added or updated code comments

as suggested by @terrelln,
to make the code of the optimal parser a bit more understandable.

diff --git a/lib/compress/zstd_compress_internal.h b/lib/compress/zstd_compress_internal.h
index dae8526d4611cd1abfcd31748eec099e849d481b..087ea49dcf8810cf20b52f8edeb5a4b52f1121ef 100644 (file)
--- a/lib/compress/zstd_compress_internal.h
+++ b/lib/compress/zstd_compress_internal.h
@@ -162,7 +162,7 @@ typedef struct {
     int price;  /* price from beginning of segment to this position */
     U32 off;    /* offset of previous match */
     U32 mlen;   /* length of previous match */
-    U32 litlen; /* nb of literals after previous match */
+    U32 litlen; /* nb of literals since previous match */
     U32 rep[ZSTD_REP_NUM];  /* offset history after previous match */
 } ZSTD_optimal_t;
 

diff --git a/lib/compress/zstd_opt.c b/lib/compress/zstd_opt.c
index 25715eabba885151c7e597d40e0060a3036095fe..8e1be1cec184f63bc0d44abea80c3edc928e8219 100644 (file)
--- a/lib/compress/zstd_opt.c
+++ b/lib/compress/zstd_opt.c
@@ -1129,13 +1129,20 @@ ZSTD_compressBlock_opt_generic(ZSTD_matchState_t* ms,
                 continue;
             }
 
+            /* Match found: let's store this solution, and eventually find more candidates.
+             * During this forward pass, @opt is used to store stretches,
+             * defined as "a match followed by N literals".
+             * Note how this is different from a Sequence, which is "N literals followed by a match".
+             * Storing stretches allows us to store different match predecessors
+             * for each literal position part of a literals run. */
+
             /* initialize opt[0] */
             opt[0].mlen = 0;  /* there are only literals so far */
             opt[0].litlen = litlen;
-            /* No need to include the actual price of the literals before the segment
+            /* No need to include the actual price of the literals before the first match
              * because it is static for the duration of the forward pass, and is included
-             * in every subsequent price. We include the literal length as the cost variation
-             * of litlen depends on the value of litlen.
+             * in every subsequent price. But, we include the literal length because
+             * the cost variation of litlen depends on the value of litlen.
              */
             opt[0].price = LL_PRICE(litlen);
             ZSTD_STATIC_ASSERT(sizeof(opt[0].rep[0]) == sizeof(rep[0]));
@@ -1353,11 +1360,10 @@ _shortestPath:   /* cur, last_pos, best_mlen, best_off have to be set */
             cur -= lastStretch.litlen;
         }
 
-        /* let's write the shortest path solution
-         * solution is stored in @opt,
-         * in reverse order,
-         * starting from @storeEnd (==cur+1)
-         * (effectively partially overwriting @opt).
+        /* Let's write the shortest path solution.
+         * It is stored in @opt in reverse order,
+         * starting from @storeEnd (==cur+2),
+         * effectively partially @opt overwriting.
          * Content is changed too:
          * - So far, @opt stored stretches, aka a match followed by literals
          * - Now, it will store sequences, aka literals followed by a match