Last active
August 29, 2015 14:23
-
-
Save Turin86/e67f2f491383b30fa0dd to your computer and use it in GitHub Desktop.
Utility function to properly capitalize names
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
def capnames(line): | |
if isinstance(line, basestring): | |
was_string = isinstance(line, str) | |
if was_string: | |
line = line.decode('utf-8') | |
line = line.lower() \ | |
.replace(u'otoole', u'o\'toole') \ | |
.replace(u'\\', u'') | |
line = u'-'.join(_.strip() for _ in line.split(u'-')) | |
line = u"'".join(_.strip() for _ in line.split(u"'")) | |
line = u'ª '.join(_.strip() for _ in line.split(u'ª')) | |
line = u'º '.join(_.strip() for _ in line.split(u'º')) | |
line = u', '.join(_.strip() for _ in line.split(u',')) | |
line = u'. '.join(_.strip() for _ in line.split(u'.')) | |
names = line.split() | |
for i in range(len(names)): | |
if names[i] not in (u'de', u'del', u'el', u'la', u'las', | |
u'los', u'do', u'von', u'van', u'der', u'ben', u'ibn', u'ebn', | |
u'bin', u'of', u'y', u'i', u'eta', u'and', u'und', u'du', u'dos', | |
u'les', u'da', u'den', u'ten', u'zu', u'z', u'af', u'av', u'ver', | |
u'op', u'della', u'di', u'lo', u'zum', u'vor', u'dem', u'auf', | |
u'als', u'sich', u'ein', u'mit', u'degli', u'zur'): | |
for special_char in (u'.', u',', u'-', u"'"): | |
name_parts = names[i].split(special_char) | |
for j in range(len(name_parts)): | |
if len(name_parts[j]) > 0: | |
name_parts[j] = name_parts[j][0].upper() + name_parts[j][1:] | |
for prefix in (u'Mc',): | |
if name_parts[j].startswith(prefix): | |
name_parts[j] = prefix + name_parts[j][len(prefix)].upper() + name_parts[j][(len(prefix)+1):] | |
names[i] = special_char.join(name_parts) | |
line = u' '.join(names) | |
if was_string: | |
line = line.encode('utf-8') | |
return line |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment