Module documentation for 1.12.1
lightweight, portable, dependable accounting tools
hledger is a computer program for easily tracking money, time, or other commodities, on unix, mac and windows (and web-capable mobile devices, to some extent).
It is first a command-line tool, but there are also curses-style and web interfaces, and a Haskell library (http://hackage.haskell.org/package/hledger-lib) for building your own programs and scripts (hledger is written in Haskell). hledger was inspired by and is largely compatible with Ledger. hledger is free software available under the GNU General Public License v3+.
hledger aims to help both computer experts and regular folks to gain clarity and control in their finances and time management, but currently it is a bit more suited to techies. I use it every day to:
- track spending and income
- see time reports by day/week/month/project
- get accurate numbers for client billing and tax filing
- track invoices
Though limited in features, hledger is lightweight, usable and reliable. For some, it is a simpler, less distracting, more future-proof alternative to Quicken or GnuCash.
For more, see http://hledger.org.
Support us with a monthly donation and help us continue our activities. [Become a backer]
Become a sponsor and get your logo on our README on Github with a link to your site. [Become a sponsor]
User-visible changes in the hledger command line tool and library.
require easytest <0.3 to fix build issue
fix some CSV parse errors which weren’t in human readable format
- fix missing Commodities.txt build error
journal: subaccount-including balance assertions have been added, with syntax =* and ==* (experimental) (#290)
new commodities command lists commodity symbols
new –invert option flips sign of amounts in reports
- print, register: restore the accidentally dropped -o, -O flags (#967)
- stop depending on here to avoid haskell-src-meta/stackage blockage.
cli: reorganised commands list. Addons now have a + prefix.
cli: the command line help and manual section for all hledger’s commands are now consistent, and generated from the same source.
cli: comprehensive bash completion support is now provided (in shell-completion/). See how-to in the Cookbook. (Jakob Schöttl)
balance –budget: budget amounts now aggregate hierarchically, like account balances. Unbudgeted accounts can be shown with -E/–empty (along with zero-balance accounts), and the –show-budgeted flag has been dropped. (Dmitry Astapov)
balance: new –transpose flag switches the rows and columns of tabular balance reports (in txt and csv output formats). (Dmitry Astapov)
close: generated balance assertions now have exact amounts with all decimal digits, ignoring display precision. Also, balance assertion amounts will no longer contain prices. (#941, #824, #958)
files: now shows up in the commands list
import: be silent when there’s nothing to import
roi: percentages smaller than 0.01% are displayed as zero (Dmitry Astapov)
stats, ui: correct file order is preserved when using –auto (#949)
journal: account directive: the account name can now be followed by a comment on the same line
journal: account directive: account types for the bs/bse/cf/is commands can now be set with a
type:tag, whose value is
X(case-insensitive). The previous syntax (
account assets A) is now deprecated.
journal: account directive: account sort codes like
account 1000(introduced in 1.9, deprecated in 1.11) are no longer supported.
journal: transaction modifiers (auto postings) can affect periodic transactions (–auto can add postings to transactions generated with –forecast). (Dmitry Astapov)
journal: balance assertion errors now show exact amounts with all decimal digits. Previously it was possible, in case of a commodity directive limiting the display precision, to have a balance assertion error with asserted and actual amounts looking the same. (#941)
journal: fixed a periodic transaction parsing failure (#942) (Dmitry Astapov)
- roi: use math-functions lib instead of statistics, be more stackage nightly compatible
install script: ensure a new-enough version of stack; more informative output
build with GHC 8.6/base-4.12 (Peter Simons)
add required upper bound for statistics (Samuel May)
–anon anonymises more thoroughly (including linked original postings) (Moritz Kiefer)
unbalanced transaction errors now include location info (Mykola Orliuk)
accounts command: –drop also affects the default flat output, without needing an explicit –flat flag
accounts command: the –codes flag has been dropped
accounts command: filtering by non-account-name queries now works
add command: fix transaction rendering regression during data entry and in journal file
balance command: fix wrongful eliding of zero-balance parent accounts in tree mode (Dmitry Astapov)
journal format, bs/bse/cf/is commands: account directives can declare account types (#877) Previously you had to use one of the standard english account names (assets, liabilities..) for top-level accounts, if you wanted them to appear in the right place in the balancesheet, balancesheetequity, cashflow or incomestatement reports.
Now you can use your preferred account names, and use account directives to declare which accounting class (Asset, Liability, Equity, Revenue or eXpense) an account (and its subaccounts) belongs to, by writing one of the letters A, L, E, R, X after the account name, after two or more spaces. This syntax may change (see issue). Experimental.
