fuse & warning, add a new warning if clone have not same name between samples

Link to #4566

doc/warnings.md

@@ -58,4 +58,5 @@ Warnings which were implemented ([x]) have a fixed code that should not be chang
 ## Comparisons between several samples or patients (such as with clonedb)
 - [ ] W8x Potential contamination or public clone #1744 
 - [ ]     Known public clone 
+- [x] W81 Clone have different names between samples #4566
 

tools/fuse.py

@@ -222,6 +222,16 @@ class Window:
                 if key not in obj.d:
                     obj.d[key] = source.d[key]
 
+        # !! No name field if fuse with an empty clone
+        if "name" in self.d and "name" in other.d and self.d["name"] != other.d["name"]:
+            if obj.d["name"] == self.d["name"]:
+                name = other.d["name"]
+            else:
+                name = self.d["name"]
+            # init warn field if not already present
+            if not "warn" in obj.d:
+                obj.d["warn"] = []
+            obj.d["warn"].append( {"code":"W81", "msg":"Clone have different names between samples: %s" % name, "level": "warn"})
         return obj
         
     def get_nb_reads(self, cid, point=0):

tools/tests/data/warnings/w81_sample_1.vidjil

+{"clones": [{"germline": "IGK", "id": "clone_1", "name": "clone_name_1", "reads": [500], "top": 1}, {"germline": "IGK", "id": "clone_3", "name": "clone_name_3", "reads": [500], "top": 2}], "clusters": [], "germlines": {}, "notes": "Sample with clone 1 and 3;", "producer": "vidjil fuse", "reads": {"germline": {"IGK": [9500]}, "segmented": [9500], "total": [9500]}, "samples": {"commandline": ["cmdline_1"], "log": ["log_1"], "number": 1, "original_names": ["seq_1"], "producer": ["vidjil-algo 2018.02"], "run_timestamp": ["2018-03-28 10:26:11"], "timestamp": ["2018-04-06 13:48:48"]}, "timestamp": "2019-04-08 15:18:00", "vidjil_json_version": "2016b", "warn": []}

tools/tests/data/warnings/w81_sample_2.vidjil

+{"clones": [{"germline": "IGK", "id": "clone_1", "name": "clone_name_1", "reads": [500], "top": 1}, {"germline": "IGK", "id": "clone_2", "name": "clone_name_2", "reads": [500], "top": 2}, {"germline": "IGK", "id": "clone_3", "name": "clone_name_3b", "reads": [500], "top": 3}], "clusters": [], "germlines": {}, "notes": "Sample with clone 1-3; clone 3 have a variation in his name", "producer": "vidjil fuse", "reads": {"germline": {"IGK": [9500]}, "segmented": [9500], "total": [9500]}, "samples": {"commandline": ["cmdline_2"], "log": ["log_2"], "number": 1, "original_names": ["seq_2"], "producer": ["vidjil-algo 2018.02"], "run_timestamp": ["2018-03-28 10:26:11"], "timestamp": ["2018-04-06 13:48:48"]}, "timestamp": "2019-04-08 15:18:00", "vidjil_json_version": "2016b", "warn": []}

tools/tests/should-get-tests/fuse_warnings.should-get

+!LAUNCH: python ../../fuse.py $FUSE_OPTIONS ../data/warnings/w81_sample_1.vidjil  ../data/warnings/w81_sample_2.vidjil -o fuse_warnings.vidjil ; cat fuse_warnings.vidjil
+
+$ Fuse the three files
+2: ListWindows: .9500, 9500.
+
+
+$ Middle file has the good 40-bp windows
+1: "id": "clone_1"
+1: "id": "clone_2"
+1: "id": "clone_3"
+
+$ fields 'warn' present 3 times (sample, clone, level)
+1: "level": "warn"
+
+$ Correct warning is present
+1: W81
+1: Clone have different names between samples: clone_name_3b
\ No newline at end of file