See more at: https://github.com/espeak-ng/espeak-ng/pull/863

translate.h gives two settings to LOPT_CAPS_IN_WORD, but only one of them is actually ever used.

It is enough to have the logic in dictsource/jbo_*.

Thanks to xunsku @ irc.freenode.org

Thanks to xunsku @ irc.freenode.org

This test is related to the next two lojban commits. Thanks to xunsku @ irc.freenode.org

Version must be changed in:
- configure.ac
- src/windows/config.h
- android/jni/include/config.h
- android/build.gradle
- android/AndroidManifest.xml

It's unclear why c2 needs to be set after an entity reference.

ProcessSsmlTag()

Note the line in readclause.c:
if ((c2 == '/') || iswalpha(c2) || c2 == '!' || c2 == '?') {

It might be enough to pass everything to ProcessSsmlTag. What are the
cases that are skipped because of this?

This is a bit slower since we don't pass n_xml_buf as an argument but
rather get it with a call to wcslen. It is much cleaner though, since
the name ProcessSsmlTag() implies that all processing should be done
there.