Currently we allow unlimited account type declarations anywhere in the account tree. So you could declare a liability account somewhere under assets, and maybe a revenue account under that, and another asset account even further down. In such cases you start to see oddities like accounts appearing in multiple places in a tree-mode report. I have left it this way for now in case it helps with, eg, modelling contra accounts, or combining multiple files each with their own account type declarations. (In that scenario, if we only allowed type declarations on top-level accounts, or only allowed a single account of each type, complications seem likely.)
journal format: periodic transaction rules now require a double space separator. In periodic transaction rules which specify a transaction description or same-line transaction comment, this must be separated from the period expression by two or more spaces, to prevent ambiguous parsing. Eg this will parse correctly as “monthly” thanks to the double space:
~ monthly In 2020 we'll end this monthly transaction.
journal format: exact/complete balance assertions (Samuel May). A stronger kind of balance assertion, written with a double equals sign, asserts an account’s complete account balance, not just the balance in one commodity. (But only if it is a single-commodity balance, for now.) Eg:
1/1 (a) A 1 (a) B 1 (a) 0 = A 1 ; commodity A balance assertion, succeeds (a) 0 == A 1 ; complete balance assertion, fails
journal format: account directives now allow whitespace or a comment after the account name
journal format: using ~ for home directory in include directives now works (#896) (Mykola Orliuk)
journal format: prevent misleading parse error messages with cyclic include directives (#853) (Alex Chen)
journal format: transaction modifier multipliers handle total-priced amounts correctly (#928). Multipliers (*N) in transaction modifier rules did not multiply total-priced amounts properly. Now the total prices are also multiplied, keeping the transaction balanced.
journal format: do amount inference/balance assignments/assertions before transaction modifiers (#893, #908) (Jesse Rosenthal) Previously, transaction modifier (auto postings) rules were applied before missing amounts were inferred. This meant amount multipliers could generate too many missing-amount postings, making the transaction unbalanceable (#893).
Now, missing amount inference (and balance assignments, and balance assertions, which are interdependent) are done earlier, before transaction modifier rules are applied (#900, #903).
Also, we now disallow the combination of balance assignments and transaction modifier rules which both affect the same account, which could otherwise cause confusing balance assertion failures (#912). (Because assignments now generate amounts to satisfy balance assertions before transaction modifier rules are applied (#908).)
journal format: periodic transaction rules are now aware of Y default year directives. (#892) Ie when a default year Y is in effect, they resolve partial or relative dates using Y/1/1 as the reference date, rather than today’s date.
- fix wrong transaction rendering in balance assertion errors and when using the add command
The default display order of accounts is now influenced by the order of account directives. Accounts declared by account directives are displayed first (top-most), in declaration order, followed by undeclared accounts in alphabetical order. Numeric account codes are no longer used, and are ignored and considered deprecated.
So if your accounts are displaying in a weird order after upgrading, and you want them alphabetical like before, just sort your account directives alphabetically.
Account sorting (by name, by declaration, by amount) is now more robust and supported consistently by all commands (accounts, balance, bs..) in all modes (tree & flat, tabular & non-tabular).
close: new –opening/–closing flags to print only the opening or closing transaction
files: a new command to list included files
prices: query arguments are now supported. Prices can be filtered by date, and postings providing transaction prices can also be filtered.
rewrite: help clarifies relation to print –auto (#745)
roi: a new command to compute return on investment, based on hledger-irr
test: has more verbose output, more informative failure messages, and no longer tries to read the journal
csv: We use a more robust CSV lib (cassava) and now support non-comma separators, eg –separator ‘;’ (experimental, this flag will probably become a CSV rule) (#829)
csv: interpolated field names in values are now properly case insensitive, so this works:
fields ...,Transaction_Date,... date %Transaction_Date
journal: D (default commodity) directives no longer break multiplier amounts in transaction modifiers (AKA automated postings) (#860)
journal: “Automated Postings” have been renamed to “Transaction Modifiers”.
journal: transaction comments in transaction modifier rules are now parsed correctly. (#745)
journal: when include files form a cycle, we give an error instead of hanging.
upper-case day/month names in period expressions no longer give an error (#847, #852)
journal: many parse error messages have become more informative, and some now show the source line and error location.
journal: ;tag: is no longer parsed as a tag named “;tag” (#655)
journal: transaction price amounts having their own price amounts is now a parse error
journal: amounts with space as digit group separator and trailing whitespace now parse correctly (#780)
journal: in amounts containing digits and a single space, the space is now interpreted as a digit group separator, not a decimal separator (#749)
journal: in commodity/format/D directives, the amount must now include a decimal separator.
When more precise control is needed over number parsing, our recommended solution is commodity directives. Commodity directives that don’t specify the decimal separator leave things ambiguous, increasing the chance of misparsing numbers. In some cases it could cause amounts with a decimal point to be parsed as if with a digit group separator, so 1.234 became 1234.
It seems the simple and really only way to do this reliably is to require an explicit decimal point character. Most folks probably do this already. Unfortunately, it makes another potential incompatiblity with ledger and beancount journals. But the error message will be clear and easy to work around.
journal: directives currently have diverse and somewhat tricky semantics, especially with multiple files. The manual now describes their behaviour precisely.
apply accountdirectives now affect
journal: periodic transactions can now have all the usual transaction fields (status mark, code, description, comment), for generating more expressive forecast transactions.
journal: forecast transactions now have the generating period expression attached as a tag named “recur”.
