/******************************************************************************

 *

 * 

 *

 *

 * Copyright (C) 1997-2008 by Dimitri van Heesch.

 *

 * Permission to use, copy, modify, and distribute this software and its

 * documentation under the terms of the GNU General Public License is hereby 

 * granted. No representations are made about the suitability of this software 

 * for any purpose. It is provided "as is" without express or implied warranty.

 * See the GNU General Public License for more details.

 *

 * Documents produced by Doxygen are derivative works derived from the

 * input used in their production; they are not affected by this license.

 *

 */

%{

#include <qfile.h>

#include <qstring.h>

#include <qstack.h>

#include <qdict.h>

#include "doctokenizer.h"

#include "cmdmapper.h"

#include "config.h"

#include "message.h"

#include "section.h"

#include "membergroup.h"

#include "definition.h"

#include "doxygen.h"

#include "portable.h"

#define YY_NEVER_INTERACTIVE 1

//--------------------------------------------------------------------------

// context for tokenizer phase

static int g_commentState;

TokenInfo *g_token = 0;

static int g_inputPos = 0;

static const char *g_inputString;

static QString g_fileName;

static bool g_insidePre;

// context for section finding phase

static Definition  *g_definition;

static MemberGroup *g_memberGroup;

static QCString     g_secLabel;

static QCString     g_secTitle;

static SectionInfo::SectionType g_secType;

static QCString     g_endMarker;

struct DocLexerContext

{

  TokenInfo *token;

  int rule;

  int inputPos;

  const char *inputString;

  YY_BUFFER_STATE state;

};

static QStack<DocLexerContext> g_lexerStack;

//--------------------------------------------------------------------------

void doctokenizerYYpushContext()

{

  DocLexerContext *ctx = new DocLexerContext;

  ctx->rule = YY_START;

  ctx->token = g_token;

  ctx->inputPos = g_inputPos;

  ctx->inputString = g_inputString;

  ctx->state = YY_CURRENT_BUFFER;

  g_lexerStack.push(ctx);

  yy_switch_to_buffer(yy_create_buffer(doctokenizerYYin, YY_BUF_SIZE));

}

bool doctokenizerYYpopContext()

{

  if (g_lexerStack.isEmpty()) return FALSE;

  DocLexerContext *ctx = g_lexerStack.pop();

  g_inputPos = ctx->inputPos;

  g_inputString = ctx->inputString;

  yy_delete_buffer(YY_CURRENT_BUFFER);

  yy_switch_to_buffer(ctx->state);

  BEGIN(ctx->rule);

  delete ctx;

  return TRUE;

}

//--------------------------------------------------------------------------

const char *tokToString(int token)

{

  switch (token)

  {

    case 0:              return "TK_EOF";

    case TK_WORD:        return "TK_WORD";

    case TK_LNKWORD:     return "TK_LNKWORD";

    case TK_WHITESPACE:  return "TK_WHITESPACE";

    case TK_LISTITEM:    return "TK_LISTITEM";

    case TK_ENDLIST:     return "TK_ENDLIST";

    case TK_COMMAND:     return "TK_COMMAND";

    case TK_HTMLTAG:     return "TK_HTMLTAG";

    case TK_SYMBOL:      return "TK_SYMBOL";

    case TK_NEWPARA:     return "TK_NEWPARA";

    case TK_RCSTAG:      return "TK_RCSTAG";

    case TK_URL:         return "TK_URL";

  }

  return "ERROR";

}

static int computeIndent(const char *str,int length)

{

  int i;

  int indent=0;

  int tabSize=Config_getInt("TAB_SIZE");

  for (i=0;i<length;i++)

  {

    if (str[i]=='\t')

    {

      indent+=tabSize - (indent%tabSize);

    }

    else if (str[i]=='\n')

    {

      indent=0;

    }

    else

    {

      indent++;

    }

  }

  return indent;

}

//--------------------------------------------------------------------------

static void processSection()

{

  //printf("%s: found section/anchor with name `%s'\n",g_fileName.data(),g_secLabel.data());

  QCString file;

  if (g_memberGroup)

  {

    file = g_memberGroup->parent()->getOutputFileBase();

  }

  else if (g_definition)

  {

    file = g_definition->getOutputFileBase();

