aboutsummaryrefslogtreecommitdiff
path: root/svg-flatten/src/test/svg_tests.py
diff options
context:
space:
mode:
Diffstat (limited to 'svg-flatten/src/test/svg_tests.py')
-rw-r--r--svg-flatten/src/test/svg_tests.py2
1 files changed, 1 insertions, 1 deletions
diff --git a/svg-flatten/src/test/svg_tests.py b/svg-flatten/src/test/svg_tests.py
index 48ee8fd..7f1adb0 100644
--- a/svg-flatten/src/test/svg_tests.py
+++ b/svg-flatten/src/test/svg_tests.py
@@ -45,7 +45,7 @@ class SVGRoundTripTests(unittest.TestCase):
# Our stroke join test shows a discrepancy in miter handling between resvg and gerbolyze. Gerbolyze's miter join is
# the one from Clipper, which unfortunately cannot be configured. resvg uses one looking like that from the SVG 2
# spec. Gerbolyze's join is legal by the 1.1 spec since this spec does not explicitly define the miter offset. It
- # only contains a blurry picture, and that picture looks like what gerbolyze produces.
+ # only contains an unclear picture, and that picture looks approximately like what gerbolyze produces.
test_mean_default = 0.02
test_mean_overrides = {