added field/prefixes for author and title + command line query language

src/internfile/internfile.cpp

@@ -1,5 +1,5 @@
 #ifndef lint
-static char rcsid[] = "@(#$Id: internfile.cpp,v 1.24 2007-01-15 13:06:38 dockes Exp $ (C) 2004 J.F.Dockes";
+static char rcsid[] = "@(#$Id: internfile.cpp,v 1.25 2007-01-17 13:53:40 dockes Exp $ (C) 2004 J.F.Dockes";
 #endif
 /*
  *   This program is free software; you can redistribute it and/or modify
@@ -236,6 +236,7 @@ static inline bool getKeyValue(const map<string, string>& docdata,
 }
 
 static const string keyab("abstract");
+static const string keyau("author");
 static const string keycs("charset");
 static const string keyct("content");
 static const string keyfn("filename");
@@ -251,6 +252,7 @@ bool FileInterner::dijontorcl(Rcl::Doc& doc)
     Dijon::Filter *df = m_handlers.back();
     const std::map<std::string, std::string>& docdata = df->get_meta_data();
 
+    getKeyValue(docdata, keyau, doc.author);
     getKeyValue(docdata, keyoc, doc.origcharset);
     getKeyValue(docdata, keyct, doc.text);    
     getKeyValue(docdata, keytt, doc.title);

src/internfile/mh_html.cpp

@@ -122,6 +122,7 @@ bool MimeHandlerHtml::next_document()
     m_metaData["charset"] = "utf-8";
     m_metaData["title"] = result.title;
     m_metaData["keywords"] = result.keywords;
+    m_metaData["author"] = result.author;
     m_metaData["modificationdate"] = result.dmtime;
     m_metaData["sample"] = result.sample;
     m_metaData["mimetype"] = "text/plain";

src/internfile/mh_mail.cpp

@@ -1,5 +1,5 @@
 #ifndef lint
-static char rcsid[] = "@(#$Id: mh_mail.cpp,v 1.28 2007-01-13 10:28:37 dockes Exp $ (C) 2005 J.F.Dockes";
+static char rcsid[] = "@(#$Id: mh_mail.cpp,v 1.29 2007-01-17 13:53:40 dockes Exp $ (C) 2005 J.F.Dockes";
 #endif
 /*
  *   This program is free software; you can redistribute it and/or modify
@@ -217,6 +217,9 @@ bool MimeHandlerMail::processMsg(Binc::MimePart *doc, int depth)
     if (doc->h.getFirstHeader("From", hi)) {
 	rfc2047_decode(hi.getValue(), transcoded);
 	text += string("From: ") + transcoded + string("\n");
+	if (depth == 1) {
+	    m_metaData["author"] = transcoded;
+	}
     }
     if (doc->h.getFirstHeader("To", hi)) {
 	rfc2047_decode(hi.getValue(), transcoded);
@@ -245,7 +248,7 @@ bool MimeHandlerMail::processMsg(Binc::MimePart *doc, int depth)
     }
     text += '\n';
 
-    LOGDEB2(("MimeHandlerMail::rocessMsg:ismultipart %d mime subtype '%s'\n",
+    LOGDEB2(("MimeHandlerMail::processMsg:ismultipart %d mime subtype '%s'\n",
 	    doc->isMultipart(), doc->getSubType().c_str()));
     walkmime(doc, depth);
 

src/internfile/myhtmlparse.cpp

@@ -154,6 +154,11 @@ MyHtmlParser::opening_tag(const string &tag, const map<string,string> &p)
 			    string tmp = i->second;
 			    decode_entities(tmp);
 			    keywords += tmp;
+			} else if (name == "author") {
+			    if (!author.empty()) author += ' ';
+			    string tmp = i->second;
+			    decode_entities(tmp);
+			    author += tmp;
 			} else if (name == "date") {
 			    // Yes this doesnt exist. It's output by filters
 			    // And the format isn't even standard http/html
@@ -168,19 +173,6 @@ MyHtmlParser::opening_tag(const string &tag, const map<string,string> &p)
 				dmtime = ascuxtime;
 			    }
 			} 
-#if 0 // We're not a robot, so we don't care about robots metainfo
-			else if (name == "robots") {
-			    string val = i->second;
-			    decode_entities(val);
-			    lowercase_term(val);
-			    if (val.find("none") != string::npos ||
-				val.find("noindex") != string::npos) {
-				indexing_allowed = false;
-				LOGDEB1(("myhtmlparse: robots/noindex\n"));
-				throw false;
-			    }
-			}
-#endif // 0
 		    } else if ((j = p.find("http-equiv")) != p.end()) {
 			string hequiv = j->second;
 			lowercase_term(hequiv);

src/internfile/myhtmlparse.h

@@ -37,7 +37,7 @@ class MyHtmlParser : public HtmlParser {
     bool in_body_tag; 
     bool in_pre_tag;
     bool pending_space;
-    string title, sample, keywords, dump, dmtime;
+    string title, sample, keywords, dump, dmtime, author;
     string ocharset; // This is the charset our user thinks the doc was
     string charset; // This is the charset it was supposedly converted to
     string doccharset; // Set this to value of charset parameter in header