# GRC changes
# 1 - handle comments not on line boundaries
# 2 - handle <define_symbol> and symbols in <headx> (etc) and handle <reference>
# 3 - handle <dec_security>
# 4 - handle <preface> <subhead1> <subhead2>
# 5 - handle <quote> <endquote> <keyword>
# 6 - handle <byline> <print_date> <copyright_date>
# 7 - handle tags within <headx> tags (can't make <title> work)
# 8 - fix <code_example> to handle "immediate" form
# 9 - add a load more tags to explicitly ignore
# 10 - add forward references, add <newterm>, add footnotes
# 11 - add alphabetic list type and handle unrecognised list types
# 12 - add space when using getline to handle long tag arguments
# 13 - handle figure tags
# 14 - handle <argument> <literal> <syntax>
# 15 - add example tags and <line_art>
# 16 - add contents support
# tomc@osi.curtin.edu.au (Tom Crawley) changes
# 17 - handle <include>

BEGIN   {
	    EOF = 0;
            title_page = "";
            copyright = "";
            footnotes = "";
            in_comment = 0; in_code_example = 0;
            IGNORECASE = 1; product = "";
            contents_needed = 0; contents = "<h1><a name=\"ContentsListing\">Contents</a></h1>\n";
            contents_level = 0; contents_list_type = "ul";
}

END     {
            if (title_page != "") print title_page;
            if (copyright_page != "") print copyright_page;
            for ( ; contents_level > 0; contents_level--) {
                contents = contents "</" contents_list_type ">\n";
            }
            if (contents_needed) print contents;
            if (footnotes != "") {
                print "<h1>Footnotes</h1>";
                print "<dl>";
                print footnotes;
                print "</dl>";
            }
}
#
# The main function has degenerated into a call to the recursive
# function do_file.
#
{
	do_file("");
	exit 0;
}
#
# The real work is done in the do_line function
# filename is used to pass the file name to the recursive do_file function
#
function do_line(filename) {
        oline = "";
        line = $0;
        if(line == "") # skip empty line
	    return;
# optimisation -- ignore complete line if it starts in a comment and does not end the comment
        if (in_comment) {
            if (index(tolower(line), "<endcomment>") == 0)
                return;
        }
        while (line != "") {
            lab = index(line, "<");
            if (lab == 0)
                break;
            rab = index(substr(line, lab + 1, length(line) - lab), ">") + lab;
            if (lab > 1 && !in_comment) {
                text = substr(line, 1, lab - 1);
                if (in_code_example == 0)
                    gsub("\\\\[A-Za-z0-9_]*", " ", text);
                oline = oline text;
                text = "";
            }
            tag = tolower(substr(line, lab + 1, rab - lab - 1));
            line = substr(line, rab + 1, length(line) - rab);
            if (tag == "")
                continue;
            if (substr(line, 1, 1) == "(") {
                balance = 1;
                text = "";
                line = substr(line, 2, length(line) - 1);
                rp = index(line, ")");
                lp = index(line, "(");
                if (rp == 0 || (lp > 0 && lp < rp)) {
                    parens = 1;
                    while (parens > 0) {
                        if (line == "") {
                            read_line(filename);
			    if (EOF)
				return;
                            line = " " $0;
                        }
                        c = substr(line, 1, 1);
                        if (c == ")") {
                            if (parens == 1) {
                                line = substr(line, 2, length(line) - 1);
                                break;
                            }
                            parens--;
                        } else if (c == "(") {
                            parens++;
                        }
                        text = text c;
                        line = substr(line, 2, length(line) - 1);
                    }
                } else {
                    text = substr(line, 1, rp - 1);
                    line = substr(line, rp + 1, length(line) - rp);
                }
            } else {
                text = "";
            }
            if (in_comment && tag != "endcomment") {
            } else if (substr(tag,1,1) == "/" || substr(tag,1,2) == "a ") {
# these have already been converted to html
                oline = oline "<" tag ">";
            } else if (tag == "p" || tag == "b" || tag == "/b") {
                oline = oline "<" tag ">";
            } else if (tag == "emphasis") {
                n = split(text, parts, "\\");
                if (n > 1) {
                    parts[2] = tolower(parts[2]);
