Configurable build attributes

Configurable attributes, commonly known as select(), is a Bazel feature that lets users toggle the values of BUILD rule attributes at the command line.

This can be used, for example, for a multiplatform library that automatically chooses the appropriate implementation for the architecture, or for a feature-configurable binary that can be customized at build time.

Example

# myapp/BUILD

cc_binary(
    name = "mybinary",
    srcs = ["main.cc"],
    deps = select({
        ":arm_build": [":arm_lib"],
        ":x86_debug_build": [":x86_dev_lib"],
        "//conditions:default": [":generic_lib"],
    }),
)

config_setting(
    name = "arm_build",
    values = {"cpu": "arm"},
)

config_setting(
    name = "x86_debug_build",
    values = {
        "cpu": "x86",
        "compilation_mode": "dbg",
    },
)

This declares a cc_binary that “chooses” its deps based on the flags at the command line. Specficially, deps becomes:

select() serves as a placeholder for a value that will be chosen based on configuration conditions, which are labels referencing config_setting targets. By using select() in a configurable attribute, the attribute effectively adopts different values when different conditions hold.

Matches must be unambiguous: either exactly one condition must match or, if multiple conditions match, one’s values must be a strict superset of all others’. For example, values = {"cpu": "x86", "compilation_mode": "dbg"} is an unambiguous specialization of values = {"cpu": "x86"}. The built-in condition //conditions:default automatically matches when nothing else does.

While this example uses deps, select() works just as well on srcs, resources, cmd, and most other attributes. Only a small number of attributes are non-configurable, and these are clearly annotated. For example, config_setting’s own values attribute is non-configurable.

select() and dependencies

Certain attributes change the build parameters for all transitive dependencies under a target. For example, genrule’s tools changes --cpu to the CPU of the machine running Bazel (which, thanks to cross-compilation, may be different than the CPU the target is built for). This is known as a configuration transition.

Given

#myapp/BUILD

config_setting(
    name = "arm_cpu",
    values = {"cpu": "arm"},
)

config_setting(
    name = "x86_cpu",
    values = {"cpu": "x86"},
)

genrule(
    name = "my_genrule",
    srcs = select({
        ":arm_cpu": ["g_arm.src"],
        ":x86_cpu": ["g_x86.src"],
    tools = select({
        ":arm_cpu": [":tool1"],
        ":x86_cpu": [":tool2"],
    }),
)

cc_binary(
    name = "tool1",
    srcs = select({
        ":arm_cpu": ["armtool.cc"],
        ":x86_cpu": ["x86tool.cc"],
    }),
)

running

$ bazel build //myapp:my_genrule --cpu=arm

on an x86 developer machine binds the build to g_arm.src, tool1, and x86tool.cc. Both of the selects attached to my_genrule use my_genrule’s build parameters, which include --cpu=arm. The tools attribute changes --cpu to x86 for tool1 and its transitive dependencies. The select on tool1 uses tool1’s build parameters, which include --cpu=x86.

Configuration conditions

Each key in a configurable attribute is a label reference to a config_setting or constraint_value.

config_setting is just a collection of expected command line flag settings. By encapsulating these in a target, it’s easy to maintain “standard” conditions users can reference from multiple places.

constraint_value provides support for multi-platform behavior.

Built-in flags

Flags like --cpu are built into Bazel: the build tool natively understands them for all builds in all projects. These are specified with config_setting’s values attribute:

config_setting(
    name = "meaningful_condition_name",
    values = {
        "flag1": "value1",
        "flag2": "value2",
        ...
    },
)

flagN is a flag name (without --, so "cpu" instead of "--cpu"). valueN is the expected value for that flag. :meaningful_condition_name matches if every entry in values matches. Order is irrelevant.

valueN is parsed as if it was set on the command line. This means:

• values = { "compilation_mode": "opt" } matches bazel build -c opt
• values = { "force_pic": "true" } matches bazel build --force_pic=1
• values = { "force_pic": "0" } matches bazel build --noforce_pic

config_setting only supports flags that affect target behavior. For example, --show_progress isn’t allowed because it only affects how Bazel reports progress to the user. Targets can’t use that flag to construct their results. The exact set of supported flags isn’t documented. In practice, most flags that “make sense” work.

