Properly deal with files encoded as UTF-8 with BOM. Fixes #179.

5 files changed, 19 insertions, 1 deletions

diff --git a/AUTHORS.txt b/AUTHORS.txt
index f59ab18b..edb6be94 100644
--- a/AUTHORS.txt
+++ b/AUTHORS.txt
@@ -7,6 +7,7 @@ Chris Adams
 Geoff Bache
 Titus Brown
 Brett Cannon
+Pablo Carballo
 Guillaume Chazarain
 David Christian
 Danek Duvall
diff --git a/CHANGES.txt b/CHANGES.txt
index 4a89b3d0..404947f3 100644
--- a/CHANGES.txt
+++ b/CHANGES.txt
@@ -2,6 +2,15 @@
 Change history for Coverage.py
 ------------------------------
 
+Version 3.5.3b1
+---------------
+
+- Files encoded as UTF-8 with a BOM are now properly handled, fixing
+  `issue 179`_.  Thanks, Pablo Carballo.
+
+.. _issue 179: https://bitbucket.org/ned/coveragepy/issue/179/htmlreporter-fails-when-source-file-is
+
+
 Version 3.5.2 --- 4 May 2012
 ----------------------------
 
diff --git a/coverage/__init__.py b/coverage/__init__.py
index b3e79c82..f4e17b29 100644
--- a/coverage/__init__.py
+++ b/coverage/__init__.py
@@ -5,7 +5,7 @@ http://nedbatchelder.com/code/coverage
 
 """
 
-__version__ = "3.5.2"       # see detailed history in CHANGES.txt
+__version__ = "3.5.3b1"       # see detailed history in CHANGES.txt
 
 __url__ = "http://nedbatchelder.com/code/coverage"
 if max(__version__).isalpha():
diff --git a/coverage/html.py b/coverage/html.py
index f39bf949..b0eff5f2 100644
--- a/coverage/html.py
+++ b/coverage/html.py
@@ -148,6 +148,10 @@ class HtmlReporter(Reporter):
         # later to properly write the HTML.
         if sys.version_info < (3, 0):
             encoding = source_encoding(source)
+            # Some UTF8 files have the dreaded UTF8 BOM. If so, junk it.
+            if encoding.startswith("utf-8") and source[:3] == "\xef\xbb\xbf":
+                source = source[3:]
+                encoding = "utf-8"
 
         # Get the numbers for this file.
         nums = analysis.numbers
diff --git a/coverage/parser.py b/coverage/parser.py
index d380eda1..f868d357 100644
--- a/coverage/parser.py
+++ b/coverage/parser.py
@@ -35,6 +35,10 @@ class CodeParser(object):
                     "No source for code: %r: %s" % (self.filename, err)
                     )
 
+        # Scrap the BOM if it exists.
+        if self.text and ord(self.text[0]) == 0xfeff:
+            self.text = self.text[1:]
+
         self.exclude = exclude
 
         self.show_tokens = False