                    if (parts[2] == "italic") {
                        oline = oline "<em>";
                        line = parts[1] "</em>" line;
                    } else if (parts[2] == "bold") {
                        oline = oline "<b>";
                        line = parts[1] "</b>" line;
                    } else {
                        oline = oline "<em>";
                        line = text "</em>" line;
                    }
                } else {
                    oline = oline "<em>";
                    line = text "</em>" line;
                }
            } else if (tag == "le") {
                if (list_type[nlists] == "ul" || list_type[nlists] == "ol")
                    oline = oline "<li>";
                else
                    oline = oline "<dt>";
            } else if (tag == "keep") {
                line = text " " line;
            } else if (tag == "center_line") {
                line = text "<p>" line ;
            } else if (tag == "ultrix_keyword") {
                oline = oline "<b>";
                line = text "</b>" line;
            } else if (tag == "chapter" || tag == "appendix") {
                n = split(text, parts, "\\");
                if (n > 1) {
                    parts[2] = tolower(parts[2]);
                    oline = oline "<h1><a name=\"" parts[2] "\"> ";
                    line = parts[1] "</a></h1>" line;
                    symbol[parts[2]] = "<a href=\"#" parts[2] "\">" parts[1] "</a>";
                    for ( ; contents_level < 1; contents_level++) {
                        contents = contents "<" contents_list_type ">\n";
                    }
                    for ( ; contents_level > 1; contents_level--) {
                        contents = contents "</" contents_list_type ">\n";
                    }
                    contents = contents "<li><a href=\"#" parts[2] "\">" parts[1] "</a>\n";
                } else {
                    oline = oline "<h1>";
                    line = text "</h1>" line;
                    for ( ; contents_level < 1; contents_level++) {
                        contents = contents "<" contents_list_type ">\n";
                    }
                    for ( ; contents_level > 1; contents_level--) {
                        contents = contents "</" contents_list_type ">\n";
                    }
                    contents = contents "<li>" text "\n";
                }
            } else if (tag == "head1") {
                n = split(text, parts, "\\");
                if (n > 1) {
                    parts[2] = tolower(parts[2]);
                    oline = oline "<h2><a name=\"" parts[2] "\"> ";
                    line = parts[1] "</a></h2>" line;
                    symbol[parts[2]] = "<a href=\"#" parts[2] "\">" parts[1] "</a>";
                    for ( ; contents_level < 2; contents_level++) {
                        contents = contents "<" contents_list_type ">\n";
                    }
                    for ( ; contents_level > 2; contents_level--) {
                        contents = contents "</" contents_list_type ">\n";
                    }
                    contents = contents "<li><a href=\"#" parts[2] "\">" parts[1] "</a>\n";
                } else {
                    oline = oline "<h2>";
                    line = text "</h2>" line;
                    for ( ; contents_level < 2; contents_level++) {
                        contents = contents "<" contents_list_type ">\n";
                    }
                    for ( ; contents_level > 2; contents_level--) {
                        contents = contents "</" contents_list_type ">\n";
                    }
                    contents = contents "<li>" text "\n";
                }
            } else if (tag == "head2") {
                n = split(text, parts, "\\");
                if (n > 1) {
                    parts[2] = tolower(parts[2]);
                    oline = oline "<h3><a name=\"" parts[2] "\"> ";
                    line = parts[1] "</a></h3>" line;
                    symbol[parts[2]] = "<a href=\"#" parts[2] "\">" parts[1] "</a>";
                    for ( ; contents_level < 3; contents_level++) {
                        contents = contents "<" contents_list_type ">\n";
                    }
                    for ( ; contents_level > 3; contents_level--) {
                        contents = contents "</" contents_list_type ">\n";
                    }
                    contents = contents "<li><a href=\"#" parts[2] "\">" parts[1] "</a>\n";
                } else {
                    oline = oline "<h3>";
                    line = text "</h3>" line;
                    for ( ; contents_level < 3; contents_level++) {
