renun/koreader
1
0
Fork 0
mirror of https://github.com/koreader/koreader.git synced 2025-08-10 00:52:38 +00:00
Code Issues Actions Packages Projects Releases Wiki Activity

bug fix: use individual gsubs

Browse Source
This commit is contained in:
Commodore64user
2025-07-31 14:17:54 +01:00
parent 400c774bcc
commit 6501c52fa2
1 changed files with 6 additions and 3 deletions
Download Patch File Download Diff File Expand all files Collapse all files

9
frontend/apps/reader/modules/readerdictionary.lua
View File

@@ -852,9 +852,12 @@ function ReaderDictionary:cleanSelection(text, is_sane)
text = text:gsub("\u{2019}", "'") -- Right single quotation mark
-- Strip punctuation characters around selection
text = util.stripPunctuation(text)
-- In some dictionaries, both interpuncts (·) and pipes (|) are used to delimiter syllables.
-- Up arrows (↑), are used in some dictionaries to indicate related words.
text = text:gsub("[·|↑]", "")
-- Note: although it seems innocuous to use a character class [·|↑] to perform a single gsub,
-- doing so will cause byte corruption in some languages (e.g. Greek).
-- In some dictionaries, both interpuncts and pipes are used to delimiter syllables.
text = text:gsub("·", "") -- interpunct
text = text:gsub("|", "") -- pipe
text = text:gsub("↑", "") -- and up arrow, used in some dictionaries to indicate related words
-- Strip some common english grammatical construct
text = text:gsub("'s$", '') -- english possessive
-- Strip some common french grammatical constructs