X-Git-Url: https://git.lyx.org/gitweb/?a=blobdiff_plain;f=src%2FLexer.cpp;h=664e1b9bfb6c10867b6981479a50d4d8b2c80fe2;hb=55a3dd7b346d29a52ba305a4558e9e380ef50f47;hp=5136c672073d6dc451360a22f6b088915c1a66f3;hpb=e7645f24391648fb08cf6b6d808fcabec76be744;p=lyx.git

diff --git a/src/Lexer.cpp b/src/Lexer.cpp
index 5136c67207..664e1b9bfb 100644
--- a/src/Lexer.cpp
+++ b/src/Lexer.cpp
@@ -4,7 +4,7 @@
  * Licence details can be found in the file COPYING.
  *
  * \author Alejandro Aguilar Sierra
- * \author Lars Gullik Bjønnes
+ * \author Lars Gullik Bjønnes
  * \author Jean-Marc Lasgouttes
  * \author John Levon
  *
@@ -20,6 +20,7 @@
 #include "support/FileName.h"
 #include "support/filetools.h"
 #include "support/gzstream.h"
+#include "support/lassert.h"
 #include "support/lstrings.h"
 #include "support/lyxalgo.h"
 #include "support/types.h"
@@ -173,7 +174,9 @@ void Lexer::Pimpl::printError(string const & message) const
 {
 	string const tmpmsg = subst(message, "$$Token", getString());
 	lyxerr << "LyX: " << tmpmsg << " [around line " << lineno
-		<< " of file " << to_utf8(makeDisplayPath(name)) << ']' << endl;
+		<< " of file " << to_utf8(makeDisplayPath(name))
+		<< " current token: '" << getString() << "'"
+		<< " context: '" << context << "']" << endl;
 }
 
 
@@ -247,9 +250,10 @@ bool Lexer::Pimpl::setFile(FileName const & filename)
 				"file or stream already set.");
 		gz_.open(filename.toFilesystemEncoding().c_str(), ios::in);
 		is.rdbuf(&gz_);
-		name = filename.absFilename();
+		name = filename.absFileName();
 		lineno = 0;
-		return gz_.is_open() && is.good();
+		if (!gz_.is_open() || !is.good())
+			return false;
 	} else {
 		LYXERR(Debug::LYXLEX, "lyxlex: UNcompressed");
 
@@ -260,12 +264,25 @@ bool Lexer::Pimpl::setFile(FileName const & filename)
 			LYXERR(Debug::LYXLEX, "Error in Lexer::setFile: "
 				"file or stream already set.");
 		}
-		fb_.open(filename.toFilesystemEncoding().c_str(), ios::in);
+		fb_.open(filename.toSafeFilesystemEncoding().c_str(), ios::in);
 		is.rdbuf(&fb_);
-		name = filename.absFilename();
+		name = filename.absFileName();
 		lineno = 0;
-		return fb_.is_open() && is.good();
+		if (!fb_.is_open() || !is.good())
+			return false;
 	}
+
+	// Skip byte order mark.
+	if (is.peek() == 0xef) {
+		is.get();
+		if (is.peek() == 0xbb) {
+			is.get();
+			LASSERT(is.get() == 0xbf, /**/);
+		} else
+			is.unget();
+	}
+
+	return true;
 }
 
 
@@ -311,11 +328,6 @@ bool Lexer::Pimpl::next(bool esc /* = false */)
 		is.get(cc);
 		c = cc;
 
-		// skip ','s
-		if (esc && c == ',')
-			continue;
-
-		
 		if (c == commentChar) {
 			// Read rest of line (fast :-)
 #if 1
@@ -380,13 +392,13 @@ bool Lexer::Pimpl::next(bool esc /* = false */)
 			break;
 		}
 
-		if (!esc && c == ',')
+		if (c == ',')
 			continue;              /* Skip ','s */
 
-			// using relational operators with chars other
-			// than == and != is not safe. And if it is done
-			// the type _have_ to be unsigned. It usually a
-			// lot better to use the functions from cctype
+		// using relational operators with chars other
+		// than == and != is not safe. And if it is done
+		// the type _have_ to be unsigned. It usually a
+		// lot better to use the functions from cctype
 		if (c > ' ' && is)  {
 			buff.clear();
 
@@ -404,7 +416,7 @@ bool Lexer::Pimpl::next(bool esc /* = false */)
 			status = LEX_TOKEN;
 		}
 
-		if (!esc && c == '\r' && is) {
+		if (c == '\r' && is) {
 			// The Windows support has lead to the
 			// possibility of "\r\n" at the end of
 			// a line.  This will stop LyX choking
@@ -461,7 +473,7 @@ bool Lexer::Pimpl::eatLine()
 		is.get(cc);
 		c = cc;
 		//LYXERR(Debug::LYXLEX, "Lexer::EatLine read char: `" << c << '\'');
-		if (c != '\r')
+		if (c != '\r' && is)
 			buff.push_back(c);
 	}
 
@@ -502,7 +514,7 @@ bool Lexer::Pimpl::nextToken()
 		char cc = 0;
 		is.get(cc);
 		c = cc;
-		if (c >= ' ' && is) {
+		if ((c >= ' ' || c == '\t') && is) {
 			buff.clear();
 
 			if (c == '\\') { // first char == '\\'
@@ -516,7 +528,7 @@ bool Lexer::Pimpl::nextToken()
 					buff.push_back(c);
 					is.get(cc);
 					c = cc;
-				} while (c >= ' ' && c != '\\' && is);
+				} while ((c >= ' ' || c == '\t') && c != '\\' && is);
 			}
 
 			if (c == '\\')
@@ -639,6 +651,7 @@ void Lexer::setCommentChar(char c)
 	pimpl_->setCommentChar(c);
 }
 
+
 int Lexer::lex()
 {
 	return pimpl_->lex();
@@ -765,6 +778,7 @@ bool Lexer::getBool() const
 	pimpl_->printError("Bad boolean `$$Token'. "
 				 "Use \"false\" or \"true\"");
 	lastReadOk_ = false;
+	return false;
 }
 
 
@@ -902,6 +916,18 @@ string Lexer::quoteString(string const & arg)
 }
 
 
+// same for docstring
+docstring Lexer::quoteString(docstring const & arg)
+{
+	docstring res;
+	res += '"';
+	res += subst(subst(arg, from_ascii("\\"), from_ascii("\\\\")), 
+		     from_ascii("\""), from_ascii("\\\""));
+	res += '"';
+	return res;
+}
+
+
 Lexer & Lexer::operator>>(char const * required)
 {
 	string token;