Separate numbers from text
21 views (last 30 days)
Show older comments
Deepa Maheshvare
on 9 Sep 2019
Commented: Walter Roberson
on 22 Feb 2020
I have the follwing character array
C = {'LAN310'}
{'SHA550'}
Expercted result:
C = [ 310 550]
From one of the answers posted in the previous posts, I tried the following
D = regexprep(C,'[\d"]','');
The above replaces the numbers with null, but I would like to replace characters with null. Any suggestions?
0 Comments
Accepted Answer
Walter Roberson
on 9 Sep 2019
Edited: Walter Roberson
on 9 Sep 2019
D = str2double( regexrep(C, '\D', '') );
Note that this will replace all + - . and e E d D used as exponents, and that if there are multiple numbers separated by non-numbers then this will jam them together.
Another approach is
D = str2double( C(ismember(C, '0':'9')) );
0 Comments
More Answers (1)
madhan ravi
on 9 Sep 2019
Edited: madhan ravi
on 9 Sep 2019
C = str2double(regexprep(C,'[a-zA-Z^()]','')).'
1 Comment
Walter Roberson
on 22 Feb 2020
There is a maxim in computer reliability and security engineering:
"Do not validate input by rejecting characters or patterns: only ever validate input by accepting characters or patterns that you know you can deal with."
When you work by rejecting things, then there is the possibility that you forgot to reject something, or that later some new method will be introduced that permits the user to slip new values past your validation. For example the pattern [a-zA-Z^()] does not reject & or ; or # or % so it would be happy to pass through A or %41 and a later step might do the translation to A or A .
There was a whole spate of security problems when unicode encoding was added to the HTML standard: code that validated by rejecting the old escape sequences stopped working when the libraries added a new escape sequence.
See Also
Categories
Find more on Characters and Strings in Help Center and File Exchange
Products
Community Treasure Hunt
Find the treasures in MATLAB Central and discover how the community can help you!
Start Hunting!