docs/versions/9.3.0/reference/glossary.mdx

---
title: 'Bazel Glossary'
---

### Action {#action}

A command to run during the build, for example, a call to a compiler that takes
[artifacts](#artifact) as inputs and produces other artifacts as outputs.
Includes metadata like the command line arguments, action key, environment
variables, and declared input/output artifacts.

**See also:** [Rules documentation](/versions/9.3.0/extending/rules#actions)

### Action cache {#action-cache}

An on-disk cache that stores a mapping of executed [actions](#action) to the
outputs they created. The cache key is known as the [action key](#action-key). A
core component for Bazel's incrementality model. The cache is stored in the
output base directory and thus survives Bazel server restarts.

### Action graph {#action-graph}

An in-memory graph of [actions](#action) and the [artifacts](#artifact) that
these actions read and generate. The graph might include artifacts that exist as
source files (for example, in the file system) as well as generated
intermediate/final artifacts that are not mentioned in `BUILD` files. Produced
during the [analysis phase](#analysis-phase) and used during the [execution
phase](#execution-phase).

### Action graph query (aquery) {#action-graph-query}

A [query](#query-concept) tool that can query over build [actions](#action).
This provides the ability to analyze how [build rules](#rule) translate into the
actual work builds do.

### Action key {#action-key}

The cache key of an [action](#action). Computed based on action metadata, which
might include the command to be executed in the action, compiler flags, library
locations, or system headers, depending on the action. Enables Bazel to cache or
invalidate individual actions deterministically.

### Analysis phase {#analysis-phase}

The second phase of a build. Processes the [target graph](#target-graph)
specified in [`BUILD` files](#build-file) to produce an in-memory [action
graph](#action-graph) that determines the order of actions to run during the
[execution phase](#execution-phase). This is the phase in which rule
implementations are evaluated.

### Artifact {#artifact}

A source file or a generated file. Can also be a directory of files, known as
[tree artifacts](#tree-artifact).

An artifact may be an input to multiple actions, but must only be generated by
at most one action.

An artifact that corresponds to a [file target](#target) can be addressed by a
label.

### Aspect {#aspect}

A mechanism for rules to create additional [actions](#action) in their
dependencies. For example, if target A depends on B, one can apply an aspect on
A that traverses *up* a dependency edge to B, and runs additional actions in B
to generate and collect additional output files. These additional actions are
cached and reused between targets requiring the same aspect. Created with the
`aspect()` Starlark Build API function. Can be used, for example, to generate
metadata for IDEs, and create actions for linting.

**See also:** [Aspects documentation](/versions/9.3.0/extending/aspects)

### Aspect-on-aspect {#aspect-on-aspect}

A composition mechanism whereby aspects can be applied to the results
of other aspects. For example, an aspect that generates information for use by
IDEs can be applied on top of an aspect that generates `.java` files from a
proto.

For an aspect `A` to apply on top of aspect `B`, the [providers](#provider) that
`B` advertises in its [`provides`](/versions/9.3.0/rules/lib/globals#aspect.provides) attribute
must match what `A` declares it wants in its [`required_aspect_providers`](/versions/9.3.0/rules/lib/globals#aspect.required_aspect_providers)
attribute.

### Attribute {#attribute}

A parameter to a [rule](#rule), used to express per-target build information.
Examples include `srcs`, `deps`, and `copts`, which respectively declare a
target's source files, dependencies, and custom compiler options. The particular
attributes available for a given target depend on its rule type.

### .bazelrc {#bazelrc}

Bazel’s configuration file used to change the default values for [startup
flags](#startup-flags) and [command flags](#command-flags), and to define common
groups of options that can then be set together on the Bazel command line using
a `--config` flag. Bazel can combine settings from multiple bazelrc files
(systemwide, per-workspace, per-user, or from a custom location), and a
`bazelrc` file may also import settings from other `bazelrc` files.

### Blaze {#blaze}

The Google-internal version of Bazel. Google’s main build system for its
mono-repository.

