[*] Move strings to auROXTL/Strings/* (excl iterator)

Include/auROXTL/Strings/auCodepointsUTF8.hpp

@@ -0,0 +1,149 @@
+/***
+    Copyright (C) 2024 Jamie Reece Wilson (a/k/a "Reece"). All rights reserved.
+
+    File: auCodepointsUTF8.hpp
+    Date: 2024-09-09
+    File: auStringUtils.hpp
+    Date: 2022-2-1
+    File: AuroraUtils.hpp
+    File: auROXTLUtils.hpp
+    Date: 2021-6-9
+    Author: Reece
+    Purpose: Introduces UTF-8 utilities and the historically missing c++ string methods.
+             AuCodepoints**** supports the initial UTF-8 specifications with codepoints up to 2^31 combinations (illegal UTF-16)
+
+    Implements: AuStringContains, AuEndsWith, AuStartsWith, AuReplaceAll, AuSplitString (views), AuSplitStringLegacy (returns an array of strings instead of views)
+                AuToLower(char), AuToUpper(char), AuToLower(view), AuToUpper(view).
+    Implements: AuCodepointsTransform, AuCodepointsTransformASCIIOp, AuCodepointsForEach, AuCodepointsToLower, AuCodepointsToUpper,
+                AuCodepointsCount, AuCodepointsNextLength, AuCodepointsDecodeOne, AuCodepointsDecode, AuCodepointsEncodeInto,
+                AuCodepointsGetByteOffset(CodepointOffset_t),AuCodepointsGetByteLength(CodepointOffset_t),
+                AuCodepointsFindByteOffset[Unsafe], AuCodepointsFindCodepointOffset(view, CodepointOffset_t), AuCodepointsFindCodepointOffset(CodepointByteOffset_t),
+                AuCodepointsContains,
+                AuCodepointsReplaceAll, AuCodepointsSplitString (views),
+                AuCodepointsFindPreviousValidByteOffsetFromOffset, AuCodepointsFindPreviousValidByteOffsetFromByteOffset
+                AuCodepointsIsEqualIgnoreCase, AuCodepointsStartsWithEqualIgnoreCase, AuCodepointsEndsWithEqualIgnoreCase,
+                AuCodepointsReverseIterate, AuCodepointsReverseIterateSubStrPrefixView, AuCodepointsReverseIterateSubStrSuffixView
+
+    Warning:   By codepoints, we mean UTF32.
+               Aurora uses UTF8 strings everywhere by convention.
+
+    Warning:   For translating between locales (including utf8-32), defer to AuLocale (Aurora::Locale) in the Aurora Runtime.
+               Although, you can decode and encode UTF8 to 32 here (AuCodepointsDecodeOne, AuCodepointsDecode, AuCodepointsEncodeInto).
+***/
+#pragma once
+
+/// Try decode UTF32 codepoint from UTF8 sequence
+static constexpr AuOptional<AuUInt32>  AuCodepointsDecodeOne(const AuROString &in);
+
+/// Try decode all UTF32 codepoints from UTF8 sequence
+static AuList<AuUInt32>                AuCodepointsDecode(const AuROString &in);
+
+/// Try encode one UTF8 codepoint into a string buffer
+static void                            AuCodepointsEncodeInto(AuUInt32 uCodepoint, 
+                                                              AuString &out);
+
+/// Similar to AuCodepointsIsEqualIgnoreCase, translates all ASCII English characters to their AuToLower counterparts.
+/// No localization complexities are, will, or should be involved.
+static auline AuString                 AuCodepointsToLower(const AuROString &in);
+
+/// Similar to AuCodepointsIsEqualIgnoreCase, translates all ASCII English characters to their AuToUpper counterparts.
+/// No localization complexities are, will, or should be involved.
+static auline AuString                 AuCodepointsToUpper(const AuROString &in);
+
+/// Counts the UTF8/UTF32 codepoints in a byte sequence
+static auline constexpr CodepointOffset_t     AuCodepointsCount(const AuROString &in);
+
+/// Counts the bytes required to iterate over a UTF8 encoded codepoint
+static auline constexpr CodepointByteOffset_t AuCodepointsNextLength(const AuROString &in);
+
+/// Iterates over a UTF8 sequence using OP.
+/// If op returns void, AuCodepointsForEach will continue until EOS or invalid squence.
+/// If op returns bool, AuCodepointsForEach will continue until EOS, invalid squence, or !op.
+/// Returns false on invalid sequence or user break, otherwise returns true.
+template <class T>
+static bool                            AuCodepointsForEach(T op, const AuROString &in);
+
+/// Translates the in UTF8 sequence using a transformer of U32(*fMyUTF32Translator)(U32)
+template <class T>
+static AuString                        AuCodepointsTransform(T op, const AuROString &in);
+
+/// Translates the in UTF8 sequence using a transformer of U8(*fMyASCIITranslator)(U8)
+template <class T>
+AuString                               AuCodepointsTransformASCIIOp(T op, const AuROString &in);
+
+/// Performs a memcmp on unknown chunks, performs a memcmp for each UTF8 sequence of bytes, and performs a AuToLower(a) != AuToLower(b) operation on any ASCII English characters.
+/// Also see: AuCodepointsStartsWithEqualIgnoreCase, AuCodepointsEndsWithEqualIgnoreCase
+static bool                            AuCodepointsIsEqualIgnoreCase(const AuROString &inA,
+                                                                     const AuROString &inB);
+
+/// Performs a memcmp on unknown chunks, performs a memcmp for each UTF8 sequence of bytes, and performs a AuToLower(a) != AuToLower(b) operation on any ASCII English characters.
+/// Also see: AuCodepointsIsEqualIgnoreCase, AuCodepointsEndsWithEqualIgnoreCase
+static bool                            AuCodepointsStartsWithEqualIgnoreCase(const AuROString &inA,
+                                                                             const AuROString &inB);
