From 73951757bb0a5bb13958d9b23f869d72aa4ed465 Mon Sep 17 00:00:00 2001
From: "Veronica K. B. Olsen" <1619840+vkbo@users.noreply.github.com>
Date: Mon, 28 Oct 2019 11:05:47 +0100
Subject: [PATCH] Added call to escape unicode in LaTeX export. Fails with a
 flag if unsuccessful

---
 nw/convert/text/tolatex.py | 34 +++++++++++++++-------------------
 requirements.txt           |  1 +
 2 files changed, 16 insertions(+), 19 deletions(-)

diff --git a/nw/convert/text/tolatex.py b/nw/convert/text/tolatex.py
index 2002f8d9..345a988e 100644
--- a/nw/convert/text/tolatex.py
+++ b/nw/convert/text/tolatex.py
@@ -12,6 +12,7 @@
 
 import textwrap
 import logging
+import codecs
 import re
 import nw
 
@@ -23,20 +24,7 @@ class ToLaTeX(Tokenizer):
 
     def __init__(self, theProject, theParent):
         Tokenizer.__init__(self, theProject, theParent)
-        return
-
-    def doAutoReplace(self):
-        Tokenizer.doAutoReplace(self)
-
-        repDict = {
-            "\u2013" : "--",
-            "\u2014" : "---",
-            "\u2500" : "---",
-            "\u2026" : "...",
-        }
-        xRep = re.compile("|".join([re.escape(k) for k in repDict.keys()]), flags=re.DOTALL)
-        self.theText = xRep.sub(lambda x: repDict[x.group(0)], self.theText)
-
+        self.texCodecFail = False
         return
 
     def doConvert(self):
@@ -122,17 +110,17 @@ class ToLaTeX(Tokenizer):
 
             elif tType == self.T_HEAD1:
                 self.theResult += begText
-                self.theResult += "{\\Huge %s}\n" % tText
+                self.theResult += "{\\Huge %s}\n" % self._escapeUnicode(tText)
                 self.theResult += endText
 
             elif tType == self.T_HEAD2:
-                self.theResult += "\\chapter*{%s}\n\n" % tText
+                self.theResult += "\\chapter*{%s}\n\n" % self._escapeUnicode(tText)
 
             elif tType == self.T_HEAD3:
-                self.theResult += "\\section*{%s}\n\n" % tText
+                self.theResult += "\\section*{%s}\n\n" % self._escapeUnicode(tText)
 
             elif tType == self.T_HEAD4:
-                self.theResult += "\\subsection*{%s}\n\n" % tText
+                self.theResult += "\\subsection*{%s}\n\n" % self._escapeUnicode(tText)
 
             elif tType == self.T_SEP:
                 self.theResult += begText
@@ -140,7 +128,7 @@ class ToLaTeX(Tokenizer):
                 self.theResult += endText
 
             elif tType == self.T_TEXT:
-                thisPar.append(tText)
+                thisPar.append(self._escapeUnicode(tText))
 
             elif tType == self.T_PBREAK:
                 self.theResult += "\\newpage\n\n"
@@ -153,4 +141,12 @@ class ToLaTeX(Tokenizer):
 
         return
 
+    def _escapeUnicode(self, theText):
+        try:
+            import latexcodec
+            return codecs.encode(theText, "ulatex")
+        except:
+            self.texCodecFail = True
+            return theText
+
 # END Class ToLaTeX
diff --git a/requirements.txt b/requirements.txt
index a83517dd..14cd2a0c 100644
--- a/requirements.txt
+++ b/requirements.txt
@@ -3,3 +3,4 @@ appdirs
 lxml
 pyenchant
 pycountry
+latexcodec