From d99c1032da71df2c0f2a6e61573244217275496a Mon Sep 17 00:00:00 2001
From: Jamie Reece Wilson <me@reece.sx>
Date: Thu, 22 Aug 2024 19:13:28 +0100
Subject: [PATCH] [*] Regression: AuROXTL string view has a different
 interpretation of what remove prefix/suffix should be. Fix SplitNewlines [*]
 Harden SplitString & SplitStringDelm, do not splice UTF8 sequences

---
 Include/Aurora/Parse/LineParser.hpp | 64 ++++++++++++++++++-----------
 1 file changed, 39 insertions(+), 25 deletions(-)

diff --git a/Include/Aurora/Parse/LineParser.hpp b/Include/Aurora/Parse/LineParser.hpp
index 049a1800..36a959e7 100644
--- a/Include/Aurora/Parse/LineParser.hpp
+++ b/Include/Aurora/Parse/LineParser.hpp
@@ -9,64 +9,71 @@
 
 namespace Aurora::Parse
 {
-    static AuROString SplitNewlines(const AuROString &in, AuFunction<void(const AuROString &)> lineCallback, bool returnRemaining)
+    static AuROString SplitNewlines(const AuROString &in,
+                                    AuConsumer<const AuROString &> lineCallback,
+                                    bool bReturnRemaining = false)
     {
         AuMach index = 0, startIdx = 0;
 
-        while ((index = in.find("\n", startIdx)) != AuString::npos)
+        while ((index = in.Find("\n", startIdx)) != AuString::npos)
         {
-            auto line = in.substr(startIdx, index - startIdx);
+            auto line = in.Substr(startIdx, index - startIdx);
             startIdx = index + 1;
 
-            if (line[line.size() - 1] == '\r')
+            if (line[line.Size() - 1] == '\r')
             {
-                line.remove_suffix(1);
+                line = line.RemoveSuffix(1);
             }
 
             lineCallback(line);
 
-            if (startIdx >= in.size())
+            if (startIdx >= in.Size())
             {
                 break;
             }
         }
 
-        if (returnRemaining)
+        if (bReturnRemaining)
         {
-            return in.substr(startIdx);
+            return in.Substr(startIdx);
         }
         else
         {
-            lineCallback(in.substr(startIdx));
+            lineCallback(in.Substr(startIdx));
             return {};
         }
     }
 
-    static void SplitNewlines(const AuROString &in, AuFunction<void(const AuROString &)> lineCallback)
-    {
-        SplitNewlines(in, lineCallback, false);
-    }
-    
     static AuList<AuROString> SplitString(const AuROString &in, AuUInt16 characters)
     {
         AuList<AuROString> ret;
-        for (auto i = 0u; i < in.size(); i += characters)
+        for (AuUInt i = 0u; i < in.Size(); )
         {
-            auto start = i;
-            auto end   = AuMin(AuUInt32(in.size()), AuUInt32(i + characters));
-            auto len   = end - start;
-            ret.push_back(in.substr(start, len));
+            AuUInt start, end, len;
+
+            start = i;
+            end   = AuMin(AuUInt(in.size()), AuUInt(i + characters));
+            if (end - 1)
+            {
+                auto uLastCodepointIndex  = AuCodepointsFindPreviousValidByteOffsetFromByteOffset(in, end);
+                auto uLastByteOffset      = AuCodepointsNextLength(in.Substr(uLastCodepointIndex));
+                end = uLastCodepointIndex + uLastByteOffset;
+            }
+            len   = end - start;
+            i    += len;
+
+            ret.push_back(in.Substr(start, len));
         }
         return ret;
     }
     
-    static AuString SplitStringDelm(const AuROString &in, const AuString &delm, AuUInt16 characters)
+    static AuString SplitStringDelm(const AuROString &in, const AuROString &delm, AuUInt16 characters)
     {
         AuString ret;
-        ret.reserve(in.size());
-        for (auto i = 0u; i < in.size(); i += characters)
+        ret.reserve(in.Size());
+        for (AuUInt i = 0u; i < in.Size(); )
         {
-            AuUInt32 start, end, len;
+            AuUInt start, end, len;
 
             if (i != 0)
             {
@@ -74,10 +81,17 @@ namespace Aurora::Parse
             }
 
             start = i;
-            end   = AuMin(AuUInt32(in.size()), AuUInt32(i + characters));
+            end   = AuMin(AuUInt(in.Size()), AuUInt(i + characters));
+            if (end - 1)
+            {
+                auto uLastCodepointIndex  = AuCodepointsFindPreviousValidByteOffsetFromByteOffset(in, end);
+                auto uLastByteOffset      = AuCodepointsNextLength(in.Substr(uLastCodepointIndex));
+                end = uLastCodepointIndex + uLastByteOffset;
+            }
             len   = end - start;
+            i    += len;
 
-            ret.insert(ret.size(), in.substr(start, len));
+            ret.insert(ret.size(), in.Substr(start, len));
         }
         return ret;
     }