v3.1.1

For use with Blacklab 3.x

Fixes

• fix document metadata when using the default xslt stylesheets

v3.1.0

For use with Blacklab 3.x

Features

• Enable more powerful search field customization. Insert a completely custom piece of HTML for a specific search field (or combination of fields). See the Readme for more information (Section Write your own (annotation) search fields)
• Extend customJs and customCss configuration options to allow multiple entries, and to allow specifying on which page the file should be loaded. See search.xml for usage.
• docker improvements

Fixes

• fix the link to a document's contents when running dev builds
• fix crash when trying to show document metadata without values

Miscellaneous improvements

• Number of words shown before/after hits can now be customized (only through customjs)

v3.0.2

For use with Blacklab 3.0

Features

• Docker support
• Debug mode in the frontend. Debug mode is enabled by default when serving javascript from webpack-dev-server, in production builds it can be enabled by entering debug.enable() in the browser console. It will render internal IDs instead of display names, and show the query parameters sent to BlackLab. When viewing a document, it will show all metadata, as well as a link to view part of the document (if allowed by BlackLab).
• Improved pagination when viewing long documents.
  Pagination can now be configured per corpus (via search.xml), instead of only globally.
• Result can now be sorted by clicking the column titles
• Results can now be sorted in ascending order (last results first).
• Autocomplete fields now only autocomplete after the last space you typed, unless you enter phrase mode by surrounding a group of multiple words with double quotes (").
• Allow customizing how a document's title is rendered; see readme
• Add ability to insert section titles in between filters. See readme
• Add ability to disable export buttons. See readme
• Add ability to customize the error message when no results could be found. See readme
• Improved date filter. See readme.
• Plugins for concordances. We use this for the audio player in our CGN corpus, as well as to make copying concordances into another tool easy.
• (optional) plausible.io integration

Fixes

• Part-of-speech/tagset now correctly handles values with special characters (.|\)
• QueryBuilder now also correctly handles values with special characters in dropdowns.
• Lexicon service now properly clears results when entering special characters
• Grouping on word before/after now only uses 1 word, instead of 5
• Fix loading indicater getting lost sometimes.
• Fix wrong tooltip value when hovering over the total results widget
• Fix exports missing metadata fields in some cases
• "Within" selector can now always be deselected.
• Filter out some illegal html characters in malformed documents so they no longer trigger an internal server error.

Miscellaneous improvements

• List of available corpora on the homepage is now sorted alphabetically
• Update the build tool versions
• When a document in an index can't be opened due to errors in the associated xslt, the error is now shown in the browser.
• Dropdown menus no longer run off the page
• Long lists of values in the filters sections can now be searched.
• Hide "contents" tab on document pagewhen contentViewable is false
• Add /config page, returns a JSON with some of the config parameters. May be useful for external tools to integrate better with corpus-frontend. For now it only returns the used blacklab server instance url.
• Filter dropdowns are now sorted alphabetically
• The breadcrumbs at the top of the results now show when sampling is enabled (settings->samplenum)

v2.1.0

For use with Blacklab 2.0

Features

• New word detail tooltips in documents! See the docs for how to configure them.
• Added option to set the displayed annotation in the left/hit/right context columns
• Added option to display context columns unescaped. This allows you to display and style xml to enable various inline features.
• Added option to disable/hide the query builder
• Added support for opening the dev version on other computers on the network

Fixes

• Fixed misplaced spinner while lexicon service is updating
• Fixed within never showing any available options (even when the corpus does contain inline tags!)
• Fixed disabling page size not working for links in the results table
• Fixed annotations without forward index being allowed in the results table (they would never show any data - as a forward index is required for that)
• Fixed inability to clear shown annotations/metadata fields shown in the result tables (using custom js)
• Fixed dropdowns failing to close properly in IE11