                        contents = contents "<" contents_list_type ">\n";
                    }
                    for ( ; contents_level > 3; contents_level--) {
                        contents = contents "</" contents_list_type ">\n";
                    }
                    contents = contents "<li>" text "\n";
                }
            } else if (tag == "head3") {
                n = split(text, parts, "\\");
                if (n > 1) {
                    parts[2] = tolower(parts[2]);
                    oline = oline "<h4><a name=\"" parts[2] "\"> ";
                    line = parts[1] "</a></h4>" line;
                    symbol[parts[2]] = "<a href=\"#" parts[2] "\">" parts[1] "</a>";
#                   for ( ; contents_level < 4; contents_level++) {
#                       contents = contents "<" contents_list_type ">\n";
#                   }
#                   for ( ; contents_level > 4; contents_level--) {
#                       contents = contents "</" contents_list_type ">\n";
#                   }
#                   contents = contents "<li><a href=\"#" parts[2] "\">" parts[1] "</a>\n";
                } else {
                    oline = oline "<h4>";
                    line = text "</h4>" line;
#                   for ( ; contents_level < 4; contents_level++) {
#                       contents = contents "<" contents_list_type ">\n";
#                   }
#                   for ( ; contents_level > 4; contents_level--) {
#                       contents = contents "</" contents_list_type ">\n";
#                   }
#                   contents = contents "<li>" text "\n";
                }
            } else if (tag == "product") {
                product = text " ";
            } else if (tag == "title") {
                print "<title>" product text "</title>";
                print "<h1><a name=\"TitleHeading\">" product text "</h1>";
                for ( ; contents_level < 1; contents_level++) {
                    contents = contents "<" contents_list_type ">\n";
                }
                for ( ; contents_level > 1; contents_level--) {
                    contents = contents "</" contents_list_type ">\n";
                }
                contents = contents "<li><a href=\"#TitleHeading\">" product text "</a>\n";
            } else if (tag == "list") {
                text = tolower(text);
                if (text == "unnumbered") {
                    list_type[++nlists] = "ul";
                } else if (text == "numbered" || text == "alphabetic") {
                    list_type[++nlists] = "ol";
                } else if (text == "simple") {
                    list_type[++nlists] = "dl";
                } else {
                    oline = oline "<!-- unrecognised list type " text " -->"
# unrecognised list types are treated as unordered
                    list_type[++nlists] = "ul";
                    }
                oline = oline "<" list_type[nlists] ">";
            } else if (tag == "endlist") {
                oline = oline "</" list_type[nlists--] ">";
            } else if (tag == "comment") {
                if (text == "")
                    in_comment = 1;
            } else if (tag == "endcomment") {
                in_comment = 0;
            } else if (tag == "include") {
                do_file(text);
            } else if (tag == "code_example") {
                if (text != "") {
                    oline = oline "<code>";
                    line = text "</code>" line;
                } else {
                    oline = oline "<tt><blockquote><pre>";
                    in_code_example = 1;
                }
            } else if (tag == "line_art" || tag == "syntax" || tag == "exc") {
                oline = oline "<tt><blockquote><pre>";
                in_code_example = 1;
            } else if (tag == "endcode_example" || tag == "endsyntax" || tag == "extext" || tag == "endline_art") {
                oline = oline "</pre></blockquote></tt>";
                in_code_example = 0;
            } else if (tag == "title_page") {
                on_title_page = 1;
                title_page = "<h1><a name=\"TitlePage\">Title Page</a></h1>\n";
            } else if (tag == "endtitle_page") {
                on_title_page = 0;
                print "To see the title page, <a href=\"#TitlePage\">click here</a>.\n";
            } else if (tag == "copyright_page") {
                on_copyright_page = 1;
                copyright_page = "<h1><a name=\"CopyrightInfo\">Copyright Information</a></h1>\n";
            } else if (tag == "endcopyright_page") {
                on_copyright_page = 0;
                print "To see the copyright information, <a href=\"#CopyrightInfo\">click here</a>.<p>\n";
            } else if (tag == "note" || tag == "example_sequence") {
