X-Git-Url: https://git.lyx.org/gitweb/?a=blobdiff_plain;f=src%2FEncoding.cpp;h=ad380d78039d69f83ffa048bb14b3ed9a54c51b5;hb=2e433654c7f3db02bfd0e4dd09eb04cd8ab44bb9;hp=e0b3b70789e144a7c80a69507615d3c4f3e45ad2;hpb=c8d1d0d33c0c899f17eca14c52a26e781cdc0204;p=lyx.git

diff --git a/src/Encoding.cpp b/src/Encoding.cpp
index e0b3b70789..ad380d7803 100644
--- a/src/Encoding.cpp
+++ b/src/Encoding.cpp
@@ -25,8 +25,9 @@
 
 #include <boost/cstdint.hpp>
 
-#include <sstream>
+#include <iterator>
 #include <algorithm>
+#include <sstream>
 
 using namespace std;
 using namespace lyx::support;
@@ -58,7 +59,7 @@ MathAlphaSet mathalpha;
 /// The highest code point in UCS4 encoding (1<<20 + 1<<16)
 char_type const max_ucs4 = 0x110000;
 
-} // namespace anon
+} // namespace
 
 
 EncodingException::EncodingException(char_type c)
@@ -188,9 +189,16 @@ bool Encoding::encodable(char_type c) const
 {
 	// assure the used encoding is properly initialized
 	init();
-
 	if (iconvName_ == "UTF-8" && package_ == none)
 		return true;
+	// platex does not load inputenc: force conversion of supported characters
+	if (package_ == Encoding::japanese
+	    && ((0xb7 <= c && c <= 0x05ff) // Latin-1 Supplement ... Hebrew
+			|| (0x1d00 <= c && c <= 0x218f) // Phonetic Extensions ... Number Forms
+			|| (0x2193 <= c && c <= 0x2aff) // Arrows ... Supplemental Mathematical Operators
+			|| (0xfb00 <= c && c <= 0xfb4f) // Alphabetic Presentation Forms
+			|| (0x1d400 <= c && c <= 0x1d7ff))) // Mathematical Alphanumeric Symbols
+		return false;
 	if (c < start_encodable_ && !isForced(c))
 		return true;
 	if (encodable_.find(c) != encodable_.end())
@@ -222,8 +230,8 @@ pair<docstring, docstring> Encoding::latexString(docstring const & input, bool d
 	docstring uncodable;
 	bool terminate = false;
 	for (size_t n = 0; n < input.size(); ++n) {
+		char_type const c = input[n];
 		try {
-			char_type const c = input[n];
 			pair<docstring, bool> latex_char = latexChar(c);
 			docstring const latex = latex_char.first;
 			if (terminate && !prefixIs(latex, '\\')
@@ -240,7 +248,8 @@ pair<docstring, docstring> Encoding::latexString(docstring const & input, bool d
 			result += latex;
 			terminate = latex_char.second;
 		} catch (EncodingException & /* e */) {
-			LYXERR0("Uncodable character in latexString!");
+			LYXERR0("Uncodable character <" << docstring(1, c) 
+					<< "> in latexString!");
 			if (dryrun) {
 				result += "<" + _("LyX Warning: ")
 					   + _("uncodable character") + " '";
@@ -263,11 +272,15 @@ vector<char_type> Encoding::symbolsList() const
 	vector<char_type> symbols;
 	for (char_type c = 0; c < start_encodable_; ++c)
 		symbols.push_back(c);
-	//add all encodable characters
+	// add all encodable characters
 	copy(encodable_.begin(), encodable_.end(), back_inserter(symbols));
-	// now the ones from the unicodesymbols file
-	for (pair<char_type, CharInfo> const & elem : unicodesymbols)
-		symbols.push_back(elem.first);
+	// now the ones from the unicodesymbols file that are not already there
+	for (pair<char_type, CharInfo> const & elem : unicodesymbols) {
+		if (find(symbols.begin(), symbols.end(), elem.first) == symbols.end())
+			symbols.push_back(elem.first);
+	}
+	// finally, sort the vector
+	sort(symbols.begin(), symbols.end());
 	return symbols;
 }
 
@@ -301,10 +314,11 @@ bool Encodings::latexMathChar(char_type c, bool mathmode,
 		if (!encoding || command.empty()) {
 			command = it->second.textcommand();
 			needsTermination = !it->second.textnotermination();
-			addTextCmd(c);
 		}
 		if (mathmode)
 			addMathSym(c);
+		else
+			addTextCmd(c);
 	}
 	return use_math;
 }
@@ -576,21 +590,29 @@ string const Encodings::TIPAShortcut(char_type c)
 }
 
 
-bool Encodings::isKnownScriptChar(char_type const c, string & preamble)
+string const Encodings::isKnownScriptChar(char_type const c)
 {
 	CharInfoMap::const_iterator const it = unicodesymbols.find(c);
 
 	if (it == unicodesymbols.end())
-		return false;
+		return string();
+	// FIXME: parse complex textpreamble (may be list or alternatives,
+	// 		  e.g., "subscript,textgreek" or "textcomp|textgreek")
+	if (it->second.textpreamble() == "textgreek"
+		|| it->second.textpreamble() == "textcyrillic")
+		return it->second.textpreamble();
+	return string();
+}
 
-	if (it->second.textpreamble() != "textgreek" && it->second.textpreamble() != "textcyr")
-		return false;
 
-	if (preamble.empty()) {
-		preamble = it->second.textpreamble();
-		return true;
-	}
-	return it->second.textpreamble() == preamble;
+bool Encodings::fontencSupportsScript(string const & fontenc, string const & script)
+{
+	if (script == "textgreek")
+		return (fontenc == "LGR" || fontenc == "TU");
+	if (script == "textcyrillic")
+		return (fontenc == "T2A" || fontenc == "T2B" || fontenc == "T2C"
+				|| fontenc == "X2" || fontenc == "TU");
+	return false;
 }
 
 
@@ -710,16 +732,16 @@ void Encodings::read(FileName const & encfile, FileName const & symbolsfile)
 				flags |= CharInfoForce;
 				forced.insert(symbol);
 			} else if (prefixIs(flag, "force=")) {
-				vector<string> encodings =
+				vector<string> encs =
 					getVectorFromString(flag.substr(6), ";");
-				for (size_t i = 0; i < encodings.size(); ++i)
-					forcedselected[encodings[i]].insert(symbol);
+				for (size_t i = 0; i < encs.size(); ++i)
+					forcedselected[encs[i]].insert(symbol);
 				flags |= CharInfoForceSelected;
 			} else if (prefixIs(flag, "force!=")) {
-				vector<string> encodings =
+				vector<string> encs =
 					getVectorFromString(flag.substr(7), ";");
-				for (size_t i = 0; i < encodings.size(); ++i)
-					forcednotselected[encodings[i]].insert(symbol);
+				for (size_t i = 0; i < encs.size(); ++i)
+					forcednotselected[encs[i]].insert(symbol);
 				flags |= CharInfoForceSelected;
 			} else if (flag == "mathalpha") {
 				mathalpha.insert(symbol);