def analyseFile(self, fileName, encoding=None, dialect_suggestion=None, limit=None):
if encoding is None:
encoding = "utf8"
if self.DEBUG_PARSER:
self.registerMessage("Analysing file: {0}, encoding: {1}".format(fileName, encoding))
with open(fileName, 'rbU') as byte_file_obj:
# I can't imagine this having any problems
byte_sample = byte_file_obj.read(1000)
byte_file_obj.seek(0)
if dialect_suggestion:
csvdialect = UnicodeCsvDialectUtils.get_dialect_from_suggestion(dialect_suggestion)
else:
try:
csvdialect = unicodecsv.Sniffer().sniff(byte_sample)
assert csvdialect.delimiter ==',', "sanity test"
except Exception:
csvdialect = UnicodeCsvDialectUtils.default_dialect
if self.DEBUG_PARSER:
self.registerMessage(UnicodeCsvDialectUtils.dialect_to_str(csvdialect))
unicodecsvreader = unicodecsv.reader(byte_file_obj, dialect=csvdialect, encoding=encoding, strict=True)
return self.analyseRows(unicodecsvreader, fileName)
return None
def exportItems(self, filePath, colNames, dialect = None, encoding="utf8"):
assert filePath, "needs a filepath"
assert colNames, "needs colNames"
assert self.objects, "meeds items"
with open(filePath, 'w+') as outFile:
if dialect is None:
csvdialect = UnicodeCsvDialectUtils.act_out
else:
csvdialect = UnicodeCsvDialectUtils.get_dialect_from_suggestion(dialect)
# unicodecsv.register_dialect('act_out', delimiter=',', quoting=unicodecsv.QUOTE_ALL, doublequote=False, strict=True, quotechar="\"", escapechar="`")
if self.DEBUG_ABSTRACT:
self.registerMessage(UnicodeCsvDialectUtils.dialect_to_str(csvdialect))
dictwriter = unicodecsv.DictWriter(
outFile,
dialect=csvdialect,
fieldnames=colNames.keys(),
encoding=encoding,
extrasaction='ignore',
)
dictwriter.writerow(colNames)
dictwriter.writerows(self.objects)
self.registerMessage("WROTE FILE: %s" % filePath)