                oline = oline "<blockquote><b>";
                if (text != "")
                    line = text "</b><p>" line;
                else
                    line = "Note: </b>" line;
            } else if (tag == "endnote" || tag == "endexample_sequence") {
                if (in_code_example) {
                    oline = oline "</pre></blockquote></tt>";
                    in_code_example = 0;
                }
                oline = oline "</blockquote>";
            } else if (tag == "table") {
                for (idx in table_data)
                    delete table[idx];
            } else if (tag == "table_heads") {
            } else if (tag == "table_row") {
            } else if (tag == "table") {
            } else if (tag == "table_end") {
            } else if (tag == "table_attributes") {
            } else if (tag == "table_setup") {
            } else if (tag == "table_row_break") {
            } else if (tag == "table_key") {
            } else if (tag == "endtable_key") {
            } else if (tag == "table_unit_heads") {
            } else if (tag == "table_unit") {
            } else if (tag == "endtable_unit") {
            } else if (tag == "endtable") {
            } else if (tag == "define_symbol" || tag == "define_book_name") {
                n = split(text, parts, "\\");
                if (n > 1) {
                    parts[1] = tolower(parts[1]);
                    symbol[parts[1]] = parts[2];
                    }
            } else if (tag == "reference") {
                split(text, parts, "\\");
                parts[1] = tolower(parts[1]);
                if (parts[1] in symbol) {
                    line = symbol[parts[1]] line;
                } else {
# we haven't seen the symbol yet -- assume it is a forward reference
                    oline = oline "<a href=\"#" parts[1] "\">click here</a>"
                }
            } else if (tag == "set_online_topic") {
            } else if (tag == "define") {
            } else if (tag == "document_attributes") {
            } else if (tag == "enddocument_attributes") {
            } else if (tag == "set_page_numbering") {
            } else if (tag == "revision") {
            } else if (tag == "dec_security") {
                    oline = oline "Digital security classification is <strong>" text "</strong>.<p>"
            } else if (tag == "front_matter") {
            } else if (tag == "preface") {
                n = split(text, parts, "\\");
                if (n > 1) {
                    parts[2] = tolower(parts[2]);
                    oline = oline "<h2><a name=\"" parts[2] "\">Preface</a></h2>";
                    symbol[parts[2]] = "<a href=\"#" parts[2] "\">Preface</a>";
                    for ( ; contents_level < 2; contents_level++) {
                        contents = contents "<" contents_list_type ">\n";
                    }
                    for ( ; contents_level > 2; contents_level--) {
                        contents = contents "</" contents_list_type ">\n";
                    }
                    contents = contents "<li><a href=\"#" parts[2] "\">Preface</a>\n";
                } else {
                    oline = oline "<h2>Preface</h2>";
                    for ( ; contents_level < 2; contents_level++) {
                        contents = contents "<" contents_list_type ">\n";
                    }
                    for ( ; contents_level > 2; contents_level--) {
                        contents = contents "</" contents_list_type ">\n";
                    }
                    contents = contents "<li>Preface\n";
                }
            } else if (tag == "subhead1") {
                n = split(text, parts, "\\");
                if (n > 1) {
                    parts[2] = tolower(parts[2]);
                    oline = oline "<h5><a name=\"" parts[2] "\"> ";
                    line = parts[1] "</a></h5>" line;
                    symbol[parts[2]] = "<a href=\"#" parts[2] "\">" parts[1] "</a>";
#                   for ( ; contents_level < 5; contents_level++) {
#                       contents = contents "<" contents_list_type ">\n";
#                   }
#                   for ( ; contents_level > 5; contents_level--) {
#                       contents = contents "</" contents_list_type ">\n";
#                   }
#                   contents = contents "<li><a href=\"#" parts[2] "\">" parts[1] "</a>\n";
                } else {
                    oline = oline "<h5>";
                    line = text "</h5>" line;
