| 1234567891011121314151617181920212223242526272829303132333435363738394041424344454647484950515253545556575859606162 |
- import csv
- import typing
- from os import path
- from pycs.interfaces.LabelProvider import LabelProvider
- from .RowWrapper import RowWrapper
- class Provider(LabelProvider):
- def __init__(self, root_folder, configuration):
- self.csv_path = path.join(root_folder, configuration['filename'])
- self.csv_minimum_rarity = configuration['minimumRarity']
- self.csv_all_hierarchy_levels = configuration['includeAllHierarchyLevels']
- def close(self):
- pass
- def get_labels(self) -> typing.List[dict]:
- result = []
- with open(self.csv_path, mode='r', newline='', encoding='utf8') as csv_file:
- # skip first line which contains column names
- csv_file.readline()
- # read csv line by line
- reader = csv.reader(csv_file, delimiter='\t')
- entries = list(map(RowWrapper, reader))
- # filter
- if self.csv_minimum_rarity is not None:
- entries = filter(lambda row: row.rarity_is_larger_than(self.csv_minimum_rarity),
- entries)
- # create result set
- for entry in entries:
- entry = entry.__dict__
- parent_reference = None
- # add hierarchy
- if self.csv_all_hierarchy_levels:
- hierarchy_levels = ('superfamily', 'family', 'subfamily', 'tribe', 'genus')
- else:
- hierarchy_levels = ('family', 'genus')
- for tax in hierarchy_levels:
- if entry[tax] is not None:
- reference, name = entry[tax].lower(), entry[tax]
- result.append(self.create_label(reference, name, parent_reference))
- parent_reference = reference
- # add element
- if entry['kr_number'].isnumeric():
- name = f'{entry["genus"]} {entry["species"]} ({entry["kr_number"]})'
- reference = entry['kr_number']
- else:
- name = f'{entry["genus"]} {entry["species"]}'
- reference = name.lower()
- result.append(self.create_label(reference, name, parent_reference))
- return result
|