### BUILD File {#build-file}

A `BUILD` file is the main configuration file that tells Bazel what software
outputs to build, what their dependencies are, and how to build them. Bazel
takes a `BUILD` file as input and uses the file to create a graph of dependencies
and to derive the actions that must be completed to build intermediate and final
software outputs. A `BUILD` file marks a directory and any sub-directories not
containing a `BUILD` file as a [package](#package), and can contain
[targets](#target) created by [rules](#rule). The file can also be named
`BUILD.bazel`.

### BUILD.bazel File {#build-bazel-file}

See [`BUILD` File](#build-file). Takes precedence over a `BUILD` file in the same
directory.

### .bzl File {#bzl-file}

A file that defines rules, [macros](#macro), and constants written in
[Starlark](#starlark). These can then be imported into [`BUILD`
files](#build-file) using the `load()` function.

{/*  TODO: ### Build event protocol  */}

{/*  TODO: ### Build flag  */}

### Build graph {#build-graph}

The dependency graph that Bazel constructs and traverses to perform a build.
Includes nodes like [targets](#target), [configured
targets](#configured-target), [actions](#action), and [artifacts](#artifact). A
build is considered complete when all [artifacts](#artifact) on which a set of
requested targets depend are verified as up-to-date.

### Build setting {#build-setting}

A Starlark-defined piece of [configuration](#configuration).
[Transitions](#transition) can set build settings to change a subgraph's
configuration. If exposed to the user as a [command-line flag](#command-flags),
also known as a build flag.

### Clean build {#clean-build}

A build that doesn't use the results of earlier builds. This is generally slower
than an [incremental build](#incremental-build) but commonly considered to be
more [correct](#correctness). Bazel guarantees both clean and incremental builds
are always correct.

### Client-server model {#client-server-model}

The `bazel` command-line client automatically starts a background server on the
local machine to execute Bazel [commands](#command). The server persists across
commands but automatically stops after a period of inactivity (or explicitly via
bazel shutdown). Splitting Bazel into a server and client helps amortize JVM
startup time and supports faster [incremental builds](#incremental-build)
because the [action graph](#action-graph) remains in memory across commands.

### Command {#command}

Used on the command line to invoke different Bazel functions, like `bazel
build`, `bazel test`, `bazel run`, and `bazel query`.

### Command flags {#command-flags}

A set of flags specific to a [command](#command). Command flags are specified
*after* the command (`bazel build <command flags>`). Flags can be applicable to
one or more commands. For example, `--configure` is a flag exclusively for the
`bazel sync` command, but `--keep_going` is applicable to `sync`, `build`,
`test` and more. Flags are often used for [configuration](#configuration)
purposes, so changes in flag values can cause Bazel to invalidate in-memory
graphs and restart the [analysis phase](#analysis-phase).

### Configuration {#configuration}

Information outside of [rule](#rule) definitions that impacts how rules generate
[actions](#action). Every build has at least one configuration specifying the
target platform, action environment variables, and command-line [build
flags](#command-flags). [Transitions](#transition) may create additional
configurations, such as for host tools or cross-compilation.

**See also:** [Configurations](/versions/9.3.0/extending/rules#configurations)

{/*  TODO: ### Configuration fragment  */}

### Configuration trimming {#config-trimming}

The process of only including the pieces of [configuration](#configuration) a
target actually needs. For example, if you build Java binary `//:j` with C++
dependency `//:c`, it's wasteful to include the value of `--javacopt` in the
configuration of `//:c` because changing `--javacopt` unnecessarily breaks C++
build cacheability.

### Configured query (cquery) {#configured-query}

A [query](#query-concept) tool that queries over [configured
targets](#configured-target) (after the [analysis phase](#analysis-phase)
completes). This means `select()` and [build flags](#command-flags) (such as
`--platforms`) are accurately reflected in the results.