+
+/// Also see: AuCodepointsIsEqualIgnoreCase, AuCodepointsStartsWithEqualIgnoreCase
+static bool                            AuCodepointsEndsWithEqualIgnoreCase(const AuROString &inA,
+                                                                           const AuROString &inB);
+
+/// Returns the byte offset of the codepoint index or AuROString::npos
+static auline constexpr CodepointByteOffset_t AuCodepointsGetByteOffset(const AuROString &in,
+                                                                        CodepointOffset_t uCodepointIndex);
+
+/// Returns the length of the codepoint index in bytes or AuROString::npos
+static auline constexpr CodepointByteOffset_t AuCodepointsGetByteLength(const AuROString &in,
+                                                                        CodepointOffset_t uCodepointIndex);
+
+/// Finds subpattern in value at codepoint offset, returning bool
+static bool                            AuCodepointsContains(const AuROString &value,
+                                                            const AuROString &subpattern,
+                                                            CodepointOffset_t uStartPosition = {});
+
+/// Performs a terribly inefficient find in sequence operation.
+/// Returns the byte offset for find in in starting at uStartPosition bytes, or AuROString::npos
+/// Assuming the in sequence is valid, you can use other traditional binary methods.
+/// UTF8 sets the higher most bits to signify multibyte sequence position; you can arbitrarily scan for a complete sequences of UTF8 characters in a legal buffer without any special logic.
+static CodepointByteOffset_t           AuCodepointsFindByteOffset(const AuROString &in,
+                                                                  const AuROString &find,
+                                                                  CodepointByteOffset_t uStartPosition = {});
+
+/// AuCodepointsFindByteOffset same as above (AuCodepointsFindByteOffset), slightly less dumb, but still too bloated.
+/// The main difference is that uStartPosition is trusted to be a valid start offset.
+/// Worst thing about this is, assuming a valid string view, we don't need to worry about testing the validity of the previous bytes (<uStartPosition), or even the current codepoint byte offset (uStartPosition).
+static constexpr CodepointByteOffset_t AuCodepointsFindByteOffsetUnsafe(const AuROString &in,
+                                                                        const AuROString &find,
+                                                                        CodepointByteOffset_t uStartPosition);
+
+/// An inefficient seek backwards to byte offset given an arbitrary codepoint offset.
+/// Returns byte offset for codepoint offset - 1 or AuROString::npos
+static constexpr CodepointByteOffset_t AuCodepointsFindPreviousValidByteOffsetFromOffset(const AuROString &in,
+                                                                                         CodepointOffset_t uStartPosition = {});
+
+/// An efficient seek backwards to byte offset given an arbitrary codepoint byte offset.
+/// Returns byte offset for codepoint byte offset - 1 or AuROString::npos
+static constexpr CodepointByteOffset_t AuCodepointsFindPreviousValidByteOffsetFromByteOffset(const AuROString &in,
+                                                                                             CodepointByteOffset_t uStartPosition = {});
+
+static constexpr CodepointOffset_t     AuCodepointsFindCodepointOffset(const AuROString &in,
+                                                                       CodepointByteOffset_t uBytePosition);
+
+/// Finds subpattern in value at codepoint offset, returning a codepoint offset or AuROString::npos
+static CodepointOffset_t               AuCodepointsFindCodepointOffset(const AuROString &in,
+                                                                       const AuROString &find,
+                                                                       CodepointOffset_t uStartPosition = {});
+
+/// For a given valid UTF8 string view, provides the delta byte offset or AuROString::npos, to delete the last most codepoint
+static constexpr AuUInt                AuCodepointsReverseIterate(const AuROString &string);
+
+/// For a given valid UTF8 string view, to delete the last most codepoint, this provides a view with the suffix removed or an empty view. 
+static constexpr AuROString            AuCodepointsReverseIterateSubStrPrefixView(const AuROString &string);
+
+/// For a given valid UTF8 string view, to delete the last most codepoint, this provides a view with the just suffix to be removed or an empty view.
+static constexpr AuROString            AuCodepointsReverseIterateSubStrSuffixView(const AuROString &string);
+
+static AuString &                      AuCodepointsReplaceAll(AuString &str,
+                                                              const AuROString &from,
+                                                              const AuROString &to);
+
+static AuList<AuROString>              AuCodepointsSplitString(const AuROString &str,
+                                                               const AuROString &delim,
+                                                               bool bIgnoreEmpty = true);