  }

  else

  {

    warn(g_fileName,yylineno,"Found section/anchor %s without context\n",g_secLabel.data()); 

  }

  SectionInfo *si=0;

  if ((si=Doxygen::sectionDict.find(g_secLabel)))

  {

    si->fileName = file;

    //si = new SectionInfo(file,g_secLabel,g_secTitle,g_secType);

    //Doxygen::sectionDict.insert(g_secLabel,si);

  }

}

static void handleHtmlTag()

{

  QCString tagText=yytext;

  g_token->attribs.clear();

  g_token->endTag = FALSE;

  g_token->emptyTag = FALSE;

  // Check for end tag

  int startNamePos=1;

  if (tagText.at(1)=='/') 

  {

    g_token->endTag = TRUE;

    startNamePos++;

  }

  // Parse the name portion

  int i = startNamePos;

  for (i=startNamePos; i < yyleng; i++)

  {

    // Check for valid HTML/XML name chars (including namespaces)

    char c = tagText.at(i);

    if (!(isalnum(c) || c=='-' || c=='_' || c==':')) break;

  }

  g_token->name = tagText.mid(startNamePos,i-startNamePos);

  // Parse the attributes. Each attribute is a name, value pair

  // The result is stored in g_token->attribs.

  int startName,endName,startAttrib,endAttrib;

  while (i<yyleng)

  {

    char c=tagText.at(i);

    // skip spaces

    while (i<yyleng && isspace(c)) { c=tagText.at(++i); }

    // check for end of the tag

    if (c == '>') break;

    // Check for XML style "empty" tag.

    if (c == '/') 

    {

      g_token->emptyTag = TRUE;

      break;

    }

    startName=i;

    // search for end of name

    while (i<yyleng && !isspace(c) && c!='=') { c=tagText.at(++i); }

    endName=i;

    HtmlAttrib opt;

    opt.name  = tagText.mid(startName,endName-startName).lower(); 

    // skip spaces

    while (i<yyleng && isspace(c)) { c=tagText.at(++i); } 

    if (tagText.at(i)=='=') // option has value

    {

      c=tagText.at(++i);

      // skip spaces

      while (i<yyleng && isspace(c)) { c=tagText.at(++i); } 

      if (tagText.at(i)=='\'') // option '...'

      {

        c=tagText.at(++i);

	startAttrib=i;

	// search for matching quote 

        while (i<yyleng && c!='\'') { c=tagText.at(++i); } 

	endAttrib=i;

        if (i<yyleng) c=tagText.at(++i);

      }

      else if (tagText.at(i)=='"') // option "..."

      {

        c=tagText.at(++i);

	startAttrib=i;

	// search for matching quote 

        while (i<yyleng && c!='"') { c=tagText.at(++i); } 

	endAttrib=i;

        if (i<yyleng) c=tagText.at(++i);

      }

      else // value without any quotes

      {

	startAttrib=i;

	// search for separator or end symbol

        while (i<yyleng && !isspace(c) && c!='>') { c=tagText.at(++i); } 

	endAttrib=i;

	if (i<yyleng) c=tagText.at(++i);

      }

      opt.value  = tagText.mid(startAttrib,endAttrib-startAttrib); 

    }

    else // start next option

    {

    }

    //printf("=====> Adding option name=<%s> value=<%s>\n",

    //    opt.name.data(),opt.value.data());

    g_token->attribs.append(&opt);

  }

}

static QString stripEmptyLines(const char *s)

{

  int result=0,p=0;

  for (;;)

  {

    int c;

    while ((c=s[p]) && isspace(c)) p++;

    if (s[p]=='\n') result=++p; else break;

  }

  return &s[result];

}

//--------------------------------------------------------------------------

#undef  YY_INPUT

#define YY_INPUT(buf,result,max_size) result=yyread(buf,max_size);

static int yyread(char *buf,int max_size)

{

  int c=0;

  const char *src=g_inputString+g_inputPos;

  while ( c < max_size && *src ) *buf++ = *src++, c++;

  g_inputPos+=c;

  return c;

}

//--------------------------------------------------------------------------

%}

CMD   ("\\"|"@")

WS    [ \t\r\n]

NONWS [^ \t\r\n]

