This makes budget reports more intuitive. It is a temporary hack which
can misorder columns in some cases (if actual and budget activity
occur in a different range of columns). We should redo this in a more
principled way.
Using "actual" and "budget" terminology. "real" is already used in the
context of real/virtual postings. "actual" should be familiar from
unit testing (expected & actual)..
Per feedback from peti. These files will no longer be installed to
cabal's data directory, where it might seem as if they are required at
runtime, which they are not. Packagers will still find them in the
tarball and can put them in the proper place.
Tabular reports from bal, bs etc. have until now been rendered on
the terminal with one final blank line, for readability.
This change drops the blank line.
This is consistent with the non-tabular balance and register output
(not print, which is a special case), and with most unix commands.
The real reason for it, which I admit is flimsy, is that I can now omit
the final delimiter (>=0) when using shelltestrunner 1.9's new format,
making functional tests easier to maintain and more readable.
If there's opposition, this could be reverted.
Previously, if you specified no report interval, the text output of
these commands was a simple report like the original balance command,
with amounts on the left and account names on the right. Also,
balances used arithmetic sign like the balance command.
Now it always draws a table, with account names in the left
column, and shows balances with normal-positive sign, consistent with
the multicolumn reports. Less code, fewer bugs.
Income, liability and equity balances, which until now we have
always displayed as negative numbers, are now shown as normally positive
by these reports.
Negative numbers now indicate a contra-balance (eg an overdrawn
checking account), a net loss, a negative net worth, etc.
This makes these reports more like conventional financial statements,
and easier to read and share with normal people.
For multi-column balance report, if there are no transactions in the
given period for budgeted account, display [0% of <budget>] for
consistency.
If balance is a mix of commodities, convert to cost basis for the
purposes of computing percent of balance spent.
Budget goals specified with periodic transactions (as with
hledger-budget) can now be displayed in balance report (but not in bs/is/cf).
--budget shows the target amount and percentage alongside the actual
amount, per account and period.
Unbudgeted accounts will be hidden, unless --show-unbudgeted is used.
Budgeted accounts are displayed folded (depth-clipped) at a depth
matching the budget specification. Unbudgeted accounts, if shown, are
displayed at their usual depth (in full detail, or according to --depth).
Ledger-style automated postings, previously supported only by
hledger-budget, have landed as a first-class feature. The --auto
flag activates them, so that any postings they generate are
included in reports.
Ledger-style periodic transactions, previously supported only by
hledger-budget, have landed as a first-class feature. The --forecast
flag activates them, so that any transactions they generate are
included in reports.
Compound balance commands like these can now be aware of normal account
balance sign, and sort negative balances accordingly.
This also adds utility-ht as a dependency, only for the uncurry function
right now but it looks potentially useful to have.
Builtin commands are now gathered more tightly in a single module,
Hledger.Cli.Commands, reducing duplication and facilitating change.
The tests command was difficult and has been dropped for now.
The obsolete convert/info/man commands have been dropped.
cli: refactor: a proper commands list, better Main/Commands separation
The legacy "convert" command has been dropped.
The activity command's module is now named consistently.
Older megaparsec is still supported.
Also cleans up our custom parser types,
and some text (un)packing is done in different places
(possible performance impact).
You can select a docs format/viewer with one of the `--info`, `--man`, `--pager`, `--cat` flags.
Otherwise, it will use info if available, otherwise man if available,
otherwise $PAGER if defined, otherwise less if available, otherwise it
prints on stdout (and always prints on stdout when piped).
Preferring info over man might not suit everyone.
See the issue and linked mail list discussion. Ambiguity between the
uncleared state, and the "not cleared" --uncleared flag causes confusion
and friction. At this point it seems best to break with Ledger and
past hledger, pick a new name and drop --uncleared to put an end to it.
The balance command now shows negative amounts in red, when it thinks
ANSI codes are supported, ie when TERM is not "dumb" and stdout is not
being redirected or piped somewhere.
* Add an option to use unicode in balance tables
fixes#522
* Add a test for unicode tables
* Document --pretty-tables
* Support --pretty-tables in BalanceView
* factored out multi-column balance reporting into table creation and string rendering
* preliminary multicolumn balance reporting for BalanceView
* added -T and -A options for balance views
* support for overriding balanceview defaults
* fixed unecessary whitespace stripping to make tree view work
* no need for ViewPatterns in BalanceView
* fixed regression where balancesheet didn't ignore the start date when in single column mode
* removed trailing whitespace to pass tests
* handling warnings in Balance.hs
* force -E to line up lines for bs/is/cf
I was working through the documentation, and this wouldn't work as-is. I noticed that there is an additional argument `assrt` which is not used here. Passing `True` causes this example to work.
