Looking for help, advice, or support? Having problems getting YCM to work?
First carefully read the installation instructions for your OS.
We recommend you use the supplied install.py
- the "full" installation guide
is for rare, advanced use cases and most users should use install.py
.
If the server isn't starting and you're getting a "YouCompleteMe unavailable" error, check the Troubleshooting guide.
Next, check the User Guide section on the semantic completer that you are using. For C/C++/Objective-C/Objective-C++/CUDA, you must read this section.
Finally, check the FAQ.
If, after reading the installation and user guides, and checking the FAQ, you're still having trouble, check the contacts section below for how to get in touch.
Please do NOT go to #vim on Freenode for support. Please contact the YouCompleteMe maintainers directly using the contact details below.
Please note that the below instructions suggest using Vundle. Currently there are problems with Vundle, so here are some alternative instructions using Vim packages.
- Intro
- Installation
- Quick Feature Summary
- User Guide
- General Usage
- Client-Server Architecture
- Completion String Ranking
- General Semantic Completion
- Signature Help
- Semantic Highlighting
- Inlay Hints
- C-family Semantic Completion
- Java Semantic Completion
- C# Semantic Completion
- Python Semantic Completion
- Rust Semantic Completion
- Go Semantic Completion
- JavaScript and TypeScript Semantic Completion
- Semantic Completion for Other Languages
- LSP Configuration
- Writing New Semantic Completers
- Diagnostic Display
- Symbol Search
- Type/Call Hierarchy
- Commands
- Functions
- Autocommands
- Options
- FAQ
- Contributor Code of Conduct
- Contact
- License
- Sponsorship
YouCompleteMe is a fast, as-you-type, fuzzy-search code completion, comprehension and refactoring engine for Vim.
It has several completion engines built-in and supports any protocol-compliant Language Server, so can work with practically any language. YouCompleteMe contains:
- an identifier-based engine that works with every programming language,
- a powerful clangd-based engine that provides native semantic code completion for C/C++/Objective-C/Objective-C++/CUDA (from now on referred to as "the C-family languages"),
- a Jedi-based completion engine for Python 2 and 3,
- an OmniSharp-Roslyn-based completion engine for C#,
- a Gopls-based completion engine for Go,
- a TSServer-based completion engine for JavaScript and TypeScript,
- a rust-analyzer-based completion engine for Rust,
- a jdt.ls-based completion engine for Java.
- a generic Language Server Protocol implementation for any language
- and an omnifunc-based completer that uses data from Vim's omnicomplete system to provide semantic completions for many other languages (Ruby, PHP, etc.).
Here's an explanation of what happened in the last GIF demo above.
First, realize that no keyboard shortcuts had to be pressed to get the list of completion candidates at any point in the demo. The user just types and the suggestions pop up by themselves. If the user doesn't find the completion suggestions relevant and/or just wants to type, they can do so; the completion engine will not interfere.
When the user sees a useful completion string being offered, they press the TAB key to accept it. This inserts the completion string. Repeated presses of the TAB key cycle through the offered completions.
If the offered completions are not relevant enough, the user can continue typing to further filter out unwanted completions.
A critical thing to notice is that the completion filtering is NOT based on
the input being a string prefix of the completion (but that works too). The
input needs to be a subsequence match of a completion. This is a fancy way
of saying that any input characters need to be present in a completion string in
the order in which they appear in the input. So abc
is a subsequence of
xaybgc
, but not of xbyxaxxc
. After the filter, a complicated sorting system
ranks the completion strings so that the most relevant ones rise to the top of
the menu (so you usually need to press TAB just once).
All of the above works with any programming language because of the identifier-based completion engine. It collects all of the identifiers in the current file and other files you visit (and your tags files) and searches them when you type (identifiers are put into per-filetype groups).
The demo also shows the semantic engine in use. When the user presses .
, ->
or ::
while typing in insert mode (for C++; different triggers are used for
other languages), the semantic engine is triggered (it can also be triggered
with a keyboard shortcut; see the rest of the docs).
The last thing that you can see in the demo is YCM's diagnostic display features (the little red X that shows up in the left gutter; inspired by Syntastic) if you are editing a C-family file. As the completer engine compiles your file and detects warnings or errors, they will be presented in various ways. You don't need to save your file or press any keyboard shortcut to trigger this, it "just happens" in the background.
And that's not all...
YCM might be the only Vim completion engine with the correct Unicode support. Though we do assume UTF-8 everywhere.
YCM also provides semantic IDE-like features in a number of languages, including:
- displaying signature help (argument hints) when entering the arguments to a function call (Vim only)
- finding declarations, definitions, usages, etc. of identifiers, and an interactive symbol finder
- displaying type information for classes, variables, functions etc.,
- displaying documentation for methods, members, etc. in the preview window, or in a popup next to the cursor (Vim only)
- fixing common coding errors, like missing semi-colons, typos, etc.,
- semantic renaming of variables across files,
- formatting code,
- removing unused imports, sorting imports, etc.