Include/auROXTL/Strings/auStringUtils.hpp

@@ -0,0 +1,138 @@
+/***
+    Copyright (C) 2022-2024 J Reece Wilson (a/k/a "Reece"). All rights reserved.
+
+    File: auStringUtils.hpp
+    Date: 2024-09-09
+    File: auStringUtils.hpp
+    Date: 2022-2-1
+    File: AuroraUtils.hpp
+    File: auROXTLUtils.hpp
+    Date: 2021-6-9
+    Author: Reece
+***/
+#pragma once
+
+/// Similar to AuCodepointsIsEqualIgnoreCase, translates all ASCII English characters to their AuToLower counterparts.
+/// No localization complexities are, will, or should be involved.
+static AuString AuCodepointsToLower(const AuROString &in);
+
+/// Similar to AuCodepointsIsEqualIgnoreCase, translates all ASCII English characters to their AuToUpper counterparts.
+/// No localization complexities are, will, or should be involved.
+static AuString AuCodepointsToUpper(const AuROString &in);
+
+static auline constexpr bool AuIsAlpha(char c)
+{
+    return (c) && (((unsigned char)c | 0x20) - 'a' < 26);
+}
+
+static auline constexpr char AuToLower(char c)
+{
+    return AuIsAlpha(c) ? c | 0x20 : c;
+}
+
+static auline constexpr char AuToUpper(char c)
+{
+    return AuIsAlpha(c) ? c & ~0x20 : c;
+}
+
+static auline constexpr bool AuStringContains(const AuROString &value, const AuROString &subpattern)
+{
+    return value.find(subpattern) != AuROString::npos;
+}
+
+static auline constexpr bool AuEndsWith(AuROString const &value, AuROString const &ending)
+{
+    if (ending.size() > value.size())
+    {
+        return false;
+    }
+    else
+    {
+        return std::equal(ending.rbegin(), ending.rend(), value.rbegin());
+    }
+}
+
+static auline constexpr bool AuStartsWith(AuROString const &value, AuROString const &starting)
+{
+#if defined(AU_STRING_IS_TINYUTF_EXPERIMENT)
+    return value.starts_with(starting);
+#else
+    return value.rfind(starting, 0) == 0;
+#endif
+}
+
+/// Alias for AuCodepointsToLower
+static AuString AuToLower(const AuROString &in)
+{
+    return AuCodepointsToLower(in);
+}
+
+/// Alias for AuCodepointsToUpper
+static AuString AuToUpper(const AuROString &in)
+{
+    return AuCodepointsToUpper(in);
+}
+
+static AuString &AuReplaceAll(AuString &str,
+                              const AuROString &from,
+                              const AuROString &to)
+{
+    AuUInt uStartPosition {};
+    while ((uStartPosition = str.find(from, uStartPosition)) != AuROString::npos)
+    {
+        str.replace(uStartPosition, from.length(), to);
+        uStartPosition += to.length();
+    }
+    return str;
+}
+
+// i told myself not to copy this, required a split function twice, now here we are :D
+static AuList<AuROString> AuSplitString(const AuROString &str,
+                                        const AuROString &delim,
+                                        bool bIgnoreEmpty = true)
+{
+    AuList<AuROString> tokens;
+    AuUInt uPrev {}, uPos {};
+    tokens.reserve(str.size() / 16);
+    do
+    {
+        uPos = str.find(delim, uPrev);
+        if (uPos == AuROString::npos)
+        {
+            uPos = str.length();
+        }
+        auto token = str.substr(uPrev, uPos - uPrev);
+        if ((!token.empty()) && bIgnoreEmpty)
+        {
+            tokens.push_back(token);
+        }
+        uPrev = uPos + delim.length();
+    }
+    while (uPos < str.length() && uPrev < str.length());
+    return tokens;
+}
+
+static AuList<AuString> AuSplitStringLegacy(const AuROString &str,
+                                            const AuROString &delim,
+                                            bool bIgnoreEmpty = true)
+{
+    AuList<AuString> tokens;
+    AuUInt prev = 0, pos = 0;
+    tokens.reserve(str.size() / 16);
+    do
+    {
+        pos = str.find(delim, prev);
+        if (pos == AuROString::npos)
+        {
+            pos = str.length();
+        }
+        auto token = str.substr(prev, pos - prev);
+        if ((!token.empty()) && bIgnoreEmpty)
+        {
+            tokens.push_back(AuString(token));
+        }
+        prev = pos + delim.length();
+    }
+    while (pos < str.length() && prev < str.length());
+    return tokens;
+}

Include/auROXTL/auUTF8StringView.hpp

@@ -23,8 +23,8 @@ static auline constexpr bool AuStartsWith(AuROString const &value, AuROString co
 
 #include "auMemoryUtils.hpp"
 
-#include "Iterators/auROString.hpp"
-#include "Iterators/auRONString.hpp"
+#include "Strings/auROString.hpp"
+#include "Strings/auRONString.hpp"
 
 inline constexpr bool operator==(const AuROString &lhs,
                                  const AuROString &rhs) noexcept