fix(snippet): return one candidate per snippet region in multi-component scans

inc/scanoss.h

@@ -33,7 +33,7 @@
 #define WFP_LN 4
 #define WFP_REC_LN 18
 
-#define SCANOSS_VERSION "6.0.1-crc64"
+#define SCANOSS_VERSION "6.0.2-crc64"
 
 /* Log files */
 #define SCAN_LOG "/tmp/scanoss_scan.log"

src/match.c

@@ -91,7 +91,9 @@ match_data_t * match_data_copy(match_data_t * in)
     out->line_ranges = strdup(in->line_ranges);
     out->oss_ranges = strdup(in->oss_ranges);
     out->matched_percent = in->matched_percent;
+    out->lines_matched = in->lines_matched;
     out->snippet_ids = strdup(in->snippet_ids);
+    out->scan_ower = in->scan_ower; /* keep owning scan so downstream (e.g. print_json_match) can deref it */
     strcpy(out->source_md5, in->source_md5);
     return out;
 }
@@ -1099,6 +1101,18 @@ void match_select_best(scan_data_t *scan)
 	}
 }
 
+/**
+ * @brief Order multi-snippet matches by relevance (matched lines, descending).
+ *        Used to rank the per-region candidates so the most-covered block is the
+ *        principal match (match_index 0) and the rest follow as alternatives.
+ */
+bool more_lines_matched(match_data_t *a, match_data_t *b)
+{
+	if (!a)
+		return true;
+	return (b->lines_matched > a->lines_matched);
+}
+
 match_list_t *match_select_m_component_best(scan_data_t *scan)
 {
 	scanlog("<<<select_best_match_M: %d>>>>\n", scan->max_snippets_to_process);
@@ -1119,7 +1133,7 @@ match_list_t *match_select_m_component_best(scan_data_t *scan)
 		dup_match->component_list.match_ref = dup_match;
 		component_list_add(&dup_match->component_list, dup_comp, NULL, false);
 		dup_match->component_list.max_items = 1; // harcoded to show only fist component in report.
-		if (!match_list_add(final, dup_match, find_oldest_match, true))
+		if (!match_list_add(final, dup_match, more_lines_matched, true))
 			match_data_free(dup_match);
 	}
 

src/purl_scan.c

@@ -542,8 +542,15 @@ static int snippet_scan_stream(FILE *in)
 			extract_csv(size_field, (char *) rec, 1, sizeof(size_field));
 			file_size = strtoull(size_field, NULL, 10);
 
+			/* snippet-scan is a multi-candidate discovery endpoint: it must expose
+			   every non-overlapping snippet region, not just the single best block.
+			   The engine-wide default (SCAN_MAX_SNIPPETS_DEFAULT = 1) would collapse
+			   biggest_snippet() to one group, so unless the caller explicitly raised
+			   -M we open it up to MAX_MULTIPLE_COMPONENTS groups. */
+			int snippet_max = scan_max_snippets > 1 ? scan_max_snippets : MAX_MULTIPLE_COMPONENTS;
+
 			scan = scan_data_init(file_path,
-			                      scan_max_snippets,
+			                      snippet_max,
 			                      scan_max_components,
 			                      scan_adjust_tolerance,
 			                      scan_ranking_threshold,

src/snippet_selection.c

@@ -158,28 +158,41 @@ void biggest_snippet(scan_data_t *scan)
 			match_new->lines_matched = matched_lines;
 			//match_new->hits = hits;
 
-			do /*Check if there is already a list for this line ranges */
+			/* Place the match in the list of the snippet region it belongs to.
+			 * First look for an existing list whose ranges intersect this match
+			 * (same region -> alternative candidate). If none intersects, open a
+			 * new list for this region while there is room; otherwise fall back to
+			 * the last list. */
+			bool assigned = false;
+			for (i = 0; i < scan->matches_list_array_index; i++)
 			{
-				if (!scan->matches_list_array[scan->matches_list_array_index] && scan->matches_list_array_index < scan->max_snippets_to_process)
+				if (match_list_eval(scan->matches_list_array[i], match_new, ranges_intersection))
 				{
-					scan->matches_list_array[scan->matches_list_array_index] = match_list_init(true, 1);	/*create the list if it doesnt exist*/
-					scan->matches_list_array_index++;
-					if(!match_list_add(scan->matches_list_array[i], match_new, hit_test, true))
-					{
-						match_data_free(match_new); 
-					}
+					if (!match_list_add(scan->matches_list_array[i], match_new, hit_test, true))
+						match_data_free(match_new);
+					assigned = true;
 					break;
 				}
-				if (match_list_eval(scan->matches_list_array[i], match_new, ranges_intersection) || i == scan->max_snippets_to_process -1)
+			}
+
+			if (!assigned)
+			{
+				if (scan->matches_list_array_index < scan->max_snippets_to_process)
 				{
-					if(!match_list_add(scan->matches_list_array[i], match_new, hit_test, true))
-					{
-						match_data_free(match_new); 
-					}
-					break;
+					/* New non-overlapping region: create its own list and add the match there */
+					int new_index = scan->matches_list_array_index;
+					scan->matches_list_array[new_index] = match_list_init(true, 1);
+					scan->matches_list_array_index++;
+					if (!match_list_add(scan->matches_list_array[new_index], match_new, hit_test, true))
+						match_data_free(match_new);
 				}
-				i++;
-			} while(i < scan->matches_list_array_index); /*Check if there is already a list for this line ranges */
+				else
+				{
+					/* No room left for new regions: force into the last list */
+					if (!match_list_add(scan->matches_list_array[scan->max_snippets_to_process - 1], match_new, hit_test, true))
+						match_data_free(match_new);
+				}
+			}
 		}
 	}
 	/*just for loging*/

tests/snippet_multicomponent/main.c.wfp

@@ -0,0 +1,110 @@
+file=3bb103a46f94ed298d8c37b4a80d973c,9918,main.c
+7=e795bac8
+10=6e7bf056
+14=16579738
+25=23d6d030
+28=d32e499d
+30=09e0aecd,b7908937,21949e96
+33=d400f39f
+37=ef7d6cba,894d3fa4,2d1597e8
+41=c848f202
+44=be39f156
+59=905e02fb
+61=b7908937
+67=18b3beaf
+76=491ac1ea
+81=06fbfde3
+83=95c75ef3
+84=d400f39f
+88=b036ba15
+95=c848f202,be39f156
+111=c711d2e1
+112=447c4417
+114=9a106044,12ab20bb,0ce7ece1
+115=151c9088,d434c84e,bac49cc6
+116=2a6b8b81,1f38d8be
+117=7001190b
+119=25c9f85f,a6453c32,e9e65a61
+120=4186b77b,ef9fe105
+124=da55d5b9,b0924155
+128=b378b9e0
+129=addfb6b0
+134=0dbb599f,ec33702f
+135=80a48c6a
+136=f4347ad7,d5fbdf3a
+137=7ce444f8
+139=137b0602,f12edbd6
+142=3dd980ae,ac0b0110
+144=a76b34a5
+148=e0fa1c18,c741c7b2
+151=193cc2c3
+152=79eee8b8,b2e74c6e
+156=7b280e25,d864c721
+157=216caf87
+158=a6a3bbe1,b66372cb
+159=04ca6169,a7452328
+162=75fa5176
+163=d0cb6c6b,abafa52d
+167=9396f071
+168=30792053
+169=4b7b9a0a
+171=e6489ee3,8c952f0c
+172=b9cded56
+174=2d95465b
+176=11781789
+178=bd14a25f,f67df255
+180=46491e29
+183=c29128bc
+184=5625dd8a,30792053
+185=4b7b9a0a
+187=dd6588d4,0a041ed0
+188=341c8c04
+189=e3cc063a
+193=bea1b86c,802ddd80
+196=f537a752,5218fbf7
+198=f21eefbd
+199=c2b9b9f9,9f6d2f6f
+201=0724e239
+203=1ec0629c
+206=8dbf6363
+207=771caaeb
+212=35ac838c
+214=f94d5b4c,a169fcbf
+216=4e6c1cbe
+217=4666a13c
+219=077eccf8
+222=0ae64d50
+224=9712f0ff,4d24ea64
+228=78ef004d,46c56394,07bc7452
+230=97b52f82
+232=e3af2023
+236=161a335f
+237=f6a61105
+241=273539a2,08dcf25a,ac7340d8
+246=d0c75748
+248=061baf8c
+250=f941efa7
+253=36d3594e
+255=af8cf5fb,40b333f4
+259=8b1ff8bf,6cf0d5ac
+260=b803e9eb
+264=3cc8ad46
+265=74d8a41f
+269=0547b4df,b846d5c2
+272=3926ed53,4bf276e9,29af2a76
+274=fdb2863f,6b99be79
+275=6bc888a0
+278=05bf9646
+281=48848bae
+284=4d6fcc13
+285=cfc7ff90
+288=ab7fc8cd,fb85472f
+291=64f5cde0,0d2eca0b
+292=791fbabd
+296=07cfc5fe
+297=357fa0d4,87642ec3
+299=16e81aa2
+301=0acd0d5d,085f6061,c727d6af
+303=ac932df0
+306=0698ae44
+308=3dbbeabb,5919152f