floraachy
/
vChewing-macOS
1
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity

UPE // Regex Optimization.

This commit is contained in:
ShikiSuen 2022-03-05 11:13:37 +08:00
parent 07fcb9ff49
commit a82deb75a1
1 changed files with 7 additions and 9 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

16
UserPhraseEditor/StringExtension.swift
View File

@ -31,15 +31,13 @@ extension String {
var strProcessed = self
// 預處理格式
strProcessed = strProcessed.replacingOccurrences(of: " #MACOS", with: "") // 去掉 macOS 標記
strProcessed = strProcessed.replacingOccurrences(of: "　", with: " ") // CJKWhiteSpace (\x{3000}) to ASCII Space
strProcessed = strProcessed.replacingOccurrences(of: " ", with: " ") // NonBreakWhiteSpace (\x{A0}) to ASCII Space
strProcessed = strProcessed.replacingOccurrences(of: "\t", with: " ") // Tab to ASCII Space
strProcessed.regReplace(pattern: "\\f", replaceWith: "\n") // Form Feed to LF
strProcessed = strProcessed.replacingOccurrences(of: "\r", with: "\n") // CR to LF
strProcessed.regReplace(pattern: " +", replaceWith: " ") // 統整連續空格為一個 ASCII 空格
strProcessed.regReplace(pattern: "\\n+", replaceWith: "\n") // 統整連續 LF 為一個 LF
strProcessed = strProcessed.replacingOccurrences(of: " \n", with: "\n") // 去除行尾空格
strProcessed = strProcessed.replacingOccurrences(of: "\n ", with: "\n") // 去除行首空格
// CJKWhiteSpace (\x{3000}) to ASCII Space
// NonBreakWhiteSpace (\x{A0}) to ASCII Space
// Tab to ASCII Space
// 統整連續空格為一個 ASCII 空格
strProcessed.regReplace(pattern: #"( +|　+| +|\t+)+"#, replaceWith: " ")
strProcessed.regReplace(pattern: #"(\f+|\r+)+"#, replaceWith: "\n") // CR & Form Feed to LF
strProcessed.regReplace(pattern: #"(\n+| \n+|\n+ )"#, replaceWith: "\n") // 去除行尾行首空格與重複行
if strProcessed.prefix(1) == " " { // 去除檔案開頭空格
strProcessed.removeFirst()
}