X-Git-Url: https://git.lyx.org/gitweb/?a=blobdiff_plain;f=src%2FEncoding.cpp;h=47e27eb97d9c94e6909e8367618f5c6f41c197e8;hb=02d2e4aa3279b872c9786718dab4f95fd44e900a;hp=5c0681903593b475b4cb8e7f38b1ffec1e6e044a;hpb=7dfc6c7f8b1a0f71a5b6f2b4f153c73513739eae;p=lyx.git

diff --git a/src/Encoding.cpp b/src/Encoding.cpp
index 5c06819035..47e27eb97d 100644
--- a/src/Encoding.cpp
+++ b/src/Encoding.cpp
@@ -23,10 +23,9 @@
 #include "support/textutils.h"
 #include "support/unicode.h"
 
-#include <boost/cstdint.hpp>
-
-#include <iterator>
 #include <algorithm>
+#include <cstdint>
+#include <iterator>
 #include <sstream>
 
 using namespace std;
@@ -189,9 +188,16 @@ 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())
@@ -223,8 +229,8 @@ 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, '\\')
@@ -241,7 +247,8 @@ 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") + " '";
@@ -267,7 +274,7 @@ vector<char_type> Encoding::symbolsList() const
 	// add all encodable characters
 	copy(encodable_.begin(), encodable_.end(), back_inserter(symbols));
 	// now the ones from the unicodesymbols file that are not already there
-	for (pair<char_type, CharInfo> const & elem : unicodesymbols) {
+	for (auto const & elem : unicodesymbols) {
 		if (find(symbols.begin(), symbols.end(), elem.first) == symbols.end())
 			symbols.push_back(elem.first);
 	}
@@ -306,10 +313,11 @@ 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;
 }
@@ -580,36 +588,29 @@ string const Encodings::TIPAShortcut(char_type c)
 	return string();
 }
 
-// Return true, if `c` is a supported Greek or Cyrillic letter.
-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;
-
-	if (it->second.textpreamble() != "textgreek"
-	    && it->second.textpreamble() != "textcyrillic")
-		return false;
-
-	if (preamble.empty()) {
-		preamble = it->second.textpreamble();
-		return true;
-	}
-	return it->second.textpreamble() == preamble;
+		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();
 }
 
 
-bool Encodings::needsScriptWrapper(string const & script, string const & fontenc)
+bool Encodings::fontencSupportsScript(string const & fontenc, string const & script)
 {
-	// Note: the wrapper is not required with Unicode font encoding "TU".
-	// However, this function is not called with non-TeX (Unicode) fonts.
 	if (script == "textgreek")
-		return (fontenc != "LGR");
-	if (script == "textcyrillic") {
-		return (fontenc != "T2A" && fontenc != "T2B"
-			&& fontenc != "T2C" && fontenc != "X2");
-	}
+		return (fontenc == "LGR" || fontenc == "TU");
+	if (script == "textcyrillic")
+		return (fontenc == "T2A" || fontenc == "T2B" || fontenc == "T2C"
+				|| fontenc == "X2" || fontenc == "TU");
 	return false;
 }
 
@@ -635,9 +636,9 @@ Encodings::fromLyXName(string const & name, bool allowUnsafe) const
 {
 	EncodingList::const_iterator const it = encodinglist.find(name);
 	if (it == encodinglist.end())
-		return 0;
+		return nullptr;
 	if (!allowUnsafe && it->second.unsafe())
-		return 0;
+		return nullptr;
 	return &it->second;
 }
 
@@ -661,7 +662,7 @@ Encodings::fromLaTeXName(string const & n, int const & p, bool allowUnsafe) cons
 		if ((it->second.latexName() == name) && (it->second.package() & p)
 				&& (!it->second.unsafe() || allowUnsafe))
 			return &it->second;
-	return 0;
+	return nullptr;
 }
 
 
@@ -673,7 +674,7 @@ Encodings::fromIconvName(string const & n, int const & p, bool allowUnsafe) cons
 		if ((it->second.iconvName() == n) && (it->second.package() & p)
 				&& (!it->second.unsafe() || allowUnsafe))
 			return &it->second;
-	return 0;
+	return nullptr;
 }
 
 
@@ -701,7 +702,7 @@ void Encodings::read(FileName const & encfile, FileName const & symbolsfile)
 		istringstream is(symbolslex.getString());
 		// reading symbol directly does not work if
 		// char_type == wchar_t.
-		boost::uint32_t tmp;
+		uint32_t tmp;
 		if(!(is >> hex >> tmp))
 			break;
 		symbol = tmp;