Revision 5167
Added by Aaron Marcuse-Kubitza about 12 years ago
| lib/tnrs.py | ||
|---|---|---|
| 66 | 66 |
|
| 67 | 67 |
padding = ' !pad ' # prepend to empty and whitespace-only strings |
| 68 | 68 |
encode_map = [ |
| 69 |
('!', ' !exc '), # escape char
|
|
| 70 |
('\t', ' !tab '),
|
|
| 71 |
('\n', ' !nl '),
|
|
| 72 |
('\r', ' !cr '),
|
|
| 73 |
('%', ' !pct '),
|
|
| 74 |
(';', ' !sem '),
|
|
| 75 |
('\\', ' !bsl '),
|
|
| 69 |
('!', ' !exc '), # our escape char
|
|
| 70 |
('\t', ' !tab '), # TNRS replaces with " "
|
|
| 71 |
('\n', ' !nl '), # used to separate multiple names
|
|
| 72 |
('\r', ' !cr '), # used to separate multiple names
|
|
| 73 |
('%', ' !pct '), # TNRS URL-decodes it in matched fields
|
|
| 74 |
(';', ' !sem '), # changes TNRS response format
|
|
| 75 |
('\\', ' !bsl '), # TNRS removes it
|
|
| 76 | 76 |
('_', ' !und '), # TNRS replaces with " "
|
| 77 | 77 |
] |
| 78 | 78 |
decode_map = strings.flip_map(encode_map) |
Also available in: Unified diff
tnrs.py: encode_map: Documented why each character needs to be encoded