For example, here's a demo of signature help:
Below we can see YCM being able to do a few things:
- Retrieve references across files
- Go to declaration/definition
- Expand
auto
in C++ - Fix some common errors, and provide refactorings, with
FixIt
- Not shown in the GIF are
GoToImplementation
andGoToType
for servers that support it.
And here's some documentation being shown in a hover popup, automatically and manually:
Features vary by file type, so make sure to check out the file type feature summary and the full list of completer subcommands to find out what's available for your favourite languages.
You'll also find that YCM has filepath completers (try typing ./
in a file)
and a completer that integrates with UltiSnips.
Runtime | Min Version | Recommended Version (full support) | Python |
---|---|---|---|
Vim | 9.1.0016 | 9.1.0016 | 3.8 |
Neovim | 0.5 | Vim 9.1.0016 | 3.8 |
Our policy is to support the Vim version that's in the latest LTS of Ubuntu.
Vim must have a working Python 3 runtime.
For Neovim users, our policy is to require the latest released version. Currently, Neovim 0.5.0 is required. Please note that some features are not available in Neovim, and Neovim is not officially supported.
YCM has two components: A server and a client. Both the server and client require Python 3.8 or later 3.x release.
For the Vim client, Vim must be, compiled with --enable-shared
(or
--enable-framework
on macOS). You can check if this is working with :py3 import sys; print( sys.version)
. It should say something like 3.8.2 (...)
.
For Neovim, you must have a python 3.8 runtime and the Neovim python
extensions. See Neovim's :help provider-python
for how to set that up.
For the server, you must run the install.py
script with a python 3.8 (or
later) runtime. Anaconda etc. are not supported. YCM will remember the runtime
you used to run install.py
and will use that when launching the server, so if
you usually use anaconda, then make sure to use the full path to a real cpython3,
e.g. /usr/bin/python3 install.py --all
etc.
Our policy is to support the python3 version that's available in the latest Ubuntu LTS (similar to our Vim version policy). We don't increase the Python runtime version without a reason, though. Typically, we do this when the current python version we're using goes out of support. At that time we will typically pick a version that will be supported for a number of years.
In order to provide the best possible performance and stability, ycmd has updated its code to C++17. This requires a version bump of the minimum supported compilers. The new requirements are:
Compiler | Current Min |
---|---|
GCC | 8 |
Clang | 7 |
MSVC | 15.7 (VS 2017) |
YCM requires CMake 3.13 or greater. If your CMake is too old, you may be able to
simply pip install --user cmake
to get a really new version.
When enabling language support for a particular language, there may be runtime requirements, such as needing a very recent Java Development Kit for Java support. In general, YCM is not in control of the required versions for the downstream compilers, though we do our best to signal where we know them.
- Install YCM plugin via Vundle
- Install CMake, MacVim and Python 3; Note that the pre-installed macOS system Vim is not supported (due to it having broken Python integration).
$ brew install cmake python go nodejs
-
Install mono from Mono Project (NOTE: on Intel Macs you can also
brew install mono
. On arm Macs, you may require Rosetta) -
For Java support you must install a JDK, one way to do this is with Homebrew:
$ brew install java
$ sudo ln -sfn $(brew --prefix java)/libexec/openjdk.jdk /Library/Java/JavaVirtualMachines/openjdk.jdk
-
Pre-installed macOS system Vim does not support Python 3. So you need to install either a Vim that supports Python 3 OR MacVim with Homebrew:
- Option 1: Installing a Vim that supports Python 3
brew install vim
- Option 2: Installing MacVim
brew install macvim
-
Compile YCM.
-
For Intel and arm64 Macs, the bundled libclang/clangd work:
cd ~/.vim/bundle/YouCompleteMe python3 install.py --all
-
If you have troubles with finding system frameworks or C++ standard library, try using the homebrew llvm:
brew install llvm cd ~/.vim/bundle/YouCompleteMe python3 install.py --system-libclang --all
And edit your vimrc to add the following line to use the Homebrew llvm's clangd:
" Use homebrew's clangd let g:ycm_clangd_binary_path = trim(system('brew --prefix llvm')).'/bin/clangd'
-
-
For using an arbitrary LSP server, check the relevant section
These instructions (using install.py
) are the quickest way to install
YouCompleteMe, however they may not work for everyone. If the following
instructions don't work for you, check out the full installation
guide.
A supported Vim version with Python 3 is required. MacVim is a good option, even if you only use the terminal. YCM won't work with the pre-installed Vim from Apple as its Python support is broken. If you don't already use a Vim that supports Python 3 or MacVim, install it with Homebrew. Install CMake as well:
brew install vim cmake
OR
brew install macvim cmake
Install YouCompleteMe with Vundle.
Remember: YCM is a plugin with a compiled component. If you update YCM
using Vundle and the ycm_core
library APIs have changed (happens
rarely), YCM will notify you to recompile it. You should then rerun the install
process.
