Autocomplete Items Order Issue
-
@valery-bogomolov said in Autocomplete Items Order Issue:
Thank you for the great work
Assuming that means is sorts as you expect now?
-
This post is deleted! -
@michael-vincent Ok, now it works. I compared it with the old autocomplete version and the results seem to be the same.
Only
fx
icon is everywhere. There is afunc="yes"
, orfunc="no"
attribute in XML that describes the KeyWord, and thefunc
attribute is not always required. There could be some language expression likereturn
, but it’s notfx
, right?Best regards.
-
The “fx” distinguishes the option comes from an autocomplete/*.xml file versus a word from the current file when both word and API autocomplete is active.
Scintilla autocomplete has a lot to be desired and Notepad++ does the best it can with it, but plugins can use autocomplete too (e.g., QuickText, TagLEET, I have PythonScripts that do it to provide Python-IDE-like features with jedi). Only you can’t merge the lists - only 1 autocomplete is active at any time. So which suggestion is coming from what source?
TagLEET global tags file:
TagLEET local tags file:
QuickText:
PythonScript custom Python IDE:
It’d be nice to see them all in the same list, but alas, each time a character is typed and a best match is found, whatever Notepad++ / Plugin matches best displays it’s abbreviated list. The Icons help determine where in the heck this suggestion is coming from.
This is no different than most other editors that offer autocomplete (VS Code example)
Cheers.
-
This post is deleted! -
Okay, to differentiate which is document words and which are autocomplete words seems to make sense, unless you factor in that if you don’t want the words from the document being introduced and only use it for the autocomplete list, the
fx
is actually redundant to that purpose. Now, if you want to show an icon for words from the document, that triggers if you switch that option on, that makes more sense. Since if the word is from an autocomplete it won’t matter, it’s a different procedure for you to autocomplete a function, and the keyword is just that, so you select it whether it has been in the document previously or not, but if it’s only from the document, that would make more sense to highlight that with a different character glyph. Just my two cents worth. :)Wow, forum is slow and double posting posts. :(
-
@michael-vincent
fx
means function of x in math… I would use something from what Notepad++ already has in its UI. For example, the cube icon from the Function List panel with light grey color, so not to distract the user and black color for the font in results, as it has right now :)The example icon from the Function List panel
and I would prefer to use unicode character instead of png, so it could be scalable and faster to load :)
P.s.: btw, I have also issue with NppGTags autocomplete as it shows me gibberish in results. It’s my favorite plugin :( Maybe, the engoding of generated tags file is wrong… no ideas how to fix
Cheers.
-
@valery-bogomolov said in Autocomplete Items Order Issue:
I would use something from what Notepad++ already has in its UI
I did:
@valery-bogomolov said in Autocomplete Items Order Issue:
and I would prefer to use unicode character instead of png, so it could be scalable and faster to load :)
No characters, it’s not a PNG. Scintilla has only 2 options.
@valery-bogomolov said in Autocomplete Items Order Issue:
P.s.: btw, I have also issue with NppGTags autocomplete as it shows me gibberish in results
I would contact the author @pnedev and open an issue. He’s been very responsive fixing the Compare Plugin.
Cheers.
-
My function list icon looks like this, not yours:
-
@alan-kilborn said in Autocomplete Items Order Issue:
My function list icon looks like this, not yours:
You’ll need to restart Notepad++.
-
This thread is becoming rather mixed-up.
Please open an issue in NppGTags repo to discuss its autocomplete problem there.At first look it seems that the parser you are using (GTags maybe) is not recognizing some symbols in your codebase. AFAIR, the parser won’t be able to parse Unicode right if you are using it.