Build command

Overview

Stack's primary command is build. This page describes its interface. The goal of the interface is to do the right thing for simple input, and allow flexibility for more complicated goals.

See the introductory part of Stack's user's guide for an introduction to the command.

Synonyms

The synonym commands for build are:

Synonym command	Equivalent build command flag
stack test	stack build --test
stack bench	stack build --bench
stack haddock	stack build --haddock
stack install	stack build --copy-bins

The advantage of the synonym commands is that they are convenient and short. The advantage of the flags is that they compose. For example, stack build --test --copy-bins will build libraries, executables, and test suites, run the test suites, and then copy the executables to your local bin path (more on this below).

Components

Every Cabal package is made up of one or more components. It can have an optional library component, one or more optional executable components, one or more optional test suite components, and one or more optional benchmark components.

Stack allows you to identify a specific component to be built. For example, stack build mypackage:test:mytests will build (and run - see further below) the mytests component of the mypackage package. mytests must be a test suite component.

By default, if a test suite component is targeted, the component is built and run. The running behaviour can be disabled with the --no-run-tests flag. Similarly, if a benchmark component is targeted, it is built and run unless the running behaviour is disabled with the --no-run-benchmarks flag.

This ability to specify a component applies only to a local package. With dependencies, Stack will always build the library (if present) and all executables (if any), and ignore test suites and benchmarks. If you want more control over a package, you must add it to your packages setting in your project-level configuration file (stack.yaml).

Target syntax

stack build takes a list of one or more optional targets to be built. The supported syntaxes for targets are:

• package, e.g. stack build foobar, is the most commonly used target. It will try to find the package in the following locations: local packages, extra deps, snapshots, and package index (e.g. Hackage). If it's found in the package index, then the latest version of that package from the index is implicitly added to your extra dependencies.

  If the package is a local package, the library and executable components are selected to be built. If the --test and --bench flags are set, then all of the test suite and benchmark components, respectively, are selected to be built.

• package identifier, e.g. stack build foobar-1.2.3, is usually used to include specific package versions from the package index. If the version selected conflicts with an existing local package or extra dep, then Stack fails with an error. Otherwise, this is the same as using stack build foobar, except instead of using the latest version from the package index, the version specified is used.

• component. Instead of referring to an entire package and letting Stack decide which components to build, you select individual components from inside a package. This can be done for more fine-grained control over which test suites to run, or to have a faster compilation cycle. There are multiple ways to refer to a specific component (provided for convenience):

  • packagename:comptype:compname is the most explicit. The available comptypes are exe, test, and bench.

    Note

    When any exe component is specified, all of the package's executable components will be built. This is due to limitations in all currently released versions of Cabal. See issue#1046

  • packagename:compname allows you to leave out the component type, as that will (almost?) always be redundant with the component name. For example, stack build mypackage:mytestsuite.

  • :compname is a useful shortcut, saying "find the component in all of the local packages." This will result in an error if multiple packages have a component with the same name. To continue the above example, stack build :mytestsuite.

• directory, e.g. stack build foo/bar, will find all local packages that exist in the given directory hierarchy and then follow the same procedure as passing in package names as mentioned above. There's an important caveat here: if your directory name is parsed as one of the above target types, it will be treated as that. Explicitly starting your target with ./ can be a good way to avoid that, e.g. stack build ./foo.

  Note

  stack build . will target local packages in the current working directory or its subdirectories.

stack build with no targets specified will build all local packages.

Command stack ide targets to get a list of the available targets in your project.

Controlling what gets built

Stack will automatically build the necessary dependencies. See the introductory part of Stack's user's guide for information about how these dependencies get specified.

In addition to specifying targets, you can also control what gets built, or retained, with the following flags:

The stack build --bench flag

Pass the flag to add benchmark components to the targets, if specific components are not identified.

The stack build --dependencies-only flag

Pass the flag to skip building the targets. The flag --only-dependencies has the same effect.

The stack build --[no-]dry-run flag

Default: Disabled

Set the flag to build nothing and output information about the build plan.

The stack build --flag option

stack build --flag <package_name>:[-]<flag_name> sets (or unsets) the specified Cabal flag for the specified package.

This option can be specified multiple times to set (or unset) multiple Cabal flags.

The same Cabal flag name can be set (or unset) for multiple packages (at the command line only) with:

stack build --flag *:[-]<flag)name>

Note

Currently you needs to list all of your modules that interpret flags in the other-modules section of a Cabal file. Cabal (the tool) has a different behavior currently and doesn't require that the modules be listed. This may change in a future release.

The stack build --[no-]force-dirty flag

Default: Disabled

Set the flag to force rebuild of packages even when it doesn't seem necessary based on file dirtiness.

The stack build --[no-]haddock flag

Default: Disabled

Set the flag to build Haddock documentation. This may cause a lot of packages to get re-built, so that the documentation links work.

The stack build --[no-]keep-going flag

Default (stack build): Disabled

Default (stack test or stack bench): Enabled

Set the flag to continue building packages even after some build step fails. The packages which depend upon the failed build won't get built.

