summaryrefslogtreecommitdiff
diff options
context:
space:
mode:
authorKUDr <kudr@openttd.org>2006-11-24 21:37:08 +0000
committerKUDr <kudr@openttd.org>2006-11-24 21:37:08 +0000
commitf1a7a34f6e33410ee76949fb9c2a99034e7020a5 (patch)
treec16eb2877dc1a570904f28233d4c2d0346ed677e
parent1955d1acd6b7fa24e47784a1c282c288b6cee8f4 (diff)
downloadopenttd-f1a7a34f6e33410ee76949fb9c2a99034e7020a5.tar.xz
(svn r7255) -Comments added for CBlobBaseSimple, CBlobT classes
-rw-r--r--yapf/blob.hpp95
1 files changed, 85 insertions, 10 deletions
diff --git a/yapf/blob.hpp b/yapf/blob.hpp
index f05e6a77d..1a20f3ac2 100644
--- a/yapf/blob.hpp
+++ b/yapf/blob.hpp
@@ -3,6 +3,10 @@
#ifndef BLOB_HPP
#define BLOB_HPP
+/** Type-safe version of memcpy().
+ * @param d destination buffer
+ * @param s source buffer
+ * @param num_items number of items to be copied (!not number of bytes!) */
template <class Titem_>
FORCEINLINE void MemCpyT(Titem_* d, const Titem_* s, int num_items = 1)
{
@@ -18,51 +22,94 @@ FORCEINLINE void MemCpyT(Titem_* d, const Titem_* s, int num_items = 1)
* module (DLL) where the blob was allocated
* - no configurable allocation policy (how big blocks should be allocated)
* - no extra ownership policy (i.e. 'copy on write') when blob is copied
- * - no thread synchronization at all */
+ * - no thread synchronization at all
+ *
+ * Internal member layout:
+ * 1. The only class member is pointer to the first item (see union ptr_u).
+ * 2. Allocated block contains the blob header (see CHdr) followed by the raw byte data.
+ * Always, when it allocates memory the allocated size is:
+ * sizeof(CHdr) + <data capacity>
+ * 3. Two 'virtual' members (m_size and m_max_size) are stored in the CHdr at beginning
+ * of the alloated block.
+ * 4. The pointer (in ptr_u) points behind the header (to the first data byte).
+ * When memory block is allocated, the sizeof(CHdr) it added to it.
+ * 5. Benefits of this layout:
+ * - items are accessed in the simplest possible way - just dereferencing the pointer,
+ * which is good for performance (assuming that data are accessed most often).
+ * - sizeof(blob) is the same as the size of any other pointer
+ * 6. Drawbacks of this layout:
+ * - the fact, that pointer to the alocated block is adjusted by sizeof(CHdr) before
+ * it is stored can lead to several confusions:
+ * - it is not common pattern so the implementation code is bit harder to read
+ * - valgrind can generate warning that allocated block is lost (not accessible)
+ * */
class CBlobBaseSimple {
protected:
+ /** header of the allocated memory block */
struct CHdr {
- int m_size; // actual blob size in bytes
- int m_max_size; // maximum (allocated) size in bytes
+ int m_size; ///< actual blob size in bytes
+ int m_max_size; ///< maximum (allocated) size in bytes
};
+ /** type used as class member */
union {
- int8 *m_pData;
- CHdr *m_pHdr_1;
+ int8 *m_pData; ///< pointer to the first byte of data
+ CHdr *m_pHdr_1; ///< pointer just after the CHdr holding m_size and m_max_size
} ptr_u;
public:
- static const int Ttail_reserve = 4; // four extra bytes will be always allocated and zeroed at the end
+ static const int Ttail_reserve = 4; ///< four extra bytes will be always allocated and zeroed at the end
+ /** default constructor - initializes empty blob */
FORCEINLINE CBlobBaseSimple() { InitEmpty(); }
+ /** copy constructor */
FORCEINLINE CBlobBaseSimple(const CBlobBaseSimple& src)
{
InitEmpty();
AppendRaw(src);
}
+ /** destructor */
FORCEINLINE ~CBlobBaseSimple() { Free(); }
protected:
+ /** initialize the empty blob by setting the ptr_u.m_pHdr_1 pointer to the static CHdr with
+ * both m_size and m_max_size containing zero */
FORCEINLINE void InitEmpty() { static CHdr hdrEmpty[] = {{0, 0}, {0, 0}}; ptr_u.m_pHdr_1 = &hdrEmpty[1]; }
+ /** initialize blob by attaching it to the given header followed by data */
FORCEINLINE void Init(CHdr* hdr) { ptr_u.m_pHdr_1 = &hdr[1]; }
+ /** blob header accessor - use it rather than using the pointer arithmetics directly - non-const version */
FORCEINLINE CHdr& Hdr() { return ptr_u.m_pHdr_1[-1]; }
+ /** blob header accessor - use it rather than using the pointer arithmetics directly - const version */
FORCEINLINE const CHdr& Hdr() const { return ptr_u.m_pHdr_1[-1]; }
+ /** return reference to the actual blob size - used when the size needs to be modified */
FORCEINLINE int& RawSizeRef() { return Hdr().m_size; };
public:
+ /** return true if blob doesn't contain valid data */
FORCEINLINE bool IsEmpty() const { return RawSize() == 0; }
+ /** return the number of valid data bytes in the blob */
FORCEINLINE int RawSize() const { return Hdr().m_size; };
+ /** return the current blob capacity in bytes */
FORCEINLINE int MaxRawSize() const { return Hdr().m_max_size; };
+ /** return pointer to the first byte of data - non-const version */
FORCEINLINE int8* RawData() { return ptr_u.m_pData; }
+ /** return pointer to the first byte of data - const version */
FORCEINLINE const int8* RawData() const { return ptr_u.m_pData; }
#if 0 // reenable when needed
+ /** return the 32 bit CRC of valid data in the blob */
FORCEINLINE uint32 Crc32() const {return CCrc32::Calc(RawData(), RawSize());}
#endif //0
+ /** invalidate blob's data - doesn't free buffer */
FORCEINLINE void Clear() { RawSizeRef() = 0; }
+ /** free the blob's memory */
FORCEINLINE void Free() { if (MaxRawSize() > 0) {RawFree(&Hdr()); InitEmpty();} }
+ /** copy data from another blob - replaces any existing blob's data */
FORCEINLINE void CopyFrom(const CBlobBaseSimple& src) { Clear(); AppendRaw(src); }
+ /** overtake ownership of data buffer from the source blob - source blob will become empty */
FORCEINLINE void MoveFrom(CBlobBaseSimple& src) { Free(); ptr_u.m_pData = src.ptr_u.m_pData; src.InitEmpty(); }
+ /** swap buffers (with data) between two blobs (this and source blob) */
FORCEINLINE void Swap(CBlobBaseSimple& src) { int8 *tmp = ptr_u.m_pData; ptr_u.m_pData = src.ptr_u.m_pData; src.ptr_u.m_pData = tmp; }
+ /** append new bytes at the end of existing data bytes - reallocates if necessary */
FORCEINLINE void AppendRaw(int8 *p, int num_bytes)
{
assert(p != NULL);
@@ -73,6 +120,7 @@ public:
}
}
+ /** append bytes from given source blob to the end of existing data bytes - reallocates if necessary */
FORCEINLINE void AppendRaw(const CBlobBaseSimple& src)
{
if (!src.IsEmpty())
@@ -164,6 +212,13 @@ public:
}
};
+/** Blob - simple dynamic Titem_ array. Titem_ (template argument) is a placeholder for any type.
+ * Titem_ can be any integral type, pointer, or structure. Using Blob instead of just plain C array
+ * simplifies the resource management in several ways:
+ * 1. When adding new item(s) it automatically grows capacity if needed.
+ * 2. When variable of type Blob comes out of scope it automatically frees the data buffer.
+ * 3. Takes care about the actual data size (number of used items).
