htmlplus.dtd

harvest是一个下载html网页得机器人

  <!ENTITY parent SDATA "parent" -- parent of current document -->
  <!ENTITY next SDATA "next" -- next document in current sequence -->
  <!ENTITY previous SDATA "previous" -- previous document in current sequence -->
  <!ENTITY home SDATA "home" -- home document -->
  <!ENTITY toc SDATA "toc" -- table of contents -->
  <!ENTITY glossary SDATA "glossary" -- glossary of terms etc. -->
  <!ENTITY index SDATA "index" -- searchable index -->
  <!ENTITY summary SDATA "summary" -- summary -->

  <!ENTITY calculator SDATA "calculator" -- A calculator -->
  <!ENTITY caution SDATA "caution" -- Warnign sign -->
  <!ENTITY clock SDATA "clock" -- A clock -->
  <!ENTITY compressed.document SDATA "compressed.document">
  <!ENTITY diskette SDATA "diskette" -- A diskette -->
  <!ENTITY display SDATA "display" -- A computer screen -->
  <!ENTITY fax SDATA "fax" -- A fax machine -->
  <!ENTITY mail.in SDATA "mail.in" -- mail-in tray -->
  <!ENTITY mail.out SDATA "mail.out" -- mail-out tray -->
  <!ENTITY mouse SDATA "mouse" -- mouse/pointing device -->
  <!ENTITY printer SDATA "printer" -- hardcopy device -->
  <!ENTITY tn3270 SDATA "tn3270" --tn3270 terminal session -->
  <!ENTITY trash SDATA "trash" -- waste paper basket -->
  <!ENTITY uuencoded.document SDATA "uuencoded.document" -- uuencoded data -->
]]>

<!-- maths symbols when needed -->
<![ %HTML.math [
    <!ENTITY % ISOtech PUBLIC "ISO 8879-1986//ENTITIES General Technical//EN">
    %ISOtech;

    <!ENTITY % ISOgrk3 PUBLIC "ISO 8879-1986//ENTITIES Greek Symbols//EN">
    %ISOgrk3;

    <!ENTITY % ISOamso PUBLIC "ISO 8879-1986//ENTITIES Added Math Symbols: Ordinary//EN">
    %ISOamso;

    <!ENTITY % ISOamsr PUBLIC "ISO 8879-1986//ENTITIES Added Math Symbols: Relations//EN">
    %ISOamsr;

    <!ENTITY % ISOamsc PUBLIC "ISO 8879-1986//ENTITIES Added Math Symbols: Delimiters//EN">
    %ISOamsc;

    <!-- misc. from ISO Binary and Large operators -->

    <!ENTITY thinsp SDATA "[thinsp]"--=thin space (1/6 em)-->
    <!ENTITY coprod SDATA "[coprod]"--/coprod L: coproduct operator-->
    <!ENTITY prod   SDATA "[prod  ]"--/prod L: product operator-->
    <!ENTITY sum    SDATA "[sum   ]"--/sum L: summation operator-->
]]>

<!-- Basic types of elements:
  <!ELEMENT tagname - - CONTENT> elements needing end tags
  <!ELEMENT tagname - O CONTENT> elements with optional end tags
  <!ELEMENT tagname - O EMPTY> elements without content or end tags

The content definition is:
       -  an entity definition as defined above
       -  a tagname
       -  (brackets enclosing the above)
These may be combined with the operators:
  A*      A occurs zero or more times
  A+      A occurs one or more times
  A|B     implies either A or B
  A?      A occurs zero or one times
  A,B     implies first A then B
  A&B     either or both A and B (in either order A B or B A)
-->

<!ELEMENT HTMLPLUS O O (HEAD, BODY)>
<!ATTLIST HTMLPLUS
        version CDATA #IMPLIED -- the HTML+ version number --
        charset CDATA   #IMPLIED -- eg "ISO-2022-JP" for japanese -->

<!ELEMENT HEAD - O (%setup;) -- delimits document wide properties -->
<!ELEMENT BODY - O (PROLOGUE?, %divisions;, EPILOGUE?)>

<!ELEMENT PROLOGUE O O (%main;)+>
<!ELEMENT EPILOGUE O O (%main;)+>

<!--
  Browsers *must* tolerate missing DIVn tags, e.g. the presence of an
  <H1> tag implies a DIV1 element enclosing it and the following text.
  The SGML standard unfortunately doesn't permit such inferences due
  to a decision made to simplify writing general SGML parsers.

