[*] Comments

Include/Aurora/Exit/IExitSubscriber.hpp

@@ -13,10 +13,7 @@ namespace Aurora::Exit
     {
         Threading::Threads::IAuroraThread *pCaller;
     };
-    
-    /**
-     * @brief User definedable callback
-    */
+
     AUKN_INTERFACE(IExitSubscriber,
         AUI_METHOD(void, OnTrigger, (ETriggerLevel, level, const ExitInvoker *, pInvoker))
     );

Include/Aurora/IO/FS/Overlapped.hpp

@@ -7,6 +7,8 @@
     Note: Defer to Async.hpp for creation of overlapped stream objects
           This header defines FS.hpp-like APIs that run overlapped or on a
           worker thread or two.
+
+          **IMPL IN PROGRESS**
 ***/
 #pragma once
 
@@ -49,7 +51,7 @@ namespace Aurora::IO::FS
       *  @brief If running under AuAsync runner, we use the local thread for the entire IO operation.
       *         In the future, whence I can be bothered to finish the APC api, *all* overlapped operations
       *         will be processed on the local thread.
-      *         Specifying true forces all IO to occour on a preallocated pool of IO workers
+      *         Specifying true forces all IO to occur on a preallocated pool of IO workers
       * @warning this api is thread-local
       * @warning this api does not change the behaviour of overlapped file streamss
      */

Include/Aurora/IO/Protocol/IProtocolInterceptorEx.hpp

@@ -12,8 +12,9 @@ namespace Aurora::IO::Protocol
     AUKN_INTERFACE(IProtocolInterceptorEx,
         /**
          * @brief
-         * @return false restores the read head, will retrieve further callbacks
-         *         true  nothing
+         * @return false restores the read head of pReadInByteBuffer. future callbacks will still be made.
+         *         true  nothing; pReadInByteBuffer's heads will remain set by the user and the tick is propagated down the stack.
+         * @warning the error flags of both bytebuffers can seize up the protocol stack by design.
         */
         AUI_METHOD(bool, OnDataAvailable, (const AuSPtr<Memory::ByteBuffer> &, pReadInByteBuffer,
                                            const AuSPtr<Memory::ByteBuffer> &, pWriteOutByteBuffer))

Include/Aurora/IO/Protocol/IProtocolPiece.hpp

@@ -17,8 +17,7 @@ namespace Aurora::IO::Protocol
         virtual void Remove() = 0;
 
         /**
-         * @brief Fetches an IStreamWriter representation of the next piece
-         *        in the stack
+         * @brief Fetches an IStreamWriter representation of the next piece in the stack
          * @return 
         */
         virtual AuSPtr<IStreamWriter> ToNextWriter() = 0;

Include/Aurora/IO/Protocol/IProtocolStack.hpp

@@ -37,15 +37,19 @@ namespace Aurora::IO::Protocol
 
         /**
          * @brief Same as AppendInterceptor, except that DoTick will repeat until the input is fully consumed.
-         *        This allows us to tick processors under our frame, followed by a retick if there is more data available.
-         *        Such paradigm contrasts the default stream-specific/non-framing behaviour where the processor is
-         *        responsible for consuming all non-segmented data (ala everything except the start of the final unfinished frame).
+         * 
+         *        This allows us to tick processors under our frame, followed by a retick if there is more data available to us.
+         *        Such paradigm contrasts the default unframed/raw stream behaviour where the processor is responsible for consuming all
+         *        available data. 
          *        With such limitations, it's impossible to build a protocol where a piece in the middle
-         *        1) can/should only flush 1 frame to the next processor
-         *        2) reads only a portion of pReadInByteBuffer
-         *        This function sets an internal flag that indicates re-ticks from the root are to be expected to complete
-         *        a tick. Otherwise, the read head of the inbound frame of a persumed stream may only skip back to frame zero
-         *        on error OR frame +1 on success. With this flag, frame current to frame end will be ticked.
+         *        1) can/should only flush 1 frame to the following interceptor 
+         *         (this is a problem when the frame does not contain a length prefix and the next interceptor is a frame processor)
+         *        2) only reads a portion of pReadInByteBuffer
+         *         (the current interceptor can only process a frame of data per tick)
+         * 
+         *        This function sets an internal flag that indicates re-ticks from the root single frame processor are to be expected
+         *        in order to complete a single ::DoTick tick. With this flag enabled, multiple tick may occur as opposed to a single
+         *        callback per ::DoTick().
          * @param pInterceptorEx
          * @return
         */
@@ -76,8 +80,6 @@ namespace Aurora::IO::Protocol
         /**
          * @brief Sends one tick down the protocol stack, regardless of how much data is written into the
          *        next piece/interceptor, and regardless of if another read tick is required.
-         *        Latterly, you are responsible for consuming all available bytes in your non-framed
-         *        (~AppendSingleFrameProcessor[Ex]) interceptors.
         */
         virtual void DoTick() = 0;
 
@@ -96,7 +98,6 @@ namespace Aurora::IO::Protocol
 
     /**
      * @brief Creates a protocol stack backed by the memory of a pipe.
-     *        DoTick must be called each time the pipe has data.
      * @param pWork 
      * @return 
     */