BLANK [ \t\r]

ID    "$"?[a-z_A-Z][a-z_A-Z0-9]*

MAILADR  [a-z_A-Z0-9.+-]+"@"[a-z_A-Z0-9-]+("."[a-z_A-Z0-9\-]+)+[a-z_A-Z0-9\-]+

OPTSTARS ("//"{BLANK}*)?"*"*{BLANK}*

LISTITEM {BLANK}*{OPTSTARS}"-"("#")?{WS}

ENDLIST  {BLANK}*{OPTSTARS}"."{BLANK}*\n

ATTRIB   {ID}{WS}*("="{WS}*(("\""[^\"]*"\"")|("'"[^\']*"'")|[^ \t\r\n'"><]+))?

URLCHAR   [a-z_A-Z0-9\!\~\,\:\;\'\$\?\@\&\%\#\.\-\+\/\=]

URLMASK   (([a-z_A-Z][^\>\"\n]*{URLCHAR})|({URLCHAR}+))([({]{URLCHAR}*[)}])?

FILESCHAR [a-z_A-Z0-9\\:\\\/\-\+]

FILEECHAR [a-z_A-Z0-9\-\+]

HFILEMASK ("."{FILESCHAR}*{FILEECHAR}+)*

FILEMASK  ({FILESCHAR}*{FILEECHAR}+("."{FILESCHAR}*{FILEECHAR}+)*)|{HFILEMASK}

LINKMASK  [^ \t\n\r\\@<&${}]+("("[^\n)]*")")?({BLANK}*("const"|"volatile"){BLANK}+)? 

VERBATIM  "verbatim"{BLANK}*

SPCMD1    {CMD}([a-z_A-Z0-9]+|{VERBATIM})

SPCMD2    {CMD}[\\@<>&$#%~"]

SPCMD3    {CMD}form#[0-9]+

INOUT	  "in"|"out"|("in"{BLANK}*","{BLANK}*"out")|("out"{BLANK}*","{BLANK}*"in")

PARAMIO   {CMD}param{BLANK}*"["{BLANK}*{INOUT}{BLANK}*"]"

TEMPCHAR  [a-z_A-Z0-9,: \t\*\&]

FUNCCHAR  [a-z_A-Z0-9,:\<\> \t\*\&]

SCOPESEP  "::"|"#"|"."

TEMPLPART "<"{TEMPCHAR}*">"

SCOPEPRE  {ID}{TEMPLPART}?{SCOPESEP}

SCOPEKEYS ":"({ID}":")*

SCOPECPP  {SCOPEPRE}*(~)?{ID}("<"{TEMPCHAR}*">")?

SCOPEOBJC {SCOPEPRE}?{ID}{SCOPEKEYS}?

SCOPEMASK {SCOPECPP}|{SCOPEOBJC}

FUNCARG   "("{FUNCCHAR}*")"({BLANK}*("volatile"|"const"){BLANK})?

OPNEW     {BLANK}+"new"({BLANK}*"[]")?

OPDEL     {BLANK}+"delete"({BLANK}*"[]")?

OPNORM    {OPNEW}|{OPDEL}|"+"|"-"|"*"|"/"|"%"|"^"|"&"|"|"|"~"|"!"|"="|"<"|">"|"+="|"-="|"*="|"/="|"%="|"^="|"&="|"|="|"<<"|">>"|"<<="|">>="|"=="|"!="|"<="|">="|"&&"|"||"|"++"|"--"|","|"->*"|"->"|"[]"|"()"

OPCAST    {BLANK}+[^<(\r\n.,][^(\r\n.,]*

OPMASK    ({BLANK}*{OPNORM}{FUNCARG})

OPMASKOPT ({BLANK}*{OPNORM}{FUNCARG}?)|({OPCAST}{FUNCARG})

LNKWORD1  ("::"|"#")?{SCOPEMASK}

CVSPEC    {BLANK}*("const"|"volatile")

LNKWORD2  ({SCOPEPRE}*"operator"{OPMASK})|({SCOPEPRE}"operator"{OPMASKOPT})|(("::"|"#"){SCOPEPRE}*"operator"{OPMASKOPT})

LNKWORD3  ([0-9a-z_A-Z\-]+("/"|"\\"))*[0-9a-z_A-Z\-]+("."[0-9a-z_A-Z]+)+

CHARWORDQ [^ \t\n\r\\@<>()\[\]:;\?{}&%$#,."]