journal: periodic transactions now start on the first instance of the recurring date, rather than the day after the last regular transaction (#750)
journal: periodic transaction rules now allow period expressions relative to today’s date
csv: amount-in/amount-out errors are more detailed
balance: –drop is now ignored when not in flat mode, rather than producing a corrupted report (#754)
budget: –drop now preserves the top-level account in –budget reports
register: in CSV output, the code field is now included (#746)
smart dates now allow the YYYYMM format, and are better documented
use hledger-lib 1.10
use hledger-lib 1.9.1
budget (balance –budget): monthly columns are displayed in the proper order. This fixes a regression in 1.9.
budget: budgets can be built from periodic transactions with different intervals again. In 1.9, budgets were restricted to a single interval, but this was a mistake. This restores the 1.5 behaviour.
budget: budget reports are more intuitive and much less likely to produce no output.
budget: when no report interval is specified, a budget report for the whole journal period is shown.
budget: periodic transactions and the requested report period can each have their own start/end dates, and the resulting report will span the union of those periods, showing zeroes where data is missing.
budget: total row and total/average columns are now calculated correctly
budget: actual, percentage, and goal amounts are now aligned in columns for better readability (usually, unless numbers get huge).
budget: combining –budget and –sort-amount is not yet supported and now gives an error.
csv: handle “-%amount” in a rule when the CSV amount is parenthesised (#736)
journal: automated postings are now generated early, before journal finalisation, so they are present for amount inference, transaction balancing, and balance assertions (#729)
journal: automated postings are now inserted right after the posting that triggered them (#729)
cli: command-line account aliases are now applied early, before journal finalisation, so they are equivalent to alias directives in the journal (#730)
journal: inferred amounts now have the appropriate standard amount style applied (setting the precision correctly, eg). (#737)
journal: when checking for balanced transactions, amount styles declared with commodity directives are also used (previously only inferred amount styles were).
support ghc 8.4, latest deps
journal: account directives can define a numeric account code to customize sorting. bal/bs/cf/is will sort accounts by account code, if any, then account name.
journal: support scientific number notation (#704, #706)
csv: reading a CSV file containing no records is no longer an error
cli: when the system text encoding is UTF-8, ignore any UTF-8 BOM prefix found when reading files. (Paypal’s new CSV has this BOM prefix, causing a confusing parse error.)
cli: tabular reports no longer have a trailing blank line added. (This allows omitting the “>=0” delimiters in our functional tests, making them easier to read and maintain.)
acc: the accounts command now has –declared and –used flags
bal: the –invert flag flips all signs
bal: –drop now works with CSV output
bal/bs/bse/cf/is: show overall report span in title
bal/bs/bse/cf/is: show short month names as headings in monthly reports
bal/bs/bse/cf/is: these commands can now generate HTML output
bal/bs/is/cf: drop short name and indent fields from multicolumn CSV
bs/bse/cf/is: these, the “financial statement” commands, now show normal income, liability and equity balances as positive numbers. Negative numbers now indicate a contra-balance (eg an overdrawn checking account), a net loss, or a negative net worth. This makes these reports more like conventional financial statements, and easier to read and share with others. (Other commands, like balance, have not changed.) (experimental)
bs/cf/is: always show a tabular report, even with no report interval. Previously you would get a simple borderless report like the original balance command. Less code, fewer bugs.
bs/bse/cf/is: in CSV output, don’t repeat the headings row for each subreport
budget: warn that CSV output with bal –budget is unimplemented
budget: bal –budget shows budget goals even with no or zero actual amounts. Makes budget reports more intuitive, at the cost of a temporary hack which may misorder columns in some cases (if actual and budget activity occur in a different range of columns).
budget: –budget uses only periodic txns with the selected interval.
Budgets with different interval, eg a daily and weekly budget, are independent.
budget: show mostly fixed-width columns for readability
budget: fix bug where a budget report could include budget goals ending on the day before the report start date (splitSpan issue)
close: the equity command has been renamed to close. It now ignores any begin date (it always closes historical end balances). It also ignores –date2.
–auto adds Ledger-style automated postings to transactions (Dmitry Astapov, Mykola Orliuk)
–forecast generates Ledger-style periodic transactions in the future (Dmitry Astapov, Mykola Orliuk)
-V/–value uses today’s market prices by default, not those of last transaction date. #683, #648
add: suggest implied (parent) and declared (by account directives) account names also
bal: –budget shows performance compared to budget goals defined with periodic transactions. Accounts with budget goals are displayed folded (depth-clipped) at a depth matching the budget specification. Unbudgeted accounts are hidden, or with –show-unbudgeted, shown at their usual depth. (Dmitry Astapov)
import: the output of –dry-run is now valid journal format
print: -B shows converted amounts again, as in 1.1, even without -x. #551 (Mykola Orliuk, Simon Michael)
tag: the first argument now filters tag names, additional arguments filter transactions (#261)
remove upper bounds on all but hledger* and base (experimental)
cli: a @FILE argument reads flags & args from FILE, one per line
cli: reorganized commands list, added some new command aliases:
- accounts: a
- balance: b
- print: p, txns
- register: r
cli: accept -NUM as a shortcut for –depth=NUM (eg: -2)
cli: improve command-line help for –date2 (#604)
cli: make –help and -h the same, drop –man and –info for now (#579)
help: offers multiple formats, accepts topic substrings. The separate info/man commands have been dropped. help now chooses an appropriate documentation format as follows:
- it uses info if available,
- otherwise man if available,
- otherwise $PAGER if defined,
- otherwise less if available,
- otherwise it prints on stdout
- (and it always prints on stdout when piped).
You can override this with the
bal/bs/cf/is: –sort-amount/-S sorts by largest amount instead of account name
bs/cf/is: support –output-file and –output-format=txt|csv The CSV output should be reasonably ok for dragging into a spreadsheet and reformatting.
bal/bs/cf/is: consistent double space between columns, consistent single final blank line. Previously, amounts wider than the column headings would be separated by only a single space.
bs/is: don’t let an empty subreport disable the grand totals (fixes #588)
cf: exclude asset accounts with “:fixed” in their name (Christian G. Warden, Simon Michael, #584)
new balancesheetequity command: like balancesheet but also shows equity accounts (Nicholas Niro)
new import command: adds new transactions seen in one or more input files to the main journal file
print: –new shows only transactions added since last time (saves state in .latest.JOURNALFILE file)
new tags command: lists tags in matched transactions
most addons formerly shipped in bin/ are now builtin commands. These include: check-dates, check-dupes, equity, prices, print-unique, register-match, rewrite.
refactor: new Commands module and subdirectory. Builtin commands are now gathered more tightly in a single module, Hledger.Cli.Commands, facilitating change. The legacy “convert” command has been dropped.
refactor: BalanceView -> CompoundBalanceCommand
deps: drop support for directory < 1.2
deps: allow ansi-terminal 0.7
deps: drop oldtime flag, require time 1.5+
deps: simplify shakespeare bounds
deps: remove ghc < 7.6 support
bs/is: don’t let an empty subreport disable the grand totals (#588)
allow megaparsec 6 (#594)
allow megaparsec-6.1 (Hans-Peter Deifel)
restore upper bounds on hledger packages
The “uncleared” transaction/posting status, and associated UI flags
and keys, have been renamed to “unmarked” to remove ambiguity and
confusion. This means that we have dropped the
-U flag now matches only unmarked things and not pending
ones. See the issue and linked mail list discussion for more
Also the -P short flag has been added for –pending, and the -U/-P/-C flags can be combined.
bs/is: fix “Ratio has zero denominator” error (#535)
bs/is/cf: fix –flat (#552) (Justin Le, Simon Michael)
bal/bs/is/cf: show negative amounts in red (Simon Michael, Justin Le). These commands now shows negative amounts in red, when hledger detects that ANSI codes are supported, (ie when TERM is not “dumb” and stdout is not being redirected or piped).
print: show pending mark on postings (fixes #563). A pending mark on postings is now displayed, just like a cleared mark. Also there will now be a space between the mark and account name.
print: amounts are now better aligned, eg when there are posting status marks or virtual postings
“hledger” and “hledger -h” now print a better organised commands list and general usage message respectively (#297).
The common reporting flags can now be used anywhere on the command line.
Fixed deduplication of addons in commands list.
Fixed ugly stack traces in command line parse error messages.
The -V/–value flag is now a global report flag, so it works with balance, print, register, balancesheet, incomestatement, cashflow, etc. (Justin Le)
--pivot global reporting option replaces all account names with
the value of some other field or tag. It has been improved, eg:
- we don’t add the field/tag name name as a prefix
- when pivoting on a tag, if the tag is missing we show a blank (rather than showing mixed tag values and account names)
- a pipe character delimiter may be used in descriptions to get a more accurate
and useful payee report (
hledger balance --pivot payee)
Easier installation: move add-ons and example scripts to bin/, convert to stack scripts, add a build script to install all deps, add some functional tests, test add-ons with Travis CI, add installation docs to download page.
Improved docs: all addons now contain their own documentation. Most of them (all but hledger-budget) use a new reduced-boilerplate declaration format and can show short (-h) and long (–help) command line help. (Long help is declared with pre and postambles to the generated options help, short help is that truncated at the start of the hledger common flags.)
hledger now shows a cleaner list of addon commands, showing only the
compiled version of an addon when both source and compiled versions
are in $PATH. (Addons with .exe extension or no extension are
considered compiled. Modification time is not checked, ie, an old
compiled addon will override a newer source version. If there are
three or more versions of an addon, all are shown. )
New addons added/included:
- autosync - example symlink to ledger-autosync
- budget - experimental budget reporting command supporting Ledger-like periodic transactions and automated transactions (Mykola Orliuk)
- chart - pie-chart-generating prototype, a repackaging of the old hledger-chart tool
- check - more powerful balance assertions (Michael Walker)
- check-dupes - find accounts sharing the same leaf name (Stefano Rodighiero)
- prices - show all market price records (Mykola Orliuk)
- register-match - a helper for ledger-autosync’s deduplication, finds best match for a transaction description
The equity command now always generates a valid journal transaction, handles prices better, and adds balance assertions (Mykola Orliuk).
The rewrite command is more robust and powerful (Mykola Orliuk):
in addition to command-line rewrite options, it understands rewrite rules defined in the journal, similar to Ledger’s automated transactions (#99). Eg:
= ^income (liabilities:tax) *.33 = expenses:gifts budget:gifts *-1 assets:budget *1
it can generate diff output, allowing easier review of the proposed changes, and safe modification of original journal files (preserving file-level comments and directives). Eg:
hledger-rewrite --diff Agency --add-posting 'Expenses:Taxes *0.17' | patch
rewrites can affect multiple postings in a transaction, not just one.
posting-specific dates are handled better
A new –pretty-tables option uses unicode characters for rendering table borders in multicolumn reports (#522) (Moritz Kiefer)
These commands are now more powerful, able to show multicolumn reports and generally having the same features as the balance command. (Justin Le)
balancesheet has always ignored a begin date specified with a
-p option; now it also ignores a begin date specified with a
query. (Related discussion at #531)
The output of print is now always a valid journal (fixes #465) (Mykola Orliuk).
print now tries to preserves the format of implicit/explicit balancing
amounts and prices, by default. To print with all amounts explicit,
use the new
--explicit/-x flag (fixes #442). (Mykola Orliuk)
Don’t lose the commodity of zero amounts/zero balance assertions (fixes #475) (Mykola Orliuk)
Fix a regression in the readability of option parsing errors (#478) (Hans-Peter Deifel)
Fix an example in Cli/Main.hs (Steven R. Baker)
Allow megaparsec 5.2 (#503)
with -V, don’t ignore market prices in the future (#453, #403)
with -V and multiple same-date market prices, use the last parsed not the highest price (#403)
fix non-existent “oldtime” dependency (#431)
extra/hledger-equity.hs now generates valid journal format when there are multiple commodities
allow megaparsec 5.0 or 5.1
fix benchmark build failure (#423)
suggest only one commodity at a time as default amount (#383)
(since we currently can’t input more than one at a time)
added –change flag for consistency
-H/–historical now also affects single-column balance reports with a start date (#392).
This has the same effect as just omitting the start date, but adds consistency.
in CSV output, render amounts in one-line format (#336)
- fix an infinite loop (#393)
- in CSV output, fix and rename the transaction id field
fix a sorting regression with –date2 (#326)
–average/-A is now affected by –historical/-H
added –cumulative flag for consistency
in CSV output, include the transaction id and rename the total field (#391)
- fixed an issue with ordering of include files
–pivot option added, groups postings by tag instead of account (#323) (Malte Brandy)
–anon option added, obfuscates account names and descriptions (#265) (Brian Scott)
(Only affects the hledger tool, for now.)
try to clarify balance/register’s various report modes,
kinds of “balance” displayed, and related options and language.
with multiple –change/–cumulative/–historical flags, use the last one instead of complaining
don’t add the “d” suffix when displaying day periods
added GHC 8 support, dropped GHC 7.6 and 7.8 support.
GHC 7.8 support could be restored with small code changes and a maintainer.
a cabal.project file has been added (Moritz Kiefer)
use hpack for maintaining cabal files (#371).
Instead of editing cabal files directly, we now edit the less verbose and less redundant package.yaml files and let stack (or hpack) update the cabal files. We commit both the .yaml and .cabal files.
clean up some old cabal flags
tools/simplebench has been spun off as the quickbench package.
add Appveyor CI builds, provide up-to-date binaries for Windows
extra: add a bunch of CSV rules examples
the website is simpler, clearer, and more mobile-friendly.
Docs are now collected on a single page and organised by type: getting started, reference, more.
reference docs have been split into one manual for each executable and file format.
This helps with maintenance and packaging and also should make it easier to see what’s available and to read just what you need.
manuals are now provided in html, plain text, man and info formats
generated from the same source by a new Shake-based docs build system. (#292)
versioned manuals are provided on the website, covering recent releases and the latest dev version (#385, #387)
manuals are built in to the hledger executables, allowing easy offline reading on all platforms.
PROG -h shows PROG's command-line usage PROG --help shows PROG's manual (fixed width) PROG --man shows PROG's manual with man (formatted/paged) PROG --info shows PROG's manual with info (hypertext) hledger help [TOPIC] shows any manual hledger man [TOPIC] shows any manual with man hledger info [TOPIC] shows any manual with info
the general and reporting options are now listed in all executable manuals.
We assume any of them which are unsupported are harmlessly ignored.
demo.hledger.org is using beancount’s example journal.
This is the somewhat realistic example journal from the beancount project, tweaked for hledger.
minor copyedits (jungle-boogie)
parsing multiple input files is now robust.
When multiple -f options are provided, we now parse each file individually rather than just concatenating them, so they can have different formats (#320). Note this also means that directives (like `Y` or `alias`) no longer carry over from one file to the next.
-I has been added as the short flag for –ignore-assertions
(this is different from Ledger’s CLI, but useful for hledger-ui).
parsing an argument-less –debug option is more robust
- Regular expression account aliases are now fast enough that you can use lots of them without slowing things down. They now take O(aliases x accounts) time, instead of O(aliases x transactions); also, regular expressions are no longer recompiled unnecessarily.
Each hledger package now includes one or more man pages, generated from markdown by the mighty pandoc. Currently there are six: one for each main executable and each input file format. Currently these somewhat duplicate the manual on the website; this will be resolved somehow. (#282).
The site is now built with hakyll-std, a generic hakyll script.
hledger once again has a HCAR entry.
The hledger cabal files are now generated from package.yaml files by hpack, in principle, removing a lot of error-prone duplication and boilerplate. (In practice, both files are being updated manually for the moment, until hpack supports flags and conditional blocks.)
Time/allocation and heap profiling is working again, and easier:
make quickprof-CMDgenerates a profile for CMD, which runs against one of the sample journals. (CMD must be one word, enclosing in double quotes isn’t working here for some reason).
make quickheap-CMDgenerates a heap profile for CMD, in hledgerprof.ps, and tries to open it in a viewer (currently the mac-friendly “open” executable, so you may need to adjust this in the makefile). As with quickprof, CMD must be one word and runs against one of the sample journals.
make hledgerprofbuilds the hledgerprof executable used for time/allocation profiling.
make hledgercovbuilds the hledgercov executable used for coverage reports.
Travis CI now tests the build on each github push and announces status changes by email and on #hledger.
Dates must now begin with a digit (not /, eg).
The comment directive longer requires an end comment, and will extend to the end of the file(s) without it.
Output (balance reports, register reports, print output etc.) containing wide characters, eg chinese/japanese/korean characters, should now align correctly, when viewed in apps and fonts that show wide characters as double width (#242).
The argument for –depth or depth: must now be positive.
Journal entries are now written with all amounts explicit, to avoid losing price info (#283).
Fixed a bug which sometimes (when the same letter pair was repeated) caused it not to pick the most similar past transaction for defaults.
There is now a -V/–value flag to report current market value (as in Ledger). It converts all reported amounts using their “default market price”. “Market price” is the new name for “historical prices”, defined with the P directive. The default market price for a commodity is the most recent one found in the journal on or before the report end date.
Unlike Ledger, hledger’s -V uses only the market prices recorded with P directives; it does not use the “transaction prices” recorded as part of posting amounts (which are used by -B/–cost). Also, using both -B and -V at the same time is supported.
Fixed a bug in amount normalization which caused amount styles (commodity symbol placement, decimal point character, etc.) to be lost in certain cases (#230, #276).
The balance command’s –format option can now adjust the rendering style of multi-commodity amounts, if you begin the format string with one of:
%_ - renders amounts on multiple lines, bottom-aligned (the default) %^ - renders amounts on multiple lines, top-aligned %, - renders amounts on one line, comma-separated
The balance report’s final total (and the line above it) now adapt themselves to a custom –format.
The –match option prints the journal entry that best matches a description (ie whose description field is most similar to the value given, and if there are several equally similar, the most recent). 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 from a guessed or partial description.
print now always right-aligns the amounts in an entry, even when they are wider than 12 characters. (If there is a price, it’s considered part of the amount for right-alignment.)
- Amount columns now resize automatically, using more space if it’s needed and available.
Account aliases are once again non-regular-expression-based, by default. (#252)
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 still available; they must be enclosed in forward slashes. (Ledger effectively ignores these.)
We now parse, and also print, journal entries with no postings, as proposed on the mail lists. These are not well-formed General Journal entries/transactions, but here is my rationale:
- Ledger and beancount parse them
- if they are parsed, they should be printed
- they provide a convenient way to record (and report) non-transaction events
- they permit more gradual introduction and learning of the concepts. So eg a beginner can keep a simple journal before learning about accounts and postings.
Trailing whitespace after a
commentdirective is now ignored.
- The -f/file option may now be used multiple times. This is equivalent to concatenating the input files before running hledger. The add command adds entries to the first file specified.
real: (no argument) is now a synonym for real:1
tag: now matches tag names with a regular expression, like most other queries
empty: is no longer supported, as it overlaps a bit confusingly with amt:0. The –empty flag is still available.
You can now match on pending status (#250)
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 relevant search query terms 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.
- activity no longer excludes 0-amount postings by default.
- Don’t show quotes around the journal file path in the “Creating…” message, for consistency with the subsequent “Adding…” message.
- Accounts beginning with “debt” or now also recognised as liabilities.
We now limit the display precision of inferred prices. (#262)
When a transaction posts to two commodities without specifying the conversion price, we generate a price which makes it balance (cf http://hledger.org/manual.html#prices). The print command showed this with full precision (so that manual calculations with the displayed numbers would look right), but this sometimes meant we showed 255 digits (when there are multiple postings in the commodity being priced, and the averaged unit price is an irrational number). In this case we now set the price’s display precision to the sum of the (max) display precisions of the commodities involved. An example:
hledgerdev -f- print <<< 1/1 c C 10.00 c C 11.00 d D -320.00 >>> 2015/01/01 c C 10.00 @ D 15.2381 c C 11.00 @ D 15.2381 d D -320.00 >>>=0
There might still be cases where this will show more price decimal places than necessary.
We now show inferred unit prices with at least 2 decimal places.
When inferring prices, if the commodities involved have low display precisions, we don’t do a good job of rendering accurate-looking unit prices. Eg if the journal doesn’t use any decimal places, any inferred unit prices are also displayed with no decimal places, which makes them look wrong to the user. Now, we always give inferred unit prices a minimum display precision of 2, which helps a bit.
