Apply an operation to both MixedAmount in an BalanceData.

Merge two BalanceData, using the given operation to combine their amounts.

A partition of time into one or more contiguous periods, plus a historical period that precedes them. Note DayPartition does not store per-period data - only the periods' start/end dates.

Construct a DayPartition from a non-empty list of period boundary dates (start dates plus a final exclusive end date).

>>> boundariesToDayPartition (fromGregorian 2025 01 01 :| [fromGregorian 2025 02 01])
DayPartition {dayPartitionToPeriodData = PeriodData{ pdpre = 2024-12-31, pdperiods = fromList [(2025-01-01,2025-01-31)]}}

Construct a DayPartition from a list of period boundary dates (start dates plus a final exclusive end date), if it's a non-empty list.

Convert DayPartition to a non-empty list of period start and end dates (both inclusive). Each end date will be one day before the next period's start date.

Convert DayPartition to a list (which will always be non-empty) of period start and end dates (both inclusive). Each end date will be one day before the next period's start date.

Convert DayPartition to a list of DateSpans. Each span will end one day before the next span begins (the span's exclusive end date will be equal to the next span's start date).

Return the union of two DayPartitions if that is a valid DayPartition, or Nothing otherwise.

Get this DayPartition's overall start date and end date (both inclusive).

Find the start and end dates of the period within a DayPartition which contains a given day. If the day is after the end of the last period, it is assumed to be within the last period. If the day is before the start of the first period (ie, in the historical period), only the historical period's end date is returned.

Split a DateSpan into a DayPartition consisting of consecutive exact spans of the specified Interval, or Nothing if the span is invalid. If no interval is specified, the original span is returned. If the original span is the null date span, ie unbounded, Nothing is returned. If the original span is empty, eg if the end date is <= the start date, Nothing is returned.

Date adjustment

Some intervals respect the "adjust" flag (years, quarters, months, weeks, every Nth weekday of month seem to be the ones that need it). This will move the start date earlier, if needed, to the previous natural interval boundary (first of year, first of quarter, first of month, monday, previous Nth weekday of month). Related: #1982 #2218

The end date is always moved later if needed to the next natural interval boundary, so that the last period is the same length as the others.

Examples

>>> let t i y1 m1 d1 y2 m2 d2 = fmap dayPartitionToNonEmpty . splitSpan True i $ DateSpan (Just $ Flex $ fromGregorian y1 m1 d1) (Just $ Flex $ fromGregorian y2 m2 d2)
>>> t NoInterval 2008 01 01 2009 01 01
Just ((2008-01-01,2008-12-31) :| [])
>>> t (Quarters 1) 2008 01 01 2009 01 01
Just ((2008-01-01,2008-03-31) :| [(2008-04-01,2008-06-30),(2008-07-01,2008-09-30),(2008-10-01,2008-12-31)])
>>> splitSpan True (Quarters 1) nulldatespan
Nothing
>>> t (Days 1) 2008 01 01 2008 01 01  -- an empty datespan
Nothing
>>> t (Quarters 1) 2008 01 01 2008 01 01
Nothing
>>> t (Months 1) 2008 01 01 2008 04 01
Just ((2008-01-01,2008-01-31) :| [(2008-02-01,2008-02-29),(2008-03-01,2008-03-31)])
>>> t (Months 2) 2008 01 01 2008 04 01
Just ((2008-01-01,2008-02-29) :| [(2008-03-01,2008-04-30)])
>>> t (Weeks 1) 2008 01 01 2008 01 15
Just ((2007-12-31,2008-01-06) :| [(2008-01-07,2008-01-13),(2008-01-14,2008-01-20)])
>>> t (Weeks 2) 2008 01 01 2008 01 15
Just ((2007-12-31,2008-01-13) :| [(2008-01-14,2008-01-27)])
>>> t (MonthDay 2) 2008 01 01 2008 04 01
Just ((2008-01-02,2008-02-01) :| [(2008-02-02,2008-03-01),(2008-03-02,2008-04-01)])
>>> t (NthWeekdayOfMonth 2 4) 2011 01 01 2011 02 15
Just ((2010-12-09,2011-01-12) :| [(2011-01-13,2011-02-09),(2011-02-10,2011-03-09)])
>>> t (DaysOfWeek [2]) 2011 01 01 2011 01 15
Just ((2010-12-28,2011-01-03) :| [(2011-01-04,2011-01-10),(2011-01-11,2011-01-17)])
>>> t (MonthAndDay 11 29) 2012 10 01 2013 10 15
Just ((2012-11-29,2013-11-28) :| [])

Get the natural start for the given interval that falls on or before the given day, when applicable. Works for Weeks, Months, Quarters, Years, eg.

Construct a PeriodData from a historical data value and a list of (period start, period data) pairs.

Convert PeriodData to a historical data value and a list of (period start, period data) pairs.

Get the data for the period containing the given SmartInterval, and that period's start date. If the day is after the end of the last period, it is assumed to be within the last period. If the day is before the start of the first period (ie, in the historical period), return Nothing.

Get the data for the period containing the given SmartInterval, and that period's start date. If the day is after the end of the last period, it is assumed to be within the last period. If the day is before the start of the first period (ie, in the historical period), return the data for the historical period and no start date.

Set historical or period data in the appropriate location in a PeriodData.

Merge two PeriodData, using the given operation to combine their data values.

This will drop keys if they are not present in both PeriodData.

Merge two PeriodData, using the given operations for combining data that's only in the first, only in the second, or in both, respectively.

Pad out the date map of a PeriodData so that every key from another PeriodData is present.

Year of Common Era (when positive).

The status of a transaction or posting, recorded with a status mark (nothing, !, or *). What these mean is ultimately user defined.

An account within a hierarchy, with references to its parent and subaccounts if any, and with per-report-period data of type a. Only the name is required; the other fields may or may not be present.

Data that's useful in "balance" reports: subaccount-exclusive and -inclusive amounts, typically representing either a balance change or an end balance; and a count of postings.

A journal, containing general ledger transactions; also directives and various other things. This is hledger's main data model.

During parsing, it is used as the type alias ParsedJournal. The jparse* fields are mainly used during parsing and included here for convenience. The list fields described as "in parse order" are usually reversed for efficiency during parsing. After parsing, "journalFinalise" converts ParsedJournal to a finalised Journal, which has all lists correctly ordered, and much data inference and validation applied.

A Ledger has the journal it derives from, and the accounts derived from that. Accounts are accessible both list-wise and tree-wise, since each one knows its parent and subs; the first account is the root of the tree and always exists.

A general container for storing data values associated with zero or more contiguous report (sub)periods, and with the (open ended) pre-report period. The report periods are typically all the same length, but need not be.

Report periods are represented only by their start dates, used as the keys of a Map.

A periodic transaction rule, describing a transaction that recurs.

A transaction modifier rule. This has a query which matches postings in the journal, and a list of transformations to apply to those postings or their transactions. Currently there is one kind of transformation: the TMPostingRule, which adds a posting ("auto posting") to the transaction, optionally setting its amount to the matched posting's amount multiplied by a constant.

The "display precision" for a hledger amount, by which we mean the number of decimal digits to display to the right of the decimal mark.