**See also:** [cquery documentation](/versions/9.3.0/query/cquery)

### Configured target {#configured-target}

The result of evaluating a [target](#target) with a
[configuration](#configuration). The [analysis phase](#analysis-phase) produces
this by combining the build's options with the targets that need to be built.
For example, if `//:foo` builds for two different architectures in the same
build, it has two configured targets: `<//:foo, x86>` and `<//:foo, arm>`.

### Correctness {#correctness}

A build is correct when its output faithfully reflects the state of its
transitive inputs. To achieve correct builds, Bazel strives to be
[hermetic](#hermeticity), reproducible, and making [build
analysis](#analysis-phase) and [action execution](#execution-phase)
deterministic.

### Dependency {#dependency}

A directed edge between two [targets](#target). A target `//:foo` has a *target
dependency* on target `//:bar` if `//:foo`'s attribute values contain a
reference to `//:bar`. `//:foo` has an *action dependency* on `//:bar` if an
action in `//:foo` depends on an input [artifact](#artifact) created by an
action in `//:bar`.

In certain contexts, it could also refer to an _external dependency_; see
[modules](#module).

### Depset {#depset}

A data structure for collecting data on transitive dependencies. Optimized so
that merging depsets is time and space efficient, because it’s common to have
very large depsets (hundreds of thousands of files). Implemented to
recursively refer to other depsets for space efficiency reasons. [Rule](#rule)
implementations should not "flatten" depsets by converting them to lists unless
the rule is at the top level of the build graph. Flattening large depsets incurs
huge memory consumption. Also known as *nested sets* in Bazel's internal
implementation.

**See also:** [Depset documentation](/versions/9.3.0/extending/depsets)

### Disk cache {#disk-cache}

A local on-disk blob store for the remote caching feature. Can be used in
conjunction with an actual remote blob store.

### Distdir {#distdir}

A read-only directory containing files that Bazel would otherwise fetch from the
internet using repository rules. Enables builds to run fully offline.

### Dynamic execution {#dynamic-execution}

An execution strategy that selects between local and remote execution based on
various heuristics, and uses the execution results of the faster successful
method. Certain [actions](#action) are executed faster locally (for example,
linking) and others are faster remotely (for example, highly parallelizable
compilation). A dynamic execution strategy can provide the best possible
incremental and clean build times.

### Execution phase {#execution-phase}

The third phase of a build. Executes the [actions](#action) in the [action
graph](#action-graph) created during the [analysis phase](#analysis-phase).
These actions invoke executables (compilers, scripts) to read and write
[artifacts](#artifact). *Spawn strategies* control how these actions are
executed: locally, remotely, dynamically, sandboxed, docker, and so on.

### Execution root {#execution-root}

A directory in the [workspace](#workspace)’s [output base](#output-base)
directory where local [actions](#action) are executed in
non-[sandboxed](#sandboxing) builds. The directory contents are mostly symlinks
of input [artifacts](#artifact) from the workspace. The execution root also
contains symlinks to external repositories as other inputs and the `bazel-out`
directory to store outputs. Prepared during the [loading phase](#loading-phase)
by creating a *symlink forest* of the directories that represent the transitive
closure of packages on which a build depends. Accessible with `bazel info
execution_root` on the command line.

### File {#file}

See [Artifact](#artifact).

### Hermeticity {#hermeticity}

A build is hermetic if there are no external influences on its build and test
operations, which helps to make sure that results are deterministic and
[correct](#correctness). For example, hermetic builds typically disallow network
access to actions, restrict access to declared inputs, use fixed timestamps and
timezones, restrict access to environment variables, and use fixed seeds for
random number generators

### Incremental build {#incremental-build}

An incremental build reuses the results of earlier builds to reduce build time
and resource usage. Dependency checking and caching aim to produce correct
results for this type of build. An incremental build is the opposite of a clean
build.

{/*  TODO: ### Install base  */}

### Label {#label}

An identifier for a [target](#target). Generally has the form
`@repo//path/to/package:target`, where `repo` is the (apparent) name of the
[repository](#repository) containing the target, `path/to/package` is the path
to the directory that contains the [`BUILD` file](#build-file) declaring the
target (this directory is also known as the [package](#package)), and `target`
is the name of the target itself. Depending on the situation, parts of this
syntax may be omitted.