- Postings with no amounts could give a runtime error in some obscure case, now fixed.
stats now supports -o/–outputfile, like register/balance/print.
An O(n^2) performance slowdown has been fixed, it’s now much faster on large journals.
+--------------------------------------++--------+--------+ | || 0.25 | 0.26 | +======================================++========+========+ | -f data/100x100x10.journal stats || 0.10 | 0.16 | | -f data/1000x1000x10.journal stats || 0.45 | 0.21 | | -f data/10000x1000x10.journal stats || 58.92 | 2.16 | +--------------------------------------++--------+--------+
The June 30 day span was not being rendered correctly; fixed. (#272)
The bench script invoked by “cabal bench” or “stack bench” now runs some simple benchmarks.
You can get more accurate benchmark times by running with –criterion. This will usually give much the same numbers and takes much longer.
Or with –simplebench, it benchmarks whatever commands are configured in bench/default.bench. This mode uses the first “hledger” executable in $PATH.
The deprecated shakespeare-text dependency has been removed more thoroughly.
- timelog: support the description field (#247)
GHC 7.10 compatibility (#239)
build with terminfo support on POSIX systems by default
On non-windows systems, we now build with terminfo support by default, useful for detecting terminal width and other things.
This requires the C curses dev libaries, which makes POSIX installation slightly harder; if it causes problems you can disable terminfo support with the new
cursescabal flag, eg: cabal install -f-curses … (or cabal might try this automatically, I’m not sure).
register: use the full terminal width, respect COLUMNS, allow column width adjustment
On POSIX systems, register now uses the full terminal width by default. Specifically, the output width is set from:
- a –width option
- or a COLUMNS environment variable (NB: not the same as a bash shell var)
- or on POSIX (non-windows) systems, the current terminal width
- or the default, 80 characters.
Also, register’s –width option now accepts an optional description column width following the overall width (–width WIDTH[,DESCWIDTH]). This also sets the account column width, since the available space (WIDTH-41) is divided up between these two columns. Here’s a diagram:
<--------------------------------- width (W) ----------------------------------> date (10) description (D) account (W-41-D) amount (12) balance (12) DDDDDDDDDD dddddddddddddddddddd aaaaaaaaaaaaaaaaaaa AAAAAAAAAAAA AAAAAAAAAAAA
$ 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
balance: new -T/–row-total and -A/–average options
In multicolumn balance reports, -T/–row-total now shows a row totals column and -A/–average shows a row averages column. This helps eg to see monthly average expenses (hledger bal ^expenses -MA).
NB our use of -T deviates from Ledger’s UI, where -T sets a custom final total expression.
balance: -N is now short for –no-total
balance: fix partially-visible totals row with –no-total
A periodic (not using –cumulative or –historical) balance report with –no-total now hides the totals row properly.
journal, csv: comment lines can also start with *
As in Ledger. This means you can embed emacs org/outline-mode nodes in your journal file and manipulate it like an outline.
journal: fix balance accumulation across assertions (#195)
A sequence of balance assertions asserting first one commodity, then another, then the first again, was not working.
timelog: show hours with two decimal places instead of one (#237)
in weekly reports, simplify week 52’s heading like the others
disallow trailing garbage in a number of parsers
Trailing garbage is no longer ignored when parsing the following: balance –format option, register –width option, hledger-rewrite options, hledger add’s inputs, CSV amounts, posting amounts, posting dates in tags.
allow utf8-string-1 (fpco/stackage/#426)
- fix redundant compilation when cabal installing the hledger packages
- switch to Decimal for representing amounts (#118)
- report interval headings (eg in balance, register reports) are shown compactly when possible
- general speedups
- detect decimal point and digit groups more robustly (#196)
- check that transaction dates are followed by whitespace or newline
- check that dates use a consistent separator character
- balance assertions now are specific to a single commodity, like Ledger (#195)
- support multi-line comments using “comment”, “end comment” directives, like Ledger
- reading CSV data from stdin now works better
- the rules file include directive is now relative to the current file’s directory (#198)
- the original order of same-day transactions is now usually preserved (if the records appear to be in reverse date order, we reverse them before finally sorting by transaction date)
- CSV output is now built in to the balance, print, and register commands, controlled by -O/–output-format (and -o/–output-file, see below)
- the –width and –debug options now require their argument (#149)
- when an option is repeated, the last value takes precedence (#219). This is helpful eg for customising your reporting command aliases on the fly.
- smart dates (used in -p/-b/-e/date:/date2:) now must use a consistent separator character, and must be parseable to the end
- output destination and format selection is now built in to the
balance, print and register commands, controlled by -o/–output-file
and -O/–output-format options. Notes:
- -o - means stdout
- an output file name suffix matching a supported format will also set the output format, unless overridden by –output-format
- commands’ supported output formats are listed in their command-line help. Two formats are currently available: txt (the default) and csv.
