def read_file(filename):
with codecs.open(filename, mode='r') as file:
reader = csv.reader(file)
first_line = reader.next()
# maintain stack of taxa that are parents of the current taxon
stack = []
# name of parent of root taxon should be in cell A1
root_name = first_line[0]
if root_name != '':
root_parent = Taxon.filter(Taxon.valid_name == root_name)[0]
stack.append(root_parent)
# current valid taxon (for synonyms)
current_valid = None
# whether current taxon should be marked as root of a page
is_page_root = True
error_occurred = False
for row in reader:
try:
# ignore blank rows
if row[14] == '' and row[8] == '':
continue
data = parse_row(row)
# deal with "nHT", which is a pain. Some of these may need to be manually
# readded to the DB
if data['kind'] == 'nHT':
if data['valid_name'] in ignored_nHT:
continue
else:
raise Exception("Unrecognized nHT: " + str(data))
# nsgen is i.s., just ignore
if data['kind'] == 'nsgen':
continue
if data['status'] == STATUS_VALID:
# get stuff off the stack
rank = data['rank']
if rank == ROOT:
current_valid = Taxon.create(
valid_name=data['valid_name'], age=data['age'],
rank=data['rank'], is_page_root=True
)
else:
# TODO: make this somehow unranked-clade-aware
while rank >= stack[-1].rank:
stack.pop()
# create new Taxon
current_valid = Taxon.create(
valid_name=data['valid_name'], age=data['age'],
rank=data['rank'], parent=stack[-1], is_page_root=is_page_root
)
if is_page_root:
is_page_root = False
stack.append(current_valid)
# create new Name
data['taxon'] = current_valid
if data['status'] == STATUS_DUBIOUS:
# current system is inadequate for properly marking si species
# assume there's only genera and species
if ' ' in data['valid_name']:
data['group'] = GROUP_SPECIES
else:
data['group'] = GROUP_GENUS
# si species don't have a meaningful "valid name", but preserve what's there now
data['data']['si_valid_name'] = data['valid_name']
else:
# this will be wrong in the few cases where a high-ranked name is listed
# as a synonym of a family-group name. Don't see a way to correct that
# programmatically.
data['group'] = helpers.group_of_rank(current_valid.rank)
if data['status'] == STATUS_SYNONYM:
if data['kind'] == 'synHT':
valid_name = data['valid_name'].split(' ', 1)[1]
else:
valid_name = data['valid_name']
if valid_name != current_valid.valid_name:
raise Exception("Valid name of synonym does not match: " + data['valid_name'] + " and " + current_valid.valid_name)
# shorten root name for family-group names
if data['group'] == GROUP_FAMILY:
data['root_name'] = helpers.strip_rank(data['root_name'], current_valid.rank)
del data['kind']
data['data'] = json.dumps(remove_null(data['data']))
# Detect whether a name object is already present (Principle of Coordination)
nm = None
if data['status'] == STATUS_VALID:
root_name = data['root_name']
if current_valid.rank == FAMILY:
nm = detect_super(root_name + 'oidea', SUPERFAMILY)
elif current_valid.rank == SUBFAMILY:
nm = detect_super(root_name + 'idae', FAMILY)
elif current_valid.rank == TRIBE:
nm = detect_super(root_name + 'inae', SUBFAMILY)
elif current_valid.rank == SUBTRIBE:
nm = detect_super(root_name + 'ini', TRIBE)
elif current_valid.rank == SUBGENUS:
nm = detect_super(root_name, GENUS)
elif current_valid.rank == SPECIES:
spg_name = helpers.spg_of_species(current_valid.valid_name)
nm = detect_super(spg_name, SPECIES_GROUP)
elif current_valid.rank == SUBSPECIES and helpers.is_nominate_subspecies(current_valid.valid_name):
sp_name = helpers.species_of_subspecies(current_valid.valid_name)
nm = detect_super(sp_name, SPECIES)
if nm is not None:
del data['taxon']
nm.add_additional_data(data)
# nm's Taxon should be the lowest-ranking one
nm.taxon = current_valid
# create a new Name if none was found
if nm is None:
nm = Name.create(**data)
# set base_name field
if data['status'] == STATUS_VALID:
current_valid.base_name = nm
if 'additional_synonyms' in data:
group = helpers.group_of_rank(current_valid.rank)
for synonym in data['additional_synonyms']:
Name.create(taxon=current_valid, root_name=synonym, group=group, status=STATUS_SYNONYM)
except Exception:
traceback.print_exc()
error_occurred = True
# ignore error and happily go on with the next
return not error_occurred
def create_root(): Taxon.create(rank=ROOT, valid_name='root', is_page_root=True)