WORD1     "%"?{CHARWORDQ}+|"{"|"}"|("\""[^"\n]*\n?[^"\n]*"\"")

WORD2     "."|","|"("|")"|"["|"]"|":"|";"|"\?"

WORD1NQ   "%"?{CHARWORDQ}+

WORD2NQ   "."|","|"("|")"|"["|"]"|":"|";"|"\?"

HTMLTAG   "<"(("/")?){ID}({WS}+{ATTRIB})*{WS}*(("/")?)">" 

HTMLKEYL  "strong"|"center"|"table"|"caption"|"small"|"code"|"dfn"|"var"|"img"|"pre"|"sub"|"sup"|"tr"|"td"|"th"|"ol"|"ul"|"li"|"tt"|"kbd"|"em"|"hr"|"dl"|"dt"|"dd"|"br"|"i"|"a"|"b"|"p"

HTMLKEYU  "STRONG"|"CENTER"|"TABLE"|"CAPTION"|"SMALL"|"CODE"|"DFN"|"VAR"|"IMG"|"PRE"|"SUB"|"SUP"|"TR"|"TD"|"TH"|"OL"|"UL"|"LI"|"TT"|"KBD"|"EM"|"HR"|"DL"|"DT"|"DD"|"BR"|"I"|"A"|"B"|"P"

HTMLKEYW  {HTMLKEYL}|{HTMLKEYU}

LABELID   [a-z_A-Z][a-z_A-Z0-9\-]*

REFWORD2  ("#"|"::")?({ID}{TEMPLPART}?("."|"#"|"::"|"-"|"/"))*({ID}(":")?){FUNCARG}?

REFWORD   {LABELID}|{REFWORD2}

%option noyywrap

%option yylineno

%x St_Para

%x St_Comment

%x St_Title

%x St_TitleN

%x St_TitleQ

%x St_TitleA

%x St_TitleV

%x St_Code

%x St_XmlCode

%x St_HtmlOnly

%x St_ManOnly

%x St_LatexOnly

%x St_XmlOnly

%x St_Verbatim

%x St_Dot

%x St_Msc

%x St_Param

%x St_XRefItem

%x St_XRefItem2

%x St_File

%x St_Pattern

%x St_Link

%x St_Ref

%x St_Ref2

%x St_IntRef

%x St_Text

%x St_SkipTitle

%x St_Sections

%s St_SecLabel1

%s St_SecLabel2

%s St_SecTitle

%x St_SecSkip

%%

<St_Para>\r               /* skip carriage return */

<St_Para>^{LISTITEM}      { /* list item */ 

                         QString text=yytext;

			 int dashPos = text.findRev('-');

			 g_token->isEnumList = text.at(dashPos+1)=='#';

			 g_token->indent     = computeIndent(yytext,dashPos);

                         return TK_LISTITEM;

                       }

<St_Para>{BLANK}*\n{LISTITEM}     { /* list item on next line */ 

                         QString text=yytext;

			 text=text.right(text.length()-text.find('\n')-1);

			 int dashPos = text.findRev('-');

			 g_token->isEnumList = text.at(dashPos+1)=='#';

			 g_token->indent     = computeIndent(text,dashPos);

                         return TK_LISTITEM;

                       }

<St_Para>^{ENDLIST}       { /* end list */ 

                         int dotPos = QString(yytext).findRev('.');

			 g_token->indent     = computeIndent(yytext,dotPos);

                         return TK_ENDLIST;

                       }

<St_Para>{BLANK}*\n{ENDLIST}      { /* end list on next line */ 

                         QString text=yytext;

			 text=text.right(text.length()-text.find('\n')-1);

                         int dotPos = text.findRev('.');

			 g_token->indent     = computeIndent(text,dotPos);

                         return TK_ENDLIST;

                       }

<St_Para>"{"{BLANK}*"@link" {

  			 g_token->name = "javalink";

			 return TK_COMMAND;

