problems splitting string

stephan_007

hy everyone! i do have a problem. I have to split strings they are like

"variable1", "variable_1", "variable"

i do only have to parse the first line, but sometimes the first line looks like

"variable11", "variable_11", "variable1"

in this line the last "1" represents a color information. I do only have to parse the first line, in both cases the result should be variable1, variable_1, variable example: my input is

variable1, variable_1, variable
variable11, variable_11, variable1

=> result of first line: variable1, variable_1, variable if the input is

variable11, variable_11, variable1
variable1, variable_1, variable

=> result of first line: variable1, variable_1, variable does anyone have an idea of how to code this? It's quit tricky. I tried myself for some hours but found no solution which works propperly. Because I do not know if the first element has two "11" or just one! The string could also be in this order:

variable, variable1, variable_1

Maybe someone of you could help me please. thanks. Stephan.

led mike

stephan_007 wrote:

does anyone have an idea of how to code this?

Yes. Use the keyboard to type syntactically correct code that represents the logic needed to satisfy your requirements.

led mike

Guffa

You have to specify which parts of the string that can vary, and how. For example, the repeated part "variable" in your example, is that always the specific word "variable", or could it be anything? If you have an actual example of some real data, that could help when trying to understand what you are trying to do.

Despite everything, the person most likely to be fooling you next is yourself.

WBurgMo

Assuming you parse the variable names into a string array, then for each element of the array examine the last character. If is numeric then examine the next to last character. If that character is also numeric then strip the last character. Of course this assumes that your color info is only a single digit.

stephan_007

well the input could be anything. the only difference between the two lines is, one has an additional "1" at the end of every fieldname. so if you remove the last "1" from the second line, then you will get the other. e.g. name, address1, address2, postalcode, city, phone_1, phone_2 name1, address11, addrress21, postalcode1, city1, phone_11, phone_21 So as you might have realized the second line is the first one with an additional "1" at the end of each entry. i guess meanwhile i found a solution to this problem: first check if every entry ends with 1, then it's the of the second type. if it is of the second type, then remove the last "1" by using Substring with the whole string without the last character (length-1). The last operation will change name1 to name and so on. But if the first operation returns, not every item has a "1" at it's end, then its of type line 1 and nothing has to be done. in code its something likestring[] Checkforcolorsign (string[] input) { int counter = 0; int i = 0; // check for "1" at the end (colorinfo) while (i < input.length - 1) { if (input[i].EndsWith("1")) { counter++; } } // if all elements had a "1" then it is a colorinfo => remove it if (counter == input.length) //as many hits as elements { for(i=0; i < input.length - 1; i++) { // remove the last character input[i] = input[i].SubString(0, input[i].Length - 1); } } return input; }
Maybe this code will illustrate what I originally wanted to ask. I hope this code really does what I expect it to. so if the input is e.g. name1, address11, addrress21, postalcode1, city1, phone_11, phone_21 then the final result should be name, address1, address2, postalcode, city, phone_1, phone_2

stephan_007

that was an idea i already had, but it did not work. e.g. if the input is of this kind name1, address11, address21, postalcode1, city1 and the other line (without the colorsign "1" added) looks like name, address1, address2, postalcode, city it would work for the address elements but not for the other ones. i already posted some code which came to my mind when thinking about it without sitting in front of the comp. i have to check the syntax and if it works, meaning if it changes the line with the "1"s and doesn't change the line without. because that's what I want to do. the "task" is to read fieldvalues because i do have something like name|address1|address2|postalcode|city Mr. X, nowhere, here and there, 1234, noname city name1|address11|address21|postalcode1|city ..... but the two lines could be the otherway round as well, meaning name1|address11|address21|postalcode1|city Mr. X, nowhere, here and there, 1234, noname city name|address1|address2|postalcode|city and because every datafieldline is the same like the associated valuelines (the position of the fields not the values!), i do only read the first fieldline and use the indexes for every dataline. the fieldvalues are used for further operations (the ones without the "1" at the end). that's why i have to remove it.

Guffa

As far as I can tell, the code will work just fine. You don't have to return the array, though. You are changing the items in the array.

void Checkforcolorsign (string[] input) {
int counter = 0;
// check for "1" at the end (colorinfo)
foreach (string s in input) {
if (s.EndsWith("1")) {
counter++;
}
}
// if all elements had a "1" then it is a colorinfo => remove it
if (counter == input.length) //as many hits as elements {
for (int i = 0; i < input.length - 1; i++) {
// remove the last character
input[i] = input[i].SubString(0, input[i].Length - 1);
}
}
}

Despite everything, the person most likely to be fooling you next is yourself.