+ * 4. Dynamically constructs only used items (as opposite of static array which constructs all items) */
template <class Titem_, class Tbase_ = CBlobBaseSimple>
class CBlobT : public CBlobBaseSimple {
// make template arguments public:
@@ -173,15 +228,25 @@ public:
static const int Titem_size = sizeof(Titem);
+ /** Default constructor - makes new Blob ready to accept any data */
FORCEINLINE CBlobT() : Tbase() {}
+ /** Copy constructor - make new blob to become copy of the original (source) blob */
FORCEINLINE CBlobT(const Tbase& src) : Tbase(src) {assert((RawSize() % Titem_size) == 0);}
+ /** Destructor - ensures that allocated memory (if any) is freed */
FORCEINLINE ~CBlobT() { Free(); }
+ /** Check the validity of item index (only in debug mode) */
FORCEINLINE void CheckIdx(int idx) { assert(idx >= 0); assert(idx < Size()); }
+ /** Return pointer to the first data item - non-const version */
FORCEINLINE Titem* Data() { return (Titem*)RawData(); }
+ /** Return pointer to the first data item - const version */
FORCEINLINE const Titem* Data() const { return (const Titem*)RawData(); }
+ /** Return pointer to the idx-th data item - non-const version */
FORCEINLINE Titem* Data(int idx) { CheckIdx(idx); return (Data() + idx); }
+ /** Return pointer to the idx-th data item - const version */
FORCEINLINE const Titem* Data(int idx) const { CheckIdx(idx); return (Data() + idx); }
+ /** Return number of items in the Blob */
FORCEINLINE int Size() const { return (RawSize() / Titem_size); }
+ /** Free the memory occupied by Blob destroying all items */
FORCEINLINE void Free()
{
assert((RawSize() % Titem_size) == 0);
@@ -193,12 +258,15 @@ public:
}
Tbase::Free();
}
+ /** Grow number of data items in Blob by given number - doesn't construct items */
FORCEINLINE Titem* GrowSizeNC(int num_items) { return (Titem*)GrowRawSize(num_items * Titem_size); }
+ /** Grow number of data items in Blob by given number - constructs new items (using Titem_'s default constructor) */
FORCEINLINE Titem* GrowSizeC(int num_items)
{
Titem* pI = GrowSizeNC(num_items);
for (int i = num_items; i > 0; i--, pI++) new (pI) Titem();
}
+ /** Destroy given number of items and reduce the Blob's data size */
FORCEINLINE void ReduceSize(int num_items)
{
assert((RawSize() % Titem_size) == 0);
@@ -211,18 +279,21 @@ public:
// remove them
ReduceRawSize(num_items * Titem_size);
}
+ /** Append one data item at the end (calls Titem_'s default constructor) */
FORCEINLINE Titem* AppendNew()
{
- Titem& dst = *GrowSizeNC(1);
- Titem* pNewItem = new (&dst) Titem();
+ Titem& dst = *GrowSizeNC(1); // Grow size by one item
+ Titem* pNewItem = new (&dst) Titem(); // construct the new item by calling in-place new operator
return pNewItem;
}
+ /** Append the copy of given item at the end of Blob (using copy constructor) */
FORCEINLINE Titem* Append(const Titem& src)
{
- Titem& dst = *GrowSizeNC(1);
- Titem* pNewItem = new (&dst) Titem(src);
+ Titem& dst = *GrowSizeNC(1); // Grow size by one item
+ Titem* pNewItem = new (&dst) Titem(src); // construct the new item by calling in-place new operator with copy ctor()
return pNewItem;
}
+ /** Add given items (ptr + number of items) at the end of blob */
FORCEINLINE Titem* Append(const Titem* pSrc, int num_items)
{
Titem* pDst = GrowSizeNC(num_items);
@@ -231,6 +302,7 @@ public:
while (pDst < pDstEnd) new (pDst++) Titem(*(pSrc++));
return pDstOrg;
}
+ /** Remove item with the given index by replacing it by the last item and reducing the size by one */
FORCEINLINE void RemoveBySwap(int idx)
{
CheckIdx(idx);
@@ -238,6 +310,7 @@ public:
Titem* pRemoved = Data(idx);
RemoveBySwap(pRemoved);
}
+ /** Remove item given by pointer replacing it by the last item and reducing the size by one */
FORCEINLINE void RemoveBySwap(Titem* pItem)
{
Titem* pLast = Data(Size() - 1);
@@ -252,6 +325,8 @@ public:
// and reduce the raw blob size
ReduceRawSize(Titem_size);
}
+ /** Ensures that given number of items can be added to the end of Blob. Returns pointer to the
+ * first free (unused) item */
FORCEINLINE Titem* MakeFreeSpace(int num_items) { return (Titem*)MakeRawFreeSpace(num_items * Titem_size); }
};