NOTE: If you want C-family completion, you MUST have the latest Xcode
installed along with the latest Command Line Tools (they are installed
automatically when you run clang
for the first time, or manually by running
xcode-select --install
)
Compiling YCM with semantic support for C-family languages through clangd:
cd ~/.vim/bundle/YouCompleteMe
./install.py --clangd-completer
Compiling YCM without semantic support for C-family languages:
cd ~/.vim/bundle/YouCompleteMe
./install.py
The following additional language support options are available:
- C# support: install by downloading the Mono macOS package
and add
--cs-completer
when callinginstall.py
. - Go support: install Go and add
--go-completer
when callinginstall.py
. - JavaScript and TypeScript support: install Node.js 18+ and npm and
add
--ts-completer
when callinginstall.py
. - Rust support: add
--rust-completer
when callinginstall.py
. - Java support: install JDK 17 and add
--java-completer
when callinginstall.py
.
To simply compile with everything enabled, there's a --all
flag. So, to
install with all language features, ensure xbuild
, go
, node
and npm
tools are installed and in your PATH
, then simply run:
cd ~/.vim/bundle/YouCompleteMe
./install.py --all
That's it. You're done. Refer to the User Guide section on how to use YCM. Don't forget that if you want the C-family semantic completion engine to work, you will need to provide the compilation flags for your project to YCM. It's all in the User Guide.
YCM comes with sane defaults for its options, but you still may want to take a look at what's available for configuration. There are a few interesting options that are conservatively turned off by default that you may want to turn on.
The following assume you're using Ubuntu 24.04.
- Install YCM plugin via Vundle
- Install CMake, Vim and Python
apt install build-essential cmake vim-nox python3-dev
- Install mono-complete, go, node, java, and npm
apt install mono-complete golang nodejs openjdk-17-jdk openjdk-17-jre npm
- Compile YCM
cd ~/.vim/bundle/YouCompleteMe
python3 install.py --all
- For plugging an arbitrary LSP server, check the relevant section
These instructions (using install.py
) are the quickest way to install
YouCompleteMe, however they may not work for everyone. If the following
instructions don't work for you, check out the full installation
guide.
Make sure you have a supported version of Vim with Python 3 support and a supported compiler. The latest LTS of Ubuntu is the minimum platform for simple installation. For earlier releases or other distributions, you may have to do some work to acquire the dependencies.
If your Vim version is too old, you may need to compile Vim from source (don't worry, it's easy).
Install YouCompleteMe with Vundle.
Remember: YCM is a plugin with a compiled component. If you update YCM
using Vundle and the ycm_core
library APIs have changed (which happens rarely), YCM
will notify you to recompile it. You should then rerun the installation process.
Install development tools, CMake, and Python headers:
- Fedora-like distributions:
sudo dnf install cmake gcc-c++ make python3-devel
- Ubuntu LTS:
sudo apt install build-essential cmake3 python3-dev
Compiling YCM with semantic support for C-family languages through clangd:
cd ~/.vim/bundle/YouCompleteMe
python3 install.py --clangd-completer
Compiling YCM without semantic support for C-family languages:
cd ~/.vim/bundle/YouCompleteMe
python3 install.py
The following additional language support options are available:
- C# support: install Mono and add
--cs-completer
when callinginstall.py
. - Go support: install Go and add
--go-completer
when callinginstall.py
. - JavaScript and TypeScript support: install Node.js 18+ and npm and
add
--ts-completer
when callinginstall.py
. - Rust support: add
--rust-completer
when callinginstall.py
. - Java support: install JDK 17 and add
--java-completer
when callinginstall.py
.
To simply compile with everything enabled, there's a --all
flag. So, to
install with all language features, ensure xbuild
, go
, node
, and npm
tools are installed and in your PATH
, then simply run:
cd ~/.vim/bundle/YouCompleteMe
python3 install.py --all
That's it. You're done. Refer to the User Guide section on how to use YCM. Don't forget that if you want the C-family semantic completion engine to work, you will need to provide the compilation flags for your project to YCM. It's all in the User Guide.
YCM comes with sane defaults for its options, but you still may want to take a look at what's available for configuration. There are a few interesting options that are conservatively turned off by default that you may want to turn on.
- Install YCM plugin via Vundle
- Install Visual Studio Build Tools 2019
- Install CMake, Vim and Python
- Install go, node and npm
- Compile YCM
cd YouCompleteMe
python3 install.py --all
- Add
set encoding=utf-8
to your vimrc - For plugging an arbitrary LSP server, check the relevant section
These instructions (using install.py
) are the quickest way to install
YouCompleteMe, however they may not work for everyone. If the following
instructions don't work for you, check out the full installation
guide.
Important: we assume that you are using the cmd.exe
command prompt and
that you know how to add an executable to the PATH environment variable.
Make sure you have a supported Vim version with Python 3 support. You
can check the version and which Python is supported by typing :version
inside
Vim. Look at the features included: +python3/dyn
for Python 3.
Take note of the Vim architecture, i.e. 32 or
64-bit. It will be important when choosing the Python installer. We recommend
using a 64-bit client. Daily updated installers of 32-bit and 64-bit Vim with
Python 3 support are available.
Add the following line to your vimrc if not already present.:
set encoding=utf-8
This option is required by YCM. Note that it does not prevent you from editing a
file in another encoding than UTF-8. You can do that by specifying the ++enc
argument to the :e
command.
Install YouCompleteMe with Vundle.
