Don't ever create HTML filenames longer than ~200 chars. Fixes #627.

4 files changed, 37 insertions, 4 deletions

diff --git a/CHANGES.rst b/CHANGES.rst
index 97a7247..51a0019 100644
--- a/CHANGES.rst
+++ b/CHANGES.rst
@@ -13,7 +13,13 @@ Unreleased
   See :ref:`api_plugin` for details.  This solves the complex configuration
   problem described in `issue 563`_.
 
+- On Windows, the HTML report could fail when source trees are deeply nested,
+  due to attempting to create HTML filenames longer than the 250-character
+  maximum.  Now filenames will never exceed 200 characters, fixing `issue
+  627`_.  Thanks to Alex Sandro for helping with the fix.
+
 .. _issue 563: https://bitbucket.org/ned/coveragepy/issues/563/platform-specific-configuration
+.. _issue 627: https://bitbucket.org/ned/coveragepy/issues/627/failure-generating-html-reports-when-the
 
 
 .. _changes_442:
diff --git a/CONTRIBUTORS.txt b/CONTRIBUTORS.txt
index b6d7889..6c6bcd0 100644
--- a/CONTRIBUTORS.txt
+++ b/CONTRIBUTORS.txt
@@ -7,6 +7,7 @@ useful bug reports, have been made by:
 Adi Roiban
 Alex Gaynor
 Alex Groce
+Alex Sandro
 Alexander Todorov
 Andrew Hoos
 Anthony Sottile
diff --git a/coverage/files.py b/coverage/files.py
index 04ce9cb..12b89e8 100644
--- a/coverage/files.py
+++ b/coverage/files.py
@@ -3,6 +3,7 @@
 
 """File wrangling."""
 
+import hashlib
 import fnmatch
 import ntpath
 import os
@@ -75,6 +76,9 @@ def canonical_filename(filename):
     return CANONICAL_FILENAME_CACHE[filename]
 
 
+MAX_FLAT = 200
+
+@contract(filename='unicode', returns='unicode')
 def flat_rootname(filename):
     """A base for a flat file name to correspond to this file.
 
@@ -86,7 +90,11 @@ def flat_rootname(filename):
 
     """
     name = ntpath.splitdrive(filename)[1]
-    return re.sub(r"[\\/.:]", "_", name)
+    name = re.sub(r"[\\/.:]", "_", name)
+    if len(name) > MAX_FLAT:
+        h = hashlib.sha1(name.encode('UTF-8')).hexdigest()
+        name = name[-(MAX_FLAT-len(h)-1):] + '_' + h
+    return name
 
 
 if env.WINDOWS:
diff --git a/tests/test_files.py b/tests/test_files.py
index c76dbaf..84d84a4 100644
--- a/tests/test_files.py
+++ b/tests/test_files.py
@@ -1,3 +1,4 @@
+# coding: utf-8
 # Licensed under the Apache License: http://www.apache.org/licenses/LICENSE-2.0
 # For details: https://bitbucket.org/ned/coveragepy/src/default/NOTICE.txt
 
@@ -6,6 +7,8 @@
 import os
 import os.path
 
+import pytest
+
 from coverage import files
 from coverage.files import (
     TreeMatcher, FnmatchMatcher, ModuleMatcher, PathAliases,
@@ -54,9 +57,24 @@ class FilesTest(CoverageTest):
         rel = os.path.join('sub', trick, 'file1.py')
         self.assertEqual(files.relative_filename(abs_file(rel)), rel)
 
-    def test_flat_rootname(self):
-        self.assertEqual(flat_rootname("a/b/c.py"), "a_b_c_py")
-        self.assertEqual(flat_rootname(r"c:\foo\bar.html"), "_foo_bar_html")
+
+@pytest.mark.parametrize("original, flat", [
+    (u"a/b/c.py", u"a_b_c_py"),
+    (u"c:\\foo\\bar.html", u"_foo_bar_html"),
+    (u"Montréal/☺/conf.py", u"Montréal_☺_conf_py"),
+    ( # original:
+        u"c:\\lorem\\ipsum\\quia\\dolor\\sit\\amet\\consectetur\\adipisci\\velit\\sed\\quia\\non"
+        u"\\numquam\\eius\\modi\\tempora\\incidunt\\ut\\labore\\et\\dolore\\magnam\\aliquam"
+        u"\\quaerat\\voluptatem\\ut\\enim\\ad\\minima\\veniam\\quis\\nostrum\\exercitationem"
+        u"\\ullam\\corporis\\suscipit\\laboriosam\\Montréal\\☺\\my_program.py",
+        # flat:
+        u"re_et_dolore_magnam_aliquam_quaerat_voluptatem_ut_enim_ad_minima_veniam_quis_"
+        u"nostrum_exercitationem_ullam_corporis_suscipit_laboriosam_Montréal_☺_my_program_py_"
+        u"97eaca41b860faaa1a21349b1f3009bb061cf0a8"
+     ),
+])
+def test_flat_rootname(original, flat):
+    assert flat_rootname(original) == flat
 
 
 class MatcherTest(CoverageTest):