- balance assertions can be disabled with –ignore-assertions
- all matching account aliases are now applied, not just one directive and one option
- account aliases now match by case insensitive regular expressions matching anywhere in the account name
- account aliases can replace multiple occurrences of the pattern within an account name
- an account alias replacement pattern can reference matched groups with \N
- date:/date2: with a malformed date now reports an error instead of being ignored
- amt: now supports >= or <=
- clarify status: docs and behaviour; "*" is no longer a synonym for “1” (fixes #227)
- fix: in tree mode, –drop is ignored instead of showing empty account names
- a depth limit of 0 now shows summary items with account name “…”, instead of an empty report (#206)
- in multicolumn balance reports, -E now also shows posting-less accounts with a non-zero balance during the period (in addition to showing leading & trailing empty columns)
- in multicolumn reports, multi-commodity amounts are rendered on one line for better layout (#186)
- multicolumn reports’ title now includes the report span
- runs faster with large output
- supports date2:, and date:/date2: combined with –date2, better (fixes #201, #221, #222)
- a depth limit of 0 now shows summary items (see balance)
- -A/–average now implies -E/–empty
- postings with multi-commodity amounts are now top-aligned, like Ledger
- hledger-equity: fix end date in title; print closing entry too
- hledger-check-dates: added
- allow text 1.2+ (#207)
- register: also fix date sorting of postings (#184)
- register: fix a refactoring-related regression that the tests missed: if transactions were not ordered by date in the journal, register could include postings before the report start date in the output. (#184)
- add: don’t apply a default commodity to amounts on entry (#138)
- cli: options before the add-on command name are now also passed to it (#182)
- csv: allow the first name in a fields list to be empty (#178)
- csv: don’t validate fields count in skipped lines (#177)
- A # (hash) in column 0 is now also supported for starting a top-level journal comment, like Ledger.
- The “too many missing amounts” error now reminds about the 2-space rule.
- Fix: . (period) is no longer parsed as a valid amount.
- Fix: default commodity directives no longer limit the maximum display precision (#169).
- Fix: + before an amount is no longer parsed as part of the commodity (#181).
- Command-line help cleanups, layout improvements.
- Descriptions are shown for known add-ons in the command list.
- Command aliases have been simplified.
- Add-ons can now have any of these file extensions: none, hs, lhs, pl, py, rb, rkt, sh, bat, com, exe.
- Add-ons are displayed without their file extensions when possible.
- Add-ons with the same name as a built-in command or alias are ignored.
- Fix: add-on detection and invocation now works on windows.
- Fix: add-ons with digits in the name are now found.
- Fix: add-on arguments containing a single quote now work.
- Fix: when – is used to hide add-on options from the main program, it is no longer passed through as an add-on argument.
- The currency/commodity query prefix (sym:) has been renamed to cur:.
- Currency/commodity queries are applied more strongly in register and balance reports, filtering out unwanted currencies entirely. Eg hledger balance cur:’$’ now reports only the dollar amounts even if there are multi-currency transactions or postings.
- Amount queries like amt:N, amt:N, where N is not 0, now do an unsigned comparison of the amount and N. That is, they compare the absolute magnitude. To do a signed comparison instead, write N with its sign (eg amt:+N, amt:<+N, amt:>-N).
- Fix: amount queries no longer give false positives on multi-commodity amounts.
- An accounts command has been added, similar to Ledger’s, for listing account names in flat or hierarchical mode.
- Tab completion now works at all prompts, and will insert the default if the input area is empty.
- Account and amount defaults are more robust and useful.
- Transactions may also be completed by the enter key, when there are no more default postings.
- Input prompts are displayed in a different colour when supported.
- Balance reports in flat mode now always show exclusive (subaccount-excluding) balances.
- Balance reports in flat mode with –depth now aggregate deeper accounts at the depth limit instead of excluding them.
- Multicolumn reports in flat mode now support –drop.
- Multicolumn balance reports can now show the account hierarchy with –tree.
- Multicolumn report start/end dates are adjusted to encompass the displayed report periods, so the first and last periods are “full” and comparable to the others.
- Fix: zero-balance leaf accounts below a non-zero-balance parent are no longer always shown (#170).
- Fix: multicolumn reports now support –date2 (cf #174).
balancesheet, cashflow, incomestatement:
- These commands now support –flat and –drop.
- Tag queries (tag:) will now match a transaction if any of its postings match.
- The –display option has been dropped. To see an accurate running total which includes the prior starting balance, use –historical/-H (like balance).
- With a report interval, report start/end dates are adjusted to encompass the displayed periods, so the first and last periods are “full” and comparable to the others.
- Fix: –date2 now works with report intervals (fixes #174).
- Default report dates now derive from the secondary dates when –date2 is in effect.
- Default report dates now notice any posting dates outside the transaction dates’ span.
- Debug output improvements.
- New add-on example: extra/hledger-rewrite.hs, adds postings to matched entries.
- Compatible with GHC 7.2 (#155) - GHC 7.8, shakespeare 2
- display years before 1000 with four digits, not three
- avoid pretty-show to build with GHC < 7.4
- allow text 1.1, drop data-pprint to build with GHC 7.8.x
0.22.1 (2014/1/6) and older: see http://hledger.org/release-notes or doc/release-notes.md.