Remember: YCM is a plugin with a compiled component. If you update YCM
using Vundle and the ycm_core
library APIs have changed (happens
rarely), YCM will notify you to recompile it. You should then rerun the install
process.
Download and install the following software:
- Python 3. Be sure to pick the version
corresponding to your Vim architecture. It is Windows x86 for a 32-bit Vim
and Windows x86-64 for a 64-bit Vim. We recommend installing Python 3.
Additionally, the version of Python you install must match up exactly with
the version of Python that Vim is looking for. Type
:version
and look at the bottom of the page at the list of compiler flags. Look for flags that look similar to-DDYNAMIC_PYTHON3_DLL=\"python36.dll\"
. This indicates that Vim is looking for Python 3.6. You'll need one or the other installed, matching the version number exactly. - CMake. Add CMake executable to the PATH environment variable.
- Build Tools for Visual Studio 2019. During setup, select C++ build tools in Workloads.
Compiling YCM with semantic support for C-family languages through clangd:
cd %USERPROFILE%/vimfiles/bundle/YouCompleteMe
python install.py --clangd-completer
Compiling YCM without semantic support for C-family languages:
cd %USERPROFILE%/vimfiles/bundle/YouCompleteMe
python install.py
The following additional language support options are available:
- C# support: add
--cs-completer
when callinginstall.py
. Be sure that the build utilitymsbuild
is in your PATH. - Go support: install Go and add
--go-completer
when callinginstall.py
. - JavaScript and TypeScript support: install Node.js 18+ and npm and
add
--ts-completer
when callinginstall.py
. - Rust support: add
--rust-completer
when callinginstall.py
. - Java support: install JDK 17 and add
--java-completer
when callinginstall.py
.
To simply compile with everything enabled, there's a --all
flag. So, to
install with all language features, ensure msbuild
, go
, node
and npm
tools are installed and in your PATH
, then simply run:
cd %USERPROFILE%/vimfiles/bundle/YouCompleteMe
python install.py --all
You can specify the Microsoft Visual C++ (MSVC) version using the --msvc
option. YCM officially supports MSVC 15 (2017), MSVC 16 (Visual Studio 2019)
and MSVC 17 (Visual Studio 17 2022).
That's it. You're done. Refer to the User Guide section on how to use YCM. Don't forget that if you want the C-family semantic completion engine to work, you will need to provide the compilation flags for your project to YCM. It's all in the User Guide.
YCM comes with sane defaults for its options, but you still may want to take a look at what's available for configuration. There are a few interesting options that are conservatively turned off by default that you may want to turn on.
The full installation guide has been moved to the wiki.
- Super-fast identifier completer including tags files and syntax elements
- Intelligent suggestion ranking and filtering
- File and path suggestions
- Suggestions from Vim's omnifunc
- UltiSnips snippet suggestions
- Semantic auto-completion with automatic fixes
- Signature help
- Real-time diagnostic display
- Go to include/declaration/definition (
GoTo
, etc.) - Go to alternate file (e.g. associated header
GoToAlternateFile
) - Find Symbol (
GoToSymbol
), with interactive search - Document outline (
GoToDocumentOutline
), with interactive search - View documentation comments for identifiers (
GetDoc
) - Type information for identifiers (
GetType
) - Automatically fix certain errors (
FixIt
) - Perform refactoring (
FixIt
) - Reference finding (
GoToReferences
) - Renaming symbols (
RefactorRename <new name>
) - Code formatting (
Format
) - Semantic highlighting
- Inlay hints
- Type hierarchy
- Call hierarchy
- Semantic auto-completion
- Signature help
- Real-time diagnostic display
- Go to declaration/definition (
GoTo
, etc.) - Go to implementation (
GoToImplementation
) - Find Symbol (
GoToSymbol
), with interactive search - View documentation comments for identifiers (
GetDoc
) - Type information for identifiers (
GetType
) - Automatically fix certain errors (
FixIt
) - Perform refactoring (
FixIt
) - Management of OmniSharp-Roslyn server instance
- Renaming symbols (
RefactorRename <new name>
) - Code formatting (
Format
)
- Semantic auto-completion
- Signature help
- Go to definition (
GoTo
) - Find Symbol (
GoToSymbol
), with interactive search - Reference finding (
GoToReferences
) - View documentation comments for identifiers (
GetDoc
) - Type information for identifiers (
GetType
) - Renaming symbols (
RefactorRename <new name>
)
- Semantic auto-completion
- Signature help
- Real-time diagnostic display
- Go to declaration/definition (
GoTo
, etc.) - Go to type definition (
GoToType
) - Go to implementation (
GoToImplementation
) - Document outline (
GoToDocumentOutline
), with interactive search - Automatically fix certain errors (
FixIt
) - Perform refactoring (
FixIt
) - View documentation comments for identifiers (
GetDoc
) - Type information for identifiers (
GetType
) - Code formatting (
Format
) - Management of
gopls
server instance - Inlay hints
- Call hierarchy
- Semantic auto-completion with automatic import insertion
- Signature help
- Real-time diagnostic display
- Go to definition (
GoTo
,GoToDefinition
, andGoToDeclaration
are identical) - Go to type definition (
GoToType
) - Go to implementation (
GoToImplementation
) - Find Symbol (
GoToSymbol
), with interactive search - Reference finding (
