How to split string into parts using Regex-CodePudding

I have a string containing placeholders which I want replace with other strings, but I would also like to split the string whenever I encounter a placeholder.

So, by splitting I mean that

"This {0} is an example {1} with a placeholder"

should become:

parts[0] -> "This"
parts[1] -> "{0}"
parts[2] -> "is an example"
parts[3] -> "{1}"
parts[4] -> "with a placeholder"

and then the next step would be to replace the placeholders (this part is simple):

parts[0] -> "This"
parts[1] -> value[0]
parts[2] -> "is an example"
parts[3] -> value[1]
parts[4] -> "with a placeholder"

I know how to match and replace the placeholders (e.g. ({\d })), but no clue how to tell regex to "match non placeholders" and "match placeholders" at the same time.

My idea was something like: (?!{\d }) | ({\d }) but it's not working. I am doing this in JavaScript if Regex flavor is important.

If I can also replace the placeholders with a value in one step it would be neat, but I can also do this after I split.

CodePudding user response：

You might write the pattern as:

{\d }|\S.*?(?=\s*(?:{\d }|$))

The pattern matches:

{\d } Match { 1 digits and }
| Or
\S.*? Match a non whitespace char followed by any character as few as possible
(?= Positive lookahead
- \s* Match optional whitespace chars
- (?:{\d }|$) Match either { 1 digits and } or assert the end of the string
) Close the lookahead

Regex demo

To get an array with those values:

const regex = /{\d }|\S.*?(?=\s*(?:{\d }|$))/gm;
const str = `This {0} is an example {1} with a placeholder`;
console.log(str.match(regex))

CodePudding user response：

If you use parenthesis around the separator, matched results are included in the output:

let parts = str.split(/ *({\d }) */);

See this demo at tio.run - If separator occurs at start/end, just filter out empty matches.

If your goal is just to replace, it can be done in one step using replace and a callback:

str = str.replace(/{(\d )}/g, (m0,m1) => value[m1]);

Another demo at tio.run - m0 is the full match, m1 holds the capture of the first group.
Used with g (global) flag to return all possible matches in the string (not just the first).