[READY] Update ycmd
Include the following changes:
- PR Valloric/ycmd#1080: replace Boost canonical function with our own implementation;
- PR Valloric/ycmd#1104: improve filename completer;
- PR Valloric/ycmd#1121: support completion FixIts for C-family languages;
- PR Valloric/ycmd#1122: update jdt.ls to 0.26.0;
- PR Valloric/ycmd#1123: install fixed version of TypeScript in third-party folder;
- PR Valloric/ycmd#1124: only add the necessary directories to Python path.
Fixes#3173.
<!-- Reviewable:start -->
---
This change is [<img src="https://reviewable.io/review_button.svg" height="34" align="absmiddle" alt="Reviewable"/>](https://reviewable.io/reviews/valloric/youcompleteme/3174)
<!-- Reviewable:end -->
[READY] Handle null characters in completion response
The `detailed_info` field may contain null characters e.g. `\x00` in Python docstrings (though it doesn't make sense to use such characters in a docstring). Since these characters cannot be evaluated through `pyeval` (and `py3eval`), we replace them with the symbol used by Vim to display them (`^@`).
This PR also rewrites the `_ConvertCompletionDataToVimData` function by relying on two things:
- [the `insertion_text` field is always defined](600f54de20/ycmd/responses.py (L109));
- [Vim ignores fields with an empty string](833e5dab14/src/edit.c (L2748-L2755)).
I did some measurements and the performance hit is negligible.
Fixes#3207.
<!-- Reviewable:start -->
---
This change is [<img src="https://reviewable.io/review_button.svg" height="34" align="absmiddle" alt="Reviewable"/>](https://reviewable.io/reviews/valloric/youcompleteme/3208)
<!-- Reviewable:end -->
The completion info field may contain null characters e.g. \x00 in
Python docstrings. These characters cannot be evaluated so they are
removed.
Rewrite the function that convert ycmd completion to Vim completion.
When auto-wrapping is enabled, Vim wraps the current line after the
completion request is sent but before calling the completefunc. The
starting column returned by the server is invalid in that case and must
be recomputed.
[READY] Ignore ft= argument from YcmCompleter command
The `ft=` argument of the `YcmCompleter` command is supposed to allow users to specify a completer when running a command. However, this was broken a long time ago in commit 29bb90a6b4. Only the `ycm:ident` value is still supported but it serves no purpose because the identifier completer doesn't define commands. Ignore completely the `ft=` argument and move the logic to the Python layer. Remove the argument from the documentation.
<!-- Reviewable:start -->
---
This change is [<img src="https://reviewable.io/review_button.svg" height="34" align="absmiddle" alt="Reviewable"/>](https://reviewable.io/reviews/valloric/youcompleteme/3161)
<!-- Reviewable:end -->
Only the "ycm:ident" value of the ft= argument in the YcmCompleter
command is working. This value forces the use of the identifier
completer which serve no purpose because this completer doesn't define
commands. Ignore completely the ft= argument and move the logic to the
Python layer.
When the client sends a request to the server, if an extra conf file is found
that is not already white/blacklisted, the server stops processing the request
and tells the client that an unknown extra conf file has been found. The client
then asks the user if that file should be loaded or not. Depending on the
user's answer, the client sends a request to the server to load or ignore the
extra conf file. Finally, the server loads the file or adds it to the
blacklist. However, the initial request was not processed by the server and
should be sent again.
[READY] Refactor server exception handling
Having to wrap all request calls with the `HandleServerException` function is inconvenient. Handle server exceptions directly in `BaseRequest`.
<!-- Reviewable:start -->
---
This change is [<img src="https://reviewable.io/review_button.svg" height="34" align="absmiddle" alt="Reviewable"/>](https://reviewable.io/reviews/valloric/youcompleteme/2976)
<!-- Reviewable:end -->
Java completer can include FixIts which are applied when a completion
entry is selected. We use the existing mechanism implemented for c-sharp
to perform these edits using the CompleteDone autocommand.
However, the existing mechanism relies on pattern matching the source to
work out which item was completed. Vim patch 8.0.1493 introduces support
for user_data on completion items, so when available we populate it with
the completion array index of the item and use that to get the exact
element that was selected. This is both a lot faster and a lot more
accirate.
Of course when applying these 'FixIts' we don't interrupt the user with
confirmation or the quickfix list as this would just be annoying. If the
server reports that an edit must be made, we just make the edit. This is
achieved by adding a silent flag to ReplaceChunks.
This implements an asynchronous message system using a long-poll request
to the server.
The server provides an endpoint /receive_messages which blocks until
either a timeout occurs or we receive a batch of asynchronous messages.
We send this request asynchronously and poll it 4 times a second to see
if we have received any messages.
The messages may either be simply for display (such as startup progress)
or diagnostics, which override the diagnostics returned by
OnFileReqdyToParse.
In the former case, we simply display the message, accepting that this
might be overwritten by any other message (indeed, requiring this), and
for the latter we fan out diagnostics to any open buffer for the file in
question.
Unfortunately, Vim has bugs related to timers when there is something
displayed (such as a "confirm" prompt or other), so we suspend
background timers when doing subcommands to avoid vim bugs. NOTE: This
requires a new version of Vim (detected by the presence of the
particular functions used).
Reduce the time spent to build the request when there are a lot of buffers by:
- using the options property on the buffer object to get the mod variable
instead of evaluating getbufvar;
- not computing the buffer filepath if the buffer is not modified;
- passing the number of the unloaded buffer instead of its filepath on the
BufferUnload event. Getting the Python buffer object from its number is
easier than from its filepath.
Bring fully asynchronous completion by polling for completions with a timer
then calling completefunc once the completions are ready. Use the start column
returned by the server in completefunc. Immediately display the last completion
on the TextChangedI event to prevent the popup menu disappearing while waiting
for the completions. Handle the TextChangedI event not being triggered while
the completion menu is open by closing the menu when inserting a character
through the InsertCharPre event, and when deleting a character on the <BS> and
<C-h> keys.
On non-Windows platforms, a ConnectionError exception is raised instead
of a ConnectTimeout one when the connection cannot be established (e.g.
the server crashed). Since the latter is a subclass of the former, we
only need to catch ConnectionError.
[READY] Fix exception when response future is not set
An `AttributeError` exception is raised when `Done` is called before `Start` in the `CompletionRequest` and `EventNotification` classes because the `_response_future` attribute is not yet defined.
Fixes#2461.
<!-- Reviewable:start -->
---
This change is [<img src="https://reviewable.io/review_button.svg" height="34" align="absmiddle" alt="Reviewable"/>](https://reviewable.io/reviews/valloric/youcompleteme/2465)
<!-- Reviewable:end -->
[READY] Move client tests to the main tests folder
This makes it possible to configure all tests at a package level by implementing the `setUpPackage` and `tearDownPackage` functions in `python/ycm/tests/__init__.py` file.
<!-- Reviewable:start -->
---
This change is [<img src="https://reviewable.io/review_button.svg" height="34" align="absmiddle" alt="Reviewable"/>](https://reviewable.io/reviews/valloric/youcompleteme/2394)
<!-- Reviewable:end -->
Send the request as the unloaded buffer instead of the current buffer
for the BufferUnload event notification. This fixes the issue where
the filetype of the current buffer is not the same as the unloaded
buffer one, making the ycmd server uses the wrong completer when
handling the request.