Regex to filter out Datetime time from string-CodePudding

Requirement: I am receiving an email with a template and I need to filter out some text from the email. I am converting all the email body text as a string.

email text looks like this:

some body text which I don't need 

Discussion:
Tue 26/04/2022/2:48 PM UTC 10/ ABC User-

TEST description - this should be logged as a comment.  --- This is the part I need

ABC Company Australia | XYZ St | Sydney NSW 2000

I was thinking of having a regex that looks for

A word "Discussion"
Next line look for DateTime format with "Tue 26/04/2022/2:48 PM UTC 10/ ABC User-"
Pick up the next line until we find this line - "ABC Company Australia | XYZ St | Sydney NSW 2000" address

Is it possible? can someone plz help with regex?

TIA.

CodePudding user response：

You may try this regex:

Discussion.*?\n ([A-Za-z]   (?:\d{2}\/){2}\d{4}\/\d :\d   [^\n] )(.*)?ABC Company Australia \| XYZ St \| Sydney NSW 2000

Explanation:

Discussion.*?\n The regex starts from where the String Discussion begins.
.*?\n keeps looking for additional word and newlines
([A-Za-z] (?:\d{2}\/){2}\d{4}\/\d :\d [^\n] ) next it looks for the date format as you described . It will fetch all until it reaches a newline [^\n ]
(.*)?It will fetch everything from the previous date line
ABC Company Australia \| XYZ St \| Sydney NSW 2000 and will conclude matching whenever it finds this.
Here I have kept the Date format line in group 1 and the body you need in group 2

Demo

source:

const regex = /Discussion.*?\n ([A-Za-z]   (?:\d{2}\/){2}\d{4}\/\d :\d   [^\n] )(.*)?ABC Company Australia \| XYZ St \| Sydney NSW 2000/gms;

const str = `some body text which I don't need 

Discussion:
Tue 26/04/2022/2:48 PM UTC 10/ ABC User-

TEST description - this should be logged as a comment.  --- This is the part I need

ABC Company Australia | XYZ St | Sydney NSW 2000
`;

var match = regex.exec(str);
if(match!=null){
console.log(match[1]);
console.log(match[2]);
}

CodePudding user response：

If it was just about the content the OP is interested in the following regex already is sufficient enough ... /Discussion:\n[a-zA-Z]{1,3}\s \d{2}\/\d{2}\/\d{4}.*\n (?<content>.*)/

const multilineMail = `Discussion:
Tue 26/04/2022/2:48 PM UTC 10/ ABC User-

TEST description - this should be logged as a comment.  --- This is the part I need

ABC Company Australia | XYZ St | Sydney NSW 2000`;


// see ... [https://regex101.com/r/v8FXCA/3]
const regXMailContent =
  /Discussion:\n[a-zA-Z]{1,3}\s \d{2}\/\d{2}\/\d{4}.*\n (?<content>.*)/;

console.log(
  regXMailContent.exec(multilineMail)?.groups?.content
);

In case the company footer has to match exactly one has to make it part of the above regex like follows ... /Discussion:\n[a-zA-Z]{1,3}\s \d{2}\/\d{2}\/\d{4}.*\n (?<content>.*)\n ABC Company Australia \| XYZ St \| Sydney NSW 2000/

const multilineMail = `Discussion:
Tue 26/04/2022/2:48 PM UTC 10/ ABC User-

TEST description - this should be logged as a comment.  --- This is the part I need

ABC Company Australia | XYZ St | Sydney NSW 2000`;


// see ... [https://regex101.com/r/v8FXCA/4]
const regXMailContent =
  /Discussion:\n[a-zA-Z]{1,3}\s \d{2}\/\d{2}\/\d{4}.*\n (?<content>.*)\n ABC Company Australia \| XYZ St \| Sydney NSW 2000/;

console.log(
  regXMailContent.exec(multilineMail)?.groups?.content
);

If the OP wants to also save date and user one would enhance the firstly provided regex like with ...

const multilineMail = `Discussion:
Tue 26/04/2022/2:48 PM UTC 10/ ABC User-

TEST description - this should be logged as a comment.  --- This is the part I need

ABC Company Australia | XYZ St | Sydney NSW 2000`;


// see ... [https://regex101.com/r/v8FXCA/2]
const regXMailDateAndContent =
  /Discussion:\n(?<date>[a-zA-Z]{1,3}\s \d{2}\/\d{2}\/\d{4}).*\n (?<content>.*)/;

// see ... [https://regex101.com/r/v8FXCA/1]
const regXMailDateUserAndContent =
  /Discussion:\n(?<date>[a-zA-Z]{1,3}\s \d{2}\/\d{2}\/\d{4}\/[^/] )\/\s*(?<user>.*?)-?\s*\n (?<content>.*)/;

console.log(
  regXMailDateAndContent.exec(multilineMail)?.groups
);
console.log(
  regXMailDateUserAndContent.exec(multilineMail)?.groups
);

.as-console-wrapper { min-height: 100%!important; top: 0; }

But in case the to be extracted content is a multiline text the regex has to feature the company footer in order to identify the correct match. The 2ndly provided regex then changes to ... /Discussion:\n[a-zA-Z]{1,3}\s \d{2}\/\d{2}\/\d{4}.*\n (?<content>(?:.*\n)*)ABC Company Australia \| XYZ St \| Sydney NSW 2000/

const multilineMail = `Discussion:
Tue 26/04/2022/2:48 PM UTC 10/ ABC User-

TEST

description - this should be
logged as a comment.  --- This is

the part I need


ABC Company Australia | XYZ St | Sydney NSW 2000`;


// see ... [https://regex101.com/r/v8FXCA/5]
const regXMailMultilineContent =
  /Discussion:\n[a-zA-Z]{1,3}\s \d{2}\/\d{2}\/\d{4}.*\n (?<content>(?:.*\n)*)ABC Company Australia \| XYZ St \| Sydney NSW 2000/;

console.log(
  regXMailMultilineContent.exec(multilineMail)?.groups?.content
);

.as-console-wrapper { min-height: 100%!important; top: 0; }

All of the above regex patterns make use of named capturing groups.