refactor: Switch to brightway

data/Makefile

@@ -38,7 +38,7 @@ sync_datapackages:
 	@$(call DOCKER,python3 common/sync_datapackages.py)
 
 delete_database:
-	@$(call DOCKER,python3 common/delete_database.py $(DB))
+	@$(call DOCKER,python3 common/delete_database.py "$(DB)")
 
 delete_method:
 	@$(call DOCKER,python3 common/delete_methods.py)

data/common/__init__.py

@@ -5,15 +5,15 @@
 
 
 def normalization_factors(impact_defs):
-    normalization_factors = {}
+    factors = {}
     for k, v in impact_defs.items():
-        if v.get("ecoscore"):
-            normalization_factors[k] = (
-                v["ecoscore"]["weighting"] / v["ecoscore"]["normalization"]
-            )
+        if v.get("ecoscore") and "normalization" in v.get("ecoscore", {}):
+            factors[k] = v["ecoscore"]["normalization"]
+        elif v.get("pef") and "normalization" in v.get("pef", {}):
+            factors[k] = v["pef"]["normalization"]
         else:
-            normalization_factors[k] = 0
-    return normalization_factors
+            pass
+    return factors
 
 
 def spproject(activity):
@@ -28,6 +28,8 @@ def spproject(activity):
             return "ADEME UPR"
         case "Woolmark":
             return "Woolmark"
+        case "WFLDB":
+            return "WFLDB"
         case "PastoEco":
             return "AGB3.1.1 2023-03-06"
         case _:
@@ -153,6 +155,8 @@ def calculate_aggregate(process_impacts, normalization_factors):
 
 def bytrigram(definitions, bynames):
     """takes the impact definitions and some impacts by name, return the impacts by trigram"""
+    if type(bynames) is not dict:
+        return {}
     trigramsByName = {method[1]: trigram for trigram, method in definitions.items()}
     return {
         trigramsByName.get(name): amount["amount"]

data/common/export.py

@@ -225,7 +225,7 @@ def compute_impacts(frozen_processes, default_db, impacts_py):
         progress_bar(index, len(processes))
         # Don't compute impacts if its a hardcoded activity
         if process.get("impacts"):
-            logger.info(f"This process has hardcoded impacts: {process['displayName']}")
+            logger.info(f"This process has hardcoded impacts: {process['name']}")
             continue
         # simapro
         activity = cached_search(
@@ -234,7 +234,7 @@ def compute_impacts(frozen_processes, default_db, impacts_py):
         if not activity:
             raise Exception(f"This process was not found in brightway: {process}")
 
-        results = compute_simapro_impacts(activity, main_method, impacts_py)
+        results = None  # compute_simapro_impacts(activity, main_method, impacts_py)
         # WARNING assume remote is in m3 or MJ (couldn't find unit from COM intf)
         if process["unit"] == "kWh" and isinstance(results, dict):
             results = {k: v * 3.6 for k, v in results.items()}
@@ -319,8 +319,8 @@ def plot_impacts(process_name, impacts_smp, impacts_bw, folder, impacts_py):
     ]
     nf = normalization_factors(impacts_py)
 
-    simapro_values = [impacts_smp[label] * nf[label] for label in trigrams]
-    brightway_values = [impacts_bw[label] * nf[label] for label in trigrams]
+    simapro_values = [impacts_smp[label] / nf[label] for label in trigrams]
+    brightway_values = [impacts_bw[label] / nf[label] for label in trigrams]
 
     x = numpy.arange(len(trigrams))
     width = 0.35
@@ -394,7 +394,13 @@ def find_id(dbname, activity):
 
 
 def compute_simapro_impacts(activity, method, impacts_py):
-    strprocess = urllib.parse.quote(activity["name"], encoding=None, errors=None)
+    name = (
+        activity["name"]
+        if spproject(activity) != "WFLDB"
+        # TODO this should probably done through disabling a strategy
+        else f"{activity['name']}/{activity['location']} U"
+    )
+    strprocess = urllib.parse.quote(name, encoding=None, errors=None)
     project = urllib.parse.quote(spproject(activity), encoding=None, errors=None)
     method = urllib.parse.quote(main_method, encoding=None, errors=None)
     api_request = f"http://simapro.ecobalyse.fr:8000/impact?process={strprocess}&project={project}&method={method}"

data/common/import_.py

@@ -235,6 +235,7 @@ def import_simapro_csv(
 
     database.apply_strategies()
     database.statistics()
+
     # try to link remaining unlinked technosphere activities
     database.apply_strategy(
         functools.partial(
@@ -252,8 +253,8 @@ def import_simapro_csv(
 
     print("### Adding unlinked flows and activities...")
     # comment to enable stopping on unlinked activities and creating an excel file
-    database.add_unlinked_flows_to_biosphere_database(biosphere)
-    database.add_unlinked_activities()
+    # database.add_unlinked_flows_to_biosphere_database(biosphere)
+    # database.add_unlinked_activities()
 
     # stop if there are unlinked activities
     if len(list(database.unlinked)):

data/common/sync_datapackages.py

@@ -1,11 +1,20 @@
+# Configure logger
+import sys
+
 import bw2data
+from loguru import logger
+
+logger.remove()  # Remove default handler
+logger.add(sys.stderr, format="{time} {level} {message}", level="INFO")
 
