From 441e2dfa2772794673b69cc72f8561fc026d077b Mon Sep 17 00:00:00 2001 From: BonfaceKilz Date: Mon, 27 Jul 2020 18:35:45 +0300 Subject: Use the correct redis instance inside object * wqflask/base/data_set.py (DatasetType): Use object's redis instance --- wqflask/base/data_set.py | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) (limited to 'wqflask/base/data_set.py') diff --git a/wqflask/base/data_set.py b/wqflask/base/data_set.py index 5d562871..bce94aa8 100644 --- a/wqflask/base/data_set.py +++ b/wqflask/base/data_set.py @@ -93,7 +93,7 @@ Publish or ProbeSet. E.g. """ self.redis_instance = redis_instance self.datasets = {} - data = redis_instance.get("dataset_structure") + data = self.redis_instance.get("dataset_structure") if data: self.datasets = json.loads(data) else: # ZS: I don't think this should ever run unless Redis is emptied @@ -115,7 +115,7 @@ Publish or ProbeSet. E.g. except: pass - redis_instance.set("dataset_structure", json.dumps(self.datasets)) + self.redis_instance.set("dataset_structure", json.dumps(self.datasets)) # Set LOG_LEVEL_DEBUG=5 to see the following: logger.debugf(5, "datasets", self.datasets) -- cgit v1.2.3 From fea8444fb7b0224fd44711853904822ee8b43f4f Mon Sep 17 00:00:00 2001 From: BonfaceKilz Date: Mon, 27 Jul 2020 19:11:11 +0300 Subject: Abstract away redundant code into set_dataset_key method * wqflask/base/data_set.py (Dataset): - Add set_dataset_key - Remove __call__ method --- wqflask/base/data_set.py | 109 ++++++++++++++++++++--------------------------- 1 file changed, 47 insertions(+), 62 deletions(-) (limited to 'wqflask/base/data_set.py') diff --git a/wqflask/base/data_set.py b/wqflask/base/data_set.py index bce94aa8..cfba9104 100644 --- a/wqflask/base/data_set.py +++ b/wqflask/base/data_set.py @@ -117,71 +117,56 @@ Publish or ProbeSet. E.g. self.redis_instance.set("dataset_structure", json.dumps(self.datasets)) - # Set LOG_LEVEL_DEBUG=5 to see the following: - logger.debugf(5, "datasets", self.datasets) + def set_dataset_key(self, t, name): + """If name is not in the object's dataset dictionary, set it, and update + dataset_structure in Redis + + args: + t: Type of dataset structure which can be: 'mrna_expr', 'pheno', + 'other_pheno', 'geno' + name: The name of the key to inserted in the datasets dictionary + + """ + sql_query_mapping = { + 'mrna_expr': ("""SELECT ProbeSetFreeze.Id FROM """ + + """ProbeSetFreeze WHERE ProbeSetFreeze.Name = "{}" """), + 'pheno': ("""SELECT InfoFiles.GN_AccesionId """ + + """FROM InfoFiles, PublishFreeze, InbredSet """ + + """WHERE InbredSet.Name = '{}' AND """ + + """PublishFreeze.InbredSetId = InbredSet.Id AND """ + + """InfoFiles.InfoPageName = PublishFreeze.Name"""), + 'other_pheno': ("""SELECT PublishFreeze.Name """ + + """FROM PublishFreeze, InbredSet """ + + """WHERE InbredSet.Name = '{}' AND """ + + """PublishFreeze.InbredSetId = InbredSet.Id"""), + 'geno': ("""SELECT GenoFreeze.Id FROM GenoFreeze WHERE """ + + """GenoFreeze.Name = "{}" """) + } + + dataset_name_mapping = { + "mrna_expr": "ProbeSet", + "pheno": "Publish", + "other_pheno": "Publish", + "geno": "Geno", + } + + if t in ['pheno', 'other_pheno']: + name = name.replace("Publish", "") + if bool(len(g.db.execute(sql_query_mapping[t].format(name)))): + self.datasets[name] = dataset_name_mapping[t] + self.redis_instance.set("dataset_structure", json.dumps(self.datasets)) + return True + + return None def __call__(self, name): + if name not in self.datasets: - mrna_expr_query = """ - SELECT - ProbeSetFreeze.Id - FROM - ProbeSetFreeze - WHERE - ProbeSetFreeze.Name = "{0}" - """.format(name) - - results = g.db.execute(mrna_expr_query).fetchall() - if len(results): - self.datasets[name] = "ProbeSet" - redis_instance.set("dataset_structure", json.dumps(self.datasets)) - return self.datasets[name] - - group_name = name.replace("Publish", "") - - pheno_query = """SELECT InfoFiles.GN_AccesionId - FROM InfoFiles, PublishFreeze, InbredSet - WHERE InbredSet.Name = '{0}' AND - PublishFreeze.InbredSetId = InbredSet.Id AND - InfoFiles.InfoPageName = PublishFreeze.Name""".format(group_name) - - results = g.db.execute(pheno_query).fetchall() - if len(results): - self.datasets[name] = "Publish" - redis_instance.set("dataset_structure", json.dumps(self.datasets)) - return self.datasets[name] - - # ZS: For when there isn't an InfoFiles ID; not sure if this and the preceding query are both necessary - other_pheno_query = """SELECT PublishFreeze.Name - FROM PublishFreeze, InbredSet - WHERE InbredSet.Name = '{}' AND - PublishFreeze.InbredSetId = InbredSet.Id""".format(group_name) - - results = g.db.execute(other_pheno_query).fetchall() - if len(results): - self.datasets[name] = "Publish" - redis_instance.set("dataset_structure", json.dumps(self.datasets)) - return self.datasets[name] - - geno_query = """ - SELECT - GenoFreeze.Id - FROM - GenoFreeze - WHERE - GenoFreeze.Name = "{0}" - """.format(name) - - results = g.db.execute(geno_query).fetchall() - if len(results): - self.datasets[name] = "Geno" - self.redis_instance.set("dataset_structure", json.dumps(self.datasets)) - return self.datasets[name] - - # ZS: It shouldn't ever reach this - return None - else: - return self.datasets[name] + for t in ["mrna_expr", "pheno", "other_pheno", "geno"]: + # This has side-effects, with the end result being a truth-y value + if(self.set_dataset_key(t, name)): + break + return self.datasets.get(name, None) # Return None if name has not been set # Do the intensive work at startup one time only -- cgit v1.2.3