string-manipulation - removing char from string

I currently have the following string in java:

"Blah, bl开发者_如何学编程ah, blah,~Part One, Part Two~,blah blah"

I need to remove the comma between the ~ character so it reads.

"Blah, blah, blah,~Part One Part Two~,blah blah"

Can anyone help me out please?

Many thanks,

    String[] tests = {
        "a,b,c,d,e,f",
        "a,b,~c~,d,e",
        "~a,b,c,d,e~",
        "a,b,c,~d,e,f~,g,h,i,~j,k,l,~m,n,o~,q,r,~s,t,u",
    };
    for (String test : tests) {
        System.out.println(
            test.replaceAll(
                "(^[^~]*~)|([^~]*$)|([^,~]*),|([^,~]*~[^~]*~)",
                "$1$2$3$4"
            )
        );
    }

The above prints:

a,b,c,d,e,f
a,b,~c~,d,e
~abcde~
a,b,c,~def~,g,h,i,~jkl~m,n,o~qr~s,t,u

How it works

There are 4 cases:

• We're at the beginning of the string, "outside"
  • Just match until we find the first ~, so next time we'll be "inside"
  • So, (^[^~]*~)
• There are no more ~ till the end of the string
  • If there are even number of ~, we'll be "outside"
  • Just match until the end
  • So, ([^~]*$)
• If it's none of the above, we're "inside"
  • Keep finding the next comma before ~ (so we're still "inside")
    • So, ([^,~]*), (don't capture the comma!)
  • If we find ~ instead of a comma, then go out, then go back in on the next ~
    • So, ([^,~]*~[^~]*~)

In all cases, we make sure we capture enough to reconstruct the string.

References

• regular-expressions.info/Character Classes, Anchors, Grouping and backreferences

Related questions

• Remove a comma between two specific characters (PHP version)

String text = "Blah, blah, blah,~Part One, Part Two~,blah blah,~Part One, Part Two~,blah blah";
Pattern pattern = Pattern.compile("~[^~]+~");
Matcher matcher =pattern.matcher(text);
StringBuffer sb = new StringBuffer();
while(matcher.find()) {
    matcher.appendReplacement(sb, matcher.group(0).replaceAll(",", ""));
}
matcher.appendTail(sb);
text = sb.toString();

I haven't tested this but I would do something like:

string sample = "Blah, blah, blah,~Part One, Part Two~,blah blah";
Regex r = new Regex("(.+)\\~(.+),(.+)\\~(.+)","${1}~${2}${3}~${4}");
r.replaceAll(sample );

I referenced Regular Expressions in Java. In here the .+ matches to one or more any character. More such patterns can be found here.

Here is a method that would do the job:

public String deleteCharacterBetween(String deleteFrom, String betweenChar, String charToRemove) {
    int nextIndex = 0, index = 0;
    while (true) {
        index = deleteFrom.indexOf(betweenChar, nextIndex);
        nextIndex= deleteFrom.indexOf(betweenChar, index + 1);
        if (nextIndex < 0 || index < 0)
            return deleteFrom;
        String before = deleteFrom.substring(0, index);
        String toEdit = deleteFrom.substring(index, nextIndex);
        String after = deleteFrom.substring(nextIndex);
        toEdit = toEdit.replace(charToRemove, "");
        deleteFrom = before + toEdit + after;
    }
}

You can call it like this:

String a = "Blah, blah, blah,~Part One, Part Two~,blah blahBlah, blah, blah,~Part One, Part Two~,blah blah";
System.out.println(deleteCharacterBetween(a, "~", ","));