Numeric string validation problem

Question

1.00/5 (1 vote)

See more:

The following is a description of the problem : Validate if a given string is numeric.

Some examples:
"0" => true
" 0.1 " => true
"abc" => false
"1 a" => false
"2e10" => true

The algorithm I have so far is the following:

C#

public static boolean isNumber(String s) {

        if(s == null){
            return false;
        }
        s = s.toLowerCase();
        s = s.trim();
        if(s == "e" || s == "." || s == "/"){
            return false;
        }

       char[] myObject = new char[]{'0','1','2','3','4','5','6','7','8','9','.','/','e'};
        char[] myChar = s.toCharArray();

        for (int i = 0; i < myChar.length; i++) {

                int temp = 0;
               while(myChar[i] != myObject[temp])
               {
                   temp++;
                   if(temp == myObject.length){
                       return false;
                   }
               }



       }

        return true;
       }

When I used "e" as an input in my local machine it returns false which is correct; however, when I test the same code in LeetCode website it returns true. I can not figure out why this is happening. Can anyone help me please??? Here is a copy of leetcode's website

VB

Input:  "e"
Output: true
Expected:   false

Posted 28-Oct-14 7:25am

Alexander24

Updated 28-Oct-14 11:49am

Sergey Alexandrovich Kryukov

v2

Add a Solution

Comments

Richard MacCutchan 28-Oct-14 13:36pm

Go and ask the people at LeetCode.

Alexander24 28-Oct-14 13:37pm

I already asked them,is really hard to get an email back :(

Richard MacCutchan 28-Oct-14 14:10pm

The point is, how do you expect members of CodeProject to know why some other website is not correct?

[no name] 28-Oct-14 15:25pm

"e" allone will be acpted? Maybe they accept "e" as the base of natural logarithm?

Sergey Alexandrovich Kryukov 28-Oct-14 17:50pm

I renamed the title of your question, to attract more experts to it. I did not look for your post for a long time, because your title scared me off. Avoid using unrelated words in the titles of your question, otherwise many readers will never see your question page.
—SA

1 solution

Add a Solution

Add your solution here

Treat my content as plain text, not as HTML

Preview 0

…

Existing Members

Sign in to your account

...or Join us

Download, Vote, Comment, Publish.

Your Email
Password
Forgot your password?

Your Email
This email is in use. Do you need your password?
Optional Password

I have read and agree to the Terms of Service and Privacy Policy
Please subscribe me to the CodeProject newsletters

When answering a question please:

Read the question carefully.
Understand that English isn't everyone's first language so be lenient of bad spelling and grammar.
If a question is poorly phrased then either ask for clarification, ignore it, or edit the question and fix the problem. Insults are not welcome.
Don't tell someone to read the manual. Chances are they have and don't get it. Provide an answer or move on to the next question.

Let's work to help developers, not make them feel stupid.

This content, along with any associated source code and files, is licensed under The Code Project Open License (CPOL)

Sergey Alexandrovich Kryukov · Answer 1 · 2014-10-28T11:48:00

Solution 1

Your algorithm is really bad, very redundant. Besides, it is incorrect. According to you algorithm, for example, some strings containing only the characters '0' to '9' is always valid, but it cannot be a a valid number if it is too long.

You need a completely different approach. Simply call the function parseDouble with your input string: http://docs.oracle.com/javase/7/docs/api/java/lang/Double.html#parseDouble%28java.lang.String%29[^].

From the documentation referenced above, you can see that NullPointerException or NumberFormatException can be thrown. Catch them both and return false; if the exception wasn't caught, return true. On top of it, you may want to add the validation for the range of the parsed number.

But even this code would be redundant. You simply don't need a validation function in this case (unless you also validated the range or something else); you should better do two things at a time: validate and obtain the numeric value from a string. If you want to wrap it in some function which will show validation result or return a valid numeric value, do it.

