Closed gunnarhj closed 1 year ago
Initially we do not observe this problem (Debian Sid):
BUT after a reboot, same problem appeared.
Also, this seems to be hard-coded. Even after I repeatedly selected "北京“, "beijing" always comes out first.
On the other hand, entering "bj", "北京" would come out as the fist choice after two tries.
Same thing happens to shanghai (上海), but apparently not to other major cities including 深圳 and 广州。 Their Chinese names always come out as the first choice after entering their pinyin.
Similar to "beijing"/"bj", I can train "shh" to move “上海” to the first choice, but never “shanghai”。
Please turn off the English Candidate option in the User data page of the ibus-libpinyin setup dialog.
Please turn off the English Candidate option in the User data page of the ibus-libpinyin setup dialog.
Actually this is a pretty handy option, really there is no need to turn off. I would also like to thank Gunnar for "discovering" this hidden treasure. Pretty neat. Please see the following screenshot:
In fact, as a native Chinese user, I don't think I would ever input the whole pinyin "beijing" to type out “北京”. Always "bj". But I would appreciate if the first letter in Beijing and Shanghai can be capitalized.
Again, many thanks to Gunnar for discovering this hidden treasure.
Are you saying "not a bug, it's a feature"? :)
Yes, I can train it to show 北京 first when typing bj
and 上海 when typing shh
. And I can get rid of English suggestions by unchecking "English Candidate".
I see now that "English Candidate" is an option in a new "Input Modes" group of options, and that it was intruduced in commit 276c9439. So, if I understand it correctly, the changes give the users better possibilities to control the behavior. This sounds indeed as an improvement.
Thanks for clarifying! Leaving it to you to decide when this issue should be closed. Possibly it may serve as a reminder for some kind of further improvement.
Are you saying "not a bug, it's a feature"? :)
(Definitely a) Yes! Since this option can be turned off, it is definitely not a bug. And since it provides great conveniences for commingled English/Chinese writings, it is definitely a feature. An improvement!
I'm sure your knowledge of Chinese is much better than my knowledge of Swedish (which is zero), but for most native Chinese speakers, when we use the pinyin class of input methods, we almost only use combination of consonants to output words (词)or even an entire sentence (this makes Chinese pinyin a very fast input method). A Chinese so-called "word" typically comprises a plurality of Chinese characters (字). I guess only elementary school students would spell out the entire pinyin such as “beijing”, to output “北京”. But they are doing that mainly to learn pinyin (i.e., the correct pinyin of Chinese characters).
But there are definitely rooms for improvement, especially since this is only the first step in officially using an English dictionary for the ibus-libpinyin input engine.
Version: 1.13.0 Distro: Ubuntu 22.10
Forgot to say thanks Gunnar, for adding ibus-libpinyin 1.13.0 to the Ubuntu/Debian repositories. Debian Sid is a rolling release, but Ubuntu 22.04LTS users may need to upgrade to 22.10.
Also, this seems to be hard-coded. Even after I repeatedly selected "北京“, "beijing" always comes out first. ... I can train "shh" to move “上海” to the first choice, but never “shanghai”。
Is that because beijing
and shanghai
are present in data/wordlist
?
Thanks for the report!
I think we can remove "beijing", "shanghai" and other words from the data/wordlist
file to fix this issue.
Or we can move the English candidates after the sentence candidates. The English candidate can appear in the 2-4 candidate position.
I would wait until more user inputs are collected. But at the present time, is there any way to edit the English dictionary?
Don't know what I did but typing beijing “北京” becomes the first candidate. Ditto shanghai:
Debian Sid; ibus-libpinyin v. 1.13.0
is there any way to edit the English dictionary?
One way to edit the English dictionary is via the "v" operator. In the following example, I have added Beijing (first letter capitalized) to the user dictionary:
Further, some observations:
"beijing" is not an appropriate word, unless its first letter is capitalized. Thus, this is a bug, and it should be removed from the dictionary. Ditto "shanghai".
As shown in the attached screenshot, typing the correct spelling, "Beijing" (or simply the letter "B"), "Beijing" will appear as the first candidate.
Even for esoteric words (such as "Hjalmarson" :smiley: ), it's easy to add them to the ibus-libpinyin dictionary (only one training is required). (Houston is my home town, I am using it so often it's difficult to knock it down from the first candidate. )
I can even add "LibreOffice" to the dictionary; it's been a pain in the 8th to type "LibreOffice" with a capital O in the middle.
I moved the English candidates after the sentence candidates in ibus-libpinyin 1.15.0 .
Version: 1.13.0 Distro: Ubuntu 22.10
If I type
beijing
, the first candidate is just 'beijing', and '北京' is only the second candidate.In 1.12.1 '北京' shows up as the first candidate.