I'm trying to match the markdown header using regex in javascript. A markdown header is a string starting from one or more # followed by one or more spaces and some more texts, like this:
## This is the title
The tricky thing is, I'm given a multi-line string and a 0-indexed start position, I need to check if the string starting from the given position starts with a markdown header. In other words, I need to write the following function:
/**
* The function should return true in the following situations:
* - text: "abc\n# My Title", startPos = 4
* - text: "abc\n# My Title\nxyz", startPos = 4
*
* The function should return false in the following situation:
* - text "abc\n#My Title", startPos = 0
*/
function isMarkdownHeader(text, startPos) {
...
}
Here is what I've tried:
function isMarkdownHeader(text, startPos) {
const pattern = new RegExp(`^[^]{${startPos}}(^#+)(\\s+)(.*$)`, 'm');
return pattern.exec(text) != null;
}
console.log(isMarkdownHeader("abc\n## My Title\nxyz", 4)); // true
console.log(isMarkdownHeader("abc\n## My Title\nxyz", 0)); // true, which is incorrect.
The function should return false in the second invocation. But it didn't. Why? How can I make it work?
You can use
^.{4}(#+\s+.*)
And the function will look
function isMarkdownHeader(text, startPos) {
const pattern = new RegExp(`^.{${startPos}}#+\\s+.*`, "s");
return pattern.exec(text) != null;
}
isMarkdownHeader("abc\n## My Title\nxyz", 4) // true
isMarkdownHeader("abc\n## My Title\nxyz", 0) // false
NOTE: Here I have used s flag. Also, see the demo.
You can slice the input string before checking the validity.
function isMarkdownHeader(text, startPos) {
const pattern = /^#+\s+.*/;
return !!pattern.exec(text.slice(startPos));
}
isMarkdownHeader("abc\n## My Title\nxyz", 4) // true
isMarkdownHeader("abc\n## My Title\nxyz", 0) // false
isMarkdownHeader("abcd## My Title\nxyz", 4)should returnfalse. With your code, the return value istrue
That occurs because the stuff called newline and the \n character is not the same thing. \n in a string is a regular character of the newline and it becomes a newline after we print the string. So if if you want to match both of them then use
^.{4}(?:^|\n|\\n)(#+\s+.*)
(?:^|\n|\\n) pattern to the old one(?: Non-capturing group
^|\n|\\n Match start of a string or newline or \n character) Close non-capturing groupabcd\n## My Title\nxyzabcd⏎## My Title\nxyz where ⏎ is a newlineP.S. For matching abc\n## My Title\nxyz you have to use 3 as a startPos, not including the \n character.
function isMarkdownHeader(text, startPos) {
const pattern = new RegExp(`^.{${startPos}}(?:^|\\n|\\\\n)(#+\\s+.*)`, "s");
return pattern.exec(text) != null;
}
console.log(isMarkdownHeader("abc\n## My Title\nxyz", 3)) // true
console.log(isMarkdownHeader("abc\n## My Title\nxyz", 0)) // false
console.log(isMarkdownHeader("abcd## My Title\nxyz", 4)) // false
According to Regex101, the {0} quantifier causes the previous token to be ignored. There's no citation to back that, but I'll take them at their word.
That being so, you can check that your match starts at index 0 (which indicates that the string started with the correct number of ignored characters).
function isMarkdownHeader(text, startPos) {
const pattern = new RegExp(`^[^]{${startPos}}(^#+)(\\s+)(.*$)`, 'm');
const match = pattern.exec(text);
return match?.index === 0;
}
console.log(isMarkdownHeader("abc\n## My Title\nxyz", 4)); // true
console.log(isMarkdownHeader("abc\n## My Title\nxyz", 0)); // now false
You don't need the regex to start at a certain position. You can just pass the substring starting at startPos:
return pattern.exec(text.substring(startPos)) != null;