<component type="inputmethod">

  <id>typing-booster.xml</id>

  <metadata_license>CC0-1.0</metadata_license>

  <name>Typing Booster</name>

  <summary>Predictive input method</summary>

  <description>

      Ibus-typing-booster is a predictive input method and predicts

      complete words based on context and partial input.  One can then

      simply select the desired word from a list of suggestions and

      improve one’s typing speed and spelling.

  </description>

  <url type="homepage">https://mike-fabian.github.io/ibus-typing-booster/</url>

  <url type="bugtracker">https://github.com/mike-fabian/ibus-typing-booster/issues</url>

  <url type="help">https://mike-fabian.github.io/ibus-typing-booster/documentation.html</url>

  <releases>

    <release version="2.1.0" date="2018-07-18" urgency="medium">

      <description>

            emoji-picker: Show a concise description of a selected

            emoji in the header bar

            Update the setup UI when settings are changed outside of

            the setup UI

            Migrate IBusConfig to GSettings (The old settings are

            unfortunately lost, so one has to open the setup tool

            and recreate ones favourite settings).

            Read emoji data files always in UTF-8

      </description>

    </release>

    <release version="2.0.2" date="2018-06-27" urgency="medium">

      <description>

            Better Tab handling, use Tab to switch to the next

            candidate, not to commit

      </description>

    </release>

    <release version="2.0.1" date="2018-06-21" urgency="medium">

      <description>

            Update translations from zanata (ja updated)

            Update emoji annotations from CLDR

            Fix some bugs in the usage of “prefix” for prefixes other

            than “/usr” (For FreeBSD)

            Make itb_util.get_ime_help() work on FreeBS

            Update UnicodeData.txt to Unicode 11.0.0

            Remove useless 't-nil vi-base': 'vi-base.mim', from

            M17N_INPUT_METHODS

            Remove extra space in entry for sa-IAST input method to

            make it work

            Show in the setup tool in the input listbox whether

            minput_open_im() succeeded.

            Use the rocket icon emoji_u1f680.svg from the “Noto Color

            Emoji” font

      </description>

    </release>

    <release version="2.0.0" date="2018-05-28" urgency="medium">

      <description>

            Update translations because of the merge of the engines

            (de, pl, uk updated)

            Update emoji annotations from CLDR

            Do not hardcode icon names in desktop files (Resolves:

            https://github.com/mike-fabian/ibus-typing-booster/issues/17)

            Change the default for “Unicode symbols and emoji

            predictions” to “False”

            Merge all typing-booster engines into one

            Change the UI of the setup tool to make it possible to

            select multiple input methods and dictionaries

            Move the buttons to learn from a file and to delete

            learned data to the options tab

            Use the same “About” dialog in the setup tool as in emoji-picker

      </description>

    </release>

    <release version="1.5.38" date="2018-05-22" urgency="medium">

      <description>

            Refresh french translation (thanks to Thierry Thomas)

            Mark comments in the emoji-picker about dialog as translatable

      </description>

    </release>

    <release version="1.5.37" date="2018-05-14" urgency="medium">

      <description>

            Update translations from Zanata (pl and uk updated)

            Make “Add direct input” option work correctly when

            “Remember last preedit input method” option is off

            Fix test case for Korean

            Rewrite setup UI completely in Python, without using Glade

            Fix format string in debug message when a dictionary .aff

            file has no encoding (Resolves: rhbz#1575659)

            Return False in read_training_data_from_file() if file

            cannot be opened

      </description>

    </release>

    <release version="1.5.36" date="2018-04-10" urgency="medium">

      <description>

            Make the default for

            self.show_status_info_in_auxiliary_text False (Resolves:

            rhbz#1564354)

            Adapt hunspell_suggest.py to work with pyhunspell 0.5.4

      </description>

    </release>

    <release version="1.5.35" date="2018-03-07" urgency="medium">

      <description>

            Update translations from zanata (es, pl, and uk updated)

            Update UnicodeData.txt to UnicodeData-11.0.0d13.txt

            Read also the emoji names from the emoji-test.txt file

            Update Unicode emoji data to a prerelease of Unicode Emoji Data 11.

            Fix PyGTKDeprecationWarning: Using positional arguments

            with the GObject constructor has been deprecated.