Miscellaneous improvements

• Added .article class to the main content container on the article page. This should help with article-specific styling
• Added the tab's name behind metadata fields in the summary at the bottom of the filters section
• Made document titles in the results table clickable across the entire width
• Added a search box to the format pickers shown when creating a new corpus and when downloading a preset
• Simple search field now has focus when the page is first opened
• In development mode, added a link to view a document's source xml at the bottom of the article page
• In development mode, also display annotation and metadata ids in addition to their name
• Various improvements to ease of use of the lexicon service
• Improved layout in the tables showing available corpora on the home page
• Page will now display first content faster in IE11
• Fixed navbar logo wrapping on to new line when the window is small

Removed features

• Removed the links next to hits in expanded hit rows. This feature did not work properly and will be revisited in the future.

v2.0.0 - The big one

Refer to the README on how to configure your corpus. The list of changes is extensive and it's probably best to redo your config.

v2.0.0-RC3.1

Due to some recent api changes in BlackLab to improve relative frequency calculations, this version will only work with a very recent version of BlackLab (commit INL/BlackLab@76f7e41 or newer).
Attached is the latest development build.

Features

• Explore view, containing corpora, n-gram, and statistics. These don't do anything new, but offer an alternative (simpler) way to get some results.
• A part of speech editor/picker
  Use uiType pos and add a tagset file, see the README for details.
• Different ways of viewing grouped results to show relative frequencies between groups
• The QueryBuilder now works with the select uiType for annotations/attributes
• Improved tooltips when viewing documents
• Support for paginating through long documents
• Support for importing and exporting queries
• Improved performance
• Support for playing audio fragments in concordances (see README for details)
• Support for gap-filling in the cql editor
• Support for viewing document statistics, such as annotation value distributions and growth charts (see README)
• Automated corpus creation and upload page (see README)

Fixes

• JQuery is once again available on window
• Opening a corpus from /corpus-frontend/corpora/ was broken.
• Link to default about page from a corpus
• RTL text improvements
• Attach controls to their labels so that clicking the label will activate the input
• Results are locked while a query is running to prevent flooding the server with heavy requests
• uiType combobox did not work for metadata

---

Update (RC-3.1): several IE related fixes.

v2.0.0-RC2

Works with BlackLab Server v2.0.0-RC1

• Fixed leading/trailing pipes in wordlist upload
• Fixed wordlist upload behaving differently in QueryBuilder and Extended views.
• Fixed build under windows

First release candidate for v2.0.0.

Major refactoring; application is based on Vue.js now. Tested to work with BlackLab Server v2.0.0-RC1 (attached). Please report any issues you experience with this preview.

v1.2 (BlackLab 1.7.2)

Breaking changes:

• Search.xml customCss and customJs now need to be prefixed with ${request:contextPath}/, they are no longer automatically prefixed with the context root. This means that including js and css on another server is now also possible!
• Search.xml NavLink relative attribute has been deprecated in favor of ${request:contextPath}.

Features:

• Allow setting the context size around hits in the results table
• Add support for case-sensitive grouping of results
• Add support for grouping properties in the simple search view (through annotationGroups in the corpus' BlackLab configuration)
• uiType for simple search fields (thanks @eduarddrenth!)
  This means you can now specify uiType combobox to enable autocompletion, or uiType select to show a field as a dropdown.
• Wide view option, for when the results table is cramped, further work will be done on how results are displayed in the future.
• There is now a help link when creating a new import format.

Bugfixes:

• CSV exports are now saved in UTF-8 properly
• Trailing/leading whitespace in cql queries is no longer transformed into empty tokens in the querybuilder
• No longer hide all your corpora when you're at the maximum (whoops!)
• Querybuilder can now load empty tokens (a [] cql query)
• Very long queries should no longer break everything
  A caveat of this is that the query is no longer in the url when it's very long, so you cannot share links for very large queries.
• Lemma popups work again for folia documents.

v1.1 (blacklab-server 1.7)

v1.1 release, coinciding with BlackLab 1.7