One-liner clear solution for Remove Accents by suic - python coding challenges

Enable Javascript in your browser and then refresh this page, for a much enhanced experience.

One-liner solution in Clear category for Remove Accents by suic

checkio = lambda i: __import__("unicodedata").normalize('NFKD', i).encode('ASCII', 'ignore').decode() or i

Feb. 11, 2015

Comments:

veky on Feb. 22, 2015, 6:59 a.m.
Nice, but why "or i"?

suic on Feb. 22, 2015, 9:41 a.m.
Hi, because of Chinese characters: <ol> <li>normalize('NFKD', i) returns the Chinese characters without changes</li> <li>so the i.encode('ASCII', 'ignore').decode() returns an empty string.</li> <li>As the Chinese characters remain unchanged (1.) I return them instead of empty string.</li> </ol> That's the whole story :)

veky on Feb. 22, 2015, 9:48 a.m.
Ah. :-( So it won't work if there are Chinese and accented characters in the string?

suic on Feb. 22, 2015, 10:24 a.m.
I'm not sure if there are Chinese and accented characters. (I've never seen ogonek, makcen or umlaut in Chinese. :)) Apparently there's no such test case as all the existing "Chinese test cases" can be passed this way. The test suite is fairly incomplete e. g. there are no testcases for Greek, Russian, Armenian or Georgian etc. alphabet or Sanskrit.

veky on Feb. 22, 2015, 12:37 p.m.
Yes, I know about the state of tests here. :-P I'm speaking idealistically. I didn't mean accented Chinese characters. I meant a string that in one part contains Chinese characters, and in another part contains accented characters. This won't pass, if I see right.

suic on Feb. 22, 2015, 1:56 p.m.
You're right it does not work for strings containing accented and Chinese characters. I tried to do "per character" normalization but it caused more problems that it solved. There's is an issue with Python 3.3 or checkio. <pre class='brush: python'># I tried: list("loài trăn lớn") # checkio prompt: ['l', 'o', 'a', '̀', 'i', ' ', 't', 'r', 'ă', 'n', ' ', 'l', 'ơ', '́', 'n'] # Python 3.4 on my machine: ['l', 'o', 'à', 'i', ' ', 't', 'r', 'ă', 'n', ' ', 'l', 'ớ', 'n'] </pre>

veky on Feb. 22, 2015, 6:51 p.m.
Of course. So the real solution lies in the other direction, unfortunately.

robertabegg on Sept. 16, 2021, 10:12 p.m.
IMHO, although it works for the included tests, a solution should work in all expected cases.

Shelter

0 %

You ClassRoom Demo is Ready With the new tool CheckiO ClassRoom, you will be able to analyze your students' progress, have your own leaderboard, change missions order on the map, and so on.