The stack build --[no-]keep-tmp-files flag

Default: Disabled

Set the flag to keep intermediate files and build directories that would otherwise be considered temporary and deleted. It may be useful to inspect these, if a build fails. By default, they are not kept.

The stack build --only-configure flag

0.1.4.0

Pass the flag to perform only the configure step, not any builds. This is intended for tool usage. It may break when used on multiple packages at once.

Note

If there are downstream actions that require a package to be built then a full build will occur, even if the flag is passed.

The stack build --only-dependencies flag

Pass the flag to skip building the targets. The flag --dependencies-only has the same effect.

The stack build --only-snapshot flag

Pass the flag to build only snapshot dependencies, which are cached and shared with other projects.

The stack build --[no-]reconfigure flag

Default: Disabled

Set the flag to force reconfiguration even when it doesn't seem necessary based on file dirtiness. This is sometimes useful with custom Setup.hs files, in particular when they depend on external data files.

The stack build --skip option

stack build --skip <component> skips building the specified components of a local package. It allows you to skip test suites and benchmark without specifying other components (e.g. stack test --skip long-test-suite will run the tests without the long-test-suite test suite). Be aware that skipping executables won't work the first time the package is built due to an issue in Cabal.

This option can be specified multiple times to skip multiple components.

The stack build --test flag

Pass the flag to add test suite components to the targets, if specific components are not identified.

Other flags and options

There are a number of other flags accepted by stack build. Instead of listing all of them, please use stack build --help. Some particularly convenient ones worth mentioning here since they compose well with the rest of the build system as described:

The stack build --coverage flag

Pass the flag to generate a code coverage report. For further information, see the code coverage documentation.

The stack build --exec option

stack build --exec "<command> [<arguments>]" will run a command after a successful build.

The stack build --file-watch flag

Pass the flag to rebuild your project every time a file changes. By default it will take into account all files belonging to the targets you specify. See also the --watch-all flag.

The stack build --[no-]interleaved-output flag

2.1.1

Default: Enabled

Set the flag for interleaved output. With interleaved output, each line of output from each package being built (targets and dependencies) is sent to the console as it happens and output relating to different packages can be interleaved. Each line will be prefixed with the name of the relevant package. The spacing between the prefix and the output will be set based on the longest relevant package name, so that the start of the output itself aligns. For example (extract):

hpack            > build
mustache         > configure
hpack            > Preprocessing library for hpack-0.35.0..
hpack            > Building library for hpack-0.35.0..
mustache         > Configuring mustache-2.4.1...
hpack            > [ 1 of 29] Compiling Data.Aeson.Config.Key
hpack            > [ 2 of 29] Compiling Data.Aeson.Config.KeyMap
mustache         > build
hpack            > [ 3 of 29] Compiling Data.Aeson.Config.Util
mustache         > Preprocessing library for mustache-2.4.1..
mustache         > Building library for mustache-2.4.1..
hpack            > [ 4 of 29] Compiling Hpack.Haskell
hpack            > [ 5 of 29] Compiling Hpack.Utf8
mustache         > [1 of 8] Compiling Paths_mustache
hpack            > [ 6 of 29] Compiling Imports
hpack            > [ 7 of 29] Compiling Hpack.Util
mustache         > [2 of 8] Compiling Text.Mustache.Internal

Unset the flag for non-interleaved output. With non-interleaved output, the build output from GHC (as opposed to from Stack) in respect of dependencies is ignored. The behaviour then depends whether there is one target package or more than one. There can be one target if the project has a single package or if one package is targetted in a multi-package project (for example, using stack build <package_name>).

• One target package: The build output for the target package is sent to the console as it happens.

• More than one target package: The build output from GHC (as opposed to from Stack) for each target package is sent to a log file for that package, unless an error occurs. At the end of the build, the location of the directory containing the log files is reported. To also output the contents of the log files to the console at the end of the build, use Stack's dump-logs option. For further information about that option, see the YAML configuration documentation. The default dump-logs mode is to output the contents of the log files that are warnings.

The stack build --pedantic flag

Pass the flag to build your project with the GHC options -Wall and -Werror. -Wall turns on all warning options that indicate potentially suspicious code. -Werror makes any warning into a fatal error.

The stack build --watch-all flag

Pass the flag to rebuild your project every time any local file changes (from project packages or from local dependencies). See also the --file-watch flag.

Composition

To come back to the composable approach described above, consider this example (which uses the wai repository). The command:

stack build --file-watch --test --copy-bins --haddock wai-extra :warp warp:doctest --exec 'echo Yay, it worked!'

will start Stack up in file watch mode, waiting for files in your project to change. When first starting, and each time a file changes, it will do all of the following.

• Build the wai-extra package and its test suites
• Build the warp executable
• Build the warp package's doctest component (which, as you may guess, is a test site)
• Run all of the wai-extra package's test suite components and the doctest test suite component
• If all of that succeeds:
  • Copy generated executables to the local bin path
  • Run the command echo Yay, it worked!