--- Begin Message ---
|
Subject: |
24.3.50; Company and CAPF: dealing with completion values containing extra text |
|
Date: |
Sun, 26 Jan 2014 06:11:23 +0200 |
When dealing with code completion for programming languages, it's usual
that candidates have associated text - most often, as function names,
they include the argument list.
Related issues:
* When inserting the "common part", we must ignore the extra text.
* The argument list is useful, for example when the candidate is
inserted in the buffer. After it was explicitly selected, we also
insert the arguments list, but replace each argument with self-erasing
field when you type over it. We do that in company-clang and
company-eclim (and could also do in company-senamtic, were someone to
request that).
* C-like languages usually provide method overloading, and then the
argument list is a part of a method's identity. Any metadata is
associated with the tuple "method name" + arguments, so we want to
pass the method name with arguments to functions that retrieve
documentation, location, etc.
* If the completion table values only contained method names,
`delete-duplicates' would remove all but one of the methods with the
same name. But we do want to remove duplicates.
At the moment the following approach emerged (for examples, again, see
company-clang and company-eclim): the completion values include the
argument lists (but nothing extra except that), and the respective
backends define an undocumented command: `crop'.
That command is only used in two situations:
1. When `company-complete-common' is called, we insert the common part
among all candidates, but before that we call (backend-function 'crop
"common-part"), to make sure not to insert the paren or anything after
it.
2. When `company-auto-complete' feature is used, typing any character
from `company-auto-complete-chars' insert the currently selected
completion into the buffer. In that case we also only want to insert the
method name, and so backend's `crop' function is called.
When the candidate is inserted normally (by selecting it with M-p or M-n
and then pressing RET), `crop' is not called. The full candidate value
is inserted into the buffer, and then we call `post-completion' backend
command, to allow it to remove the arguments list, or do something more
advanced (see the * #2 above).
This behavior is inconsistent, and `crop' doesn't sounds too good as a
command name.
At the risk of breaking some existing code, and thanks to `crop' stil'
beging undocumented, I'd like to define a `value' command instead that
would return the "cleaned" candidate text, presumably without the
arguments list. It would get called anytime before the candidate text
gets inserted, and to get the above mentioned templatification behavior,
the `post-completion' code will need to explicitly insert the rest of
the candidate text (the full candidate will be passed as the first and
the only argument). Which will be the main difference from the backend's
writer perspective.
I see two problems:
* Including extra text in completion table seems like it won't mesh well
with the completion-at-point-functions interface, and specifically
with the completion-at-point as the frontend. How would one write a
CAPF function for clang or eclim with argument lists in mind?
* `company-update-candidates' currently calls `company--safe-candidate'
on the "common part", on the assumption that the `crop' command will
return something meaningful for a string that's not itself a
completion candidate (and currently, they all do, because they look
for `(' instead of, say, using text properties or a hash table
lookup). If the command is called `value', it would be more tempting
for the implementor to only expect it to be called on actual
candidates. And then it'll return nil or do something unexpected when
called on something like "foo(".
Thoughts?
--- End Message ---
--- Begin Message ---
|
Subject: |
Re: bug#16555: 24.3.50; Company and CAPF: dealing with completion values containing extra text |
|
Date: |
Wed, 29 Jan 2014 00:51:17 +0200 |
|
User-agent: |
Mozilla/5.0 (X11; Linux x86_64; rv:24.0) Gecko/20100101 Thunderbird/24.2.0 |
On 29.01.2014 00:04, Stefan Monnier wrote:
CAPF needs to improve the annotation
support, indeed. I don't think the current annotation-function is
sufficient, since there are different kinds of annotations. E.g. adding
"<f>" is not the same as adding "(int x, float y, Vector<String>)" for
simple reasons of screen real estate, so in some UIs you'd want to
display both, while in others you'd only want the short one.
I'm not sure differentiating between them would be beneficial. We
already have "full document" annotation (company-doc-buffer), "one-line"
annotation (company-docsig), and just "annotation" itself. If we're
going to differentiate between different kinds of short annotations,
this will make 4 different functions a backend would need to define to
describe a candidate with words.
FWIW, "(int x, float y, Vector<String>)" looks short enough to me. In
Ruby, it often looks like "(table_name, column_name, [options])", which
isn't too long and still allows completion-at-point display candidates
in two columns.
We could consider it a bug, though (that annotation-function is called with
different objects).
Not really: the "objects" you're talking about are returned by the
completion-table, i.e. they cover at most one "field" (in the
completion-boundaries sense), whereas completion-all-completions returns
strings that can span several fields, so clearly they can't always be
`eq' to something returned by the completion-table.
I see. Then I'm out of ideas here, and using text properties, as
non-obvious that is, indeed remains the best option.
The idea is rather to let the backend provide more kinds of annotations
and let the UI choose which one to use. E.g. icomplete-mode doesn't
want any annotation at all, because its screen real-estate is
very limited. So completion-all-completions can't blindly add annotations.
Thanks for pointing that out.
Come to think of it, company-backends should be able to use a hash-table
with `eq' test maybe already, or if I massage the code a little. The major
question for me was about uniqueness, and looks like, yes, doing
delete-consecutive-dups' after fetching all annotations should be fast
enough (and this approach even has some space for optimization). So that
leaves a problem with CAPF.
That sounded like "thinking out loud for myself". I don't know what you
wanted to say nor how that relates to CAPF.
It was. Sorry if it's out of place.
I filed this bug for discussing a new feature in both Company and CAPF,
and that was me summing up the (one-sided) discussion of it on the
Company side. So, closing.
FWIW,
minibuffer-completion-help uses `sort' on the "annotated completions"
and then display-completion-list uses delete-consecutive-dups (tho hand
written into the loop).
Yes, delete-consecutive-dups in Company is also inlined (in
`company-calculate-candidates'), we couldn't use the function itself
anyway, cause it's 24.4-only.
It could be an either/or specification: if annotation-function is defined,
use it, otherwise, look up the `annotation' property.
I think that in most cases the annotation function will want to do some
work rather than just return the content of the annotation (as in the
sample code in my previous message).
Hm, yes, indeed.
--- End Message ---