 PROJECT = "default"
 print("Syncing datapackages...")
 bw2data.projects.set_current(PROJECT)
 for method in bw2data.methods:
+    logger.info(f"Syncing method {method}...")
     bw2data.Method(method).process()
 
 for database in bw2data.databases:
+    logger.info(f"Syncing database {database}...")
     bw2data.Database(database).process()
 print("done")

data/docker/simapro-biosphere.json

@@ -849,11 +849,12 @@
   ["water", "Benzene, 1,2-dichloro-", "o-Dichlorobenzene"],
   ["water", "Benzo(a)anthracene", "Benz(a)anthracene"],
   ["water", "Benzo(g,h,i)perylene", "Benzo(ghi)perylene"],
-  ["water", "Cadmium", "Cadmium, ion"],
+  ["water", "Cadmium", "Cadmium II"],
   ["water", "Calcium", "Calcium, ion"],
   ["water", "Carbamic acid, butyl-, 3-iodo-2-propynyl ester", "Butylcarbamate, iodopropynyl"],
   ["water", "Carbon", "Elemental carbon"],
   ["water", "Chromium", "Chromium, ion"],
+  ["soil", "Chromium", "Chromium, ion"],
   ["water", "Copper", "Copper, ion"],
   ["soil", "Copper", "Copper, ion"],
   ["water", "Ethane, chloro-", "Monochloroethane"],

data/import_ecoinvent.py

@@ -47,6 +47,18 @@ def organic_cotton_irrigation(db):
 STRATEGIES = [organic_cotton_irrigation]
 
 
+def use_unit_processes(db):
+    """the woolmark dataset comes with dependent processes
+    which are set as system processes.
+    EI3.10 has these processes but as unit processes.
+    So we change the name such as the linking be done"""
+    for ds in db:
+        for exc in ds["exchanges"]:
+            if exc["name"].endswith("Cut-off, S"):
+                exc["name"].replace("Cut-off, S", "Cut-off, U")
+    return db
+
+
 def main():
     projects.set_current(PROJECT)
     # projects.create_project(PROJECT, activate=True, exist_ok=True)
@@ -79,6 +91,7 @@ def main():
         import_simapro_csv(
             join("..", "..", "dbfiles", WOOL),
             db,
+            first_strategies=[use_unit_processes],
             external_db="Ecoinvent 3.10",  # wool is linked with EI 3.10
             excluded_strategies=EXCLUDED,
         )

data/spapi/simapro.py

@@ -41,16 +41,19 @@ async def impact(_: Request, project: str, process: str, method: str):
             server.OpenProject(project, "")
 
         print("Computing results...")
+        # hack because the open project is "WFLDB"
+        # but the process are in the library project "World Food LCA Database"
+        tmpproject = "World Food LCA Database" if project == "WFLDB" else project
         existing = [
             e
             for e in [
-                ((i, server.FindProcess(project, i, process)[0])) for i in range(12)
+                ((i, server.FindProcess(tmpproject, i, process)[0])) for i in range(12)
             ]
             if e[1]
         ]
         found = existing[0] if len(existing) else None
         if found:
-            server.Analyse(project, found[0], process, "Methods", method, "")
+            server.Analyse(tmpproject, found[0], process, "Methods", method, "")
             results, i = {}, 0
             try:
                 # try the first and stop if it raises (typically on a Dummy process.
@@ -66,6 +69,8 @@ async def impact(_: Request, project: str, process: str, method: str):
                 results[r.IndicatorName] = {"amount": r.Amount, "unit": r.UnitName}
                 i += 1
             impacts.setdefault(f"{project}/{process}", {})
+            if not results:
+                return results
             impacts[f"{project}/{process}"][method] = results
             with open("impacts.json", "w") as fp:
                 json.dump(impacts, fp, ensure_ascii=False)

data/textile/export.py

@@ -160,16 +160,16 @@ def csv_export_impact_comparison(compared_impacts):
         )
         csv_export_impact_comparison(impacts_compared_dic)
         for process_name, values in impacts_compared_dic.items():
-            displayName = processes[process_name]["displayName"]
-            print(f"Plotting {displayName}")
+            name = processes[process_name]["name"]
+            print(f"Plotting {name}")
             if "simapro_impacts" not in values and "brightway_impacts" not in values:
-                print(f"This hardcopied process cannot be plot: {displayName}")
+                print(f"This hardcopied process cannot be plot: {name}")
                 continue
             simapro_impacts = values["simapro_impacts"]
             brightway_impacts = values["brightway_impacts"]
             os.makedirs(GRAPH_FOLDER, exist_ok=True)
             plot_impacts(
-                displayName,
+                name,
                 simapro_impacts,
                 brightway_impacts,
                 GRAPH_FOLDER,

public/data/object/processes.json

@@ -31,8 +31,8 @@
       "swe": 0,
       "tre": 0,
       "wtu": 0,
-      "ecs": 23650.509942608354,
-      "pef": 27363.305839091667
+      "ecs": 23630.543538072445,
+      "pef": 27347.019127874755
     }
   },
   {