diff options
author | dimitri <dimitri@afe2bf4a-e733-0410-8a33-86f594647bc7> | 2002-11-28 20:38:57 (GMT) |
---|---|---|
committer | dimitri <dimitri@afe2bf4a-e733-0410-8a33-86f594647bc7> | 2002-11-28 20:38:57 (GMT) |
commit | f3e7209fed9703e8be668753d1083ba335121a6e (patch) | |
tree | 229ba1f0bd2e721f4cc81307b63d8c79051d1bc4 /src/doctokenizer.l | |
parent | e952dab58265a6f21867e51f72d7d81ffe39e082 (diff) | |
download | Doxygen-f3e7209fed9703e8be668753d1083ba335121a6e.zip Doxygen-f3e7209fed9703e8be668753d1083ba335121a6e.tar.gz Doxygen-f3e7209fed9703e8be668753d1083ba335121a6e.tar.bz2 |
Release-1.3-rc1-20021128
Diffstat (limited to 'src/doctokenizer.l')
-rw-r--r-- | src/doctokenizer.l | 165 |
1 files changed, 155 insertions, 10 deletions
diff --git a/src/doctokenizer.l b/src/doctokenizer.l index 879ca2c..be5db69 100644 --- a/src/doctokenizer.l +++ b/src/doctokenizer.l @@ -27,16 +27,30 @@ #include "cmdmapper.h" #include "config.h" #include "message.h" +#include "section.h" +#include "membergroup.h" +#include "definition.h" +#include "page.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 PageInfo *g_pageInfo; +static Definition *g_definition; +static MemberGroup *g_memberGroup; +static QCString g_secLabel; +static QCString g_secTitle; +static SectionInfo::SectionType g_secType; struct DocLexerContext { @@ -189,6 +203,41 @@ static void parseHtmlAttribs(const char *att) //-------------------------------------------------------------------------- +static void processSection() +{ + //printf("found section/anchor with name `%s'\n",g_secLabel.data()); + QCString file; + if (g_memberGroup) + { + file = g_memberGroup->parent()->getOutputFileBase(); + } + else if (g_definition) + { + file = g_definition->getOutputFileBase(); + } + else if (g_pageInfo) + { + file = g_pageInfo->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))==0) + { + si = new SectionInfo(file,g_secLabel,g_secTitle,g_secType); + Doxygen::sectionDict.insert(g_secLabel,si); + } + else if (!si->generated) + { + warn(g_fileName,yylineno,"Duplicate section/anchor label %s found!\n", + g_secLabel.data()); + } +} + +//-------------------------------------------------------------------------- + #undef YY_INPUT #define YY_INPUT(buf,result,max_size) result=yyread(buf,max_size); @@ -237,14 +286,15 @@ OPMASK ({BLANK}*{OPNORM}({FUNCARG}?))|({OPCAST}{FUNCARG}) LNKWORD1 ("::"|"#")?{SCOPEMASK} CVSPEC {BLANK}*("const"|"volatile") LNKWORD2 {SCOPEPRE}*"operator"{OPMASK} -WORD1 [^ \t\n\r\\@<>{}&$#,.]+|"{"|"}" -WORD2 "."|"," -WORD1NQ [^ \t\n\r\\@<>{}&$#,."]+ -WORD2NQ "."|"," +WORD1 [^ \t\n\r\\@<>(){}&$#,.]+|"{"|"}" +WORD2 "."|","|"("|")" +WORD1NQ [^ \t\n\r\\@<>(){}&$#,."]+ +WORD2NQ "."|","|"("|")" HTMLTAG "<"(("/")?){ID}({WS}+{ATTRIB})*">" HTMLKEYL "strong"|"center"|"table"|"caption"|"small"|"code"|"dfn"|"var"|"img"|"pre"|"sub"|"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"|"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\-]* %option noyywrap %option yylineno @@ -269,6 +319,12 @@ HTMLKEYW {HTMLKEYL}|{HTMLKEYU} %x St_Ref2 %x St_IntRef %x St_Text +%x St_SkipTitle + +%x St_Sections +%s St_SecLabel1 +%s St_SecLabel2 +%s St_SecTitle %% <St_Para>\r /* skip carriage return */ @@ -315,12 +371,14 @@ HTMLKEYW {HTMLKEYL}|{HTMLKEYU} g_token->name = yytext+1; return TK_COMMAND; } -<St_Para>("http:"|"https:"|"ftp:"|"file:"|"news:"){URLMASK} { +<St_Para>("http:"|"https:"|"ftp:"|"file:"|"news:"){URLMASK} { // URL g_token->name=yytext; + g_token->isEMailAddr=FALSE; return TK_URL; } -<St_Para>[a-z_A-Z0-9.