From 764fa6ebba367c344dae178816fc9566c203a17c Mon Sep 17 00:00:00 2001
From: Jaro <jaro@argonaut-constellation.org>
Date: Sun, 11 Jun 2023 05:38:10 +0200
Subject: [PATCH] Internally support text breaks in ascending order.

Intended as a workaround to `breaksRight` not including the status of
the end-of-text break.
---
 .../Text/ParagraphLayout/Internal/Break.hs    | 22 +++++++++++++++++++
 1 file changed, 22 insertions(+)

diff --git a/src/Data/Text/ParagraphLayout/Internal/Break.hs b/src/Data/Text/ParagraphLayout/Internal/Break.hs
index c03285d..49d9934 100644
--- a/src/Data/Text/ParagraphLayout/Internal/Break.hs
+++ b/src/Data/Text/ParagraphLayout/Internal/Break.hs
@@ -8,6 +8,7 @@
 module Data.Text.ParagraphLayout.Internal.Break
     ( LineBreak (..)
     , locale
+    , breaksAsc
     , breaksDesc
     , subOffsetsDesc
     )
@@ -19,7 +20,9 @@ import Data.Text.ICU
     ( Break
     , Breaker
     , LocaleName (Locale)
+    , breaks
     , breaksRight
+    , brkBreak
     , brkPrefix
     , brkStatus
     )
@@ -49,6 +52,16 @@ locale lang lb = Locale $ (clean lang) ++ (lbKeyword lb)
         -- This filter is here just to stop syntactically incorrect input.
         clean = takeWhile (`elem` ['A' .. 'Z'] ++ ['a' .. 'z'] ++ "_-")
 
+-- | List of all breaks in the given text, with offsets in ascending order,
+-- including the status of the break if applicable.
+--
+-- Excludes the start of the text (with offset 0).
+--
+-- Includes the end of the text (with offset equal to the text length)
+-- as the last list item.
+breaksAsc :: Breaker a -> Text -> [(Int, a)]
+breaksAsc breaker input = map brkEndOffsetStatus $ breaks breaker input
+
 -- | List of all breaks in the given text, with offsets in descending order,
 -- including the status of the break if applicable.
 --
@@ -61,12 +74,21 @@ breaksDesc breaker input = map brkStartOffsetStatus $ breaksRight breaker input
 brkStartOffsetStatus :: Break a -> (Int, a)
 brkStartOffsetStatus brk = (brkStartOffset brk, brkStatus brk)
 
+brkEndOffsetStatus :: Break a -> (Int, a)
+brkEndOffsetStatus brk = (brkEndOffset brk, brkStatus brk)
+
 -- | The ICU library returns "breaks" as slices of text with two boundaries.
 -- This gives the smaller of the two distances from the start of the text
 -- to the boundaries of the break.
 brkStartOffset :: Break a -> Int
 brkStartOffset brk = lengthWord8 (brkPrefix brk)
 
+-- | The ICU library returns "breaks" as slices of text with two boundaries.
+-- This gives the larger of the two distances from the start of the text
+-- to the boundaries of the break.
+brkEndOffset :: Break a -> Int
+brkEndOffset brk = lengthWord8 (brkPrefix brk) + lengthWord8 (brkBreak brk)
+
 -- | Given a list of offsets into a text in descending order, produce a list of
 -- corresponding offsets into a slice of the text starting at a given offset.
 subOffsetsDesc :: Int -> [(Int, a)] -> [(Int, a)]
-- 
2.30.2