  		       }

<St_Para>"{"{BLANK}*"@inheritDoc"{BLANK}*"}" {

  			 g_token->name = "inheritdoc";

			 return TK_COMMAND;

		       }

<St_Para>"\\_fakenl"     { // artificial new line

                             yylineno++; 

  		          }

<St_Para>{SPCMD3}      {

  			 g_token->name = "form";

			 bool ok;

			 g_token->id = QString(yytext).right(yyleng-6).toInt(&ok);

			 ASSERT(ok);

			 return TK_COMMAND;

  		       }

<St_Para>{SPCMD1}      |

<St_Para>{SPCMD2}      { /* special command */

                         g_token->name = yytext+1;

			 g_token->name = g_token->name.stripWhiteSpace();

			 g_token->paramDir=TokenInfo::Unspecified;

                         return TK_COMMAND;

  		       }

<St_Para>{PARAMIO}     { /* param [in,out] command */

  			 g_token->name = "param";

			 QString s(yytext);

			 bool isIn  = s.find("in")!=-1;

			 bool isOut = s.find("out")!=-1;

			 if (isIn)

			 {

			   if (isOut)

			   {

			     g_token->paramDir=TokenInfo::InOut;

			   }

			   else

			   {

			     g_token->paramDir=TokenInfo::In;

			   }

			 }

			 else if (isOut)

			 {

			   g_token->paramDir=TokenInfo::Out;

			 }

			 else

			 {

			   g_token->paramDir=TokenInfo::Unspecified;

			 }

			 return TK_COMMAND;

  		       }

<St_Para>("http:"|"https:"|"ftp:"|"file:"|"news:"){URLMASK} { // URL

                         g_token->name=yytext;

			 g_token->isEMailAddr=FALSE;

			 return TK_URL;

                       }

<St_Para>{MAILADR}     { // Mail address

                         g_token->name=yytext;

			 g_token->isEMailAddr=TRUE;

			 return TK_URL;

                       }

<St_Para>"$"{ID}":"[^\n$]+"$" { /* RCS tag */

                         QString tagName(yytext+1);

			 int index=tagName.find(':');

  			 g_token->name = tagName.left(index);

			 g_token->text = tagName.mid(index+1,tagName.length()-index-2);

			 return TK_RCSTAG;

  		       }

<St_Para,St_HtmlOnly>"$("{ID}")"   { /* environment variable */

                         QCString name = &yytext[2];

			 name = name.left(name.length()-1);

			 QCString value = portable_getenv(name);

			 for (int i=value.length()-1;i>=0;i--) unput(value.at(i));

                       }

<St_Para>{HTMLTAG}     { /* html tag */ 

                         handleHtmlTag();

                         return TK_HTMLTAG;

                       }

<St_Para,St_Text>"&"{ID}";" { /* special symbol */ 

                         g_token->name = yytext;

                         return TK_SYMBOL;

                       }

  /********* patterns for linkable words ******************/

<St_Para>{ID}/"<"{HTMLKEYW}">" { /* this rule is to prevent opening html 

				  * tag to be recognized as a templated classes 

				  */ 

                         g_token->name = yytext;

                         return TK_LNKWORD;

  			}

<St_Para>{LNKWORD1}/"<br>"           | // prevent <br> html tag to be parsed as template arguments

<St_Para>{LNKWORD1}                  |

<St_Para>{LNKWORD1}{FUNCARG}         |

<St_Para>{LNKWORD2}                  |

<St_Para>{LNKWORD3}    {

                         g_token->name = yytext;

                         return TK_LNKWORD;

  		       }

<St_Para>{LNKWORD1}{FUNCARG}{CVSPEC}[^a-z_A-Z0-9] {

                         g_token->name = yytext;

                         g_token->name = g_token->name.left(g_token->name.length()-1);

			 unput(yytext[yyleng-1]);

                         return TK_LNKWORD;

                       }

  /********* patterns for normal words ******************/

<St_Para,St_Text>{WORD1} |

<St_Para,St_Text>{WORD2} { /* function call */ 

                         if (yytext[0]=='%') // strip % if present

			   g_token->name = &yytext[1];

			 else

                           g_token->name = yytext;

                         return TK_WORD;

			 /* the following is dummy code to please the 

			  * compiler, removing this results in a warning 

			  * on my machine