Commit Graph

178 Commits

Author SHA1 Message Date
Simon Michael
35c1c9b6a2 pkg: progress towards supporting GHC 9.2 and newer libs (#1774)
hledger-lib builds, hledger's deps don't (shakespeare).
2021-12-06 12:32:50 -10:00
Simon Michael
66619803b7 ;pkg: bump version to 1.24.99 2021-12-01 22:16:37 -10:00
Simon Michael
387325b59e ;pkg: bump version to 1.24 2021-12-01 17:16:28 -10:00
Stephen Morgan
ff0132df28 dev: Use realLength from doclayout instead of strWidth and textWidth. (#895)
This gives us more accurate string length calculations. In particular,
it handles emoji and other scripts properly.
2021-11-11 18:29:50 -10:00
Stephen Morgan
1bc04685b7 pkg: Drop base-compat-batteries dependency.
Our supported stackage versions are now new enough that we don't need
any of the compatibility features anymore.
2021-10-31 07:56:07 -10:00
Simon Michael
69905dbc25 ;pkg: allow megaparsec 9.2 2021-10-03 22:55:10 -10:00
Simon Michael
8934c115bd ;pkg: bump version to 1.23.99 2021-09-24 12:22:15 -10:00
Simon Michael
19950df745 ;pkg: bump version to 1.23 2021-09-21 15:34:23 -10:00
Simon Michael
f51ea92cfc deps: require base >=4.11, prevent red squares on hackage matrix
We officially support GHC 8.6+ (and 8.8+ for hledger-web) now.
Hackage matrix builder shows all packages building successfully
with GHC 8.4+, somehow, so we'll adjust the base bound to
allow that but prevent any attempts to build with older GHCs,
2021-08-03 21:05:02 -10:00
Simon Michael
2a39497e21 pkg: add tested-with GHC 9.0.1 2021-08-03 20:52:05 -10:00
Simon Michael
6665ddfb9b ;pkg: bump version to 1.22.99 2021-08-03 00:24:20 -10:00
Simon Michael
9aac520edd deps: allow megaparsec 9.1 2021-07-20 20:38:39 -10:00
Simon Michael
544450f557 ;bump version to 1.22 2021-06-28 22:37:47 -10:00
Simon Michael
58b481ca5b stack: updated tested-with to 8.6+ 2021-06-03 14:07:39 -10:00
Simon Michael
ba1e91c302 drop support for GHC 8.0
Prior to this commit,
- hledger still builds with GHC 8.0
- hledger-ui does if you use the build plan specified by stack8.0.yaml,
  but you are likely to hit problems if you let cabal pick one
  (https://github.com/jtdaugherty/vty/issues/198 and others)
- hledger-web might, if you could find the right build plan

The hassles are enough and GHC 8.0 is old enough (first released in
2016) that I'm letting it go; 8.2 is the new minimum version for all
hledger packages.

This allows a bunch of cleanups to conditional imports, which I leave
for later.

Also, updated the tested-with minor versions.
2021-04-04 07:54:22 -10:00
Simon Michael
4e644840bc lib, etc: add now-required lower bound on containers (#1514) 2021-03-29 08:19:28 -07:00
Simon Michael
36cbc2b068 bump base upper bound to allow GHC 9.0 2021-03-12 06:58:46 -08:00
Simon Michael
e050790d4c ;bump version to 1.21.99 2021-03-10 13:50:49 -08:00
Simon Michael
eeddfc2509 ;bump version to 1.21 2021-03-10 08:24:58 -08:00
Simon Michael
9087532b62 ;bump version to 1.20.99 2020-12-14 11:28:07 -08:00
Simon Michael
2501329f3c ;bump version to 1.20 2020-11-30 15:18:24 -08:00
Simon Michael
a97daaf322 lib: replace pretty-show with pretty-simple
pretty-simple, already used in .ghci, will hopefully give nicer debug
output, including for values which don't have Read-able Show output.
This should mean that we can start removing custom string-like Show
instances that were a workaround for pretty-show.

We are using the latest version (4.0.0.0) to get compact output.
Here's some old pretty-show output:

 CsvRules
   { rdirectives = [ ( "skip" , "1" ) ]
   , rcsvfieldindexes = [ ( "date" , 1 ) , ( "amount" , 2 ) ]
   , rassignments = [ ( "amount" , "%2" ) , ( "date" , "%1" ) ]
   , rconditionalblocks = []
   }

And the new pretty-simple output:

 CsvRules
   { rdirectives=
     [ ( "skip", "1" ) ]
   , rcsvfieldindexes=
     [ ( "date", 1 ), ( "amount", 2 ) ]
   , rassignments=
     [ ( "amount", "%2" ), ( "date", "%1" ) ]
   , rconditionalblocks= []
   }

Non-compact pretty-simple output would be:

 CsvRules
     { rdirectives=
         [
             ( "skip"
             , "1B"
             )
         ]
     , rcsvfieldindexes=
         [
             ( "date"
             , 1
             )
         ,
             ( "amount"
             , 2
             )
         ]
     , rassignments=
         [
             ( "amount"
             , "%2"
             )
         ,
             ( "date"
             , "%1"
             )
         ]
     , rconditionalblocks=[]
     }

Also:

- Account's Show instance no longer converts : to _ in account names

- drop unused pretty-show dependency from hledger, hledger-ui packages

- regenerate hledger-lib with the older hpack that's shipped in stack
2020-11-10 08:06:11 -08:00
Simon Michael
a151bcaec7 ; bump hledger-lib, hledger, hledger-ui, hledger-web version to 1.19.99 2020-09-07 12:16:12 -07:00
Simon Michael
8599eda37c allow megaparsec 9 2020-09-07 11:41:57 -07:00
Simon Michael
3170cc7cbd bump version to 1.19 2020-09-01 20:27:17 -07:00
Stephen Morgan
f5a7c84065 ui: Bump version of vty, present in all supported stackage, to get rid of CPP. 2020-08-29 15:08:28 -07:00
Simon Michael
f62797bdb7 ;ui: cabal: simplify dep list to help packdeps reverse deps
(cf snoyberg/packdeps#51)
2020-08-01 12:43:40 -07:00
Simon Michael
3452270241 cli: --color/--colour option; smart emacs, windows autodetection (#1296) 2020-07-14 12:08:36 -07:00
Simon Michael
24996caaeb bump to dev version 2020-06-21 10:35:52 -07:00
Simon Michael
b389ee98dd ;update tested-with 2020-06-07 11:59:36 -07:00
Simon Michael
19020939c9 ;bump versions to 1.18 2020-06-06 17:18:15 -07:00
Simon Michael
c6951bf7c6 ;doc: more tweaks to package readmes, hackage descriptions 2020-03-22 10:49:02 -07:00
Simon Michael
9ca8d7cea9 ;update tested-with 2020-03-22 09:27:50 -07:00
Simon Michael
617c2ecbc5 ;doc: consistent markdown readmes for packages 2020-03-22 09:07:47 -07:00
Simon Michael
1883ff4499 bump to dev version 2020-03-07 15:06:18 -08:00
Simon Michael
ba44d00bed ;bump version to 1.17 2020-03-01 17:29:01 -08:00
Simon Michael
889767f1bd ;bump tested-with
[ci skip]
2020-03-01 17:28:43 -08:00
Simon Michael
7011160bfd ;allow base 4.14/GHC 8.10 2020-01-26 08:12:25 -08:00
Simon Michael
445e8aa3cc add support for megaparsec 8 (#1175) 2020-01-14 09:54:06 -08:00
Simon Michael
cfb2a61ae1 ;bump to dev version; bump man page dates
[ci skip]
2020-01-05 09:04:50 -08:00
Stephen Morgan
74778efcf5 Use nubSort instead of nub . sort. 2020-01-04 08:31:10 -08:00
Simon Michael
ac360893f5 ui: bump to 1.16.1 also, requiring hledger 1.16.1 2019-12-03 08:41:44 -08:00
Simon Michael
87d0491103 ;bump version to 1.16 2019-12-01 10:03:58 -08:00
Simon Michael
08951ac3e8 ;cabal: add ghc 8.8 to tested-with
[ci skip]
2019-11-28 14:59:10 -08:00
Simon Michael
843d394077 cli, ui, web: support GHC 8.8, add stack-ghc8.8.yaml (#1090)
hledger-web needs an unreleased version of json.
2019-11-28 14:31:48 -08:00
Simon Michael
24ce2686f4 drop GHC 7.10/base 4.8 support, finally, due to MonadFail hassles
in JournalReader.hs. If you still need this, feel free to work on
those errors. But hopefully not, because dropping base 4.8 should
permit some code cleanups.
2019-09-14 03:45:37 -07:00
Simon Michael
f65fc47185 ;bump version to 1.15.99
[ci skip]
2019-09-13 08:43:33 -07:00
Simon Michael
acd7010e5a ;doc: "curses-style interface" -> "terminal interface"
[ci skip]
2019-09-10 18:14:16 -07:00
Simon Michael
499e20c0b2 lib, cli, ui: start using Control.Monad.Fail, allow base-compat 0.11
fail is moving out of Monad and into it's own MonadFail class.
This will be enforced in GHC 8.8 (I think).

base-compat/base-compat-batteries 0.11.0 have adapted to this,
and are approaching stackage nightly
(https://github.com/commercialhaskell/stackage/issues/4802).

hledger is now ready to build with base-compat-batteries 0.11.0, once
all of our deps do (eg aeson). We are still compatible with the older
0.10.x and GHC 7.10.3 as well.

For now we are using both fails:

- new fail (from Control.Monad.Fail), used in our parsers, imported
  via base-compat-batteries Control.Monad.Fail.Compat to work with
  older GHC versions.

- old fail (from GHC.Base, exported by Prelude, Control.Monad,
  Control.Monad.State.Strict, Prelude.Compat, ...), used in easytest's
  Test, since I couldn't find their existing fail implementation to update.

To reduce (my) confusion, these are imported carefully, consistently,
and qualified everywhere as Fail.fail and Prelude.fail, with clashing
re-exports suppressed, like so:

import Prelude hiding (fail)
import qualified Prelude (fail)
import Control.Monad.State.Strict hiding (fail)
import "base-compat-batteries" Prelude.Compat hiding (fail)
import qualified "base-compat-batteries" Control.Monad.Fail.Compat as Fail
2019-09-08 17:13:47 -07:00
Simon Michael
29f7654c3a bump versions to 1.15 2019-09-01 08:53:06 -07:00