htmlparser.y

将HTML转换为TXT文件的程序

/* ------------------------------------------------------------------------- */

/*
 * Copyright (c) 1999
 *      GMRS Software GmbH, Innsbrucker Ring 159, 81669 Munich, Germany.
 *      http://www.gmrs.de
 *      All rights reserved.
 *      Author: Arno Unkrig ([e-mail])
 *
 * Redistribution and use in source and binary forms, with or without
 * modification, are permitted provided that the following conditions
 * are met:
 * 1. Redistributions of source code must retain the above copyright
 *    notice, this list of conditions and the following disclaimer.
 * 2. Redistributions in binary form must reproduce the above copyright
 *    notice, this list of conditions and the following disclaimer in the
 *    documentation and/or other materials provided with the distribution.
 * 3. All advertising materials mentioning features or use of this software
 *    must display the following acknowledgement:
 *      This product includes software developed by GMRS Software GmbH.
 * 4. The name of GMRS Software GmbH may not be used to endorse or promote
 *    products derived from this software without specific prior written
 *    permission.
 *
 * THIS SOFTWARE IS PROVIDED BY GMRS SOFTWARE GMBH ``AS IS'' AND ANY
 * EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
 * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR
 * PURPOSE ARE DISCLAIMED.  IN NO EVENT SHALL GMRS SOFTWARE GMBH BE LIABLE
 * FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR
 * CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF
 * SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS
 * INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN
 * CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE)
 * ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF
 * THE POSSIBILITY OF SUCH DAMAGE.
 */

/* ------------------------------------------------------------------------- */

%name HTMLParser
%define PURE
%define DEBUG 1

%{

/* ------------------------------------------------------------------------- */

#ident "$Id: HTMLParser.y,v 1.14 1999/10/26 10:56:55 arno Exp $"

#include "html.h"
#include "HTMLParser.h"

// MIPS machines don't have "alloca()", so disable stack realloc'ing.
#ifdef mips
#define yyoverflow yyerror("parser stack overflow"), (void)
#endif

/* ------------------------------------------------------------------------- */

%}

/* ------------------------------------------------------------------------- */

%define LEX_BODY = 0
%define ERROR_BODY = 0
%define MEMBERS\
  virtual ~HTMLParser(); \
  virtual void process(const Document &) = 0;\
  virtual bool read_cdata(const char *terminal, string *) = 0;\
  int list_nesting;
%define CONSTRUCTOR_INIT : list_nesting(0)

%union {
  Document                   *document;
  Element                    *element;
  list<auto_ptr<Element> >   *element_list;
  PCData                     *pcdata;
  string                     *strinG;
  list<TagAttribute>         *tag_attributes;
  int                        inT;
  list<auto_ptr<TableRow> >  *table_rows;
  list<auto_ptr<TableCell> > *table_cells;
  ListItem                   *list_item;
  list<auto_ptr<ListItem> >  *list_items;
  Caption                    *caption;
  Heading                    *heading;
  list<auto_ptr<Option> >    *option_list;
  Option                     *option;
  DefinitionList             *definition_list;
  list<auto_ptr<DefinitionListItem> > *definition_list_item_list;
  TermName                   *term_name;
  TermDefinition             *term_definition;
  Preformatted               *preformatted;
  Address                    *address;
  list<auto_ptr<list<TagAttribute> > > *tag_attributes_list;
}

%type  <document>                 document_
%type  <pcdata>                   pcdata
%type  <pcdata>                   opt_pcdata
%type  <element_list>             body_content
%type  <heading>                  heading
%type  <heading>                  HX
%type  <inT>                      END_HX
%type  <element>                  block
%type  <element>                  block_except_p
%type  <element>                  text
%type  <element_list>             texts
%type  <element_list>             opt_texts
%type  <element>                  font
%type  <element>                  phrase
%type  <element>                  special
%type  <element>                  form
%type  <table_rows>               table_rows
%type  <table_cells>              table_cells
%type  <caption>                  caption
%type  <caption>                  opt_caption
%type  <element_list>             applet_content
%type  <definition_list>          definition_list
%type  <definition_list_item_list>definition_list_content
%type  <term_name>                term_name
%type  <term_definition>          term_definition
%type  <option_list>              select_content
%type  <option>                   option
%type  <element>                  list
%type  <list_items>               list_content
%type  <list_item>                list_item
%type  <preformatted>             preformatted
%type  <element_list>             opt_flow
%type  <element_list>             flow
%type  <element>                  flow_
%type  <element_list>             paragraph_content
%type  <address>                  address
%type  <tag_attributes_list>      map_content

%type  <tag_attributes> opt_LI
%type  <tag_attributes> opt_P

%token                  DOCTYPE
%token <strinG>         PCDATA
%token                  SCAN_ERROR