  It would be nice to allow an optional prologue and epilogue, before
  and after the divisions respectively. Unfortunately, the epilogue
  would lead to an ambiguous content model (according to sgmls).
-->

<!ELEMENT DIV1 - - (H1, (%main;)*, DIV6*, DIV5*, DIV4*, DIV3*, DIV2*)>
<!ELEMENT DIV2 - - (H2, (%main;)*, DIV6*, DIV5*, DIV4*, DIV3*)>
<!ELEMENT DIV3 - - (H3, (%main;)*, DIV6*, DIV5*, DIV4*)>
<!ELEMENT DIV4 - - (H4, (%main;)*, DIV6*, DIV5*)>
<!ELEMENT DIV5 - - (H5, (%main;)*, DIV6*)>
<!ELEMENT DIV6 - - (H6, (%main;)*)>

<!ATTLIST (DIV6|DIV5|DIV4|DIV3|DIV2|DIV1)
        id      ID      #IMPLIED -- link destination --
        charset CDATA   #IMPLIED -- eg "ISO-2022-JP" for japanese -->

<!-- Document headers -->
<!ELEMENT (%heading;) - - (%text;)+>
<!ATTLIST (%heading;)
        id      ID      #IMPLIED -- defines link destination --
        charset CDATA   #IMPLIED -- eg "ISO-2022-JP" for japanese --
        align   (left|indent|center|right|justify) left>

<!-- character emphasis -->
<!ELEMENT (%emph1;) - - (%text;)+>
<!ATTLIST (%emph1;)
        id      ID      #IMPLIED -- link destination --
        charset CDATA   #IMPLIED -- eg "ISO-2022-JP" for japanese -->

<!--
 Paragraphs which act as containers for the following text

 Browsers *must* be capable of inferring missing <P>
 start tags from the content model. Basically, if the parser
 comes across unexpected %text; then there's a missing <P>.
-->
<!ELEMENT P - O (%text;)+>
<!ATTLIST P
        id      ID      #IMPLIED -- link destination --
        charset CDATA   #IMPLIED -- eg "ISO-2022-JP" for japanese --
        nowrap (nowrap) #IMPLIED -- don't wrap words --
        nofold (nofold) #IMPLIED -- don't fold whitespace --
        align   (left|indent|center|right|justify) left>

<!ELEMENT HR - O EMPTY -- Horizontal Rule -->
<!ELEMENT BR - O EMPTY -- forced line break -->

<!ELEMENT PRE - - (%text;)+ -- preformatted fixed pitch text -->
<!ATTLIST PRE
        id      ID      #IMPLIED -- link destination --
        charset CDATA   #IMPLIED -- eg "ISO-2022-JP" for japanese -->

<!ELEMENT ADDRESS - - (P)+ -- info on author -->
<!ATTLIST ADDRESS
        id      ID      #IMPLIED -- link destination --
        charset CDATA   #IMPLIED -- eg "ISO-2022-JP" for japanese -->

<!-- Lists which can be nested -->
<!ELEMENT OL - - (LI)+ -- ordered list -->
<!ATTLIST OL
        id      ID      #IMPLIED
        start   NUMBER  1           -- sequence number for first item --
        charset CDATA   #IMPLIED    -- eg "ISO-2022-JP" for japanese --
        compact (compact) #IMPLIED  -- reduced interitem spacing -->

<!ELEMENT UL - - (LI)+ -- unordered list -->
<!ATTLIST UL
        id      ID      #IMPLIED    -- link destination --
        charset CDATA   #IMPLIED    -- eg "ISO-2022-JP" for japanese --
        compact (compact) #IMPLIED  -- reduced interitem spacing --
        plain   (plain) #IMPLIED    -- suppress bullets --
        wrap (vert|horiz|none) none -- multicolumn list wrap style -->