Source and compiled versions of all addons in bin are cluttering up
the commands list. I think this was reasonably well behaved, so I'm
re-enabling it, possibly with a slight fix (.c shadowing .exe).
Some notes:
- when there's only one addon with its base filename, its extension is not displayed (as before)
- addons with (case insensitive) .exe extension or no extension are considered compiled
- when there's exactly two addons with the same base filename, and one of them looks compiled, only that one is kept
- modification time is not checked; an old compiled addon overrides a newer source version
- when there's more than two addons with same base filename, all are kept
The previous cleanup defined long help separately from the usage text
generated by cmdargs. This meant keeping flag descriptions synced
between the two, and also the short help was often too verbose and
longer than the long help.
Now, the non-usage bits of long help are defined as pre and postambles
within the cmdargs mode, letting cmdargs generate the long help
including all flags. We derive the short help from this by truncating
at the start of the hledger common flags.
Most of the bundled addons (all but hledger-budget) now use the
new scheme and have pretty reasonable -h and --help output.
We can do more to reduce boilerplate for addon authors.
* cli: fix bug in pivot for postings without tag
Without this fix for postings without tag query checked effective
account which is always empty text ("").
* rewrite: inherit dates, change application order
For budgeting it is important to inherit actual date of posting if it
differs from date of transaction. These dates will be added
as a separate line of comment.
More natural order of rewrites is when result of first defined one is
available for all next rewrites.
* rewrite: factor out Hledger.Data.AutoTransaction
* rewrite: add diff output
With this option you can modify your original files without loosing
inter-transaction comments etc. I.e. you can run:
hledger-rewrite --diff Agency \
--add-posting 'Expenses:Taxes *0.17' \
| patch
As result multiple files should be updated.
Also it is nice to review your changes using colordiff instead of
patch.
* lib: track source lines range for journal
* doc: auto entries and diff output for rewrite
Eg. if you --pivot code, postings from transactions with no code value
were formerly displayed with the unpivoted account name, so you'd see
a mixture of account names and codes. Now we always pivot the account name,
to blank if nothing else. This probably makes reports pretty ugly, it's just another
step towards making --pivot's behaviour clearer.
* Add implicit tags code/desc/payee for --pivot
Additionally allow using of transaction inherited tags.
* Use original posting in query by account name
To be able to query on individual postings and by account name it is
useful to have access to original account name (before pivot).
Especially this is useful when all postings within transaction gets
the same pivot name due.
As a side effect we'll match by alias.
Note: to query on amt it usually expected to see matches with inferred
amounts.
* Remember original postings during infer and pivot
This includes such functions like:
- inferFromAssignment
- inferBalancingAmount
- inferBalancingPrices
- pivotPosting
* Use original postings for hledger print
- Introduce "--explicit" option for "print" command which brings back
old behavior when every inferred number being printed.
- Make "print" by default print original postings without inferred
amounts. But use effective account name to have effect from aliases.
- Instruct shell tests with an new expected output or to use
--explicit option when inferred amounts are checked.
Resolvessimonmichael/hledger#442
- try to clarify naming and meaning of balance/register report modes
and kinds of "balance" displayed. Added balance --change and
register --cumulative flags to clarify report modes.
- with multiple --change/--cumulative/--historical flags use the last
instead of complaining
- register -A is now affected by -H
- options cleanups
-H/--historical now makes a single-column balance report with a start
date show historical balances reflecting earlier postings. This is
equivalent to specifying no start date, but it's more consistent.
* Replace Parsec with Megaparsec (see #289)
This builds upon PR #289 by @rasendubi
* Revert renaming of parseWithState to parseWithCtx
* Fix doctests
* Update for Megaparsec 5
* Specialize parser to improve performance
* Pretty print errors
* Swap StateT and ParsecT
This is necessary to get the correct backtracking behavior, i.e. discard
state changes if the parsing fails.
The journal/timeclock/timedot parsers, instead of constructing (opaque)
journal update functions which are later applied to build the journal,
now construct the journal directly (by modifying the parser state). This
is easier to understand and debug. It also removes any possibility of
the journal updates being a space leak. (They weren't, in fact memory
usage is now slightly higher, but that will be addressed in other ways.)
Also:
Journal data and journal parse info have been merged into one type (for
now), and field names are more consistent.
The ParsedJournal type alias has been added to distinguish being-parsed
and finalised journals.