%token <tag_attributes> A
%token <tag_attributes> ADDRESS
%token <tag_attributes> APPLET
%token <tag_attributes> AREA
%token <tag_attributes> B
%token <tag_attributes> BASE
%token <tag_attributes> BASEFONT
%token <tag_attributes> BIG
%token <tag_attributes> BLOCKQUOTE
%token <tag_attributes> BODY
%token <tag_attributes> BR
%token <tag_attributes> CAPTION
%token <tag_attributes> CENTER
%token <tag_attributes> CITE
%token <tag_attributes> CODE
%token <tag_attributes> DD
%token <tag_attributes> DFN
%token <tag_attributes> DIR
%token <tag_attributes> DIV
%token <tag_attributes> DL
%token <tag_attributes> DT
%token <tag_attributes> EM
%token <tag_attributes> FONT
%token <tag_attributes> FORM
%token <tag_attributes> H1
%token <tag_attributes> H2
%token <tag_attributes> H3
%token <tag_attributes> H4
%token <tag_attributes> H5
%token <tag_attributes> H6
%token <tag_attributes> HEAD
%token <tag_attributes> HR
%token <tag_attributes> HTML
%token <tag_attributes> I
%token <tag_attributes> IMG
%token <tag_attributes> INPUT
%token <tag_attributes> ISINDEX
%token <tag_attributes> KBD
%token <tag_attributes> LI
%token <tag_attributes> LINK
%token <tag_attributes> MAP
%token <tag_attributes> MENU
%token <tag_attributes> META
%token <tag_attributes> NOBR
%token <tag_attributes> OL
%token <tag_attributes> OPTION
%token <tag_attributes> P
%token <tag_attributes> PARAM
%token <tag_attributes> PRE
%token <tag_attributes> SAMP
%token <tag_attributes> SCRIPT
%token <tag_attributes> SELECT
%token <tag_attributes> SMALL
%token <tag_attributes> STRIKE
%token <tag_attributes> STRONG
%token <tag_attributes> STYLE
%token <tag_attributes> SUB
%token <tag_attributes> SUP
%token <tag_attributes> TABLE
%token <tag_attributes> TD
%token <tag_attributes> TEXTAREA
%token <tag_attributes> TH
%token <tag_attributes> TITLE
%token <tag_attributes> TR
%token <tag_attributes> TT
%token <tag_attributes> U
%token <tag_attributes> UL
%token <tag_attributes> VAR

%token                  END_A
%token                  END_ADDRESS
%token                  END_APPLET
%token                  END_B
%token                  END_BIG
%token                  END_BLOCKQUOTE
%token                  END_BODY
%token                  END_CAPTION
%token                  END_CENTER
%token                  END_CITE
%token                  END_CODE
%token                  END_DD
%token                  END_DFN
%token                  END_DIR
%token                  END_DIV
%token                  END_DL
%token                  END_DT
%token                  END_EM
%token                  END_FONT
%token                  END_FORM
%token                  END_H1
%token                  END_H2
%token                  END_H3
%token                  END_H4
%token                  END_H5
%token                  END_H6
%token                  END_HEAD
%token                  END_HTML
%token                  END_I
%token                  END_KBD
%token                  END_LI
%token                  END_MAP
%token                  END_MENU
%token                  END_NOBR
%token                  END_OL
%token                  END_OPTION
%token                  END_P
%token                  END_PRE
%token                  END_SAMP
%token                  END_SCRIPT
%token                  END_SELECT
%token                  END_SMALL
%token                  END_STRIKE
%token                  END_STRONG
%token                  END_STYLE
%token                  END_SUB
%token                  END_SUP
%token                  END_TABLE
%token                  END_TD
%token                  END_TEXTAREA
%token                  END_TH
%token                  END_TITLE
%token                  END_TR
%token                  END_TT
%token                  END_U
%token                  END_UL
%token                  END_VAR

/* ------------------------------------------------------------------------- */

%start document

%% /* { */

document:
  document_ {
    process(*$1);
    delete $1;
  }
  ;

