copy missing flows from one biosphere to another

copy_missing_flows.py

#!/usr/bin/env python
# coding: utf-8

# # Copy missing biosphere3 flows
# From an ecoinvent 3.10 install to a 3.8 to create a temporary migration environment to avoid to re-build complete project from scratch.
# 

# In[1]:


import bw2data as bd


# In[2]:


list(bd.projects)


# In[3]:


bd.projects.set_current("ei38-ei39-upgradable")
list(bd.databases)


# ## EI 3.8 biosphere

# In[4]:


bd_bio3 = bd.Database("biosphere3")
print(f"There are {len(bd_bio3)} flows in biosphere3")


# ## EI 3.10 biosphere

# In[5]:


bd_ei310bio3 = bd.Database("ecoinvent-3.10-biosphere")
print(f"There are {len(bd_ei310bio3)} flows in biosphere3")


# In[6]:


r_flow_e310 = bd_ei310bio3.random()
r_flow_e310.as_dict()


# In[23]:


ei310_biosphere_codes = set([a.get('code') for a  in bd_ei310bio3])


# In[9]:


# Verify that the number of codes is equal to the number of datasets in the ei 310 biosphere
assert len(ei310_biosphere_codes) == len(bd_ei310bio3)


# In[24]:


ei38_biosphere_codes = set([a.get('code') for a in bd_bio3])


# In[11]:


assert len(ei38_biosphere_codes) == len(bd_bio3)


# In[14]:


in_both = ei310_biosphere_codes.intersection(ei38_biosphere_codes)


# In[15]:


print(f"There are {len(in_both)} in both dbs")


# In[ ]:





# `164` are _à priori_ missing when importing 3.10 with the bwio Ecospold2SingleOutput importer

# In[ ]:


[
    ('biosphere3', 'xxyy'): 1.0,
]


# In[17]:


ef_31_methods = [m for m in bd.methods if m[0] == "EF v3.1"]
ef_31_methods


# In[18]:


ef_31_cc_method_key = ('EF v3.1', 'climate change: fossil', 'global warming potential (GWP100)')


# In[20]:


ef_31_cc_method = bd.Method(ef_31_cc_method_key)
print(ef_31_cc_method)


# In the project in this notebook, we have the CFs pointing to the "last" added database using ecoinvent_interface.
# In a regular project with bw2io-legacy we have the original methods pointing to the biosphere3 database in the CFs.

# In[21]:


cfs_ef_31_cc = ef_31_cc_method.load()
cfs_ef_31_cc


# In[30]:


missing_flow_codes = ei310_biosphere_codes - ei38_biosphere_codes
print(f"There are {len(missing_flow_codes)} present in ei310, not present in ei38 biosphere db.")


# ## Copy the missing flows to biosphere3 db

# In[33]:


missing_flows = [bd_ei310bio3.get(a_code) for a_code in missing_flow_codes]


# In[34]:


missing_flows


# In[35]:


missing_flows[0].as_dict()


# In[38]:


target_db = "biosphere3"


# In[47]:


missing_flows_retargetted = {}
for a_flow in missing_flows:
    a_flow_data = a_flow.as_dict()
    a_flow_data['database'] = target_db
    missing_flows_retargetted[(target_db, a_flow_data['code'])] = {k:v for k,v in a_flow_data.items() if k not in ['database']}


# In[48]:


missing_flows_retargetted


# In[50]:


print(f"retargetted flows to add: {len(missing_flows_retargetted)}")


# In[42]:


from tqdm import tqdm


# In[43]:


original_biosphere3_data = bd_bio3.load()


# In[45]:


original_biosphere3_data


# In[57]:


original_biosphere3_data.update(missing_flows_retargetted)


# In[58]:


print(f"retargetted flows to add: {len(original_biosphere3_data)}")


# In[60]:


# Assert that we have:
394 + 4421


# In[62]:


bd_bio3.write(original_biosphere3_data)