#                   for ( ; contents_level < 5; contents_level++) {
#                       contents = contents "<" contents_list_type ">\n";
#                   }
#                   for ( ; contents_level > 5; contents_level--) {
#                       contents = contents "</" contents_list_type ">\n";
#                   }
#                   contents = contents "<li>" text "\n";
                }
            } else if (tag == "subhead2") {
                n = split(text, parts, "\\");
                if (n > 1) {
                    parts[2] = tolower(parts[2]);
                    oline = oline "<h6><a name=\"" parts[2] "\"> ";
                    line = parts[1] "</a></h6>" line;
                    symbol[parts[2]] = "<a href=\"#" parts[2] "\">" parts[1] "</a>";
#                   for ( ; contents_level < 6; contents_level++) {
#                       contents = contents "<" contents_list_type ">\n";
#                   }
#                   for ( ; contents_level > 6; contents_level--) {
#                       contents = contents "</" contents_list_type ">\n";
#                   }
#                   contents = contents "<li><a href=\"#" parts[2] "\">" parts[1] "</a>\n";
                } else {
                    oline = oline "<h6>";
                    line = text "</h6>" line;
#                   for ( ; contents_level < 6; contents_level++) {
#                       contents = contents "<" contents_list_type ">\n";
#                   }
#                   for ( ; contents_level > 6; contents_level--) {
#                       contents = contents "</" contents_list_type ">\n";
#                   }
#                   contents = contents "<li>" text "\n";
                }
            } else if (tag == "mark") {
            } else if (tag == "endmark") {
            } else if (tag == "mark_deletion") {
            } else if (tag == "endpreface") {
            } else if (tag == "endfront_matter") {
            } else if (tag == "running_title") {
            } else if (tag == "quote") {
                if (text != "") {
                    line = "\"" text "\"" line;
                } else {
                    oline = oline "<blockquote>"
                }
            } else if (tag == "endquote") {
                oline = oline "</blockquote>"
            } else if (tag == "keyword" || tag == "newterm") {
                oline = oline "<dfn>";
                line = text "</dfn>" line;
            } else if (tag == "lmf") {
            } else if (tag == "lmf_producer") {
            } else if (tag == "lmf_product") {
            } else if (tag == "lmf_release_date") {
            } else if (tag == "lmf_version_number") {
            } else if (tag == "lmf_altname") {
            } else if (tag == "lmf_info") {
            } else if (tag == "endlmf") {
            } else if (tag == "date") {
            } else if (tag == "signatures") {
            } else if (tag == "byline") {
                gsub("\\\\","<p>",text);
                oline = oline "<address><img src=\"http://src-www.pa.dec.com/icons/line1.gif\"><p>";
                line = text "<p></address>" line;
            } else if (tag == "print_date") {
                oline = oline "Prepared: ";
                line = text "<p>" line;
            } else if (tag == "copyright_date") {
                n = split(text, parts, "\\");
                oline = oline "Copyright ";
                line = parts[1] ", " parts[2] line;
            } else if (tag == "tbs") {
                oline = oline "<strong>TBS</strong>"
            } else if (tag == "cp") {
            } else if (tag == "footnote") {
                n = split(text, parts, "\\");
                parts[1] = tolower(parts[1]);
                foot_seq[parts[1]]++;
                footnotes = footnotes "<dt><a name=\"footnote-" parts[1] "-" foot_seq[parts[1]] "\">Footnote " parts[1] "</a>\n";
                footnotes = footnotes "<dd>" parts[2] "\n";
                oline = oline "<a href=\"#footnote-" parts[1] "-" foot_seq[parts[1]] "\"> footnote " parts[1] "</a>";
            } else if (tag == "footref") {
                text = tolower(text);
                oline = oline "<a href=\"#footnote-" text "-" foot_seq[text] "\"> footnote " text "</a>";
            } else if (tag == "figure") {
                figure_filespec="";
                if (text != "") {
                    n = split(text, parts, "\\");
                    if (n > 1) {
                        parts[2] = tolower(parts[2]);
                        oline = oline "<h6><a name=\"" parts[2] "\">";
                        line = parts[1] "</a></h6>" line;
                        figure_filespec = parts[2] ".gif";
                        symbol[parts[2]] = "<a href=\"#" parts[2] "\">" parts[1] "</a>";