GoToReferences
) - View documentation comments for identifiers (
GetDoc
) - Type information for identifiers (
GetType
) - Automatically fix certain errors and perform refactoring (
FixIt
) - Perform refactoring (
FixIt
) - Renaming symbols (
RefactorRename <new name>
) - Code formatting (
Format
) - Organize imports (
OrganizeImports
) - Management of
TSServer
server instance - Inlay hints
- Call hierarchy
- Semantic auto-completion
- Real-time diagnostic display
- Go to declaration/definition (
GoTo
, etc.) - Go to implementation (
GoToImplementation
) - Reference finding (
GoToReferences
) - Document outline (
GoToDocumentOutline
), with interactive search - View documentation comments for identifiers (
GetDoc
) - Automatically fix certain errors (
FixIt
) - Perform refactoring (
FixIt
) - Type information for identifiers (
GetType
) - Renaming symbols (
RefactorRename <new name>
) - Code formatting (
Format
) - Management of
rust-analyzer
server instance - Semantic highlighting
- Inlay hints
- Call hierarchy
- Semantic auto-completion with automatic import insertion
- Signature help
- Real-time diagnostic display
- Go to definition (
GoTo
,GoToDefinition
, andGoToDeclaration
are identical) - Go to type definition (
GoToType
) - Go to implementation (
GoToImplementation
) - Find Symbol (
GoToSymbol
), with interactive search - Reference finding (
GoToReferences
) - Document outline (
GoToDocumentOutline
), with interactive search - View documentation comments for identifiers (
GetDoc
) - Type information for identifiers (
GetType
) - Automatically fix certain errors including code generation (
FixIt
) - Renaming symbols (
RefactorRename <new name>
) - Code formatting (
Format
) - Organize imports (
OrganizeImports
) - Detection of Java projects
- Execute custom server command (
ExecuteCommand <args>
) - Management of
jdt.ls
server instance - Semantic highlighting
- Inlay hints
- Type hierarchy
- Call hierarchy
If the offered completions are too broad, keep typing characters; YCM will continue refining the offered completions based on your input.
Filtering is "smart-case" and "smart-diacritic" sensitive; if you are typing only lowercase letters, then it's case-insensitive. If your input contains uppercase letters, then the uppercase letters in your query must match uppercase letters in the completion strings (the lowercase letters still match both). On top of that, a letter with no diacritic marks will match that letter with or without marks:
matches | foo | fôo | fOo | fÔo |
---|---|---|---|---|
foo | ✔️ | ✔️ | ✔️ | ✔️ |
fôo | ❌ | ✔️ | ❌ | ✔️ |
fOo | ❌ | ❌ | ✔️ | ✔️ |
fÔo | ❌ | ❌ | ❌ | ✔️ |
Use the TAB key to accept a completion and continue pressing TAB to cycle through the completions. Use Shift-TAB to cycle backward. Note that if you're using console Vim (that is, not gvim or MacVim) then it's likely that the Shift-TAB binding will not work because the console will not pass it to Vim. You can remap the keys; see the Options section below.
Knowing a little bit about how YCM works internally will prevent confusion. YCM has several completion engines: an identifier-based completer that collects all of the identifiers in the current file and other files you visit (and your tags files) and searches them when you type (identifiers are put into per-filetype groups).
There are also several semantic engines in YCM. There are libclang-based and clangd-based completers that provide semantic completion for C-family languages. There's a Jedi-based completer for semantic completion for Python. There's also an omnifunc-based completer that uses data from Vim's omnicomplete system to provide semantic completions when no native completer exists for that language in YCM.
There are also other completion engines, like the UltiSnips completer and the filepath completer.
YCM automatically detects which completion engine would be the best in any situation. On occasion, it queries several of them at once, merges the outputs and presents the results to you.
YCM has a client-server architecture; the Vim part of YCM is only a thin client that talks to the ycmd HTTP+JSON server that has the vast majority of YCM logic and functionality. The server is started and stopped automatically as you start and stop Vim.
The subsequence filter removes any completions that do not match the input, but then the sorting system kicks in. It's actually very complicated and uses lots of factors, but suffice it to say that "word boundary" (WB) subsequence character matches are "worth" more than non-WB matches. In effect, this means that given an input of "gua", the completion "getUserAccount" would be ranked higher in the list than the "Fooguxa" completion (both of which are subsequence matches). Word-boundary characters are all capital characters, characters preceded by an underscore, and the first letter character in the completion string.
Valid signatures are displayed in a second popup menu and the current signature is highlighted along with the current argument.
Signature help is triggered in insert mode automatically when
g:ycm_auto_trigger
is enabled and is not supported when it is not enabled.
The signatures popup is hidden when there are no matching signatures or when you
leave insert mode. If you want to manually control when it is visible, you can
map something to <plug>YCMToggleSignatureHelp
(see below).
For more details on this feature and a few demos, check out the PR that proposed it.
The signature help popup sometimes gets in the way. You can toggle its
visibility with a mapping. YCM provides the "Plug" mapping
<Plug>(YCMToggleSignatureHelp)
for this.
For example, to hide/show the signature help popup by pressing Ctrl+l in insert
mode: imap <silent> <C-l> <Plug>(YCMToggleSignatureHelp)
.
NOTE: No default mapping is provided because insert mappings are very difficult to create without breaking or overriding some existing functionality. Ctrl-l is not a suggestion, just an example.
Semantic highlighting is the process where the buffer text is coloured according to the underlying semantic type of the word, rather than classic syntax highlighting based on regular expressions. This can be powerful additional data that we can process very quickly.
This feature is only supported in Vim.
For example, here is a function with classic highlighting:
And here is the same function with semantic highlighting:
As you can see, the function calls, macros, etc. are correctly identified.
This can be enabled globally with let g:ycm_enable_semantic_highlighting=1
or
per buffer, by setting b:ycm_enable_semantic_highlighting
.
YCM uses text properties (see :help text-prop-intro
) for semantic
highlighting. In order to customise the coloring, you can define the text
properties that are used.
If you define a text property named YCM_HL_<token type>
, then it will be used
in place of the defaults. The <token type>
is defined as the Language Server
Protocol semantic token type, defined in the LSP Spec.
Some servers also use custom values. In this case, YCM prints a warning including the token type name that you can customise.
For example, to render parameter
tokens using the Normal
highlight group,
you can do this:
call prop_type_add( 'YCM_HL_parameter', { 'highlight': 'Normal' } )
More generally, this pattern can be useful for customising the groups:
let MY_YCM_HIGHLIGHT_GROUP = {
\ 'typeParameter': 'PreProc',
\ 'parameter': 'Normal',
\ 'variable': 'Normal',
\ 'property': 'Normal',
\ 'enumMember': 'Normal',
\ 'event': 'Special',
\ 'member': 'Normal',
\ 'method': 'Normal',
\ 'class': 'Special',
\ 'namespace': 'Special',
\ }
for tokenType in keys( MY_YCM_HIGHLIGHT_GROUP )
call prop_type_add( 'YCM_HL_' . tokenType,
\ { 'highlight': MY_YCM_HIGHLIGHT_GROUP[ tokenType ] } )
endfor
NOTE: Highly experimental feature, requiring Vim 9.0.214 or later (not supported in NeoVim).
When g:ycm_enable_inlay_hints
(globally) or b:ycm_enable_inlay_hints
(for a
specific buffer) is set to 1
, then YCM will insert inlay hints as supported by
the language semantic engine.
An inlay hint is text that is rendered on the screen that is not part of the buffer and is often used to mark up the type or name of arguments, parameters, etc. which help the developer understand the semantics of the code.
Here are some examples:
- C
- TypeScript
- Go
By default, YCM renders the inlay hints with the NonText
highlight group. To
override this, define the YcmInlayHint
highlight yourself, e.g. in your
.vimrc
:
hi link YcmInlayHint Comment
Similar to semantic highlighting above, you can override specific highlighting for different inlay hint types by defining text properties named after the kind of inlay hint, for example:
call prop_type_add( 'YCM_INLAY_Type', #{ highlight: 'Comment' } )
The list of inlay hint kinds can be found in python/ycm/inlay_hints.py
g:ycm_enable_inlay_hints
orb:ycm_enable_inlay_hints
- enable/disable globally or for local bufferg:ycm_clear_inlay_hints_in_insert_mode
- set to1
to remove all inlay hints when entering insert mode and reinstate them when leaving insert mode
Inlay hints can add a lot of text to the screen and may be distracting. You can
toggle them on/off instantly, by mapping something to
<Plug>(YCMToggleInlayHints)
, for example:
nnoremap <silent> <localleader>h <Plug>(YCMToggleInlayHints)
No default mapping is provided for this due to the personal nature of mappings.
You can use Ctrl+Space to trigger the completion suggestions anywhere, even without a string prefix. This is useful to see which top-level functions are available for use.
NOTE: YCM originally used the libclang
based engine for C-family, but
users should migrate to clangd, as it provides more features and better
performance. Users who rely on override_filename
in their .ycm_extra_conf.py
will need to stay on the old libclang
engine. Instructions on how to stay on
the old engine are available on the wiki.
Some of the features of clangd:
- Project wide indexing: Clangd has both dynamic and static index support. The dynamic index stores up-to-date symbols coming from any files you are currently editing, whereas static index contains project-wide symbol information. This symbol information is used for code completion and code navigation. Whereas libclang is limited to the current translation unit(TU).
- Code navigation: Clangd provides all the GoTo requests libclang provides and it improves those using the above-mentioned index information to contain project-wide information rather than just the current TU.
- Rename: Clangd can perform semantic rename operations on the current file, whereas libclang doesn't support such functionality.
- Code Completion: Clangd can perform code completions at a lower latency
than libclang; also, it has information about all the symbols in your
project so it can suggest items outside your current TU and also provides
proper
#include
insertions for those items. - Signature help: Clangd provides signature help so that you can see the names and types of arguments when calling functions.
