Revision 1621
Added by Aaron Marcuse-Kubitza almost 13 years ago
| lib/csvs.py | ||
|---|---|---|
| 7 | 7 |
|
| 8 | 8 |
delimiters = ',\t`' |
| 9 | 9 |
|
| 10 |
def sniff(line): |
|
| 11 |
'''Automatically detects the dialect''' |
|
| 12 |
dialect = csv.Sniffer().sniff(line, delimiters) |
|
| 13 |
# TSVs usually don't quote fields (nor doublequote embedded quotes) |
|
| 14 |
if dialect.delimiter == '\t': dialect.quoting = csv.QUOTE_NONE |
|
| 15 |
else: dialect.doublequote = True # Sniffer doesn't turn this on by default |
|
| 16 |
return dialect |
|
| 17 |
|
|
| 10 | 18 |
def stream_info(stream): |
| 11 | 19 |
'''Automatically detects the dialect based on the header line |
| 12 | 20 |
@return NamedTuple {header_line, dialect}'''
|
| 13 | 21 |
info = util.NamedTuple() |
| 14 | 22 |
info.header_line = stream.readline() |
| 15 |
sniffer = csv.Sniffer() |
|
| 16 |
info.dialect = sniffer.sniff(info.header_line, delimiters) |
|
| 17 |
info.dialect.doublequote = True # Sniffer doesn't turn this on by default |
|
| 23 |
info.dialect = sniff(info.header_line) |
|
| 18 | 24 |
return info |
| 19 | 25 |
|
| 20 | 26 |
def reader_and_header(stream): |
Also available in: Unified diff
csvs.py: stream_info(): Set dialect.quoting = csv.QUOTE_NONE for TSVs because they usually don't quote fields. Factored dialect detecting code into new function sniff().