            Add “Twemoji” as a good colour emoji font to the emoji-picker font list

            Don’t show the languages en_001 and  es_419 in the browsing treeview

            Use romaji=True by default in EmojiMatcher

            Update emoji annotations from CLDR

      </description>

    </release>

    <release version="1.5.34" date="2017-10-04" urgency="medium">

      <description>

            Update translations from zanata (cs new, de updated)

            Add some tooltips

            Add an option whether to use pango font fallback to

            emoji-picker

            Update emoji annotations from CLDR

      </description>

    </release>

    <release version="1.5.33" date="2017-09-11" urgency="medium">

      <description>

            Update translations from zanata (es new)

            Install appstream metadata to /usr/share/metainfo/ 

            Update UnicodeData.txt to Unicode 10.0.0

            Fix test cases and kakasi support for the update of the

            emoji annotations from CLDR

            Update emoji annotations from CLDR

            Skip the emoji which already have skin tone modifiers in

            itb_emoji.emoji_by_label()

            Never load characters of Unicode categories “Cc”, “Co”,

            and “Cs” into the emoji dictionary

            Update emoji-data.txt to 5.0

      </description>

    </release>

    <release version="1.5.32" date="2017-04-24" urgency="medium">

      <description>

            Fix error when starting emoji-picker when the

            “recently-used” does not yet exist

            Update emojione.json to version 3.0

            Add the data from CLDR common/annotationsDerived

            Load also the CLDR annotations from “annotationsDerived”

      </description>

    </release>

    <release version="1.5.31" date="2017-04-18" urgency="medium">

      <description>

            Rename option --use_vs16 to --non_fully_qualified

            (effectivly reversing the default)

            Make description labels in info popover selectable to be

            able to copy and paste their contents

            Sort similar emoji with the same number of matching labels

            by cldr_order distance

            Map cldr subgroup 'person-sport' to emojione category

            'activity'

            Make the categorie listings and the search work right when

            using --use_vs16

            Always store only non-fully-qualified emoji or

            emoji-sequences in the internal dictionary

            Update emoji annotations from CLDR

      </description>

    </release>

    <release version="1.5.30" date="2017-03-29" urgency="medium">

      <description>

            Update translations from zanata (de, pl, uk updated)

            Use string order as a fallback to cldr_order in category listings

            The rainbow flag should be a zwj sequence

            Also display the Unicode version in the emoji info popover

            When looking up emoji or other characters via Unicode

            codepoint, ignore surrogates and private use characters

            Show the fonts really used to render an emoji in the info

            popover for the emoji

            Fix typo in translatable string

            itb_emoji.py: Use CLDR order to sort the candidates and

            the similar emoji if score is the same

      </description>

    </release>

    <release version="1.5.29" date="2017-03-21" urgency="medium">

      <description>

            Update translations from zanata (de, pl updated)

            Add a “--version” command line option to emoji-picker

             itb_emoji.py: Also read emoji-test.txt (from unicode.org)

            emoji-picker: Set default font to first available in ['Noto Color Emoji', 'Emoji One', 'Symbola']

            Small parsing improvement of emoji-sequences.txt

            Add support to either use U+FE0F VARIATION SELECTOR-16 in

            emoji sequences or not

            emoji-picker: Show “∅ Search produced empty result.” when

            nothing matches in a search

      </description>

    </release>

    <release version="1.5.28" date="2017-03-18" urgency="medium">

      <description>

            Allow query by code point even if Python’s

            unicodedata.name() does not know the character

            Also read names from emoji-sequences.txt and

            emoji-zwj-sequences.txt

            itb_emoji.py: Also read emoji-sequences.txt (from unicode.org)

            Fix positioning of info popover (fix a typo in an “if”

            statement)

            Show emoji properties from unicode.org when debugging is on

            itb_emoji.py: Also read emoji-zwj-sequences.txt (from

            unicode.org)

            Also use the emoji properties from unicode.org to decide

            whether to offer a lookup on emojipedia

            Use property “Emoji_Modifier_Base” from emoji-data.txt to

            check whether an emoji supports skin tones

            itb_emoji.py: Also read emoji-data.txt (from unicode.org)

            Tentative skin tone support for families

            Improve skin tone support: make it work for professions

            (roles) as well

            Make skin tone popover scrollable and limit its maximum

            size

      </description>

    </release>

    <release version="1.5.27" date="2017-03-17" urgency="medium">

      <description>

            Update translations from zanata (pl, uk updated)

            emoji-picker: make skin tone selection work for gendered emoji

      </description>

    </release>

    <release version="1.5.26" date="2017-03-16" urgency="medium">

      <description>

            Update translations from zanata (de updated)

            Fix display of warning message when a dictionary is not installed.

