RegEx解析日志文件中的一行

Question

I have a jenkins log file that spits the following output:

angularjs@1_4_7-ie8 found in path(s): public/components/angularjs-ie8-build/dist/angular.min.js

[INFO] Registered manifest into PaaS: https://deploy-apixyz.com/swdeploy/v2/manifests/demonodeserver/versions/1.0.0_20180628165604811

Your build metrics have been recorded with id demonodeserver-06-29T00:07:42.845Z and manifest_id demonodeserver-1.0.0_20180628165604811

I am interested in the maniest_id part demonodeserver-1.0.0_20180628165604811

I am looking to see if i can write a regex that can parse this.

Tried various methods but failed. Can someone enlighten me?

str.match(\\[demonodeserver-\\] (.*)) but this doesn't return valid result.

Answer 1

In order to extract demonodeserver-1.0.0_20180628165604811 from your logs based on your needs, you have several options:

 const log = ` angularjs@1_4_7-ie8 found in path(s): public/components/angularjs-ie8-build/dist/angular.min.js [INFO] Registered manifest into CMPaaS: https://deploy-apixyz.com/swdeploy/v2/manifests/demonodeserver/versions/1.0.0_20180628165604811 Your build metrics have been recorded with id demonodeserver-06-29T00:07:42.845Z and manifest_id demonodeserver-1.0.0_20180628165604811 `; // If you need the tag part only: const patternTagOnly = /manifest_id\\s+\\w+-([\\w:\\.]+)/m; const tag = log.match(patternTagOnly)[1]; console.log(`tag only: ${tag}`); // If you need the name part only: const patternNameOnly = /manifest_id\\s+(\\w+)-[\\w:\\.]+/m; const name = log.match(patternNameOnly)[1]; console.log(`name only: ${name}`); // If you need the name and tag part too, unseparated: const patternFull = /manifest_id\\s+(\\w+-[\\w:\\.]+)/m; const full = log.match(patternFull)[1]; console.log(`full: ${full}`); // If you need the name and tag part too, but separated: const patternFullSeparated = /manifest_id\\s+(\\w+)-([\\w:\\.]+)/m; const parts = log.match(patternFullSeparated); console.log(`full, separated: ${parts[1]} - ${parts[2]}`); 

To create/test regexes for JavaScript, check out regex101 , but make sure you choose the JavaScript Flavor regex.

Answer 2

A regex is created with / , not \\ . There's also no space in your log output (after demonodeserver-1.0.0), and to get the id I'd also specify which line exactly you're interested in. The following should do the job

 const str = `angularjs@1_4_7-ie8 found in path(s): public/components/angularjs-ie8-build/dist/angular.min.js [INFO] Registered manifest into CMPaaS: https://deploy-apixyz.com/swdeploy/v2/manifests/demonodeserver/versions/1.0.0_20180628165604811 Your build metrics have been recorded with id demonodeserver-06-29T00:07:42.845Z and manifest_id demonodeserver-1.0.0_20180628165604811`; const regex = /demonodeserver-(\\d\\.?){3}_\\w+/gm; const match = str.match(regex); console.log(match); 

Answer 3

If you're interested in the manifest_id part, match on that instead of expecting the actual id to have a constant prefix (you're not always going to be using a demo node server, right?):

 const log = `angularjs@1_4_7-ie8 found in path(s): public/components/angularjs-ie8-build/dist/angular.min.js [INFO] Registered manifest into CMPaaS: https://deploy-apixyz.com/swdeploy/v2/manifests/demonodeserver/versions/1.0.0_20180628165604811 Your build metrics have been recorded with id demonodeserver-06-29T00:07:42.845Z and manifest_id demonodeserver-1.0.0_20180628165604811`; console.log(log.match(/manifest_id (\\S+)/)[1]); 

The regex itself:

/manifest_id (\S+)/
/                 / regexes are delimited by /
 manifest_id        a literal "manifest_id "
             (   )  capture the stuff in here and put it in the match array
              \S    any non-whitespace character
                +   one or more times

String.match returns an array with the entire match at index 0 and any capture groups after that. We only care about our single group, so we grab the string at index 1.