[pkgsrc/trunk]: pkgsrc/lang/quickjs lang/quickjs: document possible cause for...

[rillig <rillig%pkgsrc.org@localhost>]

details:   https://anonhg.NetBSD.org/pkgsrc/rev/95b0600a28ba
branches:  trunk
changeset: 344135:95b0600a28ba
user:      rillig <rillig%pkgsrc.org@localhost>
date:      Fri Nov 15 20:51:23 2019 +0000

description:
lang/quickjs: document possible cause for the test patch

diffstat:

 lang/quickjs/distinfo                             |  4 ++--
 lang/quickjs/patches/patch-tests_test__builtin.js |  9 +++++++--
 2 files changed, 9 insertions(+), 4 deletions(-)

diffs (34 lines):

diff -r 62fec8deec8b -r 95b0600a28ba lang/quickjs/distinfo
--- a/lang/quickjs/distinfo     Fri Nov 15 16:00:23 2019 +0000
+++ b/lang/quickjs/distinfo     Fri Nov 15 20:51:23 2019 +0000
@@ -1,4 +1,4 @@
-$NetBSD: distinfo,v 1.2 2019/11/14 21:42:13 rillig Exp $
+$NetBSD: distinfo,v 1.3 2019/11/15 20:51:23 rillig Exp $
 
 SHA1 (quickjs-2019-10-27.tar.xz) = 6d10526eecc6bbbd13f33262f796f30090195974
 RMD160 (quickjs-2019-10-27.tar.xz) = 4cd24d0d6c6c9fd23c2bc66d1e724e7e43985b22
@@ -10,4 +10,4 @@
 SHA1 (patch-qjsc.c) = 9a70bde5e19b65942d9f23c10d0b7b3e2741bfaa
 SHA1 (patch-quickjs-libc.c) = 8a82afa9f13c9f7d53f8a94e377fbb4aa4471947
 SHA1 (patch-quickjs.c) = c411d8ed5539bcd77a504caf53541b7bddb51b5a
-SHA1 (patch-tests_test__builtin.js) = bf337da0772cbf3cd4981fe0c46be6c49dd1600f
+SHA1 (patch-tests_test__builtin.js) = 1433415f3e0e9fed061377027a78fe62f5fde8a7
diff -r 62fec8deec8b -r 95b0600a28ba lang/quickjs/patches/patch-tests_test__builtin.js
--- a/lang/quickjs/patches/patch-tests_test__builtin.js Fri Nov 15 16:00:23 2019 +0000
+++ b/lang/quickjs/patches/patch-tests_test__builtin.js Fri Nov 15 20:51:23 2019 +0000
@@ -1,8 +1,13 @@
-$NetBSD: patch-tests_test__builtin.js,v 1.1 2019/11/14 21:42:13 rillig Exp $
+$NetBSD: patch-tests_test__builtin.js,v 1.2 2019/11/15 20:51:24 rillig Exp $
 
 Disable tests that fail on NetBSD.
 
-rillig contacted Fabrice via mail on 2019-11-14.
+rillig contacted Fabrice via mail. Fabrice replied:
+> It is likely the tests fail because QuickJS relies on a the fact that
+> printf() should respect the current FP rounding mode when printing a
+> floating point number. There is no simple way to fix it, but it is not a
+> big issue. The fix will be not to depend on printf to convert floating
+> point numbers to string.
 
 --- tests/test_builtin.js.orig 2019-10-27 10:55:35.000000000 +0000
 +++ tests/test_builtin.js