            Emulate xdg.BaseDirectory.save_data_path() on systems

            which lack pyxdg

            Show the skin tone popover also on a long press gesture

            Fix pyhunspell support Resolves:

            https://github.com/mike-fabian/ibus-typing-booster/issues/5#issuecomment-286251818

      </description>

    </release>

    <release version="1.5.25" date="2017-03-13" urgency="medium">

      <description>

            Update translations from zanata (de, pl, uk updated, zh_CN new)

            Show the categories as well on right mouse click in

            emoji-picker

            Improve information displayed on right mouse click in

            emoji-picker

            html.unescape() the strings parsed from the cldr annotations

            Fix fontsize change for invisible emoji in browse flowbox

            Add an option whether the arrow keys are allowed to reopen

            a preëdit

            Add an option to work around the broken

            forward_key_event() in the Qt 4/5 im module

            Use xdg.BaseDirectory to add a USER_DATADIR to the search

            path for data for itb_emoji.py

            emoji_picker.py: Speedup: Fix wrong indentation of block

            in _fill_flowbox_browse()

            emoji-picker: Do not override the decoration layout of the

            header bar

            When an emoji with a different skin tone is selected,

            replace the original emoji immediately

            Make emoji-picker work on dark themes like Adwaita-dark as well

      </description>

    </release>

    <release version="1.5.24" date="2017-03-06" urgency="medium">

      <description>

            Update translations from zanata (de updated)

            The spin button to change the fontsize should grab focus

            without selecting

            emoji-picker: Don’t check if ibus is running, ibus does

            not need to run for emoji-picker

            Load .desktop files for emoji-picker and

            ibus-setup-typing-booster correctly under Gnome Wayland

            Show the most recently used skin tone by default

            Use Popovers for skin tones

            Don’t use HeaderBar with the default widget titlebar

            Display the detailed information of an emoji as a popover,

            not as a tooltip

            Use CSS to show light gray borders around flowbox and

            listbox children

      </description>

    </release>

    <release version="1.5.23" date="2017-02-27" urgency="medium">

      <description>

            Update translations from zanata (de, ja, pl, uk updated)

            emoji_picker.py: Add a menu button to change the font for

            the emoji

            emoji-picker: Make background colour of the flowbox

            listing the emoji white

            emoji-picker: Use “Symbola” as the default font

            Remember the font and the fontsize in a config file

            emoji_picker.py: Add a spin button to change the font size

            of the emoji

            Add option to the emoji-picker to load *all* Unicode

            characters

            UI redesign of the emoji-picker

            Return an empty list immediately if candidates() is called

            with an empty search string

            Make the fontsize for the names of the emoji in the search

            results smaller

            Make the search in emoji-picker a bit more responsive by

            using GLib.idle_add()

            Save the recently used emoji immediately, not only when

            the program quits

            Set default font size of emoji-picker to 24 instead of 16

            Set the emoji font only for the emoji, not for its name in

            the search results

            Set WM_CLASS of emoji-picker and ibus-setup-typing-booster

            correctly

            Add “Icon” and “Categories” to emoji-picker.desktop

      </description>

    </release>

    <release version="1.5.22" date="2017-02-21" urgency="medium">

      <description>

            Update translations from zanata (de, fr, pl, uk updated)

            Add an emoji-picker

            Update of en.xml from CLDR’s emoji annotations

            Fix skipping of the Korean test case when no Korean dictionary can be found

            Fix invalid xml in typing-booster.appdata.xml

      </description>

    </release>

    <release version="1.5.21" date="2017-02-07" urgency="medium">

      <description>

            Handle Return and Enter correctly when the cursor is not

            at the end of the preëdit (Resolves: rhbz#1418313)

