Uploaded image for project: 'Atlassian Intelligence'
  1. Atlassian Intelligence
  2. AI-658

Search Query to Return Result with Exact Page Name

XMLWordPrintable

    • Our product teams collect and evaluate feedback from a number of different sources. To learn more about how we use customer feedback in the planning process, check out our new feature policy.

      NOTE: This suggestion is for Confluence Cloud. Using Confluence Server? See the corresponding suggestion.

      It would be very helpful if the search query can return page(s) with the exact same title with the search query and excluding the page(s) with extra word before/after the queried phrase/word
      When searching a page with query title:"Development engineer", it should only returns the page:

      • Development engineer
        and NOT these pages:
      • Development engineer lead
      • Project Development engineer

            [AI-658] Search Query to Return Result with Exact Page Name

            Sherif Mansour added a comment -

            I think this is ultimately incorporated by CONF-10412. Marking it as tracked there and linking the two issues.

            Sherif Mansour added a comment - I think this is ultimately incorporated by CONF-10412 . Marking it as tracked there and linking the two issues.

            uvoellger added a comment -

            At least together with metadata, this feature would help to create awesome Confluence pages. Imagine Confluence is used to hold requirements for a software application. Some pages contain the roles used in this application, and other use cases. Now one could define which roles are used in which use case. At the end, it is possible to show all use cases related to a role on the role page by inserting a report with a query string similar to 

            page-type-untokenized:"Use case" and role-untokenized:"Development engineers"

            Without this feature, the whole approach does not work, as reports end with wrong results.

            This feature is also important when dealing with blue prints. Currently, a blue print page is considered as something that is independent from other pages, but I guess this will be changed dramatically in large environments. Not only the page is important but also the relation to other pages, and as relations are dynamic, they should be supported by exact reports.

            uvoellger added a comment - At least together with metadata, this feature would help to create awesome Confluence pages. Imagine Confluence is used to hold requirements for a software application. Some pages contain the roles used in this application, and other use cases. Now one could define which roles are used in which use case. At the end, it is possible to show all use cases related to a role on the role page by inserting a report with a query string similar to page-type-untokenized: "Use case " and role-untokenized: "Development engineers" Without this feature, the whole approach does not work, as reports end with wrong results. This feature is also important when dealing with blue prints. Currently, a blue print page is considered as something that is independent from other pages, but I guess this will be changed dramatically in large environments. Not only the page is important but also the relation to other pages, and as relations are dynamic, they should be supported by exact reports.

            uvoellger added a comment -

            We would like to see this feature not only for page titles (where is already exists as there is are hidden field "content-name-untokenized"), but also for page metadata. As page metadata are 3rd-party-plugins, we would not expect support of creation of according indexes. We can do this easily by ourselves.

            What we would like to see is support of untokenized query strings in the standard search UI as well as search macros (e.g. live-search).

            • Currently, all search terms in a query string are tokenized, because ConfluenceLuceneAnalyzerFactory tokenizes all fields except those in a hard-coded list, and
            • uses ConfluenceAnalyzer instances for field search which automatically performs a toLowerCase for query string field values.

            A possible solution would be to define a pattern for fields that are handled untokenized. The query would be "title-untokenized:"Development engineers"" or "page-type-untokenized:"Artifact page"" Plugin developer would be responsible to create according index fields to support untokenized search for their fields.

            Thanks,
            Uwe

            uvoellger added a comment - We would like to see this feature not only for page titles (where is already exists as there is are hidden field "content-name-untokenized"), but also for page metadata. As page metadata are 3rd-party-plugins, we would not expect support of creation of according indexes. We can do this easily by ourselves. What we would like to see is support of untokenized query strings in the standard search UI as well as search macros (e.g. live-search). Currently, all search terms in a query string are tokenized, because ConfluenceLuceneAnalyzerFactory tokenizes all fields except those in a hard-coded list, and uses ConfluenceAnalyzer instances for field search which automatically performs a toLowerCase for query string field values. A possible solution would be to define a pattern for fields that are handled untokenized. The query would be "title-untokenized:"Development engineers"" or "page-type-untokenized:"Artifact page"" Plugin developer would be responsible to create according index fields to support untokenized search for their fields. Thanks, Uwe

            uvoellger added a comment -

            Hello,

            from my point of view, I would expect exact results in case of queries like 

            title:"Development engineers"

            uvoellger added a comment - Hello, from my point of view, I would expect exact results in case of queries like title: "Development engineers"

              Unassigned Unassigned
              prompas Patrice Rompas (Inactive)
              Votes:
              1 Vote for this issue
              Watchers:
              3 Start watching this issue

                Created:
                Updated:
                Resolved: