -      Standard for ARPA Internet Text Messagesh          C.3.2.  FROM   F         The "From" field must contain machine-usable addresses  (addr-F         spec).   Multiple  addresses may be specified, but named-lists         (groups) may not.i        C.3.3.  RESENT   F         The meta-construct of prefacing field names  with  the  stringF         "Resent-"  has been added, to indicate that a message has been/         forwarded by an intermediate recipient.         C.3.4.  DESTINATION  F         A message must contain at least one destination address field.C         "To" and "CC" are required to contain at least one address.m        C.3.5.  IN-REPLY-TO  F         The field-body is no longer a comma-separated list, although a$         sequence is still permitted.        C.3.6.  REFERENCE  F         The field-body is no longer a comma-separated list, although a$         sequence is still permitted.        C.3.7.  ENCRYPTED  F         A field has been specified that permits  senders  to  indicate6         that the body of a message has been encrypted.        C.3.8.  EXTENSION-FIELD  F         Extension fields are prohibited from beginning with the  char-         acters "X-".  &      C.4.  DATE AND TIME SPECIFICATION        C.4.1.  SIMPLIFICATIONc  F         Fewer optional forms are permitted  and  the  list  of  three--         letter time zones has been shortened.          C.5.  ADDRESS SPECIFICATION            F      August 13, 1982              - 42 -                      RFC #822      r-      Standard for ARPA Internet Text Messagese          C.5.1.  ADDRESS  F         The use of quoted-string, and the ":"-atom-":" construct, haveF         been  removed.   An  address  now  is  either a single mailboxF         reference or is a named list of addresses.  The  latter  indi-#         cates a group distribution.e        C.5.2.  GROUPSJ  F         Group lists are now required to to have a name.   Group  lists         may not be nested.        C.5.3.  MAILBOX  F         A mailbox specification  may  indicate  a  person's  name,  asF         before.   Such  a  named  list  no longer may specify multiple(         mailboxes and may not be nested.        C.5.4.  ROUTE ADDRESSING.  F         Addresses now are taken to be absolute, global specifications,F         independent  of transmission paths.  The <route> construct hasF         been provided, to permit explicit specification  of  transmis-F         sion  path.   RFC  #733's  use  of multiple at-signs ("@") wasF         intended as a general syntax  for  indicating  routing  and/orF         hierarchical addressing.  The current standard separates these9         specifications and only one at-sign is permitted.n        C.5.5.  AT-SIGN  F         The string " at " no longer is used as an  address  delimiter./         Only at-sign ("@") serves the function.R        C.5.6.  DOMAINS  ;         Hierarchical, logical name-domains have been added.         C.6.  RESERVED ADDRESSl  F      The local-part "Postmaster" has been reserved, so that users  can8      be guaranteed at least one valid address at a site.                    F      August 13, 1982              - 43 -                      RFC #822      e-      Standard for ARPA Internet Text Messagesd    -      D.  ALPHABETICAL LISTING OF SYNTAX RULESM  @      address     =  mailbox                      ; one addressee=                  /  group                        ; named list A      addr-spec   =  local-part "@" domain        ; global address 4      ALPHA       =  <any ASCII alphabetic character>F                                                  ; (101-132, 65.- 90.)F                                                  ; (141-172, 97.-122.)@      atom        =  1*<any CHAR except specials, SPACE and CTLs>@      authentic   =   "From"       ":"   mailbox  ; Single authorC                  / ( "Sender"     ":"   mailbox  ; Actual submittor C                      "From"       ":" 1#mailbox) ; Multiple authorssA                                                  ;  or not senderuF      CHAR        =  <any ASCII character>        ; (  0-177,  0.-127.)<      comment     =  "(" *(ctext / quoted-pair / comment) ")"F      CR          =  <ASCII CR, carriage return>  ; (     15,      13.)      CRLF        =  CR LFpC      ctext       =  <any CHAR excluding "(",     ; => may be foldedr/                      ")", "\" & CR, & includingc(                      linear-white-space>F      CTL         =  <any ASCII control           ; (  0- 37,  0.- 31.)F                      character and DEL>          ; (    177,     127.)A      date        =  1*2DIGIT month 2DIGIT        ; day month year B                                                  ;  e.g. 20 Jun 82;      dates       =   orig-date                   ; Originala<                    [ resent-date ]               ; Forwarded;      date-time   =  [ day "," ] date time        ; dd mm yyd@                                                  ;  hh:mm:ss zzz4      day         =  "Mon"  / "Tue" /  "Wed"  / "Thu"+                  /  "Fri"  / "Sat" /  "Sun"o;      delimiters  =  specials / linear-white-space / commentn:      destination =  "To"          ":" 1#address  ; Primary/                  /  "Resent-To"   ":" 1#addressc<                  /  "cc"          ":" 1#address  ; Secondary/                  /  "Resent-cc"   ":" 1#addresss?                  /  "bcc"         ":"  #address  ; Blind carbon /                  /  "Resent-bcc"  ":"  #address F      DIGIT       =  <any ASCII decimal digit>    ; ( 60- 71, 48.- 57.)0      domain      =  sub-domain *("." sub-domain)5      domain-literal =  "[" *(dtext / quoted-pair) "]"aE      domain-ref  =  atom                         ; symbolic referenceiC      dtext       =  <any CHAR excluding "[",     ; => may be folded-/                      "]", "\" & CR, & including (                      linear-white-space>      extension-field =<                    <Any field which is defined in a document;                     published as a formal extension to this A                     specification; none will have names beginningh)                     with the string "X-">     F      August 13, 1982              - 44 -                      RFC #822 s     -      Standard for ARPA Internet Text Messagese    6      field       =  field-name ":" [ field-body ] CRLFA      fields      =    dates                      ; Creation time,dC                       source                     ;  author id & one D                     1*destination                ;  address requiredC                      *optional-field             ;  others optional '      field-body  =  field-body-contentse/                     [CRLF LWSP-char field-body]i      field-body-contents =E                    <the ASCII characters making up the field-body, assE                     defined in the following sections, and consistingc?                     