**See also**: [Labels](/versions/9.3.0/concepts/labels)

### Loading phase {#loading-phase}

The first phase of a build where Bazel executes [`BUILD` files](#build-file) to
create [packages](#package). [Macros](#macro) and certain functions like
`glob()` are evaluated in this phase. Interleaved with the second phase of the
build, the [analysis phase](#analysis-phase), to build up a [target
graph](#target-graph).

### Legacy macro {#legacy-macro}

A flavor of [macro](#macro) which is declared as an ordinary
[Starlark](#starlark) function, and which runs as a side effect of executing a
`BUILD` file.

Legacy macros can do anything a function can. This means they can be convenient,
but they can also be harder to read, write, and use. A legacy macro might
unexpectedly mutate its arguments or fail when given a `select()` or ill-typed
argument.

Contrast with [symbolic macros](#symbolic-macro).

**See also:** [Legacy macro documentation](/versions/9.3.0/extending/legacy-macros)

### Macro {#macro}

A mechanism to compose multiple [rule](#rule) target declarations together under
a single [Starlark](#starlark) callable. Enables reusing common rule declaration
patterns across `BUILD` files. Expanded to the underlying rule target
declarations during the [loading phase](#loading-phase).

Comes in two flavors: [symbolic macros](#symbolic-macro) (since Bazel 8) and
[legacy macros](#legacy-macro).

### Mnemonic {#mnemonic}

A short, human-readable string selected by a rule author to quickly understand
what an [action](#action) in the rule is doing. Mnemonics can be used as
identifiers for *spawn strategy* selections. Some examples of action mnemonics
are `Javac` from Java rules, `CppCompile` from C++ rules, and
`AndroidManifestMerger` from Android rules.

### Module {#module}

A Bazel project that can have multiple versions, each of which can have
dependencies on other modules. This is analogous to familiar concepts in other
dependency management systems, such as a Maven _artifact_, an npm _package_, a
Go _module_, or a Cargo _crate_. Modules form the backbone of Bazel's external
dependency management system.

Each module is backed by a [repo](#repository) with a `MODULE.bazel` file at its
root. This file contains metadata about the module itself (such as its name and
version), its direct dependencies, and various other data including toolchain
registrations and [module extension](#module-extension) input.

Module metadata is hosted in Bazel registries.

**See also:** [Bazel modules](/versions/9.3.0/external/module)

### Module Extension {#module-extension}

A piece of logic that can be run to generate [repos](#repository) by reading
inputs from across the [module](#module) dependency graph and invoking [repo
rules](#repository-rule). Module extensions have capabilities similar to repo
rules, allowing them to access the internet, perform file I/O, and so on.

**See also:** [Module extensions](/versions/9.3.0/external/extension)

### Native rules {#native-rules}

[Rules](#rule) that are built into Bazel and implemented in Java. Such rules
appear in [`.bzl` files](#bzl-file) as functions in the native module (for
example, `native.cc_library` or `native.java_library`). User-defined rules
(non-native) are created using [Starlark](#starlark).

### Output base {#output-base}

A [workspace](#workspace)-specific directory to store Bazel output files. Used
to separate outputs from the *workspace*'s source tree (the [main
repo](#repository)). Located in the [output user root](#output-user-root).

### Output groups {#output-groups}

A group of files that is expected to be built when Bazel finishes building a
target. [Rules](#rule) put their usual outputs in the "default output group"
(e.g the `.jar` file of a `java_library`, `.a` and `.so` for `cc_library`
targets). The default output group is the output group whose
[artifacts](#artifact) are built when a target is requested on the command line.
Rules can define more named output groups that can be explicitly specified in
[`BUILD` files](#build-file) (`filegroup` rule) or the command line
(`--output_groups` flag).

### Output user root {#output-user-root}

A user-specific directory to store Bazel's outputs. The directory name is
derived from the user's system username. Prevents output file collisions if
multiple users are building the same project on the system at the same time.
Contains subdirectories corresponding to build outputs of individual workspaces,
also known as [output bases](#output-base).

### Package {#package}

The set of [targets](#target) defined by a [`BUILD` file](#build-file). A
package's name is the `BUILD` file's path relative to the [repo](#repository)
root. A package can contain subpackages, or subdirectories containing `BUILD`
files, thus forming a package hierarchy.

### Package group {#package-group}

A [target](#target) representing a set of packages. Often used in `visibility`
attribute values.

### Platform {#platform}

A "machine type" involved in a build. This includes the machine Bazel runs on
(the "host" platform), the machines build tools execute on ("exec" platforms),
and the machines targets are built for ("target platforms").

### Provider {#provider}

A schema describing a unit of information to pass between
[rule targets](#rule-target) along dependency relationships. Typically this
contains information like compiler options, transitive source or output files,
and build metadata. Frequently used in conjunction with [depsets](#depset) to
efficiently store accumulated transitive data. An example of a built-in provider
is `DefaultInfo`.

Note: The object holding specific data for a given rule target is
referred to as a "provider instance", although sometimes this is conflated with
"provider".

**See also:** [Provider documentation](/versions/9.3.0/extending/rules#providers)

### Query (concept) {#query-concept}

The process of analyzing a [build graph](#build-graph) to understand
[target](#target) properties and dependency structures. Bazel supports three
query variants: [query](#query-command), [cquery](#configured-query), and
[aquery](#action-graph-query).

### query (command) {#query-command}

A [query](#query-concept) tool that operates over the build's post-[loading
phase](#loading-phase) [target graph](#target-graph). This is relatively fast,
but can't analyze the effects of `select()`, [build flags](#command-flags),
[artifacts](#artifact), or build [actions](#action).

**See also:** [Query how-to](/versions/9.3.0/query/guide), [Query reference](/versions/9.3.0/query/language)

### Repository {#repository}

A directory tree with a boundary marker file at its root, containing source
files that can be used in a Bazel build. Often shortened to just **repo**.

A repo boundary marker file can be `MODULE.bazel` (signaling that this repo
represents a Bazel module), `REPO.bazel`, or in legacy contexts, `WORKSPACE` or
`WORKSPACE.bazel`. Any repo boundary marker file will signify the boundary of a
repo; multiple such files can coexist in a directory.

The *main repo* is the repo in which the current Bazel command is being run.

*External repos* are defined by specifying [modules](#module) in `MODULE.bazel`
files, or invoking [repo rules](#repository-rule) in [module
extensions](#module-extension). They can be fetched on demand to a predetermined
"magical" location on disk.

Each repo has a unique, constant *canonical* name, and potentially different
*apparent* names when viewed from other repos.

**See also**: [External dependencies overview](/versions/9.3.0/external/overview)

### Repository cache {#repo-cache}

A shared content-addressable cache of files downloaded by Bazel for builds,
shareable across [workspaces](#workspace). Enables offline builds after the
initial download. Commonly used to cache files downloaded through [repository
rules](#repository-rule) like `http_archive` and repository rule APIs like
`repository_ctx.download`. Files are cached only if their SHA-256 checksums are
specified for the download.

### Repository rule {#repository-rule}

A schema for repository definitions that tells Bazel how to materialize (or
"fetch") a [repository](#repository). Often shortened to just **repo rule**.
Repo rules are invoked by Bazel internally to define repos backed by
[modules](#module), or can be invoked by [module extensions](#module-extension).
Repo rules can access the internet or perform file I/O; the most common repo
rule is `http_archive` to download an archive containing source files from the
internet.

**See also:** [Repo rule documentation](/versions/9.3.0/external/repo)

### Reproducibility {#reproducibility}

The property of a build or test that a set of inputs to the build or test will
always produce the same set of outputs every time, regardless of time, method,
or environment. Note that this does not necessarily imply that the outputs are
[correct](#correctness) or the desired outputs.