- Format Code: Clangd provides code formatting either for the selected lines or the whole file, whereas libclang doesn't have such functionality.
- Performance: Clangd has faster re-parse and code completion times compared to libclang.
On supported architectures, the install.py
script will download a suitable
clangd (--clangd-completer
) or libclang (--clang-completer
) for you.
Supported architectures are:
- Linux glibc >= 2.39 (Intel, armv7-a, aarch64) - built on ubuntu 24.04
- MacOS >=10.15 (Intel, arm64)
- For Intel, compatibility per clang.llvm.org downloads
- For arm64, macOS 10.15+
- Windows (Intel) - compatibility per clang.llvm.org downloads
clangd:
Typically, clangd is installed by the YCM installer (either with --all
or with
--clangd-completer
). This downloads a pre-built clangd
binary for your
architecture. If your OS or architecture is not supported or is too old, you can
install a compatible clangd
and use g:ycm_clangd_binary_path
to point to
it.
libclang:
libclang
can be enabled also with --all
or --clang-completer
. As with
clangd
, YCM will try and download a version of libclang
that is suitable for
your environment, but again if your environment can't be supported, you can
build or acquire libclang
for yourself and specify it when building, as:
$ EXTRA_CMAKE_ARGS='-DPATH_TO_LLVM_ROOT=/path/to/your/llvm' ./install.py --clang-completer --system-libclang
Please note that if using custom clangd
or libclang
it must match the
version that YCM requires. Currently YCM requires clang 17.0.1.
In order to perform semantic analysis such as code completion, GoTo
, and
diagnostics, YouCompleteMe uses clangd
, which makes use of
clang compiler, sometimes also referred to as LLVM. Like any compiler,
clang also requires a set of compile flags in order to parse your code. Simply
put: If clang can't parse your code, YouCompleteMe can't provide semantic
analysis.
There are 2 methods that can be used to provide compile flags to clang:
Option 1: Use a compilation database
The easiest way to get YCM to compile your code is to use a compilation
database. A compilation database is usually generated by your build system
(e.g. CMake
) and contains the compiler invocation for each compilation unit in
your project.
For information on how to generate a compilation database, see the clang documentation. In short:
- If using CMake, add
-DCMAKE_EXPORT_COMPILE_COMMANDS=ON
when configuring (or addset( CMAKE_EXPORT_COMPILE_COMMANDS ON )
toCMakeLists.txt
) and copy or symlink the generated database to the root of your project. - If using Ninja, check out the
compdb
tool (-t compdb
) in its docs. - If using GNU make, check out compiledb or Bear.
- For other build systems, check out
.ycm_extra_conf.py
below.
If no .ycm_extra_conf.py
is found,
YouCompleteMe automatically tries to load a compilation database if there is
one.
YCM looks for a file named compile_commands.json
in the directory of the
opened file or in any directory above it in the hierarchy (recursively); when
the file is found before a local .ycm_extra_conf.py
, YouCompleteMe stops
searching the directories and lets clangd take over and handle the flags.
If you don't have a compilation database or aren't able to generate one, you have to tell YouCompleteMe how to compile your code some other way.
Every C-family project is different. It is not possible for YCM to guess what compiler flags to supply for your project. Fortunately, YCM provides a mechanism for you to generate the flags for a particular file with arbitrary complexity. This is achieved by requiring you to provide a Python module that implements a trivial function that, given the file name as an argument, returns a list of compiler flags to use to compile that file.
YCM looks for a .ycm_extra_conf.py
file in the directory of the opened file or
in any directory above it in the hierarchy (recursively); when the file is
found, it is loaded (only once!) as a Python module. YCM calls a Settings
method in that module which should provide it with the information necessary to
compile the current file. You can also provide a path to a global configuration
file with the
g:ycm_global_ycm_extra_conf
option,
which will be used as a fallback. To prevent the execution of malicious code
from a file you didn't write YCM will ask you once per .ycm_extra_conf.py
if
it is safe to load. This can be disabled and you can white-/blacklist files. See
the g:ycm_confirm_extra_conf
and
g:ycm_extra_conf_globlist
options
respectively.
This system was designed this way so that the user can perform any arbitrary sequence of operations to produce a list of compilation flags YCM should hand to Clang.
NOTE: It is highly recommended to include -x <language>
flag to libclang.
This is so that the correct language is detected, particularly for header files.
Common values are -x c
for C, -x c++
for C++, -x objc
for Objective-C, and
-x cuda
for CUDA.
To give you an impression, if your C++ project is trivial, and your usual
compilation command is: g++ -Wall -Wextra -Werror -o FILE.o FILE.cc
, then the
following .ycm_extra_conf.py
is enough to get semantic analysis from
YouCompleteMe:
def Settings( **kwargs ):
return {
'flags': [ '-x', 'c++', '-Wall', '-Wextra', '-Werror' ],
}
As you can see from the trivial example, YCM calls the Settings
method which
returns a dictionary with a single element 'flags'
. This element is a list
of compiler flags to pass to libclang for the current file. The absolute path of
that file is accessible under the filename
key of the kwargs
dictionary.
