|Subject:||encode/decode fails with certain constructs|
I am building a .ucm file for converting entities (&...;) into unicode. But after running enc2xs and compiling it, the resulting encoding module won't convert all of my entities to unicode. Multiple unicode values aren't converted correct into an entity if there is a shorter form available as well. E.g. U0030+U0304 -> &0macr; doesn't work if there is a U0030 -> \x30 mapping. I have attached a minimal test case of the problems, including a test module that will fail validation on the module. To make the test work you can delete the line which maps "&acedil;" and the line mapping U0030 -> \x30. I am testing this on Debian Lenny, which contains perl 5.10 and Encode 2.23.
Message body not shown because it is not plain text.