def parse_header(self, rows):
header_row = [self.extract_cell_value(cell) for cell in rows[self.header_row_index]]
field_row = [self.extract_cell_value(cell) for cell in rows[self.field_row_index]]
type_row = [self.extract_cell_value(cell) for cell in rows[self.type_row_index]]
for col, field in enumerate(field_row):
if field == "": break
self.converters[col] = None
if field in self.field_2_col:
util.log_error("列名'%s'重复，列：%s", field, util.int_to_base26(col))
continue
self.field_2_col[field] = col
header = header_row[col] or field
type = type_row[col] or "String"
method = None
try:
method = FAST_CONVERTER[type.lower()]
except:
util.log_error("无效的类型'%s'，列：%s", type, util.int_to_base26(col))
continue
self.converters[col] = ConverterInfo((header, field, method, True))
self.sheet_types[field] = (col, field, header, type)
self.key_name = self.converters[0].field
return
def parse_arguments(self, rows):
row_index = self.argument_row_index
arg_row = [self.extract_cell_value(cell) for cell in rows[row_index]]
self.arguments = {}
for col in xrange(0, len(arg_row), 2):
header = arg_row[col]
if header is None: break
converter = xlsconfig.ARGUMENT_CONVERTER.get(header)
if converter is None: continue
field, type = converter
method = FAST_CONVERTER[type.lower()]
value = arg_row[col + 1]
ret = None
try:
ret = method(value)
except:
traceback.print_exc()
log_error("参数转换失败，(%d, %s) = [%s]", row_index + 1,
util.int_to_base26(col), value)
self.arguments[field] = ret
return
def parse_header(self, rows):
header_row = [
self.extract_cell_value(cell)
for cell in rows[self.header_row_index]
]
name_set = set()
for col, header in enumerate(header_row):
if header == "": break
converter = None
cfg = self.header_2_config.get(header)
if cfg is None:
print "警告：第(%s)列的表头'%s'没有被解析。%s" % (
util.int_to_base26(col),
header,
self.filename,
)
else:
converter = ConverterInfo(cfg)
field = converter.field
self.field_2_col[field] = col
type = tp0.type2string(converter.convert)
self.sheet_types[field] = (col, field, header, type)
self.converters[col] = converter
return
def write_types_comment(self, sheet_name):
if not self.data_module: return
module_info = self.data_module.info
sheet_types = module_info["sheet_types"].get(sheet_name)
if sheet_types is None: return
sheet_types = sheet_types.values()
sheet_types.sort(key=lambda v: v[0])
for info in sheet_types:
col, field, text, type = info
col_name = util.int_to_base26(col) if col is not None else "None"
comment = "%s\t%-20s%s" % (col_name, field, text)
self.write_comment(comment)
def do_parse(self):
import openpyxl
self.workbook = openpyxl.load_workbook(self.filename)
sheets = self.workbook.worksheets
if self.sheet_index >= len(sheets):
log_error("Excel表'%s'没有子表'%d'", self.filename, self.sheet_index)
return
table = sheets[self.sheet_index]
self.worksheet = table
rows = list(table.rows)
self.parse_arguments(rows)
self.parse_header(rows)
self.parse_defaults(rows)
if self.data_row_index >= len(rows):
return
ncols = len(self.converters)
# the remain rows is raw data.
for r in xrange(self.data_row_index, len(rows)):
cells = rows[r]
# 遇到空白行，表示解析完成
first_value = cells[0].value
if first_value == '' or first_value is None: break
current_row_data = {}
for c in xrange(ncols):
value = self.extract_cell_value(cells[c])
try:
self.convert_cell(r, c, value, current_row_data)
except:
traceback.print_exc()
util.log_error("单元格(%d, %s) = [%s] 数据解析失败", r + 1,
util.int_to_base26(c), value)
self.add_row(current_row_data)
return