That's it! This is actually enough for most projects, but for complex projects
it is not uncommon to integrate directly with an existing build system using the
full power of the Python language.
For a more elaborate example,
see ycmd's own .ycm_extra_conf.py
. You should be able to
use it as a starting point. Don't just copy/paste that file somewhere and
expect things to magically work; your project needs different flags. Hint:
just replace the strings in the flags
variable with compilation flags
necessary for your project. That should be enough for 99% of projects.
You could also consider using YCM-Generator to generate the
ycm_extra_conf.py
file.
If Clang encounters errors when compiling the header files that your file includes, then it's probably going to take a long time to get completions. When the completion menu finally appears, it's going to have a large number of unrelated completion strings (type/function names that are not actually members). This is because Clang fails to build a precompiled preamble for your file if there are any errors in the included headers and that preamble is key to getting fast completions.
Call the :YcmDiags
command to see if any errors or warnings were detected in
your file.
-
Ensure that you have enabled the Java completer. See the installation guide for details.
-
Create a project file (gradle or maven) file in the root directory of your Java project, by following the instructions below.
-
(Optional) Configure the LSP server. The jdt.ls configuration options can be found in their codebase.
-
If you previously used Eclim or Syntastic for Java, disable them for Java.
-
Edit a Java file from your project.
In order to provide semantic analysis, the Java completion engine requires knowledge of your project structure. In particular, it needs to know the class path to use, when compiling your code. Fortunately jdt.ls supports eclipse project files, maven projects and gradle projects.
NOTE: Our recommendation is to use either Maven or Gradle projects.
The native support for Java includes YCM's native real-time diagnostics display. This can conflict with other diagnostics plugins like Syntastic, so when enabling Java support, please manually disable Syntastic Java diagnostics.
Add the following to your vimrc
:
let g:syntastic_java_checkers = []
The native support for Java includes YCM's native real-time diagnostics display. This can conflict with other diagnostics plugins like Eclim, so when enabling Java support, please manually disable Eclim Java diagnostics.
Add the following to your vimrc
:
let g:EclimFileTypeValidate = 0
NOTE: We recommend disabling Eclim entirely when editing Java with YCM's
native Java support. This can be done temporarily with :EclimDisable
.
Eclipse-style projects require two files: .project and .classpath.
If your project already has these files due to previously being set up within Eclipse, then no setup is required. jdt.ls should load the project just fine (it's basically eclipse after all).
However, if not, it is possible (easy in fact) to craft them manually, though it is not recommended. You're better off using Gradle or Maven (see below).
A simple eclipse style project example can be found in
the ycmd test directory. Normally all that is required is to copy these files to
the root of your project and to edit the .classpath
to add additional
libraries, such as:
<classpathentry kind="lib" path="/path/to/external/jar" />
<classpathentry kind="lib" path="/path/to/external/java/source" />
It may also be necessary to change the directory in which your source files are located (paths are relative to the .project file itself):
<classpathentry kind="src" output="target/classes" path="path/to/src/" />
NOTE: The eclipse project and classpath files are not a public interface and it is highly recommended to use Maven or Gradle project definitions if you don't already use Eclipse to manage your projects.
Maven needs a file named pom.xml in the root of the project. Once again a simple pom.xml can be found in the ycmd source.
The format of pom.xml files is way beyond the scope of this document, but we do recommend using the various tools that can generate them for you, if you're not familiar with them already.
Gradle projects require a build.gradle. Again, there is a trivial example in ycmd's tests.
The format of build.gradle files are way beyond the scope of this document, but we do recommend using the various tools that can generate them for you if you're not familiar with them already.
Some users have experienced issues with their jdt.ls when using the Groovy language for their build.gradle. As such, try using Kotlin instead.
If you're not getting completions or diagnostics, check the server health:
- The Java completion engine takes a while to start up and parse your project.
You should be able to see its progress in the command line, and
:YcmDebugInfo
. Ensure that the following lines are present:
-- jdt.ls Java Language Server running
-- jdt.ls Java Language Server Startup Status: Ready
- If the above lines don't appear after a few minutes, check the jdt.ls and ycmd
log files using
:YcmToggleLogs
. The jdt.ls log file is called.log
(for some reason).
If you get a message about "classpath is incomplete", then make sure you have correctly configured the project files.
If you get messages about unresolved imports, then make sure you have correctly configured the project files, in particular check that the classpath is set correctly.
YCM relies on OmniSharp-Roslyn to provide completion and code navigation. OmniSharp-Roslyn needs a solution file for a C# project and there are two ways of letting YCM know about your solution files.
YCM will scan all parent directories of the file currently being edited and look
for a file with .sln
extension.
If YCM loads .ycm_extra_conf.py
which contains CSharpSolutionFile
function,
YCM will try to use that to determine the solution file. This is useful when one
wants to override the default behaviour and specify a solution file that is not
in any of the parent directories of the currently edited file. Example:
def CSharpSolutionFile( filepath ):
# `filepath` is the path of the file user is editing
return '/path/to/solution/file' # Can be relative to the `.ycm_extra_conf.py`
If the path returned by CSharpSolutionFile
is not an actual file, YCM will
fall back to the other way of finding the file.