Properly handle Unicode characters in Python source when creating the HTML report (except surrogates come out as pairs of blots in the HTML). Fixes #124 and #144.

1 files changed, 20 insertions, 0 deletions

diff --git a/test/farm/html/run_unicode.py b/test/farm/html/run_unicode.py
new file mode 100644
index 00000000..210a9113
--- /dev/null
+++ b/test/farm/html/run_unicode.py
@@ -0,0 +1,20 @@
+def html_it():
+    """Run coverage and make an HTML report for unicode.py."""
+    import coverage
+    cov = coverage.coverage()
+    cov.start()
+    import unicode
+    cov.stop()
+    cov.html_report(unicode, directory="../html_unicode")
+
+runfunc(html_it, rundir="src")
+
+# HTML files will change often.  Check that the sizes are reasonable,
+#   and check that certain key strings are in the output.
+compare("gold_unicode", "html_unicode", size_within=10, file_pattern="*.html")
+contains("html_unicode/unicode.html",
+    "<span class='str'>&quot;&#654;d&#729;&#477;b&#592;&#633;&#477;&#652;o&#596;&quot;</span>",
+    "<span class='str'>&quot;db40,dd00: x&#56128;&#56576;&quot;</span>",
+    )
+
+clean("html_unicode")