Regex Replacer Extension

deploy.ps1

@@ -52,6 +52,7 @@ foreach ($language in @{
 			"Google Translate",
 			"Lua",
 			"Regex Filter",
+			"Regex Replacer",
 			"Remove Repeated Characters",
 			"Remove Repeated Phrases",
 			"Remove Repeated Phrases 2",

extensions/CMakeLists.txt

@@ -21,6 +21,7 @@ add_library(Remove\ 30\ Repeated\ Sentences MODULE removerepeatsentence.cpp exte
 add_library(Replacer MODULE replacer.cpp extensionimpl.cpp)
 add_library(Styler MODULE styler.cpp extensionimpl.cpp)
 add_library(Thread\ Linker MODULE threadlinker.cpp extensionimpl.cpp)
+add_library(Regex\ Replacer MODULE regexreplacer.cpp extensionimpl.cpp)
 
 target_precompile_headers(Bing\ Translate REUSE_FROM pch)
 target_precompile_headers(Copy\ to\ Clipboard REUSE_FROM pch)
@@ -40,6 +41,7 @@ target_precompile_headers(Remove\ 30\ Repeated\ Sentences REUSE_FROM pch)
 target_precompile_headers(Replacer REUSE_FROM pch)
 target_precompile_headers(Styler REUSE_FROM pch)
 target_precompile_headers(Thread\ Linker REUSE_FROM pch)
+target_precompile_headers(Regex\ Replacer REUSE_FROM pch)
 
 target_link_libraries(Bing\ Translate winhttp Qt5::Widgets)
 target_link_libraries(DeepL\ Translate winhttp Qt5::Widgets)

extensions/regexreplacer.cpp

@@ -0,0 +1,53 @@
+#include "extension.h"
+#include "module.h"
+#include "blockmarkup.h"
+#include <fstream>
+
+extern const wchar_t* REGEX_REPLACER_INSTRUCTIONS;
+
+const char* REGEX_REPLACEMENTS_SAVE_FILE = "SavedRegexReplacements.txt";
+
+std::optional<std::wregex> regex;
+concurrency::reader_writer_lock m;
+
+BOOL WINAPI DllMain(HMODULE hModule, DWORD ul_reason_for_call, LPVOID lpReserved)
+{
+	switch (ul_reason_for_call)
+	{
+	case DLL_PROCESS_ATTACH:
+	{
+		if (!std::ifstream(REGEX_REPLACEMENTS_SAVE_FILE).good())
+			{
+			auto file = std::ofstream(REGEX_REPLACEMENTS_SAVE_FILE, std::ios::binary) << "\xff\xfe";
+			for (auto ch : std::wstring_view(REGEX_REPLACER_INSTRUCTIONS))
+				file << (ch == L'\n' ? std::string_view("\r\0\n", 4) : std::string_view((char*)&ch, 2));
+			_spawnlp(_P_DETACH, "notepad", "notepad", REGEX_REPLACEMENTS_SAVE_FILE, NULL); // show file to user
+		}
+	}
+	break;
+	case DLL_PROCESS_DETACH:
+	{
+	}
+	break;
+	}
+	return TRUE;
+}
+
+bool ProcessSentence(std::wstring& sentence, SentenceInfo sentenceInfo)
+{
+	//if (!sentenceInfo["current select"] || sentenceInfo["text number"] == 0 || sentence == L"") return false;
+	if (!sentenceInfo["current select"] || sentenceInfo["text number"] == 0) return false;
+
+	std::ifstream stream(REGEX_REPLACEMENTS_SAVE_FILE, std::ios::binary);
+	BlockMarkupIterator savedFilters(stream, Array<std::wstring_view>{ L"|REGEX|", L"|BECOMES|" });
+	//std::vector<std::wstring> regexes;
+	concurrency::reader_writer_lock::scoped_lock_read readLock(m);
+	while (auto read = savedFilters.Next()) {
+		const auto& [regex, replacement] = read.value();
+		if (regex == L"") continue;
+		try { ::regex = regex; }
+		catch (std::regex_error) { continue; }
+		sentence = std::regex_replace(sentence, ::regex.value(), replacement);
+	}
+	return true;
+}

text.cpp

@@ -224,6 +224,12 @@ All text in this file outside of a replacement command is ignored.
 A caret (^) acts as a wildcard that matches any other single character.
 Whitespace in original_text is ignored, but replacement_text can contain spaces, newlines, etc.
 This file must be encoded in Unicode (UTF-16 Little Endian).)";
+const wchar_t* REGEX_REPLACER_INSTRUCTIONS = LR"(This file only does anything when the "Regex Replacer" extension is used.
+Replacement commands must be formatted like this:
+|REGEX|regular_expression|BECOMES|replacement_text|END|
+All text in this file outside of a replacement command is ignored.
+This file must be encoded in Unicode (UTF-16 Little Endian).
+Learn, build, & test Regular Expressions: https://regexr.com/)";
 const char* THREAD_LINKER = u8"Thread Linker";
 const char* LINK = u8"Link";
 const char* UNLINK = u8"Unlink";
@@ -931,6 +937,12 @@ I comandi di rimpiazzo devono essere formattati cosi:
 Tutto il testo in questo file all'infuori di un comando di rimpiazzo è ignorato.
 La spaziatura nel testo_originale è ignorato, ma testo_sostituito può contenere spaziature, ritorni a capo, ecc.
 Questo file deve essere codificato in Unicode (UTF-16 Little Endian).)";
+	REGEX_REPLACER_INSTRUCTIONS = LR"(Questo file fa qualcosa solo quando l'estenzione "Regex Replacer" è utilizzata.
+I comandi di sostituzione devono essere formattati cosi:
+|ORIG|espressione_regolare|BECOMES|testo_sostituito|END|
+Tutto il testo in questo file all'infuori di un comando di sostituzione è ignorato.
+Questo file deve essere codificato in Unicode (UTF-16 Little Endian).
+Apprendere, creare e testare Espressioni Regolari: https://regexr.com/)";
 	THREAD_LINKER = u8"Collegatore di thread";
 	LINK = u8"Collegamento";
 	THREAD_LINK_FROM = u8"Numero di thread da cui collegarsi";