summaryrefslogtreecommitdiff
Commit message (Collapse)AuthorAgeFilesLines
* Optimise email CSS for speed with Premaileroptimise-email-cssSean McGivern2016-06-072-28/+167
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Remove all descendant selectors from the push email styling, to drastically reduce CPU time when inlining the CSS for syntax-highlighted diffs. Background: Premailer is a Ruby gem that inlines CSS styles from an external stylesheet before emails are sent, so that they are compatible with Gmail. At a high level, it parses the CSS files it finds, and parses the email body with Nokogiri. It then loops through the selectors in the CSS, using Nokogiri to find matching elements, and adds inline styles. (It does more than this, like merging styles applied to the same element, but that's not relevant to this issue.) Nokogiri converts CSS selectors to XPath first, like so: Nokogiri::CSS.xpath_for('foo bar') # => ["//foo//bar"] On documents with high node counts (say, a syntax-highlighted copy of jQuery), having both descendant selectors is very expensive. Both `//foo/bar` and `//bar` will be much more efficient, although neither are directly equivalent. An example, on a document containing two syntax-highlighted copies of jQuery: Benchmark.realtime { p doc.search('.o').count } # 9476 # => 0.3462457580026239 Benchmark.realtime { p doc.search('.code.white .o').count } # 9476 # => 85.51952634402551 The performance is similar for selectors which _don't_ match any elements, and as Premailer loops through all the available selectors, we want to avoid all descendant selectors in push emails. Because of the theming support in the web UI, all syntax highlighting selectors are descendant selectors of classes like `.code.white` or `.code.monokai`. There are over 60 CSS classes for syntax highlighting styles alone, all of which are expressed in the inefficient form above. In emails we always use the white theme, and were reusing the same CSS file. But in emails, we don't need to descend from `.code.white` as that will always be the theme, and we can also remove some other selectors that are only applicable to the web UI. For the remaining descendant selectors, we can convert them to child selectors, type selectors, or class selectors as appropriate. As in the example above, having no descendant selectors at all in the push email CSS can provide a drastic (and surprising) performance improvement.
* Merge branch 'master' of gitlab.com:gitlab-org/gitlab-ceDmitriy Zaporozhets2016-06-0649-146/+1889
|\
| * Merge branch '15337-yubikey-support' into 'master' Rémy Coutable2016-06-0649-146/+1889
| |\ | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Allow a U2F Device to be the Second Factor for Authentication Parent Issue: #15337 ## TODO - [ ] #15337 (!3905) FIDO/U2F 2FA using Yubikey - [x] Order a Yubikey? - [x] Do some reading to figure out what all this stuff means - [x] Look through the existing MR - [x] Browser support? - [x] Implementation - [x] User can register 2FA using their U2H device instead of authenticator - [x] Barebones flow - [x] Save the registration in the database - [x] Authentication flow - [x] First try after login/server start doesn't work - [x] User can log in using their U2F device - [x] Allow setting up authenticator if U2F is already set up (or vice versa) - [x] Change `two_factor_auths/new` to `show` - [x] `sign_requests` during registration? (Registering a device that has already been registered) - [x] 2FA skippable flow? - [x] Enforced 2FA flow (grace period?) - [x] Move the "Configure it Later" button to the right place - [x] Don't allow registration when the yubikey isn't plugged in - [x] Polish authentication flow - [x] Login should only show the 2FA method that's enabled - [x] Message to say that u2f only works on chrome, and it's recommended to