—SA

Posted 28-Oct-14 11:48am

Sergey Alexandrovich Kryukov

Updated 28-Oct-14 11:52am

v2

Comments

[no name] 28-Oct-14 18:11pm

Hello Sergey
I think
"According to you algorithm, for example, some strings containing only the characters '0' to '9' is always valid, but it cannot be a a valid number if it is too long."
contains big potential to missunderstand for non english natives like me...at least I hope I interpreted it right ;)

Sergey Alexandrovich Kryukov 28-Oct-14 19:15pm

For example, 897 is valid, but according to OP's algorithm, this is also valid:
943282570498570425974235972435092834750943827502349857230495782034958720349857342098573409587043985723408596234085972309809745723495734250938475034
But it would not parse as integer of any FPU numeric type (only as longint). :-)
Do you know how to explain it more clearly?
—SA

[no name] 28-Oct-14 19:23pm

No, but at least that explains, that I interpreted your answer in the right way.

OP did not ask to convert to int (or whatever) he/she asked only IsNumber ;)

Sergey Alexandrovich Kryukov 28-Oct-14 19:28pm

Yes, yes, but this is the whole point: it's pointless to validate if something is a number. It can be a number from one point of view and not a number from another.
Okay, another example: 1.1e99999999999999999999999999999999999999999999999999999999999999999999999999999999999999999999999999999999999999 won't parse as a number of any floating point or integer type.
Even simpler, "1e2e3e" will pass OP's algorithm, too, and it cannot be any number at all.
The validation used by OP is simply wrong, from any point of view. But more important is the rationale: if something is validated as a number, it will be later parsed to a number of one or another type. There should not be later.
Hence my solution: just a few line of code. Parse methods already do validation.
—SA

[no name] 28-Oct-14 19:44pm

I agree the OP's "isNumber" is far away from correct (one number more after your "1e2e3e" can be become again a discussion, whether it is correct or not, for me it is correct, even if it is uncommon)...
Kind regards, Bruno

But OP's question was about "e" which is still not clear for me, is it "e" allone like "e" (base Ln) or "e" in scientific notation of a number. And this for me unknown "LeetCode" accept it and he is asking why?

Sergey Alexandrovich Kryukov 28-Oct-14 20:20pm

Quite apparently, this "e" used in exponential notation: 2e3 ≡ 2000. Of course, "1e2e3e" cannot make any sense; what to discuss here? :-)
It has nothing to do with natural exponent base ~2.71...; the notation is based on decimal base.
—SA

[no name] 28-Oct-14 20:22pm

But is 1e1e2e3 really wrong? (pls. note it Ends with a number).
Bruno

Sergey Alexandrovich Kryukov 28-Oct-14 20:25pm

Of course it's wrong. How can it be correct?
—SA

[no name] 28-Oct-14 20:28pm

Because it is 1*10^1*10^2*10^3? Or is there any Limitation _defined_ using "e" as Exponent?

Ok, I agree also it becomes very theoretcial, because with my Argumentation any factor (more then that, any expression) in series can/should be possible... a thing which OP did not ask/think about

N.B. For me it is similar like --------1, where my Parser had Problems (and yes I'm convinced, my Parser implementation was wrong)

Sergey Alexandrovich Kryukov 28-Oct-14 22:25pm

What are you even talking about. Exponent notation is fixed, its generalization of 123.4234.
It could be 123.4234e12, 123.4234e-12. Like you have only one or zero points, and zero or 1 e. THIS IS NOT EXPRESSION. THIS IS LITERAL!
(Please forgive me shouting at you. I know that this is impolite. But may be you need some shock. Please, don't take it as offense. You need to learn numeric literals. This is the first thing many developers learn.)
—SA

[no name] 29-Oct-14 4:21am

You are right. Regards, Bruno

Sergey Alexandrovich Kryukov 29-Oct-14 13:38pm

Thank you for understanding. No offense, I really hope.
Cheers,
—SA

[no name] 29-Oct-14 13:45pm

No 100% no offense, never felt so. I know -by expirience!- how much I can learn from you. Kind regards, Bruno

Sergey Alexandrovich Kryukov 29-Oct-14 14:05pm

Thank you for your nice words and being so open to learning.
—SA