/*
 * Well... actually, an HTML document should look like
 *
 * <!DOCTYPE ...>
 * <HTML>
 *   <HEAD>
 *   ...
 *   </HEAD>
 *   <BODY>
 *   ...
 *   </BODY>
 * </HTML>
 *
 * but...
 *
 * (A) All seven tags are optional
 * (B) The contents of the HEAD and the BODY section can be distinuished
 * (C) Most people out there do not know which element to put before, into,
 *     or after which section...
 *
 * so... let's just forget about the structure of an HTML document, discard
 * the seven tags, and process the remainder as a series of sections.
 */

document_:
  /* empty */ {
    $$ = new Document;
    $$->body.content.reset(new list<auto_ptr<Element> >);
  }
  | document_ error {
    $$ = $1;
  }
  | document_ DOCTYPE {
    $$ = $1;
  }
  | document_ HTML {
    $$->attributes.reset($2);
    $$ = $1;
  }
  | document_ END_HTML {
    $$ = $1;
  }
  | document_ HEAD {
    delete $2;
    $$ = $1;
  }
  | document_ END_HEAD {
    $$ = $1;
  }
  | document_ TITLE opt_pcdata opt_END_TITLE {
    delete $2; // Ignore <TITLE> attributes
    ($$ = $1)->head.title.reset($3);
  }
  | document_ ISINDEX {
    ($$ = $1)->head.isindex_attributes.reset($2);
  }
  | document_ BASE {
    ($$ = $1)->head.base_attributes.reset($2);
  }
  | document_ META {
    ($$ = $1)->head.meta_attributes.reset($2);
  }
  | document_ LINK {
    ($$ = $1)->head.link_attributes.reset($2);
  }
  | document_ SCRIPT {
    auto_ptr<Script> s(new Script);
    s->attributes.reset($2);
    if (!read_cdata("</SCRIPT>", &s->text)) {
      yyerror("CDATA terminal not found");
    }
    ($$ = $1)->head.scripts.push_back(s);
  }
  | document_ STYLE {
    auto_ptr<Style> s(new Style);
    s->attributes.reset($2);
    if (!read_cdata("</STYLE>", &s->text)) {
      yyerror("CDATA terminal not found");
    }
    ($$ = $1)->head.styles.push_back(s);
  }
  | document_ BODY {
    delete $2;
    $$ = $1;
  }
  | document_ END_BODY {
    $$ = $1;
  }
  | document_ texts {
    Paragraph *p = new Paragraph;
    p->texts.reset($2);
    ($$ = $1)->body.content->push_back(auto_ptr<Element>(p));
  }
  | document_ heading {
    ($$ = $1)->body.content->push_back(auto_ptr<Element>($2));
  }
  | document_ block {
    ($$ = $1)->body.content->push_back(auto_ptr<Element>($2));
  }
  | document_ address {
    ($$ = $1)->body.content->push_back(auto_ptr<Element>($2));
  }
  ;

pcdata:
  PCDATA {
    $$ = new PCData;
    $$->text = *$1;
    delete $1;
  }
  ;

body_content:
  /* empty */ {
    $$ = new list<auto_ptr<Element> >;
  }
  | body_content error {
    $$ = $1;
  }
  | body_content texts {
    Paragraph *p = new Paragraph;
    p->texts = auto_ptr<list<auto_ptr<Element> > >($2);
    ($$ = $1)->push_back(auto_ptr<Element>(p));
  }
  | body_content heading {
    ($$ = $1)->push_back(auto_ptr<Element>($2));
  }
  | body_content block {
    ($$ = $1)->push_back(auto_ptr<Element>($2));
  }
  | body_content address {
    ($$ = $1)->push_back(auto_ptr<Element>($2));
  }
  ;

heading:
  HX paragraph_content END_HX {
            /* EXTENSION: Allow paragraph content in heading, not only texts */
    if ($1->level != $3) {
      yyerror ("Levels of opening and closing headings don't match");
    }
    $$ = $1;
    $$->content.reset($2);
  }
  ;

block:
  block_except_p {
    $$ = $1;
  }
  | P paragraph_content opt_END_P {
    Paragraph *p = new Paragraph;
    p->attributes.reset($1);
    p->texts.reset($2);
    $$ = p;
  }
  ;

paragraph_content:  /* EXTENSION: Allow blocks (except "<P>") in paragraphs. */
  /* empty */ {
    $$ = new list<auto_ptr<Element> >;
  }
  | paragraph_content error {
    $$ = $1;
  }
  | paragraph_content texts {
    $$ = $1;
    $$->splice($$->end(), *$2);
    delete $2;
  }
  | paragraph_content block_except_p {
    ($$ = $1)->push_back(auto_ptr<Element>($2));
  }
  ;

block_except_p:
  list {
    $$ = $1;
  }
  | preformatted {
    $$ = $1;
  }
  | definition_list {
    $$ = $1;
  }
  | DIV body_content END_DIV {
    Division *p = new Division;
    p->attributes.reset($1);
    p->body_content.reset($2);
    $$ = p;
  }
  | CENTER body_content opt_END_CENTER {
    Center *p = new Center;
    delete $1;       // CENTER has no attributes.
    p->body_content.reset($2);
    $$ = p;
  }
  | BLOCKQUOTE body_content END_BLOCKQUOTE {
    delete $1; // BLOCKQUOTE has no attributes!
    BlockQuote *bq = new BlockQuote;
    bq->content.reset($2);