enable otp as well. - [x] Index for key_handle - [x] Server-side errors while registering/logging in - [x] Handle non-chrome browsers - [x] Try to authenticate with a key that hasn't been registered (shouldn't work) - [x] Try the same key for multiple user accounts (should work) - [x] Fix existing tests - [x] Make sure CI is green - [x] Add tests - [x] Figure out how to fake the Yubikey - [x] Teaspoon tests for the React components - [x] Each device can only be registered once per user - [x] Feature specs - [x] Regular flows - [x] Test error cases - [x] Refactoring - [x] Refactor App ID - [x] Clean up the `show` action - [x] Annotate methods with definition of U2F - [x] Changelog - [x] Fix merge conflicts - [x] Verify flows - [x] Authenticator + no U2F - [x] U2F + no authenticator - [x] U2F + authenticator - [x] U2F + authenticator -> disable 2FA - [x] 2FA required with different grace periods - [x] Screenshots for MR - [x] Augment the [help docs](http://localhost:3000/help/profile/two_factor_authentication) - [x] Assign to endboss - [x] Ask for feedback on UI/UX - [x] Ask for feedback on copy - [x] Wait for review/merge - [x] Fix merge conflicts - [x] Wait for CI to pass - [x] Implement review comments/suggestions - [x] Move `TwoFactorAuthController#create_u2f` to a service - [x] Extra space before `Base64` in `u2f_registration` model - [x] Move `with/without_two_factor` scopes to class methods - [x] In `profiles/accounts/show`, add spaces at `{` and `}` - [x] Remove blank lines in `profiles/two_factor_auths/show` - [x] Fix typo in doc. "(universal 2nd factor )" - [x] Add "Added in 8.8" to doc - [x] In the doc, use 'Enable 2FA via mobile application' instead of 'Via Mobile Application' - [x] In the doc, use 'Enable 2FA via U2F device' instead of 'Via U2F Device - [x] Use "Two-Factor Authentication" everywhere - [x] Use `#icon` wrapper instead of `fa_stacked_icon` - [x] Check if `string` is enough for `key_handle` and `public_key` - [x] Separate `exercise` and `verify` phases of test (u2f_spec) - [x] Assert that `user_without_2fa` is _not_ in results (with_two_factor) - [x] Remove rubocop exception - [x] Refactor call to `User.with_two_factor.count` to not include `.length` - [x] Add a note that makes the "Disable" button/feature obvious - [x] Remove i18n - [x] Test in Firefox with addon (+ create new issue for support) - [x] Remove React - [x] Rewrite registration - [x] Switch underscore template to default style - [x] Rewrite authentication - [x] Move `register` haml to `u2f` dir - [x] Remove instance variables - [x] Fix tests - [x] Read SCSS guidelines - [x] Address @connorshea's comments regarding text style - [x] Make sure all classes and IDs are in line (add `js-` prefixes) - [x] Register - [x] Authenticate - [x] Refactoring? - [x] Include non-minifed version of bowser - [x] Audit log - [x] Look at the `browser` gem (and don't use bowser) - [x] Error message when on HTTP? - [x] Test on Mobile - [x] Fix merge conflicts - [x] Retest all flows - [x] Back to Rémy for review - [x] Make sure CI is green - [x] Wait for merge / more feedback - [x] Implement @rymai's changes - [x] JS/Coffeescript variables should be lowerCamelCase - [x] Spaces before/after `}` and `{` in HAML (and elsewhere) - [x] Rails view helpers in u2f HAML - [x] `%div.row.append-bottom-10` - [x] Wrap line in `without_two_factor` scope - [x] Exception-less flow in `U2F::CreateService` - [x] Fix merge conflicts - [x] Move service to model class method - [x] Fix teaspoon specs - [x] Address @rymai's suggestions about error handing - [x] Javascript error constants - [x] Fix merge conflicts - [x] One final review - [x] Test "registration with errors" flow - [x] Assign to Remy - [x] Wait for replies from @jschatz1 - [x] Address @rymai's comments - [x] Omit `%div` - [x] Scope `$.find` globally - [x] Replace `find('#element-id).click` with `click_on('Element Text') - [x] Rebase master + conflicts - [x] Look at https://news.ycombinator.com/item?id=11690774 - [x] Address @connorshea's comment regarding HTTPS on localhost - [x] Final sanity check - [x] Wait for [CI to pass](https://gitlab.com/gitlab-org/gitlab-ce/commit/c84179ad233529c33ee6ba8491cfea862c6cd864/builds) - [x] Address @rymai's next round of comments - [x] Interpolate `true` and `false` in DB scopes - [x] Why have `Gon::Base.render_data` thrice? - [x] `user_spec` should have correct spacing - [x] Use `arel_table[:id]` instead of `users.id` - [x] URL helper in `app/views/profiles/two_factor_auths/show.html.haml` - [x] Remove polyfill change - [x] Wait for [CI to pass](https://gitlab.com/gitlab-org/gitlab-ce/commit/0123ab8/builds) - [x] Address @jschatz1's comments - [x] Use `on('click', ...)` instead of `click(...)` - [x] Use `is` and `isnt` in coffeescript - [x] Use `and` and `or` in coffeescript - [x] Add `Gon::Base.render_data` to `devise_empty` (and other base layouts) - [x] Wait for [CI to pass](https://gitlab.com/gitlab-org/gitlab-ce/commit/401916397336174c582be3d3004a072f845d4b5f/builds) - [x] Wait for [build](https://gitlab.com/gitlab-org/gitlab-ce/commit/75955710ef9a5f0dcee04e8617028c0e3ea5bf50/builds) to pass - [x] Fix merge conflicts - [x] Inspect diff / workflow - [x] Assign back to @rymai - [x] Make sure [ci](https://gitlab.com/gitlab-org/gitlab-ce/commit/2c6316b29a9276ef44c7b4b39363a611bf5973a6/builds) has passed - [x] Fix merge conflicts (probably introduced by [devise upgrade](https://gitlab.com/gitlab-org/gitlab-ce/merge_requests/4216) - [x] Wait for [CI](https://gitlab.com/gitlab-org/gitlab-ce/commit/a5ef48b7aa63d0d9e45b41643043b57208eaab9f/builds) to pass - [x] Respond to @rymai's comments - [x] Use `elsif` - [x] Check if we need `and return` - [x] Only fetch key handles from the DB - [x] No annotations to models? - [x] Align hash keys in model - [x] Wait for [build](https://gitlab.com/gitlab-org/gitlab-ce/commit/e0ef504734e7f14813c73bbb79f5c5f6fae3248c/builds) to pass - [ ] Wait for merge ## Screenshots ![Screenshot_2016-05-03_09.53.04](/uploads/1af3f277efa488dc107d36e6b4b07ca4/Screenshot_2016-05-03_09.53.04.png) ![Screenshot_2016-05-03_10.19.53](/uploads/2bfc67dfb96c0e005cce033d8b456813/Screenshot_2016-05-03_10.19.53.png) ![Screenshot_2016-05-03_10.19.56](/uploads/e912abedd5b1d07d7185cee9f204c5ff/Screenshot_2016-05-03_10.19.56.png) ![Screenshot_2016-05-03_10.20.04](/uploads/9350d5c98823d1f3d4e59517dfb8910a/Screenshot_2016-05-03_10.20.04.png) ![Screenshot_2016-05-03_10.31.15](/uploads/84473dc263e0643311a39006e649035f/Screenshot_2016-05-03_10.31.15.png) ![Screenshot_2016-05-03_10.31.22](/uploads/13ce43e0d7a565000af29984667eeb08/Screenshot_2016-05-03_10.31.22.png) ![Screenshot_2016-05-03_10.31.37](/uploads/b90fbb40dbf9bbd73af324f48ffdc948/Screenshot_2016-05-03_10.31.37.png) ![Screenshot_2016-05-03_10.36.48](/uploads/41a0fbc493c6fefeafd922b3ddf2a25e/Screenshot_2016-05-03_10.36.48.png) See merge request !3905
| | * Add the U2F feature to the CHANGELOG15337-yubikey-supportTimothy Andrew2016-06-061-0/+1
| | |
| | * Add documentation for U2F registration & authentication.Timothy Andrew2016-06-063-6/+57
| | |
| | * Add feature specs covering U2F registration and authentication.Timothy Andrew2016-06-062-0/+275
| | |
| | * Add a U2F-specific audit log entry after logging in.Timothy Andrew2016-06-062-3/+36
| | | | | | | | | | | | | | | | | | - "two-factor" for OTP-based 2FA - "two-factor-via-u2f-device" for U2F-based 2FA - "standard" for non-2FA login
| | * Implement authentication (login) using a U2F device.Timothy Andrew2016-06-068-43/+230
| | | | | | | | | | | | | | | | | | - Move the `authenticate_with_two_factor` method from `ApplicationController` to the `AuthenticatesWithTwoFactor` module, where it should be.
