summaryrefslogtreecommitdiff
diff options
context:
space:
mode:
authorOlly Betts <olly@survex.com>2023-04-23 11:21:14 +1200
committerOlly Betts <olly@survex.com>2023-04-23 11:21:14 +1200
commit1657a0a1ec8dbdded01ab0a225bf290b44ef4a18 (patch)
tree0e47e17f31ddb1ecbb40546e625be9209ba1620e
parentd3b17437b12625bd31118f40e06a3bb28b6c5fda (diff)
downloadswig-1657a0a1ec8dbdded01ab0a225bf290b44ef4a18.tar.gz
Update CHANGES.current for Javascript changes
-rw-r--r--CHANGES.current22
1 files changed, 22 insertions, 0 deletions
diff --git a/CHANGES.current b/CHANGES.current
index afe7d8969..15398ef7c 100644
--- a/CHANGES.current
+++ b/CHANGES.current
@@ -7,6 +7,28 @@ the issue number to the end of the URL: https://github.com/swig/swig/issues/
Version 4.2.0 (in progress)
===========================
+2023-04-23: olly
+ [Javascript] #2453 The testsuite and examples now select which
+ Javascript engine to test based on what was detected by configure.
+ Previously they'd always test with node you specified a different
+ engine (e.g. with `ENGINE=jsc` on the make command line). Now you
+ only need to specify ENGINE if you have more than one engine
+ installed.
+
+2023-04-23: olly
+ [Javascript] Turn on C++ output when wrapping for node, like
+ we already do when wrapping for V8-without-node.
+
+ The testsuite was masking this bug by using SWIG options
+ `-v8 -DBUILDING_NODE_EXTENSION=1` rather than `-node` when testing
+ with nodejs, while the javascript examples were masking this by
+ all getting processed with -c++.
+
+ This shouldn't be an incompatible change for users, as if you're
+ wrapping a C API you'd have to be working around the problem before
+ this change (like our testsuite and examples were), and this change
+ shouldn't break your workaround - it just makes it unnecessary.
+
2023-04-21: mmomtchev
[Javascript] Fix naming of internal C++ helper for wrapping
variables for node to use the "getter" naming scheme rather