Journal is now a monoid.
stats: fixed an issue with ordering of include files
journal: fixed an issue with ordering of included same-date transactions
timeclock: sessions can no longer span file boundaries (unclocked-out
sessions will be auto-closed at the end of the file).
expandPath now throws a proper IO error (and requires the IO monad).
Commit 0a90ef4 a few weeks ago caused hledger-ui to show wrong balances
after reloading with the g key, if the journal had changed and you had
specified a query on the command line.
Now, journalReloadIfChanged reloads the full journal without filtering,
like journalReload and withJournalDo.
The hledger package tarball now contains a copy of the generated docs
from all the other packages, so that it can build without being in a
source tree.
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).
Amount display style canonicalisation code and terminology has been
clarified a bit. Individual amounts still have styles; from these we
derive the standard "commodity styles". In user docs, we might call
these "commodity formats" since a Ledger-compatible commodity directive
would use the "format" keyword.
Simple (non-multicolumn) balance reports containing wide characters
should now align correctly (in apps and fonts that show wide chars as
double width). Likewise, the print command.
Wide characters, eg chinese/japanese/korean characters, are typically
rendered wider than latin characters. In some applications (eg gnome
terminal or osx terminal) and fonts (eg monaco) they are exactly double
width. This is a start at making hledger aware of this. A register
report containing wide characters (in descriptions, account names, or
commodity symbols) should now align its columns correctly, when viewed
with a suitable font and application.
This was originally an add-on I used to guess account names for
ledger-autosync. It's nice for quickly looking up a recent transaction
with a guessed or partial description.
In some cases (when the same letter pair was repeated) it was not
picking the post similar past transaction. This was due to the
similarity algorithm using the order-dependent Data.List.intersect, when
it should use Set.intersection.
Also added a reference for the algorithm, which is known as the
Sørensen–Dice coefficient.
Make these modules' names more like the heavily-used types they
define (CliOpts, UIOpts, WebOpts). This is consistent with
RawOptions and ReportOptions, and helps with code navigation.
- register screen:
- smarter width-sensitive layout, with multi-commodity amounts on one line
- items are sorted in date order
- jumps to the latest item by default, with consistent scroll position
- more prerendering, might speed up movement/paging slightly
- themes! --theme to select, --help to list (current themes: default, terminal, greenterm)
- border tweaks - dropped side borders, added side padding
hledger-ui (formerly hledger-vty) is now built on brick, a new
declarative UI layer built on vty. This brings much new power, and
should make the UI much easier to grow and maintain.
At this point, functionality and performance are similar to the old
version. There's no journal entries screen, but entering the postings
screen jumps to the latest posting, and layout is better (multiple
commodities are rendered on one line).
Requires unreleased brick from github (HEAD or some branch) for now.
The balance command's --format option (in single-column mode) can now
adjust the rendering of multi-line strings, such as amounts with multiple
commodities. To control this, begin the format string with one of:
%_ - renders on multiple lines, bottom-aligned (the default)
%^ - renders on multiple lines, top-aligned
%, - render on one line, comma-separated
Also the final total (and the line above it) now adapt themselves to a
custom format.
The --format option's OutputFormat type was named confusingly like the
--output-format option. It has been renamed StringFormat to distinguish
it from StorageFormat (aka the data file format, referenced by
--output-format). Related code and types have been consolidated.
Also the (single-column) balance report's item rendering has had
some cleanup.
Simpler and clearer. We now have "transaction prices" (recorded as part
of transaction amounts) and "market prices" (recorded with P
directives). Both are matters of historical record, also this avoids
confusion with the balance command's "historical balances".
Initial support of market value reporting and currency conversion,
similar in spirit to Ledger's. The balance command now has a -V/--value
flag that converts all the reported amounts using their "default market
price". That is the latest market price (P directive, formerly called
"historical prices") found in the journal for their commodity that is on
or before the report end date.
Unlike Ledger, hledger's -V only uses the market prices recorded with P
directives, ignoring transaction prices recorded as part of posting
amounts (which -B/--cost uses). Using -B and -V together is allowed.
A transaction/posting status of ! (pending) was effectively equivalent
to * (cleared). Now it's a separate state, not matched by --cleared.
The new Ledger-compatible --pending flag matches it, and so does
--uncleared. The equivalent search queries are now status:*, status:!
and status: (the old status:1 and status:0 spellings are deprecated).
Since we interpret --uncleared and status: as "any state except cleared",
it's not currently possible to match things which are neither cleared
nor pending.
The regex account aliases added in 0.24 trip up people switching between
hledger and Ledger. (Also they are currently slow).
This change makes the old non-regex aliases the default; they are
unsurprising, useful, and pretty close in functionality to Ledger's.