### Rule {#rule}

A schema for defining [rule targets](#rule-target) in a `BUILD` file, such as
`cc_library`. From the perspective of a `BUILD` file author, a rule consists of
a set of [attributes](#attributes) and black box logic. The logic tells the
rule target how to produce output [artifacts](#artifact) and pass information to
other rule targets. From the perspective of `.bzl` authors, rules are the
primary way to extend Bazel to support new programming languages and
environments.

Rules are instantiated to produce rule targets in the
[loading phase](#loading-phase). In the [analysis phase](#analysis-phase) rule
targets communicate information to their downstream dependencies in the form of
[providers](#provider), and register [actions](#action) describing how to
generate their output artifacts. These actions are run in the [execution
phase](#execution-phase).

Note: Historically the term "rule" has been used to refer to a rule target.
This usage was inherited from tools like Make, but causes confusion and should
be avoided for Bazel.

**See also:** [Rules documentation](/versions/9.3.0/extending/rules)

### Rule target {#rule-target}

A [target](#target) that is an instance of a rule. Contrasts with file targets
and package groups. Not to be confused with [rule](#rule).

### Runfiles {#runfiles}

The runtime dependencies of an executable [target](#target). Most commonly, the
executable is the executable output of a test rule, and the runfiles are runtime
data dependencies of the test. Before the invocation of the executable (during
bazel test), Bazel prepares the tree of runfiles alongside the test executable
according to their source directory structure.

**See also:** [Runfiles documentation](/versions/9.3.0/extending/rules#runfiles)

### Sandboxing {#sandboxing}

A technique to isolate a running [action](#action) inside a restricted and
temporary [execution root](#execution-root), helping to ensure that it doesn’t
read undeclared inputs or write undeclared outputs. Sandboxing greatly improves
[hermeticity](#hermeticity), but usually has a performance cost, and requires
support from the operating system. The performance cost depends on the platform.
On Linux, it's not significant, but on macOS it can make sandboxing unusable.

### Skyframe {#skyframe}

[Skyframe](/versions/9.3.0/reference/skyframe) is the core parallel, functional, and incremental evaluation framework of Bazel.

{/*  TODO: ### Spawn strategy  */}

### Stamping {#stamping}

A feature to embed additional information into Bazel-built
[artifacts](#artifact). For example, this can be used for source control, build
time and other workspace or environment-related information for release builds.
Enable through the `--workspace_status_command` flag and [rules](/versions/9.3.0/extending/rules) that
support the stamp attribute.

### Starlark {#starlark}

The extension language for writing [rules](/versions/9.3.0/extending/rules) and [macros](#macro). A
restricted subset of Python (syntactically and grammatically) aimed for the
purpose of configuration, and for better performance. Uses the [`.bzl`
file](#bzl-file) extension. [`BUILD` files](#build-file) use an even more
restricted version of Starlark (such as no `def` function definitions), formerly
known as Skylark.

**See also:** [Starlark language documentation](/versions/9.3.0/rules/language)

{/*  TODO: ### Starlark rules  */}

{/*  TODO: ### Starlark rule sandwich  */}

### Startup flags {#startup-flags}

The set of flags specified between `bazel` and the [command](#query-command),
for example, bazel `--host_jvm_debug` build. These flags modify the
[configuration](#configuration) of the Bazel server, so any modification to
startup flags causes a server restart. Startup flags are not specific to any
command.

### Symbolic macro {#symbolic-macro}

A flavor of [macro](#macro) which is declared with a [rule](#rule)-like
[attribute](#attribute) schema, allows hiding internal declared
[targets](#target) from their own package, and enforces a predictable naming
pattern on the targets that the macro declares. Designed to avoid some of the
problems seen in large [legacy macro](#legacy-macro) codebases.

**See also:** [Symbolic macro documentation](/versions/9.3.0/extending/macros)

### Target {#target}

An object that is defined in a [`BUILD` file](#build-file) and identified by a
[label](#label). Targets represent the buildable units of a workspace from
the perspective of the end user.

