Fixes#7237.
Full clean, configure and build will be required for Core, Developer
after this change.
kmc is much less unwieldy a name than kmldmlc, and becomes our platform
for future compiler integration. Start of implementation for #5283 also.
Adds compilers and minimal unit tests for transform sections bksp,
finl, ordr, and tran, and cleans up problems in the corresponding
builders.
Adds a utility script for building test fixtures for manual analysis:
./build.sh build-fixtures
Refactors constants to give us design-time type safety once more.
Adds definitions for LDML keyboard transform elements `transforms`,
`backspaces`, and `reorders`, and the binary format sections `bksp`,
`elem`, `finl`, `ordr`, and `tran`.
Minimally updates the compiler so that it will build, but no other
changes to the compiler.
Compiler and Typescript updates to support building these sections will
come in the next PR.
Just reorders all work around the various sections to match the binary
format order -- that is, 'sect' first, then all other sections
alphabetically.
This also means the compiler now emits the sections in the expected
order.
No other changes to functionality.
Updates compiler, headers and documentation for the 'name' section,
which will always be present as the LDML keyboard spec requires at least
one name in the `<names>` element.
Well, almost. There are still callback side-effects for messages. These
may in future be returned as an array as well, eliminating the need for
a message callback.
But restructures each section compiler so that it is self-contained and
responsible only for writing its own section. Also tightens up the types
for the sections constants.
Last, but not least, moves responsibility for writing the output .kmx
out of compiler.ts and into its callers -- removing one dependency from
the compiler. Reading input files still needs to be done in the
compiler, as there may be multiple input files with `<import>` that we
cannot know about until we start parsing xml.
The hextobin tool is used to help us construct invalid .kmx files
without having to either hack these into the compiler, or write them
with a hex editor. It feels a little bit of overkill, but this makes it
easy to (a) work with our intended file structures, and (b) review any
changes to these invalid keyboard tests.
Note: I did do a cursory search for a hex2bin or hextobin that we could
easily use but did not find a good simple cross-platform tool that we
could consume (`xxd -r` would work but AFAICT constrains our format to a
dump which is much harder to work with and read -- I wanted to be able
to annotate with file structures.)
For an example, see ik_000_null_invalid.txt.
This means that building the invalid keyboards for tests requires node,
but that should be fine given we need node to build the valid keyboards
anyway.
I've put hextobin under core/tools but it could just as well live under
common/tools in the future, if we find it useful for other project unit
tests.