| | * Implement U2F registration.Timothy Andrew2016-06-0616-65/+1086
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | - Move the `TwoFactorAuthsController`'s `new` action to `show`, since the page is not used to create a single "two factor auth" anymore. We can have a single 2FA authenticator app, along with any number of U2F devices, in any combination, so the page will be accessed after the first "two factor auth" is created. - Add the `u2f` javascript library, which provides an API to the browser's U2F implementation. - Add tests for the JS components
| | * Render `gon` data in the page `body`, not `head`Timothy Andrew2016-06-065-2/+5
| | | | | | | | | | | | | | | | | | | | | | | | - Turbolinks caches the `head`, so `gon` updates don't show up unless the user navigates to page directly (by URL) or performs a refresh. - The solution is to render `gon` in the body instead. - Also update the syntax to the new Rails 4 (according to the gon README) syntax.
| | * Update the `browser` gem.Timothy Andrew2016-06-063-5/+5
| | | | | | | | | | | | | | | - Need the `mobile?` detection (that the new version provides) for the U2F registration/ authentication flow
| | * Add a `U2fRegistrations` table/model.Timothy Andrew2016-06-0613-22/+194
| |/ | | | | | | | | | | | | | | | | | | | | | | | | | | - To hold registrations from U2F devices, and to authenticate them. - Previously, `User#two_factor_enabled` was aliased to the `otp_required_for_login` column on `users`. - This commit changes things a bit: - `User#two_factor_enabled` is not a method anymore - `User#two_factor_enabled?` checks both the `otp_required_for_login` column, as well as `U2fRegistration`s - Change all instances of `User#two_factor_enabled` to `User#two_factor_enabled?` - Add the `u2f` gem, and implement registration/authentication at the model level.
* | Merge branch 'fix/unauthorized-access-to-build-data' into 'master' Douwe Maan2016-06-063-34/+138
|\ \ | |/ |/| | | | | | | | | | | | | Remove 'unscoped' from project builds selection This is a fix for this security bug: https://gitlab.com/gitlab-org/gitlab-ce/issues/18188 /cc @kamil @grzegorz @stanhu See merge request !1968
| * Remove 'unscoped' from project builds selectionTomasz Maczukin2016-06-043-34/+138
| |
* | Merge branch 'fix/rubocop-offense-in-specs' into 'master' Grzegorz Bizon2016-06-061-3/+2
|\ \ | | | | | | | | | | | | | | | | | | Fix rubocop offense in awardable specs Fixes failing tests on master. See merge request !4481
| * | Fix rubocop offense in awardable specsfix/rubocop-offense-in-specsGrzegorz Bizon2016-06-061-3/+2
|/ /
* | Merge branch 'rubocop/enable-ambiguous-operator-rubocop-lint' into 'master' Robert Speicher2016-06-054-8/+8
|\ \ | | | | | | | | | | | | | | | | | | | | | | | | Enable Lint/AmbiguousOperator rubocop cop Enforce using parentheses if ambiguous operators are detected, like `header *some_method` or `array.select &:even?`. See #17478 See merge request !4454
| * | Enable Lint/AmbiguousOperator rubocop coprubocop/enable-ambiguous-operator-rubocop-lintGrzegorz Bizon2016-06-034-8/+8
| | | | | | | | | | | | See #17478
* | | Fix CHANGELOG for 8.8.4 (unreleased)Robert Speicher2016-06-051-4/+2
| | | | | | | | | | | | [ci skip]
* | | Merge branch 'documentation-for-build-trace' into 'master' Achilleas Pipinellis2016-06-041-0/+24
|\ \ \ | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Document the API endpoint for gathering a build log ## What does this MR do? Adds documentation for gathering a trace build log via the API. Documenting the following: https://gitlab.com/gitlab-org/gitlab-ce/blob/de20bd5b31715f096db3fb0155c82b0eea992b6c/lib/api/builds.rb#L98-110 ## Are there points in the code the reviewer needs to double check? Please verify that a curl request using the same format as documented works ## Why was this MR needed? No documentation currently exists ## What are the relevant issue numbers? #18015 ## Screenshots (if relevant) See merge request !4467
