Title: TeamPage 6.2.29

This Traction® TeamPage release is focused on a small set of bug fixes and improvements, particularly on reducing memory usage and improving performance when handling very large entries. If your TeamPage server tends to have many very large entries - each entry with enough text to fill dozens or hundreds of pages (or more) - you will definitely want to consider upgrading to TeamPage 6.2.29 so that you can benefit from these improvements. Please read on for the full list of changes.

Download TeamPage 6.2




Bug Fixes



Search



• Fixed a bug that prevented search results retrieved from a Solr service from including the correct amount of "snippet" text with each search result. This affects customers using TeamPage's Advanced Search module. (Solr1086)

• Fixed a bug that could cause the search specified for certain section definitions to be ignored. (Server89737)

• Fixed a bug that could cause type-ahead completion results suggested in the link dialog when completing by "Content" (rather than by "Title") to be scored incorrectly and possibly discarded. Some other minor errors and inconsistencies in scoring search hits to be displayed for type-ahead completions have also been corrected. (Server89708)

Other Fixes



• Fixed a series of minor issues that could cause slightly broken HTML to be generated in certain cases due to insufficient application of entity-encoding of certain special characters. This could result in minor display anomalies in the resulting pages, although we have not received any reports of such issues "in the wild." (Server89630 / Server89686)

• Fixed a bug that prevented custom column headings from being applied to Section Table widgets in certain cases. (Server89763)

Improvements



Handling of "Very Large" Entries



This release contains several under-the-hood changes that greatly improve TeamPage's handling of entries that have many dozens or even hundreds of pages worth of text.

• The way that private drafts ("Work in Progress") are stored has been changed. Listing, saving and loading private drafts is much more efficient. The reduction in memory usage and improved performance will be most noticeable for users who have one or more very large private drafts. (Server89359 / Server89781)

• The efficiency of "snippet" generation has been greatly improved. TeamPage now generates snippets much more quickly, and when displaying results from an external search engine, always uses the snippet provided with the search result. This affects any page using either "Snippets" or "Feed" volume, including, e.g., a page of search results from the advanced search feature, or any Activity > Feed view. The performance improvements here will be most noticeable when TeamPage is displaying snippets for very large entries. (Server89559 / Server89715)

Plug-ins



• Improved the performance of the "Task Table" view from the PM Custom Dashboard plug-in. (JPBO20576)

• Fixed a bug in the "Optionally Convert Incoming Text-Only Email Messages to HTML" plug-in that could cause the text-to-HTML conversion to be applied at the wrong times. (JPBO20591)

For Developers



• The com.traction.sdk.util.SearchResultHighlighter class has been split into multiple parts. SearchResultHighlighter is now an interface with a much simpler API, and the default implementation is in com.traction.sdk.util.StandardSearchResultHighlighter.

• The SDL truncate tag has been modified to include support for a token-sensitive version of the usual truncation behavior. This is intended to be used in place of the snippet tag when token-sensitive truncation is required and there's no text search hit highlighting to be performed. (Server89641)

• All com.traction.sdk.search.Document implementations -- com.traction.sdk.Entry, File and Attachment -- now support getter and setter methods for setting and retrieving the com.traction.sdk.search.SearchResult that was used to load it, when applicable, and also providing direct access to the snippet HTML provided by that SearchResult, when available. The SearchResult interface now contains the methods getSnippetHtml() and getSnippetText() that return the snippet in HTML or text format, and the searchresult.document.snippethtml or searchresult.document.snippettext SDL tags can be used to render these values in an SDL template.

• The new entry.snippet tag can be used to retrieve a snippet for the currently scoped com.traction.sdk.Entry. By default, this will either be a snippet generated by TeamPage using the user's preferred snippet size setting (or that of the current skin, if applicable), or the that came from the external search engine result which caused the Entry to be loaded. To generate a snippet from the match com.traction.sdk.Items of the current Entry using parameters of their own choosing, SDL designers can specify much the same attributes that the snippet SDL tag supports.

• The snippet, truncate and entry.snippet tags now all support the inputFormat=, outputFormat= and collapseWhitespace= attributes. The accepted values for inputFormat= and outputFormat= are "html", "text_only" and "text_only_with_entities". These govern, respectively, what transformation will be applied to the input in order to turn it into pure text, and what transformation will be applied to the output to make it suitable for the desired context. collapseWhitespace= may be set to "true" or "false" to indicate whether the text-only version of the input should have all runs of one or more whitespace characters collapsed to a single ordinary space character, which can often be desirable.



Related Articles
references (1)
Forum6840: PM Custom Dashboard plug-in
Article: Customer5104 (permalink)
Categories: :Doc:r62, :Doc:changelog
Date: August 23, 2018; 5:18:22 PM Eastern Daylight Time

Author Name: Dave Shepperton
Author ID: shep