Custom flags

You can model your own project-specific flags with Starlark build settings. Unlike built-in flags, these are defined as build targets, so Bazel references them with target labels.

These are triggered with config_setting’s flag_values attribute:

config_setting(
    name = "meaningful_condition_name",
    flag_values = {
        "//myflags:flag1": "value1",
        "//myflags:flag2": "value2",
        ...
    },
)

Behavior is the same as for built-in flags. See here for a working example.

--define is an alternative legacy syntax for custom flags (for example --define foo=bar). This can be expressed either in the values attribute (values = {"define": "foo=bar"}) or the define_values attribute (define_values = {"foo": "bar"}). --define is only supported for backwards compatibility. Prefer Starlark build settings whenever possible.

values, flag_values, and define_values evaluate independently. The config_setting matches if all values across all of them match.

The default condition

The built-in condition //conditions:default matches when no other condition matches.

Because of the “exactly one match” rule, a configurable attribute with no match and no default condition emits a "no matching conditions" error. This can protect against silent failures from unexpected settings:

# myapp/BUILD

config_setting(
    name = "x86_cpu",
    values = {"cpu": "x86"},
)

cc_library(
    name = "x86_only_lib",
    srcs = select({
        ":x86_cpu": ["lib.cc"],
    }),
)

$ bazel build //myapp:x86_only_lib --cpu=arm
ERROR: Configurable attribute "srcs" doesn't match this configuration (would
a default condition help?).
Conditions checked:
  //myapp:x86_cpu

For even clearer errors, you can set custom messages with select()’s no_match_error attribute.

Platforms

While the ability to specify multiple flags on the command line provides flexibility, it can also be burdensome to individually set each one every time you want to build a target. Platforms let you consolidate these into simple bundles.

# myapp/BUILD

sh_binary(
    name = "my_rocks",
    srcs = select({
        ":basalt": ["pyroxene.sh"],
        ":marble": ["calcite.sh"],
        "//conditions:default": ["feldspar.sh"],
    }),
)

config_setting(
    name = "basalt",
    constraint_values = [
        ":black",
        ":igneous",
    ],
)

config_setting(
    name = "marble",
    constraint_values = [
        ":white",
        ":metamorphic",
    ],
)

# constraint_setting acts as an enum type, and constraint_value as an enum value.
constraint_setting(name = "color")
constraint_value(name = "black", constraint_setting = "color")
constraint_value(name = "white", constraint_setting = "color")
constraint_setting(name = "texture")
constraint_value(name = "smooth", constraint_setting = "texture")
constraint_setting(name = "type")
constraint_value(name = "igneous", constraint_setting = "type")
constraint_value(name = "metamorphic", constraint_setting = "type")

platform(
    name = "basalt_platform",
    constraint_values = [
        ":black",
        ":igneous",
    ],
)

platform(
    name = "marble_platform",
    constraint_values = [
        ":white",
        ":smooth",
        ":metamorphic",
    ],
)

The platform can be specified on the command line. It activates the config_settings that contain a subset of the platform’s constraint_values, allowing those config_settings to match in select() expressions.

For example, in order to set the srcs attribute of my_rocks to calcite.sh, we can simply run

bazel build //my_app:my_rocks --platforms=//myapp:marble_platform

Without platforms, this might look something like

bazel build //my_app:my_rocks --define color=white --define texture=smooth --define type=metamorphic

select() can also directly read constraint_values:

constraint_setting(name = "type")
constraint_value(name = "igneous", constraint_setting = "type")
constraint_value(name = "metamorphic", constraint_setting = "type")
sh_binary(
    name = "my_rocks",
    srcs = select({
        ":igneous": ["igneous.sh"],
        ":metamorphic" ["metamorphic.sh"],
    }),
)

This saves the need for boilerplate config_settings when you only need to check against single values.

Platforms are still under development. See the documentation for details.

Combining select()s

select can appear multiple times in the same attribute:

sh_binary(
    name = "my_target",
    srcs = ["always_include.sh"] +
           select({
               ":armeabi_mode": ["armeabi_src.sh"],
               ":x86_mode": ["x86_src.sh"],
           }) +
           select({
               ":opt_mode": ["opt_extras.sh"],
               ":dbg_mode": ["dbg_extras.sh"],
           }),
)

select cannot appear inside another select. If you need to nest selects and your attribute takes other targets as values, use an intermediate target:

sh_binary(
    name = "my_target",
    srcs = ["always_include.sh"],
    deps = select({
        ":armeabi_mode": [":armeabi_lib"],
        ...
    }),
)

sh_library(
    name = "armeabi_lib",
    srcs = select({
        ":opt_mode": ["armeabi_with_opt.sh"],
        ...
    }),
)

If you need a select to match when multiple conditions match, consider AND chaining.

OR chaining

Consider the following:

sh_binary(
    name = "my_target",
    srcs = ["always_include.sh"],
    deps = select({
        ":config1": [":standard_lib"],
        ":config2": [":standard_lib"],
        ":config3": [":standard_lib"],
        ":config4": [":special_lib"],
    }),
)

Most conditions evaluate to the same dep. But this syntax is hard to read and maintain. It would be nice to not have to repeat [":standard_lib"] multiple times.

One option is to predefine the value as a BUILD variable:

STANDARD_DEP = [":standard_lib"]

sh_binary(
    name = "my_target",
    srcs = ["always_include.sh"],
    deps = select({
        ":config1": STANDARD_DEP,
        ":config2": STANDARD_DEP,
        ":config3": STANDARD_DEP,
        ":config4": [":special_lib"],
    }),
)

This makes it easier to manage the dependency. But it still causes unnecessary duplication.

For more direct support, use one of the following:

selects.with_or

The with_or macro in Skylib’s selects module supports ORing conditions directly inside a select:

load("@bazel_skylib//lib:selects.bzl", "selects")

sh_binary(
    name = "my_target",
    srcs = ["always_include.sh"],
    deps = selects.with_or({
        (":config1", ":config2", ":config3"): [":standard_lib"],
        ":config4": [":special_lib"],
    }),
)

selects.config_setting_group

The config_setting_group macro in Skylib’s selects module supports ORing multiple config_settings:

load("@bazel_skylib//lib:selects.bzl", "selects")

config_setting(
    name = "config1",
    values = {"cpu": "arm"},
)
config_setting(
    name = "config2",
    values = {"compilation_mode": "dbg"},
)
selects.config_setting_group(
    name = "config1_or_2",
    match_any = [":config1", ":config2"],
)
sh_binary(
    name = "my_target",
    srcs = ["always_include.sh"],
    deps = select({
        ":config1_or_2": [":standard_lib"],
        "//conditions:default": [":other_lib"],
    }),
)

Unlike selects.with_or, different targets can share :config1_or_2 across different attributes.

It’s an error for multiple conditions to match unless one is an unambiguous “specialization” of the others. See here for details.

AND chaining

If you need a select branch to match when multiple conditions match, use the Skylib macro config_setting_group:

```python
config_setting(
    name = "config1",
    values = {"cpu": "arm"},
)
config_setting(
    name = "config2",
    values = {"compilation_mode": "dbg"},
)
selects.config_setting_group(
    name = "config1_and_2",
    match_all = [":config1", ":config2"],
)
sh_binary(
    name = "my_target",
    srcs = ["always_include.sh"],
    deps = select({
        ":config1_and_2": [":standard_lib"],
        "//conditions:default": [":other_lib"],
        }),
)

Unlike OR chaining, existing config_settings can’t be directly ANDed inside a select. You have to explicitly wrap them in a config_setting_group.

Custom error messages

By default, when no condition matches, the target the select() is attached to fails with the error:

ERROR: Configurable attribute "deps" doesn't match this configuration (would
a default condition help?).
Conditions checked:
  //tools/cc_target_os:darwin
  //tools/cc_target_os:android

This can be customized with the no_match_error attribute:

cc_library(
    name = "my_lib",
    deps = select(
        {
            "//tools/cc_target_os:android": [":android_deps"],
            "//tools/cc_target_os:windows": [":windows_deps"],
        },
        no_match_error = "Please build with an Android or Windows toolchain",
    ),
)

$ bazel build //myapp:my_lib
ERROR: Configurable attribute "deps" doesn't match this configuration: Please
build with an Android or Windows toolchain

Rules compatibility

Rule implementations receive the resolved values of configurable attributes. For example, given:

# myapp/BUILD

some_rule(
    name = "my_target",
    some_attr = select({
        ":foo_mode": [":foo"],
        ":bar_mode": [":bar"],
    }),
)

$ bazel build //myapp/my_target --define mode=foo

Rule implementation code sees ctx.attr.some_attr as [":foo"].

Macros can accept select() clauses and pass them through to native rules. But they cannot directly manipulate them. For example, there’s no way for a macro to convert

select({"foo": "val"}, ...)

to

select({"foo": "val_with_suffix"}, ...)

This is for two reasons.

First, macros that need to know which path a select will choose cannot work because macros are evaluated in Bazel’s loading phase, which occurs before flag values are known. This is a core Bazel design restriction that’s unlikely to change any time soon.

Second, macros that just need to iterate over all select paths, while technically feasible, lack a coherent UI. Further design is necessary to change this.

Bazel query and cquery

Bazel query operates over Bazel’s loading phase. This means it doesn’t know what command line flags a target uses since those flags aren’t evaluated until later in the build (in the analysis phase). So it can’t determine which select() branches are chosen.

Bazel cquery opeates after Bazel’s analysis phase, so it has all this information and can accurately resolve select()s.

Consider:

load("@bazel_skylib//rules:common_settings.bzl", "string_flag")

# myapp/BUILD

string_flag(
    name = "dog_type",
    build_setting_default = "cat"
)

cc_library(
    name = "my_lib",
    deps = select({
        ":long": [":foo_dep"],
        ":short": [":bar_dep"],
    }),
)

config_setting(
    name = "long",
    flag_values = {":dog_type": "dachshund"},
)

config_setting(
    name = "short",
    flag_values = {":dog_type": "pug"},
)

query overapproximtes :my_lib’s dependencies:

$ bazel query 'deps(//myapp:my_lib)'
//myapp:my_lib
//myapp:foo_dep
//myapp:bar_dep

while cquery shows its exact dependencies:

$ bazel cquery 'deps(//myapp:my_lib)' --//myapp:dog_type=pug
//myapp:my_lib
//myapp:bar_dep

FAQ

Why doesn’t select() work in macros?

select() does work in rules! See Rules compatibility for details.

The key issue this question usually means is that select() doesn’t work in macros. These are different than rules. See the documentation on rules and macros to understand the difference. Here’s an end-to-end example:

Define a rule and macro:

# myapp/defs.bzl

# Rule implementation: when an attribute is read, all select()s have already
# been resolved. So it looks like a plain old attribute just like any other.
def _impl(ctx):
    name = ctx.attr.name
    allcaps = ctx.attr.my_config_string.upper()  # This works fine on all values.
    print("My name is " + name + " with custom message: " + allcaps)

# Rule declaration:
my_custom_bazel_rule = rule(
    implementation = _impl,
    attrs = {"my_config_string": attr.string()},
)

# Macro declaration:
def my_custom_bazel_macro(name, my_config_string):
    allcaps = my_config_string.upper()  # This line won't work with select(s).
    print("My name is " + name + " with custom message: " + allcaps)

Instantiate the rule and macro:

# myapp/BUILD

load("//myapp:defs.bzl", "my_custom_bazel_rule")
load("//myapp:defs.bzl", "my_custom_bazel_macro")

my_custom_bazel_rule(
    name = "happy_rule",
    my_config_string = select({
        "//tools/target_cpu:x86": "first string",
        "//tools/target_cpu:ppc": "second string",
    }),
)

my_custom_bazel_macro(
    name = "happy_macro",
    my_config_string = "fixed string",
)

my_custom_bazel_macro(
    name = "sad_macro",
    my_config_string = select({
        "//tools/target_cpu:x86": "first string",
        "//tools/target_cpu:ppc": "other string",
    }),
)

Building fails because sad_macro can’t process the select():

$ bazel build //myapp:all
ERROR: /myworkspace/myapp/BUILD:17:1: Traceback
  (most recent call last):
File "/myworkspace/myapp/BUILD", line 17
my_custom_bazel_macro(name = "sad_macro", my_config_stri..."}))
File "/myworkspace/myapp/defs.bzl", line 4, in
  my_custom_bazel_macro
my_config_string.upper()
type 'select' has no method upper().
ERROR: error loading package 'myapp': Package 'myapp' contains errors.

Building succeeds when we comment out sad_macro:

# Comment out sad_macro so it doesn't mess up the build.
$ bazel build //myapp:all
DEBUG: /myworkspace/myapp/defs.bzl:5:3: My name is happy_macro with custom message: FIXED STRING.
DEBUG: /myworkspace/myapp/hi.bzl:15:3: My name is happy_rule with custom message: FIRST STRING.

This is impossible to change because by definition macros are evaluated before Bazel reads the build’s command line flags. That means there isn’t enough information to evaluate select()s.

Macros can, however, pass select()s as opaque blobs to rules:

# myapp/defs.bzl

def my_custom_bazel_macro(name, my_config_string):
    print("Invoking macro " + name)
    my_custom_bazel_rule(
        name = name + "_as_target",
        my_config_string = my_config_string,
    )

$ bazel build //myapp:sad_macro_less_sad
DEBUG: /myworkspace/myapp/defs.bzl:23:3: Invoking macro sad_macro_less_sad.
DEBUG: /myworkspace/myapp/defs.bzl:15:3: My name is sad_macro_less_sad with custom message: FIRST STRING.

Why does select() always return true?

Because macros (but not rules) by definition can’t evaluate select()s, any attempt to do so usually produces an error:

ERROR: /myworkspace/myapp/BUILD:17:1: Traceback
  (most recent call last):
File "/myworkspace/myapp/BUILD", line 17
my_custom_bazel_macro(name = "sad_macro", my_config_stri..."}))
File "/myworkspace/myapp/defs.bzl", line 4, in
  my_custom_bazel_macro
my_config_string.upper()
type 'select' has no method upper().

Booleans are a special case that fail silently, so you should be particularly vigilant with them:

$ cat myapp/defs.bzl
def my_boolean_macro(boolval):
  print("TRUE" if boolval else "FALSE")

$ cat myapp/BUILD
load("//myapp:defs.bzl", "my_boolean_macro")
my_boolean_macro(
    boolval = select({
        "//tools/target_cpu:x86": True,
        "//tools/target_cpu:ppc": False,
    }),
)

$ bazel build //myapp:all --cpu=x86
DEBUG: /myworkspace/myapp/defs.bzl:4:3: TRUE.
$ bazel build //mypro:all --cpu=ppc
DEBUG: /myworkspace/myapp/defs.bzl:4:3: TRUE.

This happens because macros don’t understand the contents of select(). So what they’re really evaluting is the select() object itself. According to Pythonic design standards, all objects aside from a very small number of exceptions automatically return true.

Can I read select() like a dict?

Macros can’t evaluate select(s) because macros evaluate before Bazel knows what the build’s command line parameters are. Can they at least read the select()’s dictionary to, for example, add a suffix to each value?

Conceptually this is possible, but it isn’t yet a Bazel feature. What you can do today is prepare a straight dictionary, then feed it into a select():

$ cat myapp/defs.bzl
def selecty_genrule(name, select_cmd):
  for key in select_cmd.keys():
    select_cmd[key] += " WITH SUFFIX"
  native.genrule(
      name = name,
      outs = [name + ".out"],
      srcs = [],
      cmd = "echo " + select(select_cmd + {"//conditions:default": "default"})
        + " > $@"
  )

$ cat myapp/BUILD
selecty_genrule(
    name = "selecty",
    select_cmd = {
        "//tools/target_cpu:x86": "x86 mode",
    },
)

$ bazel build //testapp:selecty --cpu=x86 && cat bazel-genfiles/testapp/selecty.out
x86 mode WITH SUFFIX

If you’d like to support both select() and native types, you can do this:

$ cat myapp/defs.bzl
def selecty_genrule(name, select_cmd):
    cmd_suffix = ""
    if type(select_cmd) == "string":
        cmd_suffix = select_cmd + " WITH SUFFIX"
    elif type(select_cmd) == "dict":
        for key in select_cmd.keys():
            select_cmd[key] += " WITH SUFFIX"
        cmd_suffix = select(select_cmd + {"//conditions:default": "default"})

    native.genrule(
        name = name,
        outs = [name + ".out"],
        srcs = [],
        cmd = "echo " + cmd_suffix + "> $@",
    )

Why doesn’t select() work with bind()?

Because bind() is a WORKSPACE rule, not a BUILD rule.

Workspace rules do not have a specific configuration, and aren’t evaluated in the same way as BUILD rules. Therefore, a select() in a bind() can’t actually evaluate to any specific branch.

Instead, you should use alias(), with a select() in the actual attribute, to perform this type of run-time determination. This works correctly, since alias() is a BUILD rule, and is evaluated with a specific configuration.

You can even have a bind() target point to an alias(), if needed.

$ cat WORKSPACE
workspace(name = "myapp")
bind(name = "openssl", actual = "//:ssl")
http_archive(name = "alternative", ...)
http_archive(name = "boringssl", ...)

$ cat BUILD
config_setting(
    name = "alt_ssl",
    define_values = {
        "ssl_library": "alternative",
    },
)

alias(
    name = "ssl",
    actual = select({
        "//:alt_ssl": "@alternative//:ssl",
        "//conditions:default": "@boringssl//:ssl",
    }),
)

With this setup, you can pass --define ssl_library=alternative, and any target that depends on either //:ssl or //external:ssl will see the alternative located at @alternative//:ssl.

Why doesn’t my select() choose what I expect?

If //myapp:foo has a select() that doesn’t choose the condition you expect, use cquery and bazel config to debug:

If //myapp:foo is the top-level target you’re building, run:

$ bazel cquery //myapp:foo <desired build flags>
//myapp:foo (12e23b9a2b534a)

If you’re building some other target //bar that depends on //myapp:foo somewhere in its subgraph, run:

$ bazel cquery 'somepath(//bar, //myapp:foo)' <desired build flags>
//bar:bar   (3ag3193fee94a2)
//bar:intermediate_dep (12e23b9a2b534a)
//myapp:foo (12e23b9a2b534a)

The (12e23b9a2b534a) next to //myapp:foo is a hash of the configuration that resolves //myapp:foo’s select(). You can inspect its values with bazel config:

$ bazel config 12e23b9a2b534a
BuildConfiguration 12e23b9a2b534a
Fragment com.google.devtools.build.lib.analysis.config.CoreOptions {
  cpu: darwin
  compilation_mode: fastbuild
  ...
}
Fragment com.google.devtools.build.lib.rules.cpp.CppOptions {
  linkopt: [-Dfoo=bar]
  ...
}
...

Then compare this output against the settings expected by each config_setting.

//myapp:foo may exist in different configurations in the same build. See the cquery docs for guidance on using somepath to get the right one.