Dev/test acceptance #1023
Closed
Dev/test acceptance #1023
Conversation
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
- move directanswers markup and script to common-partials folder - update vertical standard templates to include direct answer - update direct answer base card component to include the correct searcher (Universal or Vertical) - add and comment out directanswer config for vertical standard page NOTE: this requires [changes in SDK](yext/answers-search-ui#1596) to work J=SLAP-1674 T=manual change people vertical page to vertical-standard with directanswer configs/templates uncommented. serve test-site, and perform 'bryan phone number' on universal page and people page. See that direct answers appear for both page. When click thumbs up, see that the appropriate analytics payload is send in the request.
- add commented out `environment` in global_config.json J=SLAP-1663 TEST=manual see the environment is change based on value in global_config.json
This allows users to call window.AnswersExperience.init() from within the on-document-load partial.
I added an error message if you try to call window.AnswersExperience.init() before ManualInitializer.setup()
has been called.
While testing, I noticed a bug with our script/* partials that get inserted into inline js.
If the last line of one of these partials is a single line js comment, then the line immediately following
the partial in the template would also be commented out
i.e., the below template
```hbs
{{#*inline 'script-partial'}}// I will comment out the line after me too{{/inline}}
function () {
{{> script-partial }}
}
```
gets the output
```js
function () {
// I will comment out the line after me too}
```
You can easily test this out in the handlebars playground or in a local node script.
To fix this I added new lines to the end of script/* partials that would be affected by this.
J=SLAP-1673
TEST=manual
tested calling window.AnswersExperience.init inside on-document-load with initializeManually set to true/false, and doing an end-to-end with consumer auth
The API is allowed to send an undefined cta linkType. Setting the default to URL effectively grabs the link with no extra formatting. J=SLAP-1655 TEST=auto
Change the image formatter to use dynamic thumbnailer instead of expecting an array of thumbnails. A thumbnail array will no longer be provided after the streams migration. Dynamic thumbnailer uses the host `dynl` for images at least as large as the specified size (the smallest image with at least one dimension greater than or equal to the specified values). The host `dynm` gives the largest image that is smaller than the specified values in both dimensions. Note: for `dynl`, desired size values of 1 or smaller are a special case which indicates that dimension should be ignored and the returned thumbnail should be larger than the other dimension. J=SLAP-1670 TEST=auto, manual Check that the current profile images still appear as before and those in the streams test account now appear as well.
Update testing infrastructure to CaC approach for percy and wcag tests. - added `PageOperator` which is responsible fo automating browser navigation using the given test locations - test locations provide information on what page to navigate to and what action/query to perform on that page - update wcag and percy index.js file to use PageOperator - added some snapshots for es/ar pages that wasn't there before - remove photographer, multilangphotographer, and wcagreporter class as they are no longer needed with pageOperator in use J=1657 TEST=auto see that wcag tests run successfully on local machine see that percy snapshots are taken properly in this pr
Merge master (v1.26.0) into develop
Change the default universalLimit for all vertical page-configs to 4. J=SLAP-1681 TEST=manual Smoke test in test-site and see that faq universal results are limited to 4 when uncommented.
J=SLAP-1678 TEST=manual built vertical-grid and vertical-standard page in english checked spellcheck looks as expected for mispelled searches
- perform full HTML escape on the constructed highlighted text (will transform ", ', &, <, and >) - `substr` is deprecated, replaced with `substring` J=SLAP-1703 TEST=manual & auto update bryan's entity description to include script tags, search for 'where is joe exotic' and see that the script tags appear. (without the replacement, they do not appear) see added jest tests passed
- Add github action to create a pr with version update commit to a release/hotfix branch on create. J=SLAP-1356
- update escape-html to be devDependency in top level package - add escape-html as depeendency in static package
Change styling of `yxt-FilterOptions-options` to remove scrollbars that were appearing on filters even when all options were displayed for some cases with custom font sizes. Add a `--yxt-filters-and-sorts-line-height` variable so that customers can change the line-height of the `optionLabel` easily if the scrollbar appears with even larger custom font sizes. J=SLAP-1727 TEST=visual See that filter scrollbar was removed on a site where it was previously appearing.
- The create event does not support branch filter and tag filter so this would run on any branch creation. Update to trigger on push from release or hotfix branch. Then check if version is updated before making a new branch for pr J=none TEST=manual test in dummy repo, see version update on hotfix/release branch, but not in dev branches
J=SLAP-1678 TEST=manual rebuilt test site, looks as expected
…ight`. (#1019) The variable name was changed from `--yxt-filters-option-label-line-height` to `--yxt-filter-options-option-label-line-height`. TEST=none Simply did a find/replace all.
Fix font preloads on multilang sites and no longer generate a hash for fonts When webpack was processing the preload tags to update them to use the hash, it was dropping the relative path which jambo was adding which caused the preloads to point to invalid links on multilang sites. I couldn't find an option to adjust the path names after webpack processed it, so the simpler solution is to remove the hashes from the fonts. That way webpack will no longer need to update the font preloads and we can instead rely on the relative path provided by Jambo. There was an additional bug where we had an extra dot in the font names. (e.g. source-sans..woff), and this change fixes that as well. Because we are removing the hash, that makes it more difficult to cache bust it. If someone uses a new font file and names it exactly the same as one of the previous fonts, they cache will return the old asset. The current font names are so specific that I find this unlikely to be a problem (e.g. source-sans-pro-v14-latin-300.woff). J=SLAP-1724 TEST=manual Check the network tab for font loads on english experiences and on multilang pages. Check both universal and vertical pages and observe that the error in the console about invalid font preloads is gone for multilang sites. See that the loaded font no longer has an extra dot before the extension.
Bump version to v1.27.0 J=none TEST=none
Previously, if a `universalLimit` were specified for a vertical, we would perform client-side filtering. Instead, we should be creating a `search.universalLimit` object and passing that to `ANSWERS.init`. Not only does this remove responsibility from the front-end, but it allows the HH to display more than 10 results per vertical on the Universal page. This PR adds a `getDefaultUniversalLimit` helper that iterates through the vertical page configs and uses the `verticalsToConfig.universalLimits` to build a `search.universalLimit`. J=SLAP-1749 TEST=manual Tested the following scenarios and verified correct behavior in each: - No `universalLimit` values are set and there is no limit specified in the `global_config` or `pageSettings`. - Some `universalLimit` values are set, but not others. There is no universal limit in the `global_config` or `pageSettings`. - Some `universalLimit` values are set, but a universal limit is also present in the `global_config`.
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
No description provided.