def __store(self, item, accumulator):
if isinstance(accumulator, dict):
accumulator.update(item)
elif isinstance(accumulator, list):
accumulator.append(item)
else:
errorout(5)
return accumulator
def __brandnewAccForType(self, schema):
assert isinstance(schema, dict)
_type = self.__getType(schema)
if _type == TypeSign.ARRAY:
return []
elif _type == TypeSign.OBJ:
return {}
else:
errorout(4, _type)
def renew_acc(cls, schema):
Hoare.P(isinstance(schema, dict))
_type = XLSX.__get_type(schema)
if _type == TypeSign.ARRAY:
return []
elif _type == TypeSign.OBJ:
return {}
else:
errorout(4, _type)
def generate_json(self, sheet_name, acc=None):
"""
sheet_nameが指すsheetのJSONをaccに追加する
"""
sheets = self.__name_to_sheets()
# pyxl...Workbookで[sheet名]を持っているが、あまり高速処理向けではないため
sheet_names = list(sheets.keys())
Util.sprint('in process %s'%sheet_name, self.DEBUG)
Hoare.P(sheet_name in sheet_names, '"%s" not found in %s'%(sheet_name ,sheet_names))
root_sheet = sheets[sheet_name]
self.check_charcode(root_sheet)
columns = []
acc = [] if not acc else acc
Util.sprint('I\'ll update {}'.format(acc), self.DEBUG)
# COMBAK: 処理速度に問題が出るようであれば分散処理検討
# A1, B1...で場所を特定するか、indexで回すか
for i, row in enumerate(root_sheet.iter_rows()):
subacc = {}
if self.format:
self.__output_to_csv(self.format_output, root_sheet, self.char_encode)
for j, cell in enumerate(row):
v = cell.value # off-by-oneを気にしないといけなくなるので、col_idxではなくenumerate使う
if v is None: continue # cell check
if i == 0: # 初行 column check
# cell.commentは必ずつくが、中身がない場合はNone
if hasattr(cell, "comment") and cell.comment:
# column 準備 / schemaは遅延せずこの時点で辞書として成立している事を保証
columns.append((v, Util.runtime_dict(cell.comment.text)))
else:
self.errorout(2, 'sheet = {}, col = {}, row = {}'.format(sheet_name, j, i))
else:
# TODO: 関数へ置き換え type = array, objectのケース をカバー
# 別sheet評価
if isinstance(v, str) and v.startswith(XLSX.sheet_link_sign):
# COMBAK: sheetであることがarray, objectの必要条件になってしまっている
# primitive配列をどう表現するかによって改修が必要 __storeに包含させる？
link = v.lstrip(XLSX.sheet_link_sign)
if link in sheet_names:
col_name, col_schema = columns[j]
Util.sprint('process %s -> %s'%(col_name, link), self.DEBUG)
Util.sprint('current acc = %s'%acc, self.DEBUG)
# recursive seed
XLSX.__store(
self.generate_leaf(root_sheet.title, col_name, link, col_schema),
subacc)
else:
errorout(1, 'sheet = from %s to %s, col = %d, row = %d'%(sheet_name, link, j, i))
else:
XLSX.__store(self.type_validator(sheet_name, v, columns[j]), subacc)
# pass columns
Util.check_emptyOR(lambda x: XLSX.__store(x, acc), subacc)
# pass a row
return acc
def __store(cls, item, accumulator):
"""
評価済みが保証された値を、rootのleafに連結
accumlator: either dict or list
"""
if isinstance(accumulator, dict):
accumulator.update(item)
elif isinstance(accumulator, list):
accumulator.append(item)
else:
errorout(5)
return accumulator
def __run__(self, **kwargs):
args = kwargs['args']
self.args = args
fileloc = os.path.abspath(os.path.expanduser(args.input))
workpath = self.__treatFileTypes(fileloc)
from xlsx import XLSX
self.__print(f'Analyzing... {fileloc}')
x = XLSX(workpath, args.encoding, args.output_format)
# sys.setrecursionlimit(1024 * 8)
j = x.generateJSON(sheet_name=args.root_sheet)
with wild_open(args.output, encoding=args.encoding) as f:
try:
print(json.dumps(j, sort_keys=True, indent=args.human_readable) \
if args.human_readable > 0 else json.dumps(j), file=f)
except:
errorout(6, args.output)
else:
self.__print(f'Output json Success ➡️ {args.output}')
pass
def __run__(self, **kwargs):
args = kwargs['args']
from settings import SettingProcessor, SettingsError
# 初期化モード
setting_file = r'%s.yaml' % os.path.splitext(
os.path.expanduser(args.template_xlsx))[0]
if os.path.exists(setting_file):
print(r'Setting file found on %s' % setting_file)
else:
errorout(8, r'Please make sure [%s] location?' % setting_file)
settings = SettingProcessor(setting_file, args.template_xlsx,
args.encoding)
settings.check_settingfile()
try:
settings.create_sheets()
except SettingsError as se:
print(se)
else:
settings.save()
print('Construct xlsx file Success ➡️ %s' % args.template_xlsx)
def __run__(self, **kwargs):
args = kwargs['args']
from settings import SettingProcessor, SettingsError
# 初期化モード
setting_file = fr'{os.path.splitext(os.path.expanduser(args.template_xlsx))[0]}.yaml'
if os.path.exists(setting_file):
print(fr'Setting file found on {setting_file}')
else:
errorout(8, fr'Please make sure [{setting_file}] location?')
settings = SettingProcessor(setting_file, args.template_xlsx,
args.encoding)
settings.checkSettingFile()
try:
settings.createSheets()
except SettingsError as se:
print(se)
else:
settings.save()
print(f'Construct xlsx file Success ➡️ {args.template_xlsx}')
pass
def __run__(self, **kwargs):
args = kwargs['args']
# default option 対策 seelaso AnalyzeXSeparatedOutPath
# https://github.com/setminami/Jsonica/issues/47
refactor_check(args.__class__.__name__ == 'Namespace')
if not args.output_format:
args.output_format = ('tsv', output_delimiters[1], './output/')
self.args = args
fileloc = os.path.abspath(os.path.expanduser(args.input))
workpath = Generate.__treatFileTypes(fileloc)
from xlsx import XLSX
self._print('Analyzing... %s' % fileloc)
x = XLSX(workpath, args.encoding, args.output_format)
# sys.setrecursionlimit(1024 * 8)
j = x.generate_json(sheet_name=args.root_sheet)
with wild_open(args.output, encoding=args.encoding) as f:
try:
print(json.dumps(j, sort_keys=True, indent=args.human_readable) \
if args.human_readable > 0 else json.dumps(j), file=f)
except:
errorout(6, args.output)
else:
self._print('Output json Success ➡️ %s' % args.output)
Util.sprint('XXX %s XXX' % x.piled_schema, self.DEBUG)
def __treatFileTypes(self, file):
if file.endswith('.xlsx'):
return file
else:
errorout(7, f'{file} format is not supported yet.')
def __treatFileTypes(cls, file):
if file.endswith('.xlsx'):
return file
else:
errorout(7, '%s format is not supported yet.' % file)