I'm writing a code where I'm getting rich text and want to replace things. Here I've 2 queries.

1. How can I match *, **, and *** separately?
2. How can I conditionally replace data? i.e. if the index of *** from match is 1 I want to have it replaced by _* else *_.

Here is my sample code.

let x =
  "Lorem ***ipsum*** dolor sit amet, consectetur adipiscing elit. Quod non faceret, si in voluptate summum bonum poneret. Ratio quidem vestra sic cogit. *Qui\\-vere falsone, quaerere mittimus\\-dicitur oculis se privasse;* Sed in rebus apertissimis ***nimium longi sumus. Eiuro, inquit adridens, iniquum, hac quidem de re; **Que Manilium, ab iisque M.** Quis animo aequo videt eum, quem inpure ***ac flagitiose putet vivere? Duo Reges: constructio interrete.";

var re = new RegExp(/(\*)(\*)(\*)/g);
var results = new Array(); //this is the results you want
for (let match of x.matchAll(re)) {
  results.push(match.index);
}

results.forEach((item, idx) =>
  idx % 2 === 0
    ? (x = replaceRange(x, item, item   3, "_*"))
    : (x = replaceRange(x, item, item   3, "*_"))
);

function replaceRange(s, start, end, substitute) {
  return s.substring(0, start)   substitute   s.substring(end);
}
console.log(x);

When I run this code, it is not replacing the data as expected. Please let me know where I'm going wrong.

Thanks

CodePudding user response：

One way is to use RexExp.prototype.exec, which is stateful and keeps track of the last match so you can iterate over each match. Then use a counter to keep track of how many matches you've found so you know whether to use *_ or _*.

let x =
  "Lorem ***ipsum*** dolor sit amet, consectetur adipiscing elit. Quod non faceret, si in voluptate summum bonum poneret. Ratio quidem vestra sic cogit. *Qui\\-vere falsone, quaerere mittimus\\-dicitur oculis se privasse;* Sed in rebus apertissimis ***nimium longi sumus. Eiuro, inquit adridens, iniquum, hac quidem de re; **Que Manilium, ab iisque M.** Quis animo aequo videt eum, quem inpure ***ac flagitiose putet vivere? Duo Reges: constructio interrete.";

const regex = new RegExp(/\*\*\*/g)
let result
let matchCount = 0
while(result !== null)
  {
    result = regex.exec(x)
    if (result === null) break
    matchCount  = 1
    matchCount % 2 === 0
      ? (x = replaceRange(x, result.index, result.index   3, "_*"))
      : (x = replaceRange(x, result.index, result.index   3, "*_"))
  }

console.log('x', x)

function replaceRange(s, start, end, substitute) {
  return s.substring(0, start)   substitute   s.substring(end);
}

CodePudding user response：

You can search for the same * pattern after and before each word with Backreference, which matches the same text as matched by the 1st capturing group. For instance, with triple asterisks as in ***, the regex could be:

[^_]([*]{3})(\w )\1

Using this replacement:

 _*$2*_

Explanation:

• [^_]: avoids capturing if it starts with _
• ([*]{3}): capture group that matches exactly 3 `*
• (\w ): capture group that matches any word character at least once.
• \1: backreference to the first capture group, it will capture its same match.

Regarding your queries, it depends on the replacements. Since the provided example _* also has an asterisk, you must adapt the regex so that it won't substitute the output of the previous regex.

let x =
  "Lorem ***ipsum*** dolor sit amet, consectetur adipiscing elit. Quod non faceret, si in voluptate summum bonum poneret. Ratio quidem vestra sic cogit. *Qui\\-vere falsone, quaerere mittimus\\-dicitur oculis se privasse;* Sed in rebus apertissimis ***nimium longi sumus. Eiuro, inquit adridens, iniquum, hac quidem de re; **Que Manilium, ab iisque M.** Quis animo aequo videt eum, quem inpure ***ac flagitiose putet vivere? Duo Reges: constructio interrete.";

const re1 = new RegExp(/[^_]([*]{1})(\w )\1/g);
const re2 = new RegExp(/([*]{2})(\w )\1/g);
const re3 = new RegExp(/[^_]([*]{3})(\w )\1/g);
 
let result = x.replaceAll(re3, " _*$2*_");

console.log(result);