The new regex aliases are also available; they must be enclosed in
forward slashes. Ledger effectively ignores these, which is ok.
Also clarify docs, refactor, and use the same parser for alias
directives and alias options
Refactored and enhanced the --width option used by register (and other
commands in future). register now uses the full terminal width by
default except on windows. Specifically, the output width is set from:
1. a --width option
2. or a COLUMNS environment variable (NB: not the same as a bash shell var)
3. or on POSIX (non-windows) systems, the current terminal width
4. or the default, 80 characters.
Also, register now accepts a description column width as part of
--width's argument, comma-separated (--width W,D). This adjusts the
relative widths of register's description and account columns, which are
normally about half of (W-40):
<--------------------------------- width (W) ---------------------------------->
date (10) description (D) account (W-41-D) amount (12) balance (12)
DDDDDDDDDD dddddddddddddddddddd aaaaaaaaaaaaaaaaaaa AAAAAAAAAAAA AAAAAAAAAAAA
Examples:
$ hledger reg # use terminal width on posix
$ hledger reg -w 100 # width 100, equal description/account widths
$ hledger reg -w 100,40 # width 100, wider description
$ hledger reg -w $COLUMNS,100 # terminal width and set description width
-N doesn't seem to be used by Ledger currently; we'll
use it as shortcut for --no-total.
-T is used by Ledger to set a custom value expression for the final
total. I'm going to take it as a shortcut for --row-total instead.
A status: query term no longer accepts * as a synonym for 1,
which was a bit confusing since 1 matches both * and !.
For now, it takes a value of 1 (true) or anything else (false).
NOTE: this is important to correctly build JournalContext
NOTE: currently a list reverse must done at the end,
maybe using a Data.Queue would be more efficient.
Use two options and less surprising behaviour. Notes:
* a recognised suffix on the output file name can set the format
* but --output-format always wins
* -o - means stdout
* currently these are command-specific options on balance, print, register
* later, --output-file should become a global option
* and --output-format should be available at least on all commands which
support multiple formats. Each command may support a different set of
output formats, which should be listed in its command line help.
The -A flag now enables -E, so that with a report interval the averages
are always per-period and not per-report-line. (Without a report
interval, -E is already the default).
To avoid totally breaking the layout of multicolumn balance reports when
there are multi-commodity balances, these are now rendered on a single
line, comma separated. Simple balance reports still use the old vertical
format for now.
This is a workaround for a cmdargs limitation. Having "--debug 2"
or "--width 100" produce no output (because the number is parsed
as a separate argument) is too annoying.
Can be helpful when reading Ledger files, where assertions may have
different semantics; or for getting some answers from your journal
to help you fix your assertions.
Could be called --no-assertions, but this might create surprise when it
has an effect contrary to --no-new-accounts.
I had to add another flag throughout the parsers & journal read
functions, ok for now.
We provided a very limited implementation of --display only for one use
case: to see an accurate running balance. Now that is achieved more easily
with -H/--historical, similar to the balance command, and --display
can be dropped.
Drop the special case where we hide an add-on's source version if a
compiled version is also present. Better to be simple and explicit.
Improve robustness of command parsing, eg "hledger addon.hs"
will now work even though the command is listed as "addon".
And ignore any add-ons which would shadow a built-in command
(or any of the official command aliases displayed in the command list,
like "bal" and "reg"). Built-ins may not be replaced by an add-on.
We now hide add-ons' file extensions in the commands list except when
they are needed for disambiguation. There are three cases:
- when an add-on exists with only one file extension, the extension is
not displayed.
- when an add-on has exactly two variants, one with the .[l]hs extension
and the other with none or .exe - presumably the source and compiled
versions - we omit the source version from the list.
- otherwise when an add-on has multiple variants differing by file
extension, they are listed with their extensions displayed.
In addition to hledger-X and hledger-X.hs, other common executable file
extensions will be recognised. In particular, on windows hledger now
lists hledger-web.exe as an add-on. The current list is:
bat, com, exe, hs, lhs, pl, py, rb, rkt, sh.
Periodic, cumulative and historical multicolumn balance reports are now
generated by one code path, which helps with consistency and reducing
the bug/test surface. --tree now also works with --cumulative or
--historical.
Multicolumn balance reports can now be switched to a hierarchical view
with --tree. This is similar to the single column balance report with
--no-elide, ie we do not elide boring parents into the following line.
The --flat and --tree flags are opposites; the last one to appear on
the command line is decisive (in future, if not already).
To end a transaction, period is required because there might be another
default posting being suggested, and enter usually means "accept the
default". But having to switch back and forth from enter to period
during data entry is a bit unergonomic, requiring too much thought.
Now, in the (common) case where you have entered a balanced transaction
and there are no more default postings to suggest, enter is also allowed
to end the transaction. Experimental; this might be too inconsistent.
The code is now much more manageable, faciliating further
improvements. Completion now works at all prompts, and
will insert the default value if the input area is empty.
Account and amount defaults are more robust and useful
in various situations. There might be a slight regression
with default commodity handling.
Previously executables with eg digits in their name were ignored.
It now finds all files beginning with hledger-, optionally ending with
.hs or .lhs, and with no other dots in the name.
As before, we don't check for executable permission (performance ?).
Two new multi-column balance report modes show ending balance per
period: `--cumulative`, starting from 0, and `--historical`, starting
from the historical starting balance.
The balance command's specification has been clarified and consolidated
in the Balance.hs haddock. Reports.hs has also had haddock updates. The
old AccountsReport type is now BalanceReport, still used by
single-column balance report. The new MultiBalanceReport type is used by
the multi-column reports.
Command line processing has been overhauled and made more consistent,
and now has tests and extensive debug output. More flags now work
both before and after COMMAND: -f, --rule-file, --alias, --help,
--debug, --version. Command line help, command aliases, API docs and
code have been improved.
In general, when cabal CPP macros like min_PKG are used, they should
be made optional. If they are not present, assume a developer is
building and choose the latest libs.
As part of adding -w in december I cleaned up/adjusted register field
widths, and didn't make all the tests pass. This commit makes one more
width adjustment
(one space after the date instead of two) and fixes all tests depending on register output.
Revisiting the "directory 1.2 can be installed with older GHC, so we'd better pin it in the cabal file" change.
Requiring directory 1.2 made hledger-web uninstallable with GHC < 7.6 because current yesod depends on ghc
which depends on a particular directory version and can't be upgraded. This is a smarter fix, which should let
all hledger packages build with whatever directory version is installed.
hledger now requires directory >= 1.2. This might make installation
more complex on old GHC versions; if it turns out to be needed,
perhaps it can be conditionalised again, more accurately, in the cabal file.
- A transaction code may be entered following the date in parentheses, eg: DATE (CODE)
- A transaction comment (possibly containing tags) may be entered following the description, eg: DESC ; COMMENT
- A posting comment may be entered following the amount, eg: AMOUNT ; COMMENT
With one or more command line arguments, they will used as defaults
for the first transaction, and history will be ignored. Subsequent
transactions will use history for defaults (and now only the best match
from history is shown).
- tidier output
- show more help
- suggest . only for recording, not for quitting (though it still works)
- show each transaction after adding it (#52)
- don't parse . as a zero amount
- don't show [] when there's no default
This refactoring fixes an O(n^2) slowdown in the balance command with
large numbers of accounts. It's now speedy, and the implementation is
clearer. To facilitate this, the Account type now represents a tree of
accounts which can easily be traversed up or down (and/or flattened
into a list).
Benchmark on a 2010 macbook:
+-------------------------------------------++--------------+------------+--------+
| || before: | after: | |
| || hledger-0.18 | hledgeropt | ledger |
+===========================================++==============+============+========+
| -f data/100x100x10.journal balance || 0.21 | 0.07 | 0.09 |
| -f data/1000x1000x10.journal balance || 10.13 | 0.47 | 0.62 |
| -f data/1000x10000x10.journal balance || 40.67 | 0.67 | 1.01 |
| -f data/10000x1000x10.journal balance || 15.01 | 3.22 | 2.36 |
| -f data/10000x1000x10.journal balance aa || 4.77 | 4.40 | 2.33 |
+-------------------------------------------++--------------+------------+--------+
hledger and hledger-web were reading their version number (and program
name) from their cabal files at compile time using cabal-file-th,
which allowed the version number be maintained in one place (per
package).
This meant you had to be in same directory as the cabal file when
building, which made life more complicated, eg emacs compilation mode
could not jump to errors. Also, it slowed down building slightly, and
is a factor in hledger Debian packages being unavailable on a number
of platforms (we also use TH for report templates).
Now, the build version is set with a CPP VERSION flag, which seems
simpler overall. For cabal builds, this needs to be configured
manually in a few more places in each cabal file. For makefile builds,
it is set it to the name of the most recent darcs tag (which should be
more useful than the old behaviour). If not set, it defaults to the
blank string, useful eg for haddock. And, all makefile builds now run
from the top directory.
- use new query system for command line too, filterspec is no more
- move unit tests near the code they test, run them in bottom up order, add more
- more precise Show instances, used for debugging not ui
