So once upon a time I set out to fix the problem reported by Tobin wherein a literal block within a kerneldoc comment would be corrupted in processing. On the way, though, I got annoyed at the way I have to learn how kernel-doc works from the beginning every time I tear into it.
As a result, seven of the following eight patches just get rid of some dead code and reorganize the rest - mostly turning the 500-line process_file() function into something a bit more rational. Sphinx output is unchanged after these are applied. Then, at the end, there's a tweak to stop messing with literal blocks. If anybody was unaware that I've not done any serious Perl since the 1990's, they will certainly understand that fact now. Changes in v2: - Fix up old coding-style ugliness as I move the code - Handle ".. code-block::" literals as well. This fixed the formatting in dev-tools/kselftest.rst. Jonathan Corbet (8): docs: kernel-doc: Get rid of xml_escape() and friends docs: kernel-doc: Rename and split STATE_FIELD docs: kernel-doc: Move STATE_NORMAL processing into its own function docs: kernel-doc: Move STATE_NAME processing into its own function docs: kernel-doc: Move STATE_BODY processing to a separate function docs: kernel-doc: Move STATE_PROTO processing into its own function docs: kernel-doc: Finish moving STATE_* code out of process_file() docs: kernel-doc: Don't mangle literal code blocks in comments scripts/kernel-doc | 666 +++++++++++++++++++++++++++++------------------------ 1 file changed, 365 insertions(+), 301 deletions(-) -- 2.14.3