#                       for ( ; contents_level < 6; contents_level++) {
#                           contents = contents "<" contents_list_type ">\n";
#                       }
#                       for ( ; contents_level > 6; contents_level--) {
#                           contents = contents "</" contents_list_type ">\n";
#                       }
#                       contents = contents "<li><a href=\"#" parts[2] "\">" parts[1] "</a>\n";
                    } else {
                        oline = oline "<h6>";
                        line = text "</h6>" line;
#                       for ( ; contents_level < 6; contents_level++) {
#                           contents = contents "<" contents_list_type ">\n";
#                       }
#                       for ( ; contents_level > 6; contents_level--) {
#                           contents = contents "</" contents_list_type ">\n";
#                       }
#                       contents = contents "<li>" text "\n";
                    }
                }
            } else if (tag == "figure_attributes") {
            } else if (tag == "figure_file") {
                n = split(text, parts, "\\");
                parts[1] = tolower(parts[1]);
                if (parts[1] == "html")
                    figure_filespec = parts[2];
            } else if (tag == "endfigure") {
                if (figure_filespec != "")
                    oline = oline "<img src=\"" figure_filespec "\"><p>"
            } else if (tag == "online_chunk") {
            } else if (tag == "online_popup") {
            } else if (tag == "endonline_popup") {
            } else if (tag == "argument") {
                oline = oline "<var>";
                line = text "</var>" line;
            } else if (tag == "literal") {
                gsub("&","\\&amp;",text);
                gsub(">","\\&gt;",text);
                gsub("<","\\&lt;",text);
                oline = oline text;
            } else if (tag == "argdeflist") {
                oline = oline "<dl>";
            } else if (tag == "argitem") {
                oline = oline "<dt>";
                line = text line;
            } else if (tag == "argdef") {
                oline = oline "<dd>";
            } else if (tag == "endargdeflist") {
                oline = oline "</dl>";
            } else if (tag == "examples_intro") {
            } else if (tag == "hellipsis") {
                oline = oline " ... "
            } else if (tag == "ellipsis") {
                oline = oline "<p>.<p>.<p>.<p>"
            } else if (tag == "issue") {
            } else if (tag == "endappendix") {
            } else if (tag == "set_appendix_letter") {
            } else if (tag == "abstract") {
                oline = oline "<h2>Abstract</h2>";
                for ( ; contents_level < 2; contents_level++) {
                    contents = contents "<" contents_list_type ">\n";
                }
                for ( ; contents_level > 2; contents_level--) {
                    contents = contents "</" contents_list_type ">\n";
                }
                contents = contents "<li>Abstract\n";
            } else if (tag == "endabstract") {
            } else if (tag == "contents_file") {
                print "To see the contents listing, <a href=\"#ContentsListing\">click here</a>.\n";
                contents_needed = 1;
            } else {
                gsub(">","\\&gt;",tag);
                gsub("<","\\&lt;",tag);
                gsub(">","\\&gt;",text);
                gsub("<","\\&lt;",text);
                oline = oline "<!-- unrecognised tag " tag;
                if (text != "")
                    oline = oline "(" text ")";
                oline = oline " -->"
            }
        }
        if (line != "" && !in_comment) {
            if (in_code_example == 0)
                gsub("\\\\[A-Za-z0-9_]*", " ", line);
            oline = oline line;
        }
        if (oline != "" || in_code_example == 1) {
            if (on_title_page) {
                title_page = title_page oline "\n";
            } else if (on_copyright_page) {
                copyright_page = copyright_page oline "\n";
            } else 
                print oline;
        }
        return;
}
#
# Read a line from the current input file.
# Sets EOF when appropriate
#
function read_line(filename, result, name) {	# result and name are locals

    if (filename == "")
	result = getline;
    else
	result = getline < filename;

    if (result < 0) {
	name = (filename == "")? "standard input": filename;
	print "ERROR reading from", name;
	exit 1;
    }
    else if (result == 0)
	EOF = 1;
}
#
# Read in a file.
#
function do_file(filename) {

	if (filename == "")	# Will already be in $0
	    do_line(filename);

	while (!EOF) {
	    read_line(filename);
	    if (!EOF)
		do_line(filename);
	}

	if (filename != "")
	    close(filename);

	EOF = 0; # EOF is shared by all files
}