            Values of spin buttons should not be translatable

            Make the categories from emojione translatable

            Make emoji matching accent insensitive

            If available use pykakasi to convert Japanese emoji

            category names to hiragana

            Add some Japanese translations as a test case for pykakasi

            If available use the “pinyin” Python module to add pinyin

            to the Chinese names and keywords

            Don’t fallback to “zh” from “zh_TW”, “zh_HK”, “zh_MO” and

            “zh_Hant”

            Don’t sort the labels when listing similar emoji

            Don’t change Unicode categories to lowercase when loading,

            use the original case

            Also treat categories 'Zl' and 'Zp' as invisible and add

            Unicode code point

            When searching for similar emoji, the original emoji

            should be most similar to itself

            Fix duplicate listing of labels when looking up similar

            emoji

            Make it optionally possible to match emoji in Japanese using romaji

            itb_emoji.py: Add the code point to the name of invisible

            characters also when looking up similar characters

            Better matching of the Unicode categories

            Small performance optimization in EmojiMatcher.similar()

            Remove any U+2028 LINE SEPARATOR and U+2029 PARAGRAPH

            SEPARATOR characters from the lookup table

            Nicer display of the matching labels when looking up

            similar emoji

            Don’t strip mathematical symbols (category 'Sm') from

            tokens

            Update of en.xml from CLDR’s emoji annotations

            Update translations from zanata (de, pl updated)

      </description>

    </release>

    <release version="1.5.20" date="2017-01-25" urgency="medium">

      <description>

            Calculate the maximum word length for each dictionary

            individually

            Use .startswith instead of regexp matching when matching

            in hunspell dictionaries (speed optimization)

            Improve accent insensitive matching (“filosofičtějš”

            should also match “filosofičtější”)

            Some updates for the emoji annotations in en.xml from CLDR

      </description>

    </release>

    <release version="1.5.19" date="2017-01-19" urgency="medium">

      <description>

            Improve setup layout (thanks to Trinh Anh Ngoc <atw1990@gmail.com>)

            Add some more directories to search for dictionaries (for FreeBSD)

            Wrong variable “page_size” was used in set_lookup_table_orientation()

            Do not try to reopen the preëdit when any modifier except

            CapsLock is on (Resolves: rhbz#1414642)

      </description>

    </release>

    <release version="1.5.18" date="2017-01-17" urgency="medium">

      <description>

            Fix typo in the “Unbreak sqlite on Python 3.6” patch

            Fix the fallback to use pyhunspell-python3 when

            python3-enchant is not available

            Remove useless ibus-typing-booster.pc

      </description>

    </release>

    <release version="1.5.17" date="2017-01-13" urgency="medium">

      <description>

            Update py-compile to current upstream version

            Also use ＿ U+FF3F FULLWIDTH LOW LINE as a separator for

            emoji keywords

            Unbreak sqlite on Python 3.6 (thanks to Jan Alexander Steffens)

            Return immediately if _update_candidates() is called with

            empty input (Resolves: rhbz#1413082)

      </description>

    </release>

    <release version="1.5.16" date="2017-01-10" urgency="medium">

      <description>

            Skip Korean test case also on Arch

            Remove everything following a tab (including the tab) from

            hunspell dictionary lines (Resolves: rhbz#1411659)

            Delete a candidate correctly from the user database even

            if it starts with a prefix to be stripped from tokens

            (Resolves: rhbz#1411676)

            Trigger emoji lookup when the input starts or ends with

            '_' or ' '

            Better handling of BackSpace and Delete when reaching the

            ends of the preëdit (Resolves: rhbz#1411688)

            Search for hunspell dictionaries in a list of directories

            (Resolves:

            https://github.com/mike-fabian/ibus-typing-booster/issues/6)

      </description>

    </release>

    <release version="1.5.15" date="2016-12-08" urgency="medium">

      <description>

            Default value for self._show_status_info_in_auxiliary_text

            should be True

            Don’t use keyword arguments when instantiating

            IBus.LookupTable()

            Add an option to choose the orientation of the lookup table

            Update translations from zanata (de, pl, and uk updated)

            Update emojione.json

      </description>

    </release>

    <release version="1.5.14" date="2016-11-25" urgency="medium">

      <description>

            Reopen preëdit not only on Backspace but also on Delete

            and arrow keys