Here are hpack package.yaml files for the other hledger cabal files.
These remove a lot of human-error-prone duplication.
They are not used yet as hpack isn't quite mature enough -
when it supports flags and benchmarks we will probably switch.
hakyll-std's cabal file is now generated by hpack from an easier yaml
config file, which looks like a valuable timesaver. "make gencabalfiles"
will regenerate this cabal file (and soon the others) when needed.
Drop deprecated shakespeare-text dependency for ghc 7.10+ consistently,
not just in one place. Hopefully this will clear the warning
at packdeps.haskellers.com.
The makefile now uses stack primarily.
The following updated rules are available:
$ make
Makefile:35: -------------------- hledger make rules --------------------
Makefile:37: make [help] -- list documented rules in this makefile. make -n RULE shows more detail.
Makefile:202: (INSTALLING:)
Makefile:204: make install -- download dependencies and install hledger executables to ~/.local/bin or equivalent (with stack)
Makefile:229: (BUILDING:)
Makefile:233: make build -- download dependencies and build hledger executables (with stack)
Makefile:302: make hledgerdev -- quickly build the hledger executable (with ghc and -DDEVELOPMENT)
Makefile:487: (TESTING:)
Makefile:495: make test -- run default tests
Makefile:541: make pkgtest -- run the test suites for each package
Makefile:548: make builtintest -- run tests built in to the hledger executable (subset of pkg tests)
Makefile:572: make functest -- run hledger functional tests with the stack build
Makefile:606: make haddocktest -- run haddock and make sure it succeeds
Makefile:787: (DOCUMENTATION:)
Makefile:895: make haddock -- generate haddock docs for the hledger packages
Makefile:974: (RELEASING:)
Makefile:1053: make setversion -- update all version strings to match .version
Makefile:1056: make setversionforce -- update all version strings even if .version seems unchanged
Makefile:1080: make tagrelease -- commit a release tag based on .version for each package
Makefile:1253: (MISCELLANEOUS:)
Makefile:1255: make usage -- show size of various dirs
Makefile:1260: make stackusage -- show size of stack working dirs
Makefile:1264: make cabalusage -- show size of cabal working dirs if any
Makefile:1268: make tag -- generate tag files for source code navigation (for emacs)
Makefile:1294: make clean -- default cleanup (ghc build leftovers)
Makefile:1297: make Clean -- thorough cleanup (stack/cabal/ghc builds and tags)
Makefile:1302: make cabalCMD -- run cabal CMD inside each hledger package directory
Makefile:1306: make all"CMD" -- run CMD inside each hledger package directory
Using "hledgerdev" was a hack to help ensure that tests used a fresh
developer build by default. Now they specify "hledger" again, which fits
better with stack. It's up to the tester to make sure the desired
executable is first in PATH or specified with -w. (Note a couple of
tests currently don't obey -w and will always run "hledger", see addons.test).
We don't do a good job of calculating good-looking unit prices when the
commodity display precisions are low. Eg when a journal doesn't use any
decimal places, any inferred unit prices are shown by the print command
also with no decimal places, which makes them look wrong.
Now inferred unit prices always have a minimum display precision of 2,
which helps a bit. Could do better.
The bench script invoked by cabal bench or stack bench now links the
benchmarked code directly, avoiding any dependence on the shell and
$PATH. Also, it reports the journal parse time and the various reports
on the parsed data separately, and runs quicker.
You can get more accurate benchmark times by running with --criterion.
This will usually give much the same numbers and takes much longer.
Benchmarking via simplebench is still available, by running with
--simplebench. In this mode it benchmarks whatever commands are
configured in bench/default.bench. A disadvantage is that it uses the
first "hledger" executable in $PATH, so check which one it reports.
