commit 0eb748fd019d0f11449747ddc5711ba28f9f5aa2
Author: Daniel Swanson <popcorn.tomato.dude@gmail.com>
Date:   Thu Jun 3 12:43:24 2021 -0500

    cleverness is to be avoided (investigating #85)

diff --git a/lttoolbox/att_compiler.cc b/lttoolbox/att_compiler.cc
index 7e14624..1ad56d0 100644
--- a/lttoolbox/att_compiler.cc
+++ b/lttoolbox/att_compiler.cc
@@ -71,7 +71,8 @@ AttCompiler::convert_hfst(UString& symbol)
 bool
 AttCompiler::is_word_punct(UChar symbol)
 {
-  return u_charType(symbol) & (U_NON_SPACING_MARK | U_ENCLOSING_MARK | U_COMBINING_SPACING_MARK);
+  // this version isn't quite write, but something like it should be possible
+  //return u_charType(symbol) & (U_NON_SPACING_MARK | U_ENCLOSING_MARK | U_COMBINING_SPACING_MARK);
   // https://en.wikipedia.org/wiki/Combining_character#Unicode_ranges
   if((symbol >= 0x0300 && symbol <= 0x036F) // Combining Diacritics
   || (symbol >= 0x1AB0 && symbol <= 0x1AFF) // ... Extended
diff --git a/lttoolbox/input_file.cc b/lttoolbox/input_file.cc
index ac6eda5..e7dea2f 100644
--- a/lttoolbox/input_file.cc
+++ b/lttoolbox/input_file.cc
@@ -56,27 +56,21 @@ InputFile::internal_read()
     ubuffer[buffer_size++] = '\0';
     return;
   }
-  switch (cbuffer[0] & 0xF0) {
-  case 0xF0:
+  if ((cbuffer[0] & 0xF0) == 0xF0) {
     i += 3;
     if (fread(cbuffer+1, 1, 3, infile) != 3) {
       throw std::runtime_error("Could not read 3 expected bytes from stream");
     }
-    break;
-  case 0xE0:
+  } else if ((cbuffer[0] & 0xE0) == 0xE0) {
     i += 2;
     if (fread(cbuffer+1, 1, 2, infile) != 2) {
       throw std::runtime_error("Could not read 2 expected bytes from stream");
     }
-    break;
-  case 0xC0:
+  } else if ((cbuffer[0] & 0xC0) == 0xC0) {
     i += 1;
     if (fread(cbuffer+1, 1, 1, infile) != 1) {
       throw std::runtime_error("Could not read 1 expected byte from stream");
     }
-    break;
-  default:
-    break;
   }
   memset(ubuffer, 0, 3*sizeof(UChar));
   utf8::utf8to32(cbuffer, cbuffer+i, ubuffer);