brightway-lca / bw_hestia_bridge Goto Github PK

The biosphere flows need to be mapped to ecoinvent as well.

_hestia_api/
    __init__.py
    query.py
        search_hestia(element: str)
        get_hestia_node(node_id: str)
    credentials.py
        login_to_hestia(email: str, pwd: str)
        set_token(token: str)
utils/
    __init__.py
    version.py
    appdirs.py

appdirs (MIT) is to get folders to store the token and cache requests

I'd propose to load everything from _hestia_api into the main __init__.py.

Currently, linking strategy for "znex7uqxsfqn" fails with

Applying strategy: link_iterable_by_fields
Couldn't apply strategy link_iterable_by_fields:
	Object in source database can't be uniquely linked to target database.
Problematic dataset is:
{'cycle_id': 'znex7uqxsfqn',
 'filename': '(missing)',
 'term_id': 'excretaBeefCattleExceptFeedlotFedKgN',
 'transformation_id': '3'}
Possible targets include (at least one not shown):
[{'cycle_id': 'znex7uqxsfqn',
  'filename': '(missing)',
  'term_id': 'excretaBeefCattleExceptFeedlotFedKgN',
  'transformation_id': '3'},
 {'cycle_id': 'znex7uqxsfqn',
  'filename': '(missing)',
  'term_id': 'excretaBeefCattleExceptFeedlotFedKgN',
  'transformation_id': '3'},
 {'cycle_id': 'znex7uqxsfqn',
  'filename': '(missing)',
  'term_id': 'excretaBeefCattleExceptFeedlotFedKgN',
  'transformation_id': '3'}]

This cycle: https://www-staging.hestia.earth/cycle/s_iwp86ylxhd?dataState=recalculated#impactDriverChart

Consumes one of the outputs of this cycle: https://www-staging.hestia.earth/cycle/ribhbymfmmt3?dataState=recalculated#impactDriverChart

(though one is 2013, one is 2014, I am not sure if there could be another cycle which is more cleanly linked...)

It would be great, when turning the cycle into a process that Brightway can handle, to be able to find these kinds of links by calling the API.

We're never going to upload and this is the only thing that requires a token so I'll remove the credentials unless someone has a good reason to keep them.

• Add exchanges
• Add transport exchanges
• Add emissions (except for emissions from background)
• Add production exchanges
• Add products for production exchanges
• Make sure documentation exists for the conversions
• Add tests for conversions
• Add dummy waste treatment processes if not otherwise provided

https://pypi.org/project/hestia-api/

• Decide if we use this
• Document missing functionality, if any

So far, we only have a preliminary version for the milestone. There remain some gaps that need to be filled for Hestia flows that don't have an ecoivent equivalent.

#10

is a good starting point.

While

bhb.search_hestia({"products.term.name": "Saplings", "@type": "cycle"})

yields perfect results,

bhb.search_hestia({"products.term.name": "Soybean, seed (whole)", "@type": "impactassessment"})

yields only an emtpy list. But there is actually at least one Impact Assessment who has "Soybean, seed (whole)" as a product name. See:
https://www-staging.hestia.earth/impactassessment/bv2ktx0jjkub?dataState=recalculated

To avoid having to search for the API URL, it will be included in the source and switch between stable and staging will just be done via:

set_config("use_staging", True/False)
save_config()  # optional, to remember the config from one run to another

In the Hestia ontology, a Transformation is: "The Transformation of one Product to a new Product in the same Cycle."

This is complicated because transformations can branch (product A goes down one branch, product B down the other), and there can be allocation. Also, the branches can have multiple steps, so in theory one cycle can produce many products, all linked together in an arborescence.

This should be well tested...

When not setting use_staging to True the following occurs:

>>> import bw_hestia_bridge as bhb
>>> bhb.get_hestia_node("zcuqxv54gn2x")
Traceback (most recent call last):
  File "miniconda3\envs\hestia\lib\site-packages\IPython\core\interactiveshell.py", line 3442, in run_code
    exec(code_obj, self.user_global_ns, self.user_ns)
  File "<ipython-input-16-a0f08a0feebb>", line 1, in <module>
    bhb.get_hestia_node("zcuqxv54gn2x")
  File "hestia\bw_hestia_bridge\bw_hestia_bridge\hestia_api\querying.py", line 154, in get_hestia_node
    node_type = node_type or get_node_type(node_id)
  File "hestia\bw_hestia_bridge\bw_hestia_bridge\hestia_api\querying.py", line 169, in get_node_type
    return search_hestia({"@id": node_id}, how="exact")[0]["@type"].lower()
IndexError: list index out of range

>>> bhb.get_cycle_graph("zcuqxv54gn2x")
Traceback (most recent call last):
  File "miniconda3\envs\hestia\lib\site-packages\IPython\core\interactiveshell.py", line 3442, in run_code
    exec(code_obj, self.user_global_ns, self.user_ns)
  File "<ipython-input-17-cbfabe4f7c90>", line 1, in <module>
    bhb.get_cycle_graph("zcuqxv54gn2x")
  File "hestia\bw_hestia_bridge\bw_hestia_bridge\hestia_api\cycle_graph.py", line 48, in get_cycle_graph
    eid = elt["@id"]
TypeError: string indices must be integers

The node with id "zcuqxv54gn2x" only exists on the staging server. It would be helpful to get somehow the information as a return of the called function that the node is not present via the implicitly or explicitly chosen API.

Note: this is not urgent at all but I think we need to make a decision at some point.

At the moment, having the staging argument in the HestiaImporter together with the config is problematic as we have a class argument that permanently overrides the library's state (if you forget to set it, and were on staging, you're back on stable).

In my mind, if you're going to use staging, you can just use set_config("use_staging", True) at the beginning of your script.
You can also use save_config() to make this change permanent and not have to set it anymore or use a "use_staging" environment variable. This is documented here.
I do not believe that people are going to switch between stable and staging dynamically within a script.
If you think this is the assumption we should be working with, then probably we need to remove the config altogether.

As we've mostly been working on staging and this might be the case for quite some time, the config avoid having to provide a staging=True argument to all functions, which I personally appreciate, together with the fact that you only need one command to switch your whole script from one to the other, but maybe it's not what is actually needed, especially in the long run.

I just think we need to decide once and for all what to implement because the current mixed situation is not satisfactory.

Enable get_hestia_node to work with both a dict (returned from the search), or an id and a type argument.

Would be good to set up tests for the various functions.

This issue lists Renovate updates and detected dependencies. Read the Dependency Dashboard docs to learn more.

This repository currently has no open or pending branches.

Detected dependencies

• Check this box to trigger a request for Renovate to run again on this repository

People in the Hackathon added the EcoInvent ids in the Hestia database so we should be able to use this directly to map to EcoInvent.

• to find for a specific cycle all related hestia cycles ("activities")
• in order to reach just dependencies with emissions ("elementary flows")