-]+"@"[a-z_A-Z0-9-]+"."[a-z_A-Z0-9.-]+ { +<St_Para>[a-z_A-Z0-9.-]+"@"[a-z_A-Z0-9-]+"."[a-z_A-Z0-9.-]+ { // Mail address g_token->name=yytext; + g_token->isEMailAddr=TRUE; return TK_URL; } <St_Para>"$"{ID}":"[^\n$]+"$" { /* RCS tag */ @@ -408,8 +466,19 @@ HTMLKEYW {HTMLKEYL}|{HTMLKEYU} return TK_COMMAND; } <St_Para>({BLANK}*\n)+{BLANK}*\n { - /* start of a new paragraph */ - return TK_NEWPARA; + if (g_insidePre) + { + /* Inside a <pre>..</pre> blank lines are treated + * as whitespace. + */ + g_token->chars=yytext; + return TK_WHITESPACE; + } + else + { + /* start of a new paragraph */ + return TK_NEWPARA; + } } <St_Code>{CMD}"endcode" { return RetVal_OK; @@ -621,6 +690,56 @@ HTMLKEYW {HTMLKEYL}|{HTMLKEYU} } <St_Comment>[^-\n]+ /* inside html comment */ <St_Comment>. /* inside html comment */ + + /* State for skipping title (all chars until the end of the line) */ + +<St_SkipTitle>. +<St_SkipTitle>\n { return 0; } + + /* State for the pass used to find the anchors and sections */ + +<St_Sections>[^\n@\\]+ +<St_Sections>"@@"|"\\\\" +<St_Sections>{CMD}"anchor"{BLANK}+ { + g_secType = SectionInfo::Anchor; + BEGIN(St_SecLabel1); + } +<St_Sections>{CMD}"section"{BLANK}+ { + g_secType = SectionInfo::Section; + BEGIN(St_SecLabel2); + } +<St_Sections>{CMD}"subsection"{BLANK}+ { + g_secType = SectionInfo::Subsection; + BEGIN(St_SecLabel2); + } +<St_Sections>{CMD}"subsubsection"{BLANK}+ { + g_secType = SectionInfo::Subsubsection; + BEGIN(St_SecLabel2); + } +<St_Sections>{CMD}"paragraph"{BLANK}+ { + g_secType = SectionInfo::Paragraph; + BEGIN(St_SecLabel2); + } +<St_Sections>. +<St_Sections>\n +<St_SecLabel1>{LABELID} { + g_secLabel = yytext; + processSection(); + BEGIN(St_Sections); + } +<St_SecLabel2>{LABELID}{BLANK}+ { + g_secLabel = yytext; + g_secLabel = g_secLabel.stripWhiteSpace(); + BEGIN(St_SecTitle); + } +<St_SecTitle>[^\n]*\n { + g_secTitle = yytext; + g_secTitle = g_secTitle.stripWhiteSpace(); + processSection(); + BEGIN(St_Sections); + } + + /* Generic rules that work for all states */ <*>\n { warn(g_fileName,yylineno,"Error: Unexpected new line character"); } @@ -636,11 +755,27 @@ HTMLKEYW {HTMLKEYL}|{HTMLKEYU} //-------------------------------------------------------------------------- +void doctokenizerYYFindSections(const char *input,PageInfo *pi,Definition *d, + MemberGroup *mg) +{ + if (input==0) return; + g_inputString = input; + //printf("parsing --->`%s'<---\n",input); + g_inputPos = 0; + g_pageInfo = pi; + g_definition = d; + g_memberGroup = mg; + BEGIN(St_Sections); + doctokenizerYYlineno = 1; + doctokenizerYYlex(); +} + void doctokenizerYYinit(const char *input,const char *fileName) { g_inputString = input; - g_inputPos = 0; - g_fileName = fileName; + g_inputPos = 0; + g_fileName = fileName; + g_insidePre = FALSE; BEGIN(St_Para); } @@ -718,11 +853,21 @@ void doctokenizerYYsetStateText() BEGIN(St_Text); } +void doctokenizerYYsetStateSkipTitle() +{ + BEGIN(St_SkipTitle); +} + void doctokenizerYYcleanup() { yy_delete_buffer( YY_CURRENT_BUFFER ); } +void doctokenizerYYsetInsidePre(bool b) +{ + g_insidePre = b; +} + extern "C" { // some bogus code to keep the compiler happy void doctokenizerYYdummy() { yy_flex_realloc(0,0); } } |