| * | | Document the API endpoint for gathering a build logMark Fletcher2016-06-031-0/+24
| | | | | | | | | | | | | | | | * Resolves #18015
* | | | Merge branch 'fix-design-link-in-contributing' into 'master' Achilleas Pipinellis2016-06-041-2/+2
|\ \ \ \ | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Fix link to current design reference in the CONTRIBUTING guide ## What does this MR do? It fixes an outdated link by replacing it with the up-to-date one. See merge request !4439
| * | | | Fix link to current design reference in the CONTRIBUTING guidefix-design-link-in-contributingRémy Coutable2016-06-021-2/+2
| | | | | | | | | | | | | | | | | | | | Signed-off-by: Rémy Coutable <remy@rymai.me>
* | | | | Merge branch 'issue_13499' into 'master' Jacob Schatz2016-06-0421-133/+776
|\ \ \ \ \ | |_|_|_|/ |/| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Allow bulk-assign labels to issues - [x] Allow bulk-assignment labels. - [x] Show indeterminate state for labels present on selection. - [x] Remove existing labels from selected items if label gets unmarked. - [x] Fix conflicting tests. - [x] Write tests. Closes #13499 and #15489 See merge request !3917
| * | | | Remove extra bracketissue_13499Alfredo Sumaran2016-06-031-2/+1
| | | | |
| * | | | Fix specAlfredo Sumaran2016-06-032-7/+8
| | | | |
| * | | | Address feedbackAlfredo Sumaran2016-06-039-32/+26
| | | | |
| * | | | Update CHANGELOGAlfredo Sumaran2016-06-031-2/+1
| | | | |
| * | | | Spec label add / delete in UpdateServiceSean McGivern2016-06-032-9/+54
| | | | |
| * | | | Allow bulk-updating to remove all labelsSean McGivern2016-06-032-1/+9
| | | | | | | | | | | | | | | | | | | | | | | | | Instead of passing `remove_label_ids`, just pass an empty array for `label_ids` (and don't pass `add_label_ids` or `remove_label_ids`).
| * | | | Update CHANGELOGAlfredo Sumaran2016-06-033-2/+3
| | | | |
| * | | | EnhancementsAlfredo Sumaran2016-06-034-23/+25
| | | | |
| * | | | Update CHANGELOGAlfredo Sumaran2016-06-031-0/+1
| | | | |
| * | | | Fix specAlfredo Sumaran2016-06-031-8/+9
| | | | |
| * | | | Respond to .json onlyAlfredo Sumaran2016-06-031-1/+6
| | | | |
| * | | | Fix specAlfredo Sumaran2016-06-031-1/+1
| | | | |
| * | | | Add mising paramsAlfredo Sumaran2016-06-031-3/+6
| | | | |
| * | | | Fix failing specsAlfredo Sumaran2016-06-031-6/+8
| | | | |
| * | | | Bulk assignment testsAlfredo Sumaran2016-06-031-38/+120
| | | | |
| * | | | Remove console.logAlfredo Sumaran2016-06-031-1/+0
| | | | |
| * | | | Bulk assignment implementationAlfredo Sumaran2016-06-033-2/+111
| | | | |
| * | | | Tweaks for bulk assignmentAlfredo Sumaran2016-06-031-3/+15
| | | | |
| * | | | Expose Issue IDAlfredo Sumaran2016-06-031-1/+1
| | | | |
| * | | | Expose label IDAlfredo Sumaran2016-06-031-1/+1
| | | | |
| * | | | Add show_create param to toggle label creation optionAlfredo Sumaran2016-06-031-2/+2
| | | | |
| * | | | Add show_footer parameter for ability to hide footerAlfredo Sumaran2016-06-032-1/+3
| | | | |
| * | | | Do not trigger keyup event if we are persisting stateAlfredo Sumaran2016-06-031-3/+8
| | | | | | | | | | | | | | | | | | | | Triggering keyup will re-render the dropdown which is not needed when option persistWhenHide is true
| * | | | Remove unneeded call to keyup eventAlfredo Sumaran2016-06-031-3/+0
| | | | | | | | | | | | | | | | | | | | This was re-rendering the dropdown unnecessarily
| * | | | Only filter asynchronously if option remote is true.Alfredo Sumaran2016-06-031-11/+11
| | | | | | | | | | | | | | | | | | | | We need to update the results right away when we do bulk assignment.
| * | | | Fix statementAlfredo Sumaran2016-06-031-1/+1
| | | | |