Difference between kibana github and kibana docker oss

Elastic Stack Kibana
1

Hello!

I've noticed that there are differences between the code from github and the code inside docker oss images.

For examples:

2

Here's the version I pulled from the kibana-oss:6.2.4 image:

'use strict';

Object.defineProperty(exports, "__esModule", {
  value: true
});

var _download = require('./download');

var _bluebird = require('bluebird');

var _bluebird2 = _interopRequireDefault(_bluebird);

var _path = require('path');

var _path2 = _interopRequireDefault(_path);

var _cleanup = require('./cleanup');

var _pack = require('./pack');

var _rename = require('./rename');

var _rimraf = require('rimraf');

var _kibana = require('./kibana');

var _mkdirp = require('mkdirp');

var _mkdirp2 = _interopRequireDefault(_mkdirp);

function _interopRequireDefault(obj) { return obj && obj.__esModule ? obj : { default: obj }; }

function _asyncToGenerator(fn) { return function () { var gen = fn.apply(this, arguments); return new _bluebird2.default(function (resolve, reject) { function step(key, arg) { try { var info = gen[key](arg); var value = info.value; } catch (error) { reject(error); return; } if (info.done) { resolve(value); } else { return _bluebird2.default.resolve(value).then(function (value) { step("next", value); }, function (err) { step("throw", err); }); } } return step("next"); }); }; }

const mkdir = _bluebird2.default.promisify(_mkdirp2.default);

exports.default = (() => {
  var _ref = _asyncToGenerator(function* (settings, logger) {
    try {
      yield (0, _cleanup.cleanPrevious)(settings, logger);

      yield mkdir(settings.workingPath);

      yield (0, _download.download)(settings, logger);

      yield (0, _pack.getPackData)(settings, logger);

      yield (0, _pack.extract)(settings, logger);

      (0, _rimraf.sync)(settings.tempArchiveFile);

      (0, _kibana.existingInstall)(settings, logger);

      (0, _kibana.assertVersion)(settings);

      yield (0, _rename.renamePlugin)(settings.workingPath, _path2.default.join(settings.pluginDir, settings.plugins[0].name));

      yield (0, _kibana.rebuildCache)(settings, logger);

      logger.log('Plugin installation complete');
    } catch (err) {
      logger.error(`Plugin installation was unsuccessful due to error "${err.message}"`);
      (0, _cleanup.cleanArtifacts)(settings);
      process.exit(70); // eslint-disable-line no-process-exit
    }
  });

  function install(_x, _x2) {
    return _ref.apply(this, arguments);
  }

  return install;
})();

module.exports = exports['default'];

Is that what you're seeing too? The reason it differs from what you see on GitHub is due to Kibana's build process. All of our source files are transpiled using Babel to ensure compatibility across different environments.

3

Yes, this is what I see in files.

The problem with this difference is that the files can't be patched. (For example, I needed the pull request from https://github.com/elastic/kibana/pull/16302 to be applied on docker files, and using a simple diff doesn't work).

4

Correct, you can't patch source files into the distributable like that. If you're not able/willing to upgrade, then your only option is to create a fork Kibana, and merge the patch yourself. Once merged in your version, you can build your own OSS distribution with the patch applied.

5

This topic was automatically closed 28 days after the last reply. New replies are no longer allowed.