I was about to drop the m4 experiment, but.. not yet. I want to split up
the big man pages into chunks that are easier to deal with, so now using
m4 for that (only. Pandoc is taking care of selective content stripping
and other rendering adjustments.)
I've switched the suffix of m4-enabled files from .md.m4 to .m4.md -
less correct but activates more useful editor highlighting.
This docs page is less verbose, and should make it easier to know what
you should read next, and what you don't need to read.
The reference docs are still in flux, with three alternatives presented:
1. the old manual, 2. the new individual man pages, and 3. the new
manual which combines the man pages. At this point I think that 2 will
get top billing, 3 will also be provided, and 1 will be dropped when 2/3
are a bit better.
The new m4manpages, m4webmanpages targets generate nroff and markdown
man pages via an alternate, excitingly complicated process, involving
shake, hakyll, pandoc *and* m4. Currently just the hledger man page is
processed this way, and the output (hledger/doc/m4-hledger.1,
site/m4-hledger.1.md) is equivalent to that of the non-m4 process.
Pro: selecting and massaging web/man content may be smoother with m4
than with pandoc filters. File inclusion allows documentation to be
broken up into chunks, which may be easier to edit, reorganize and
reuse. Macros could reduce boilerplate and enable more featureful and
attractive docs.
Con: the non-m4 process was simpler, easier to for contributors to
understand and working well enough. YAGNI.
Move to a newer stackage snapshot. This helps ensure a new-enough pandoc
will more be available for doc building without needing extra
compilation. Also we get to use the latest swagger and servant-swagger
for hledger-api, but we have to go to stackage nightly. With luck there
will be a lts 6 before next hledger release.
We now parse account directives, like Ledger's. We don't do anything
with them yet. The default parent account feature must now be spelled
"apply account"/"end apply account".
Since commit 7aab544, "-f -" before the command broke command
detection, causing spurious "no such option" errors (and breaking
about 70 functional tests which neither I nor travis noticed).
