This guide is meant to provide instructions for how to enable precise code intelligence for any programming language. The general steps for enabling precise code intelligence are as follows:
- Install Sourcegraph CLI
- Install LSIF Indexer
- Generate LSIF data
- Upload LSIF data
1. Install Sourcegraph CLI
The Sourcegraph CLI (
src) is used for uploading LSIF data to your Sourcegraph instance (replace
darwin for macOS):
$ curl -L https://sourcegraph.com/.api/src-cli/src_linux_amd64 -o /usr/local/bin/src $ chmod +x /usr/local/bin/src
2. Install LSIF indexer
An LSIF indexer is a command line tool that performs code analysis of source code and outputs a file with metadata containing all the definitions, references, and hover documentation in a project in LSIF. The LSIF file is then uploaded to Sourcegraph to power code intelligence features.
Install the indexer for the required programming language of your repository by following the instructions in the indexer’s README:
3. Generate LSIF data
To generate the LSIF data for your repository run the command in the generate LSIF data step found in the README of the installed indexer.
4. Upload LSIF data
The upload step is the same for all languages. Make sure the current working directory is a path inside your repository, then use the Sourcegraph CLI to upload the LSIF file:
To a private Sourcegraph instance (on prem)
$ src -endpoint=<your sourcegraph endpoint> lsif upload -file=<LSIF file (e.g. dump.lsif)>
To cloud based Sourcegraph.com
$ src lsif upload -github-token=<your github token> -file=<LSIF file (e.g. dump.lsif)>
src-cli upload command will try to infer the repository and git commit by invoking git commands on your local clone. If git is not installed, is older than version 2.7.0 or you are running on code outside of a git clone, you will need to also specify the
-commit flags explicitly.
On successful upload you’ll see the following message:
Repository: <location of repository> Commit: <40-character commit associated with this LSIF upload> File: <LSIF data file> Root: <subdirectory in the repository where this LSIF dump was generated> LSIF dump successfully uploaded for processing. View processing status at <link to your Sourcegraph instance LSIF status>.
Automate code indexing
Now that you have successfully enabled code intelligence for your repository, you can automate source code indexing to ensure precise code intelligence stays up to date with the most recent code changes in the repository. See our continuous integration guide to setup automation.
Once LSIF data has uploaded, open the Sourcegraph UI or your code host (i.e. GitHub) and navigate to any code file that was part of the repository that was analyzed by the LSIF indexer. Hover over a symbol, variable or function name in the file, you should now see rich LSIF metadata as the source for hover-tooltips, definitions, and references.
If precise code intelligence has been correctly enabled hover text should not be decorated with a icon. This icon indicates the results are from search-based search-based code intelligence. This tooltip icon will be absent when results are precise!
To view LSIF indexer processing failures go to Repository settings > Code intelligence > Activity for this repository in your Sourcegraph instance. Failures can occur if the LSIF data is invalid (e.g., malformed indexer output), or problems were encountered during processing (e.g., system-level bug, flaky connections, etc). Try again or file an issue if the problem persists.
Possible errors that can happen during upload include:
- Clone in progress: the instance doesn’t have the necessary data to process your upload yet, retry in a few minutes
- Unknown repository (404): check your
-endpointand make sure you can view the repository on your Sourcegraph instance
- Invalid commit (404): try visiting the repository at that commit on your Sourcegraph instance to trigger an update
- Invalid auth when using Sourcegraph.com or when
true(401 for an invalid token or 404 if the repository cannot be found on GitHub.com): make sure your GitHub token is valid and that the repository is correct
- Unexpected errors (500s): file an issue