You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
The java bytecode entry point relied on things working a little
differently.
Firstly, the function name is transformed to allow simpler function
names to be passed in. As such the rebuild_entry_point cannot be
responsible for looking up the symbol. This in turn involved a new way
of identifying the mode to use (using the existing entry points mode)
Secondly, as the lanuages are destroyed after the parsing phase, we must
re-provide the message handler as this is used by the entry point
generation.
Finally, the java_bytecode_entry_point adds a new symbol to the symbol
table, so we remove all symbols that are in the scope of the original
entry point.
0 commit comments