Known and Fixed Issues

This page lists known and fixed issues for software releases of Site Search.

Note
For new features, see New Features.

December 21, 2018 (Version 1.3.5)

New known issues

There are no new known issues.

Bug fixes

  • Data source creation fails in some circumstances.

November 26, 2018 (Version 1.3.4)

New known issues

There are no new known issues.

Bug fixes

  • Request processing by the Search API doesn’t strip query filters (specified with the request parameter f) from the request.

  • When using the Push API, some fields in the resulting documents aren’t searchable.

September 20, 2018 (Version 1.3.2)

New known issues

  • In the result template editor, the id field isn’t present in the drop-down list of fields. You can add the field by entering id, and then clicking Add.

Bug fixes

  • For documents that you push to a Push Endpoint data source, Site Search no longer appends the string #-1 to the document IDs (the contents of the id field). Therefore:

    • If you display the document IDs in search results, they no longer have the string #-1 appended.

    • When using the Push API to delete a document, you don’t append the string #-1 to the document ID.

  • The result template editor now displays previews for documents in Push Endpoint data sources.

August 21, 2018 (Version 1.2.0)

New known issues

Following are new known issues in this release. Issues listed below for the March 6 and May 22, 2018 releases still apply.

Synonyms

Synonyms you specify are used for all search languages.

Facets

In the Advanced options for facet configuration, the Show More setting has no effect. If there is a Show More Label, then clicking that label always displays all facets.

Topics for a Search Box

If a topic specified for a Search Box doesn’t exist, then the Search Box won’t work.

More Like This results don’t respect topic limits that are specified for a Search Box module.

Document IDs for documents in a Push Endpoint data source

For documents that you push to a Push Endpoint data source, Site Search appends the string #-1 to the document IDs (the contents of the id field). For example, D237 would become D237#-1. `These are the effects:

  • If search results display the Document IDs, then the ID is displayed with the appended #-1.

  • When pushing document updates, you refer to the original IDs (without the #-1).

  • When deleting a document, you append the document ID with the #-1 to the Push API endpoint, with the # URL encoded as %23. For example, for a document with an id of D237, then the URL would end in D237%23-1.

    This is the syntax showing the push endpoint and the document ID:

    https://subdomain.lucidworks.cloud/pathname/api/v1/push/endpoint/id%23-1

    This is an example URL:

https://my-corp.lucidworks.cloud/fusion-search/api/v1/push/push-endpoint-prod/Rec10023%23-1

May 22, 2018 (Version 1.1.0)

New known issues

Following are new known issues in this release. Issues listed below for the March 6, 2018 initial release still apply.

User role

  • The Site Search app UI is now intended for use solely by administrators and account owners. The User role has been deprecated, but it is still present in the Site Search app UI. Don’t give a user the User role (by revoking the Admin role).

Embedded modules

If Spellchecking and Did You Mean are enabled, topic tabs don’t work in embedded result list modules.

Facets

Fields with all numeric content don’t work as facets.

Insights

  • All data is recorded and displayed in UTC time.

  • Insights charts show only the top 15 values.

  • Search query terms persist as you navigate the Admin UI. As a result, opening and closing the Admin mode and Editor windows can trigger searches that are reflected in the metrics.

  • “Top Queries” and “Top Queries that Gave No Results” are collected differently, so the counts might not be the same. This is caused by the persistence of search queries while performing administrative actions. The discrepancy is not caused by regular end user actions. In regular use, the percentage of administrative actions should be small enough to make the discrepancy negligible.

  • In the Results section of Insights, click-throughs via the name and click-throughs via the URL are treated as if they are separate documents.

  • In the Results section of Insights, if multiple web pages have the exact same document name, they are grouped together in the metrics.

  • In the Site Search app UI, right-clicking and opening a search result in a new tab/window is not tracked in Insights > Results. This is not an issue with embedded search results.

  • Pie charts of queries and results can be invisible. Whether this happens depends on the size of the browser window and on the length of chart labels. If a chart is missing, try resizing the browser window (both larger and smaller), and then refreshing the page.

  • The Most Popular Results chart isn’t visible if the name of a most-popular document contains a slash. The document is listed in the table.

API

If you don’t specify which fields to return in an API call, all fields are returned. To limit the payload of API calls, specify the fields to return using the fi request parameter.

March 6, 2018 (initial release)

Known issues

Following are known issues in this release.

Browser

  • Pages in Admin mode look best in browser windows of at least 1280 x 1024 pixels.

  • The IE11 browser might run slowly when performing administrative tasks. We suggest using Chrome or Firefox for administrative tasks.

  • In IE11 when running a search in Admin mode, you might have to triple click the search box to enter a search term.

Data sources

  • When creating data source, the focus begins on the start URL or upload file field. If you click off the field before entering an URL or file, you will get a warning message telling you that the field is required.

  • Data source topics support lower case a-z, 0-9, and underscore characters only. All other characters will be discarded. If you type a space. that saves the topic.

  • All topics appear in search results as tabs, regardless of whether any documents are associated with a topic. To remove a tab completely, delete the topic in the Topic manager. This removes the tab from all associated data sources.

  • If you map fields in a data source, the mapping might be disabled when you next edit the data source. If you ever edit a data source that has field mapping enabled, check that field mapping is still enabled before you save the data source configuration.

Indexing web pages

  • Crawling JavaScript on web pages is currently not enabled.

  • To crawl only the files in a site map, put the target website in the Start URL field, put the site map in the Site map URL field under Show More Options, and set the crawl depth to 1.

  • If you click the Display tab in a Web crawler data source while it’s trying to connect, you’ll see a message stating that your changes will be abandoned even though your settings are saved.

Indexing CSV and JSON documents

  • To update documents uploaded via CSV or JSON, each row or item must have a field called “ID” (or "id"; the case of field names doesn’t matter) so Site Search can match existing documents with new ones each time the file is uploaded.

  • If you upload a CSV or JSON file in which the same ID is listed multiple times, Site Search indexes the first instance of that ID for both initial and subsequent uploads.

  • When uploading a CSV or JSON file, if you don’t have the fields/properties name or description, then you might not initially see any search results. You must either map a fields/properties from your source file to name and/or description, or modify the result template that the data source uses to display a custom field.

  • If you upload another CSV or JSON file while one is being uploaded, then the first upload might not finish. Wait until a file is uploaded before uploading a new file.

Facets

  • When uploading JSON files, we recommend that all values be strings (in double quotation marks) if you want to use the values as facets.

  • Names of facet fields are not case sensitive, but values are. For example, “phone” and “Phone” are considered different facets.

Push Endpoint data sources

  • If you add a Push Endpoint data source, the data source is created even if you don’t explicitly click Save.

  • Push Endpoint data sources require the values of the ID property to be string, for example, “1” instead of 1.

Result templates

  • In result templates, the only way to reorder custom fields is to delete them, and then to add them back in the order in which you want them to be displayed.

  • When you add a custom field to a result template, the field name is added as well. To remove the name, click the pencil icon next to the field in the sample data, and then clear the "Label" value.

Synonyms

  • After editing synonyms, you must close the Synonym manager, and then reopen the Synonym manager for the synonyms to be reflected in search results.

  • If you edit a synonym, it is moved to the bottom of the list.

Embedding modules

  • Embedding modules doesn’t work for websites entered in the Embed website whitelist as IP addresses.

  • Embed website whitelists must start with http:// or https://.

More Like This

  • The More Like This module highlights the search term with <em></em>.

Fixed issues

Not applicable. This is the first release.