of combinations of atom, quoted-string, andTA                     specials tokens, or else consisting of texts>u@      field-name  =  1*<any CHAR, excluding CTLs, SPACE, and ":">-      group       =  phrase ":" [#mailbox] ";" 2      hour        =  2DIGIT ":" 2DIGIT [":" 2DIGIT]F                                                  ; 00:00:00 - 23:59:59F      HTAB        =  <ASCII HT, horizontal-tab>   ; (     11,       9.)F      LF          =  <ASCII LF, linefeed>         ; (     12,      10.)D      linear-white-space =  1*([CRLF] LWSP-char)  ; semantics = SPACEB                                                  ; CRLF => folding@      local-part  =  word *("." word)             ; uninterpretedA                                                  ; case-preserved D      LWSP-char   =  SPACE / HTAB                 ; semantics = SPACEA      mailbox     =  addr-spec                    ; simple address C                  /  phrase route-addr            ; name & addr-spec C      message     =  fields *( CRLF *text )       ; Everything afterdC                                                  ;  first null line C                                                  ;  is message body 6      month       =  "Jan"  /  "Feb" /  "Mar"  /  "Apr"6                  /  "May"  /  "Jun" /  "Jul"  /  "Aug"6                  /  "Sep"  /  "Oct" /  "Nov"  /  "Dec"D      msg-id      =  "<" addr-spec ">"            ; Unique message id      optional-field = 4                  /  "Message-ID"        ":"   msg-id4                  /  "Resent-Message-ID" ":"   msg-id?                  /  "In-Reply-To"       ":"  *(phrase / msg-id) ?                  /  "References"        ":"  *(phrase / msg-id)e4                  /  "Keywords"          ":"  #phrase2                  /  "Subject"           ":"  *text2                  /  "Comments"          ":"  *text3                  /  "Encrypted"         ":" 1#2wordo@                  /  extension-field              ; To be definedD                  /  user-defined-field           ; May be pre-empted1      orig-date   =  "Date"        ":"   date-time E      originator  =   authentic                   ; authenticated addr 2                    [ "Reply-To"   ":" 1#address] )D      phrase      =  1*word                       ; Sequence of words        F      August 13, 1982              - 45 -                      RFC #822 T    N-      Standard for ARPA Internet Text Messagese    C      qtext       =  <any CHAR excepting <">,     ; => may be folded ,                      "\" & CR, and including(                      linear-white-space>E      quoted-pair =  "\" CHAR                     ; may quote any char C      quoted-string = <"> *(qtext/quoted-pair) <">; Regular qtext or B                                                  ;   quoted chars.@      received    =  "Received"    ":"            ; one per relay?                        ["from" domain]           ; sending host-A                        ["by"   domain]           ; receiving host @                        ["via"  atom]             ; physical pathE                       *("with" atom)             ; link/mail protocoleB                        ["id"   msg-id]           ; receiver msg id?                        ["for"  addr-spec]        ; initial form @                         ";"    date-time         ; time received  %      resent      =   resent-authentic 8                    [ "Resent-Reply-To"  ":" 1#address] )      resent-authentic =o5                  =   "Resent-From"      ":"   mailbox 5                  / ( "Resent-Sender"    ":"   mailboxi8                      "Resent-From"      ":" 1#mailbox  )1      resent-date =  "Resent-Date" ":"   date-timeeA      return      =  "Return-path" ":" route-addr ; return addressF@      route       =  1#("@" domain) ":"           ; path-relative-      route-addr  =  "<" [route] addr-spec ">"nA      source      = [  trace ]                    ; net traversals @                       originator                 ; original mail<                    [  resent ]                   ; forwardedF      SPACE       =  <ASCII SP, space>            ; (     40,      32.)E      specials    =  "(" / ")" / "<" / ">" / "@"  ; Must be in quoted-oB                  /  "," / ";" / ":" / "\" / <">  ;  string, to useB                  /  "." / "[" / "]"              ;  within a word./      sub-domain  =  domain-ref / domain-literalpF      text        =  <any CHAR, including bare    ; => atoms, specials,@                      CR & bare LF, but NOT       ;  comments andF                      including CRLF>             ;  quoted-strings areC                                                  ;  NOT recognized.eD      time        =  hour zone                    ; ANSI and MilitaryA      trace       =    return                     ; path to sender ?                     1*received                   ; receipt tags       user-defined-field =-8                    <Any field which has not been defined<                     in this specification or published as an>                     extension to this specification; names for9                     such fields must be unique and may be 7                     pre-empted by published extensions>:(      word        =  atom / quoted-string        F      August 13, 1982              - 46 -                      RFC #822       -      Standard for ARPA Internet Text Messagese    A      zone        =  "UT"  / "GMT"                ; Universal TimenF                                                  ; North American : UTF                  /  "EST" / "EDT"                ;  Eastern:  - 5/ - 4F                  /  "CST" / "CDT"                ;  Central:  - 6/ - 5F                  /  "MST" / "MDT"                ;  Mountain: - 7/ - 6F                  /  "PST" / "PDT"                ;  Pacific:  - 8/ - 7D                  /  1ALPHA                       ; Military: Z = UT;F      <">         =  <ASCII quote mark>           ; (     42,      34.)                                                                                      F      August 13, 1982              - 47 -                      RFC #822