<!-- List items for UL and OL lists

 By default UL list items have bullets while OL list items
 are numbered in a style dependent on nesting. For extra impact
 specify an explicit image with the icon attribute or a string
 with the label attribute. This attribute can also be used with
 SGML entities to specify standard icons, e.g. label="&folder;"
-->

<!ELEMENT LI - O (DL|UL|OL|P|HR|PRE)+>
<!ATTLIST LI
        id      ID      #IMPLIED
        charset CDATA   #IMPLIED -- eg "ISO-2022-JP" for japanese --
        number  NUMBER  #IMPLIED -- sets sequence number for OL lists --
        icon     %URL;  #IMPLIED -- image for use in place of bullet --
        label   CDATA   #IMPLIED -- when you can't show the icon image -->

<!-- Definition Lists (terms + definitions) -->
<!ELEMENT DL - - (DT+,DD)+>
<!ATTLIST DL
        id      ID      #IMPLIED
        charset CDATA   #IMPLIED    -- eg "ISO-2022-JP" for japanese --
        compact (compact) #IMPLIED  -- reduced interitem spacing -->

<!ELEMENT DT - O (%text;)+          -- term text -- >
<!ELEMENT DD - O (DL|UL|OL|P|HR|PRE)+   -- definition text -- >
<!ATTLIST (DT|DD)
        id      ID      #IMPLIED
        charset CDATA   #IMPLIED -- eg "ISO-2022-JP" for japanese -->

<!-- Hypertext Links from points within document nodes

 The HREF attribute specifies the link destination as a URL or URN.
 In figures, the SHAPE attribute defines the extent of the link as
 a polygonal region, and is used with the FIG element.

 The PRINT attribute determines how the browser should deal with
 links when printing this document. This makes it possible for
 users to print a document and related subdocuments with a single
 menu action. If PRINT="Section", then the link is followed and
 printed as a follow-on section after the current document. If
 PRINT="Footnote" and the linked document is sufficiently small
 then it is included as a footnote. If PRINT="Reference" then the
 document's URL (and title) is included in a footnote or in a list
 of references at the end of the document.

 The TITLE attribute may be used for links in which the destination
 node doesn't define a title itself, e.g. non-html documents.

 The REL attribute is used to specify how the browser interprets
 the link when this document is being used as a hypertext path
 REL="Path" causes the linked document to be treated as a path
 and inserted into the current path, while REL="Node" treats
 the linked document as a node on the current path. REL="Embed"
 is a hint to embed the referenced node into the current document.

 The SIG attribute allows authors to specify a digital signature
 of linked documents to check that they haven't been changed.
 It starts with a prefix denoting the algorithm used, in particular
 SIG="md5:2l3k4j2lkj423l" denotes the MD5 signature: 2l3k4j2lkj423l
 which is encoded using the standard MIME base64 representation
-->

<!ELEMENT A - - (#PCDATA | IMG | %emph;)*>
<!ATTLIST A
        id      ID      #IMPLIED -- as target of link --
        %anchorname;             -- see HTML.obsolete --
        charset CDATA   #IMPLIED -- eg "ISO-2022-JP" for japanese --
        shape   CDATA   #IMPLIED -- list of points for shaped buttons --
        href    %URL;   #IMPLIED -- destination node --
        rel     CDATA   #IMPLIED -- forward relationship type --
        rev     CDATA   #IMPLIED --  reverse relationship type --
        print   CDATA   #IMPLIED -- reference/footnote/section --
        title   CDATA   #IMPLIED -- when otherwise unavailable --
        sig     CDATA   #IMPLIED -- MD5 digital signature -->

<!-- Other kinds of relationships between documents

 There are a set of standard RELationship types which alter the
 browser's navigation menu:

    UseIndex        searchable index
    UseGlossary     shared glossary
    Contents        shared contents page
    Previous        previous document in a hypertext path
    Next            next document in a hypertext path
    Bookmark        named with the title attribute
    Made            Defines who is the "maker" of this document
    Help            provides help on this document
    Annotation      an additional note on current document
    Reply           a note with equal footing to current document
    Subdocument     defines parent->child relationship
    Parent          defines child->parent relationship
    StyleSheet      an associated style sheet

 Bookmarks allow authors to define a set of useful links
 which are to be accessed via a menu, rather than as conventional
 in-line hypertext links. Previous and Next links are inserted
 by the browser when interpreting a separate document as a path.
 See above description of REL="Node" and REL="Path" for <A>.

 The FROM attribute makes it possible to specify annotation
 links separately from the document text flow. The FROM
 attribute specifies an ID for the source of a link, while
 the HREF attribute specifies its destination. HTTP servers
 can use the WWW-Link: header to "insert" such annotations
 into documents.
-->

<!ELEMENT LINK - O EMPTY>
<!ATTLIST LINK
        id      ID      #IMPLIED -- to allow meta info on links --
        charset CDATA   #IMPLIED -- eg "ISO-2022-JP" for japanese --
        from    IDREF   #IMPLIED -- starting point --
        href    %URL;  #REQUIRED -- destination node --
        rel     CDATA   #IMPLIED -- forward relationship type --
        rev     CDATA   #IMPLIED -- reverse relationship type --
        src     %URL;   #IMPLIED -- an image for displaying link --
        print   CDATA   #IMPLIED -- reference/footnote/section --
        title   CDATA   #IMPLIED -- when otherwise unavailable --
        seal    CDATA   #IMPLIED -- MD5 digital signature -->

<!-- Document title -->
<!ELEMENT TITLE - - (#PCDATA | %emph;)+>
<!ATTLIST TITLE
        id      ID      #IMPLIED -- link destination --
        charset CDATA   #IMPLIED -- eg "ISO-2022-JP" for japanese -->

<!-- Original document URL for resolving relative URLs  -->
<!ELEMENT BASE - O EMPTY>
<!ATTLIST BASE HREF %URL; #IMPLIED>

<!-- Signifies the document's URL accepts queries,
     and may be implied by HTTP header info -->
<!ELEMENT ISINDEX - O EMPTY>
<!ATTLIST ISINDEX href %URL; #IMPLIED -- defaults to document's URL -->

<!--
 Servers should read the document head to generate HTTP headers
 corresponding to META elements, e.g. if the document contains:

        <meta name="Expires" value="Tue, 04 Dec 1993 21:29:02 GMT">

 The server should include the HTTP date format header field:

        Expires: Tue, 04 Dec 1993 21:29:02 GMT

 Other likely names are "Summary", "Keywords", "Created",
 "Owner" (a name) and "Reply-To" (an email address)

 These elements can be used to construct IAFA style templates
 which describe the contents of the server. The templates are
 then used for automatic indexing of servers.

 The effectiveness of indexing is dependent on using well defined
 keywords drawn from standard sets. 

 The Document-Icon header is used to specify a URL for
 use as an icon in the hotlist and history list.
-->

<!ELEMENT META - O EMPTY>
<!ATTLIST META
        id      ID      #IMPLIED -- to allow meta info  --
        name    CDATA   #IMPLIED -- HTTP header e.g. "Expires" --
        value   CDATA   #IMPLIED -- associated value -->

<![ %HTML.obsolete [

<!ELEMENT (MENU|DIR) - - (LI)+ -- plain single/multicolumn lists-->
<!ATTLIST (MENU|DIR)
        compact (compact) #IMPLIED  -- reduced interitem spacing -->

<!ELEMENT BLOCKQUOTE - - (P)+ -- extended quotes -->

]]>

<![ %HTML.emph [

<!-- additional character emphasis -->
<!ELEMENT (%emph2;) - - (%text;)*>
<!ATTLIST (%emph2;)
        id      ID      #IMPLIED -- link destination --
        charset CDATA   #IMPLIED -- eg "ISO-2022-JP" for japanese -->