A somewhat better solution to the words/UTF-8 problem: do encoding last, but only on what appears to be string literals.

1 files changed, 10 insertions, 0 deletions

diff --git a/src/GF/Compile/Coding.hs b/src/GF/Compile/Coding.hs
index a226265c5..704e95201 100644
--- a/src/GF/Compile/Coding.hs
+++ b/src/GF/Compile/Coding.hs
@@ -48,3 +48,13 @@ codeSourceModule co (id,moi) = case moi of
       PSeq p q -> PSeq (codp p) (codp q)
       PAlt p q -> PAlt (codp p) (codp q)
       _ -> p
+
+-- | Run an encoding function on all string literals within the given string.
+codeStringLiterals :: (String -> String) -> String -> String
+codeStringLiterals _ [] = []
+codeStringLiterals co ('"':cs) = '"' : inStringLiteral cs
+  where inStringLiteral [] = error "codeStringLiterals: unterminated string literal"
+        inStringLiteral ('"':ds) = '"' : codeStringLiterals co ds
+        inStringLiteral ('\\':d:ds) = '\\' : co [d] ++ inStringLiteral ds
+        inStringLiteral (d:ds) = co [d] ++ inStringLiteral ds
+codeStringLiterals co (c:cs) = c : codeStringLiterals co cs