A target that is declared by instantiating a [rule](#rule) is called a [rule
target](#rule-target). Depending on the rule, these may be runnable (like
`cc_binary`) or testable (like `cc_test`). Rule targets typically depend on
other targets via their [attributes](#attribute) (such as `deps`); these
dependencies form the basis of the [target graph](#target-graph).

Aside from rule targets, there are also file targets and [package group](#package-group)
targets. File targets correspond to [artifacts](#artifact) that are referenced
within a `BUILD` file. As a special case, the `BUILD` file of any package is
always considered a source file target in that package.

Targets are discovered during the [loading phase](#loading-phase). During the
[analysis phase](#analysis-phase), targets are associated with [build
configurations](#configuration) to form [configured
targets](#configured-target).

### Target graph {#target-graph}

An in-memory graph of [targets](#target) and their dependencies. Produced during
the [loading phase](#loading-phase) and used as an input to the [analysis
phase](#analysis-phase).

### Target pattern {#target-pattern}

A way to specify a group of [targets](#target) on the command line. Commonly
used patterns are `:all` (all rule targets), `:*` (all rule + file targets),
`...` (current [package](#package) and all subpackages recursively). Can be used
in combination, for example, `//...:*` means all rule and file targets in all
packages recursively from the root of the [workspace](#workspace).

### Tests {#tests}

Rule [targets](#target) instantiated from test rules, and therefore contains a
test executable. A return code of zero from the completion of the executable
indicates test success. The exact contract between Bazel and tests (such as test
environment variables, test result collection methods) is specified in the [Test
Encyclopedia](/versions/9.3.0/reference/test-encyclopedia).

### Toolchain {#toolchain}

A set of tools to build outputs for a language. Typically, a toolchain includes
compilers, linkers, interpreters or/and linters. A toolchain can also vary by
platform, that is, a Unix compiler toolchain's components may differ for the
Windows variant, even though the toolchain is for the same language. Selecting
the right toolchain for the platform is known as toolchain resolution.

### Top-level target {#top-level-target}

A build [target](#target) is top-level if it’s requested on the Bazel command
line. For example, if `//:foo` depends on `//:bar`, and `bazel build //:foo` is
called, then for this build, `//:foo` is top-level, and `//:bar` isn’t
top-level, although both targets will need to be built. An important difference
between top-level and non-top-level targets is that [command
flags](#command-flags) set on the Bazel command line (or via
[.bazelrc](#bazelrc)) will set the [configuration](#configuration) for top-level
targets, but might be modified by a [transition](#transition) for non-top-level
targets.

### Transition {#transition}

A mapping of [configuration](#configuration) state from one value to another.
Enables [targets](#target) in the [build graph](#build-graph) to have different
configurations, even if they were instantiated from the same [rule](#rule). A
common usage of transitions is with *split* transitions, where certain parts of
the [target graph](#target-graph) is forked with distinct configurations for
each fork. For example, one can build an Android APK with native binaries
compiled for ARM and x86 using split transitions in a single build.

**See also:** [User-defined transitions](/versions/9.3.0/extending/config#user-defined-transitions)

### Tree artifact {#tree-artifact}

An [artifact](#artifact) that represents a collection of files. Since these
files are not themselves artifacts, an [action](#action) operating on them must
instead register the tree artifact as its input or output.

### Visibility {#visibility}

One of two mechanisms for preventing unwanted dependencies in the build system:
*target visibility* for controlling whether a [target](#target) can be depended
upon by other targets; and *load visibility* for controlling whether a `BUILD`
or `.bzl` file may load a given `.bzl` file. Without context, usually
"visibility" refers to target visibility.

**See also:** [Visibility documentation](/versions/9.3.0/concepts/visibility)

### Workspace {#workspace}

The environment shared by all Bazel commands run from the same [main
repository](#repository).

Note that historically the concepts of "repository" and "workspace" have been
conflated; the term "workspace" has often been used to refer to the main
repository, and sometimes even used as a synonym of "repository". Such usage
should be avoided for clarity.