From Fedora Project Wiki

Revision as of 08:53, 1 September 2021 by Mfabian (talk | contribs)
(diff) ← Older revision | Latest revision (diff) | Newer revision → (diff)

Description

This is a test case to check ibus typing booster is working as expected.

Setup

Follow up to the steps in QA:Testcase i18n input method setup to configure the input method on the desktop in general.

  • Install ibus typing booster using following command
    dnf install ibus-typing-booster
  • Installing emoji-picker is optional, but this is a nice stand-alone tool to search for emoji by browsing emoji categories or typing search strings.
    dnf install emoji-picker
  • Add the “Typing Booster” input source. Open gnome-control-center -> then on left side you click on "Keyboard" -> then on right side you you click on the + below the list of input sources -> In the dialog which pops up, select vertical 3 dots at the bottom. This is because only input methods for the most common languages are shown at the top. To browse the full list, you click on the 3 vertical dots. Then in the search box you type "Typing booster", You will find "Other". click on "Other". Then on "Other (Typing Booster)". Then on "Add" button. Now you should have an entry “Other (Typing Booster) 🚀” in the gnome panel.
  • You may also want to install hunspell dictionaries for your language because ibus-typing-booster can use them if they are installed. To install dictionaries for your language use
    dnf install hunspell-??
    where ?? is the code for your language. Alternatively you can skip installing the dictionaries in advance and test whether installing them later in the setup toool of Typing Booster using the “Install missing dictionaries” button works 😃

How to test

  • Typing Booster testing:
    • Select the (Typing Booster) input source in the Gnome panel
    • Basic language setup.:
      • You can select the settings entry for (Typing Booster) in the panel now. This starts the setup tool.
      • Check whether the default setup for the current locale looks reasonable.
      • You may try to add more or different dictionaries and input methods.
        • If dictionaries are marked as missing with “❌”, you can try the button “Install missing dictionaries” which should install all missing dictionaries (as long as a suitable hunspell-?? package exists on Fedora)
        • Good opportunity for a bonus test for the m17n input methods here: When adding different input methods, scroll through the full list and see if you can spot any “minput_open_im() failed ❌” error messages. If not, that’s good because it means that all m17n input methods can be successfully opened 👍🏼
    • Basic typing test:
      • Type something. A list of completion candidates should pop up.
      • You can select and commit a candidate by typing the number displayed in front of the candidate.
      • You can also select a candidate using the mouse.
      • Or you can select a candidate using arrow-up/arrow-down and commit with the space key.
      • Or don’t select a candidate at all and commit the preëdit (the typed string) using the space key.
    • Emoji typing test:
      • Quick emoji-mode toggling:When typing something like cat_ i.e. a word which matches an emoji followed by an underscore, you should see some cat emoji (😺 or similar) among the candidates. Scroll down the candidate list to find more emoji, right click with the mouse on an emoji candidate to see similar emoji. Right clicking on 🐱 should give something like 🐈🐈‍⬛🐶🐹🐰🦊…
      • This should work in almost all languages, for example if you are using typing-booster with a Spanish dictionary, gato_ should give you cat emojis 😸.
      • If you want to get emoji matches without having to type _ you can switch the option “☑️ Unicode symbols and emoji predictions“ on in the setup tool. Note that this slows down typing-booster a lot, therefore this option is off by default.
      • The emoji input in Typing Booster can actually be used to input arbitrary Unicode symbols. For example typing “integral_volume_” should show something like ∰ and maybe similar symbols as a candidate.
    • Compose input testing:
      • Choose a Compose key (Video showing how to choose a Compose key in Gnome3)
      • Type compose F ., this should give “Ḟ”.
      • Type compose F Tab, possible completions for the unfinished compose sequence should be shown as candidates, in this case something like Ḟ, 🖕, ffi, ffl, ₣. The first column in the candidate lists shows the keys which would have finished the compose sequence to get that candidate.
        • You can choose a candidate as usual or close the candidate list by typing Escape and finish the compose sequence by normal typing (Escape needs to be pressed twice if you selected something in the candidate list already)
  • Emoji picker testing:
    • Start
      /usr/bin/emoji-picker
      to test the emoji browsing tool.
    • You can left click on a emoji to get it into the cut and paste buffer and then paste them into some program like gedit.
    • You can right click on a emoji to get more information about that emoji.
    • On emoji which support skin tone variants,for example 👧, you can middle click or long press with the left mouse button to get the skin tone variants.
    • If an emoji is matched, it usually appears at the bottom of the candidate list (unless you typed that emoji already before, then it may get scored higher). If you see an emoji at the bottom of the candidate list and this is not the one you wanted, you can scroll the candidate list down to the next page(s), there will probably be more similar emoji. Or you can click the any emoji candidate with the right mouse button, this will give you similar emoji.

Expected Results

  • New words are getting added to the user database after commit, therefore one should notice that the system becomes more intelligent while typing because it remembers which words appear in which context
  • To speed up the learning process, one can use the ibus-typing-booster setup tool to read some text files