async def query_root_taxa() \
-> List[TaxonBO]:
"""
Return all taxa with no parent.
"""
sce = TaxonomyService()
ret = sce.query_roots()
return ret
async def taxa_tree_status(current_user: int = Depends(get_current_user)):
"""
Return the status of taxonomy tree w/r to freshness.
"""
sce = TaxonomyService()
refresh_date = sce.status(_current_user_id=current_user)
return TaxonomyTreeStatus(
last_refresh=refresh_date.isoformat() if refresh_date else None)
async def query_taxa(taxon_id: int,
_current_user: Optional[int] = Depends(get_optional_current_user)) \
-> Optional[TaxonBO]:
"""
Information about a single taxon, including its lineage.
"""
sce = TaxonomyService()
ret = sce.query(taxon_id)
return ret
async def query_taxa_in_worms(aphia_id: int,
_current_user: Optional[int] = Depends(get_optional_current_user)) \
-> Optional[TaxonBO]:
"""
Information about a single taxon in WoRMS reference, including its lineage.
"""
sce = TaxonomyService()
ret = sce.query_worms(aphia_id)
return ret
async def query_taxa_set(ids: str,
_current_user: Optional[int] = Depends(get_optional_current_user)) \
-> List[TaxonBO]:
"""
Information about several taxa, including their lineage.
The separator between numbers is arbitrary non-digit, e.g. ":", "|" or ","
"""
sce = TaxonomyService()
num_ids = _split_num_list(ids)
ret = sce.query_set(num_ids)
return ret
async def search_taxa(
query: str,
project_id: Optional[int],
current_user: Optional[int] = Depends(get_optional_current_user)):
"""
Search for taxa by name.
Queries can be 'small', i.e. of length < 3 and even zero-length.
For a public, unauthenticated call:
- zero-length and small queries always return nothing.
- otherwise, a full search is done and results are returned in alphabetical order.
Behavior for an authenticated call:
- zero-length queries: return the MRU list in full.
- small queries: the MRU list is searched, so that taxa in the recent list are returned, if matching.
- otherwise, a full search is done. Results are ordered so that taxa in the project list are in first,
and are signalled as such in the response.
"""
sce = TaxonomyService()
ret = sce.search(current_user_id=current_user,
prj_id=project_id,
query=query)
return ret
def do_match(
self
) -> Tuple[Dict[ClassifIDT, WoRMS], Dict[ClassifIDT, ClassifIDT]]:
"""
Returns: A dict with matched Phylo taxo ids->WoRMS entry. The taxa might be Morpho ones,
as the XLSX data source contains such mappings.
+ Another dict with unmatched Morpho entries.
:return:
"""
ret: Dict[ClassifIDT, WoRMS] = {}
# Do the manual (i.e. XLSX-driven) matching of what can be.
# In this part, both Morpho and Phylo taxa are matched to WoRMS.
to_worms: ToWorms = ToWorms()
to_worms.prepare()
to_worms.apply()
manual_ids = set(to_worms.done_remaps.keys()).intersection(
self.taxa_ids)
for a_manual_id in list(manual_ids):
# Get the mapping
aphia_id = to_worms.done_remaps[a_manual_id]
if aphia_id is None:
# No mapping -> re-add to auto matching to come, who knows...
manual_ids.remove(a_manual_id)
continue
worms_rec = self.session.query(WoRMS).get(aphia_id)
assert worms_rec is not None
ret[a_manual_id] = worms_rec
# Do auto matching of the rest
ids_for_auto_match = self.taxa_ids.difference(manual_ids)
# Only keep Phylo taxa, the Morpho ones will get aggregated with their nearest Phylo parent
phylo_auto_ids = TaxonomyBO.keep_phylo(self.session,
list(ids_for_auto_match))
# Lookup parents of morpho taxa
with TaxonomyService() as taxo_sce:
morpho_ids = list(ids_for_auto_match.difference(phylo_auto_ids))
# Borrow taxo tree from ToWorms which has nearly all we need.
to_worms.add_to_unieuk(morpho_ids, taxo_sce)
unieuk_per_id = to_worms.unieuk
needed_parents = []
for a_morpho_id in morpho_ids:
for a_parent_id in unieuk_per_id[a_morpho_id].id_lineage:
if a_parent_id not in unieuk_per_id:
needed_parents.append(a_parent_id)
to_worms.add_to_unieuk(needed_parents, taxo_sce)
# Build morpho -> nearest phylo mapping
phylo_per_morpho: Dict[ClassifIDT, ClassifIDT] = {}
for a_morpho_id in morpho_ids:
for a_parent_id in unieuk_per_id[a_morpho_id].id_lineage:
if unieuk_per_id[a_parent_id].type == 'P':
phylo_per_morpho[a_morpho_id] = a_parent_id
if a_parent_id not in ret:
phylo_auto_ids.add(a_parent_id)
break
# Do more matches
ret.update(WoRMSSetFromTaxaSet(self.session, list(phylo_auto_ids)).res)
# Sanity check
# for an_id in ret.keys():
# assert unieuk_per_id[an_id].type == 'P' # Not true
for a_morpho_id, a_phylo_id in phylo_per_morpho.items():
assert unieuk_per_id[a_morpho_id].type == 'M'
assert unieuk_per_id[a_phylo_id].type == 'P'
return ret, phylo_per_morpho