[WIP] Fix source mapping where utf8 char length != byte length

slither/core/source_mapping/source_mapping.py

@@ -20,6 +20,7 @@ def _compute_line(source_code, start, length):
 
             Not done in an efficient way
         """
+        source_code = source_code.encode('utf-8')
         total_length = len(source_code)
         source_code = source_code.splitlines(True)
         counter = 0
@@ -29,17 +30,18 @@ def _compute_line(source_code, start, length):
         ending_column = None
         while counter < total_length:
             # Determine the length of the line, and advance the line number
-            lineLength = len(source_code[i])
+            line_content = source_code[i]
+            line_length = len(line_content)
             i = i + 1
 
             # Determine our column numbers.
-            if starting_column is None and counter + lineLength > start:
+            if starting_column is None and counter + line_length > start:
                 starting_column = (start - counter) + 1
-            if starting_column is not None and ending_column is None and counter + lineLength > start + length:
+            if starting_column is not None and ending_column is None and counter + line_length > start + length:
                 ending_column = ((start + length) - counter) + 1
 
             # Advance the current position counter, and determine line numbers.
-            counter += lineLength
+            counter += line_length
             if counter > start:
                 lines.append(i)
 

tests/source_mapping.sol

@@ -0,0 +1,22 @@
+contract A{
+
+    // ëëëëëëëëëëëëë这这这这这这这这这这
+
+    address unused;
+
+
+    address unused2;
+
+
+    // ই এই এই এইই এই এই এইই এই এই এই
+
+    address unused3;
+
+
+    address unused4;
+
+    // 这这这这这这这这这这这这这这这这这这这这这这这这这这这这这这这这这这这这这这这这这这这这
+    address used;
+
+
+}