ASDF Manual

Table of Contents

Next: , Previous: (dir), Up: (dir)

ASDF: Another System Definition Facility

Manual for Version 3.1.5

This manual describes ASDF, a system definition facility for Common Lisp programs and libraries.

You can find the latest version of this manual at https://common-lisp.net/project/asdf/asdf.html.

ASDF Copyright © 2001-2015 Daniel Barlow and contributors.

This manual Copyright © 2001-2015 Daniel Barlow and contributors.

This manual revised © 2009-2015 Robert P. Goldman and Francois-Rene Rideau.

Permission is hereby granted, free of charge, to any person obtaining a copy of this software and associated documentation files (the “Software”), to deal in the Software without restriction, including without limitation the rights to use, copy, modify, merge, publish, distribute, sublicense, and/or sell copies of the Software, and to permit persons to whom the Software is furnished to do so, subject to the following conditions:

The above copyright notice and this permission notice shall be included in all copies or substantial portions of the Software.

THE SOFTWARE IS PROVIDED “AS IS”, WITHOUT WARRANTY OF ANY KIND, EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE.

--- The Detailed Node Listing ---

Loading ASDF

Configuring ASDF

Using ASDF

Defining systems with defsystem

The Object model of ASDF

Operations

Components

properties

Controlling where ASDF searches for systems

Configuration Directories

Introspection

Controlling where ASDF saves compiled files

Miscellaneous additional functionality

FAQ

``What has changed between ASDF 1, ASDF 2, and ASDF 3?''

Issues with installing the proper version of ASDF

Issues with configuring ASDF

Issues with using and extending ASDF to define systems

ASDF development FAQs

Next: , Previous: Top, Up: Top

1 Introduction

ASDF is Another System Definition Facility: a tool for specifying how systems of Common Lisp software are made up of components (sub-systems and files), and how to operate on these components in the right order so that they can be compiled, loaded, tested, etc. If you are new to ASDF, see the quick start guide.

ASDF presents three faces: one for users of Common Lisp software who want to reuse other people's code, one for writers of Common Lisp software who want to specify how to build their systems, and one for implementers of Common Lisp extensions who want to extend the build system. For more specifics, see Using ASDF, to learn how to use ASDF to load a system. See Defining systems with defsystem, to learn how to define a system of your own. See The object model of ASDF, for a description of the ASDF internals and how to extend ASDF.

Note that ASDF is not a tool for library and system installation; it plays a role like make or ant, not like a package manager. In particular, ASDF should not to be confused with Quicklisp or ASDF-Install, that attempt to find and download ASDF systems for you. Despite what the name might suggest, ASDF-Install is not part of ASDF, but a separate piece of software. ASDF-Install is also unmaintained and obsolete. We recommend you use Quicklisp (http://www.quicklisp.org/) instead, a Common Lisp package manager which works well and is being actively maintained. If you want to download software from version control instead of tarballs, so you may more easily modify it, we recommend clbuild (http://common-lisp.net/project/clbuild/). We recommend ~/common-lisp/ as a place into which to install Common Lisp software; starting with ASDF 3.1.2, it is included in the default source-registry configuration.

Finally, note that this manual is incomplete. All the bases are covered, but many advanced topics are only barely alluded to, and there is not much in terms of examples. The source code remains the ultimate source of information, free software systems in Quicklisp remain the best source of examples, and the mailing-list the best place to ask for help.

Next: , Previous: Introduction, Up: Top

2 Quick start summary

Next: , Previous: Quick start summary, Up: Top

3 Loading ASDF

Next: , Previous: Loading ASDF, Up: Loading ASDF

3.1 Loading a pre-installed ASDF

The recommended way to load ASDF is via: 

     (require "asdf")

All actively maintained Lisp implementations now include a copy of ASDF 3 that you can load this way using Common Lisp's require function.1

If the implementation you are using doesn't provide a recent ASDF 3, we recommend you upgrade it. If for some reason you would rather not upgrade it, we recommend you replace your implementation's ASDF. See Replacing your implementation's ASDF. If all else fails, see see Loading ASDF from source below.

If you use an actively maintained implementation that fails to provide an up-to-date enough stable release of ASDF, you may also send a bug report to your Lisp vendor and complain about it — or you may fix the issue yourself if it's free software.

As of the writing of this manual, the following implementations provide ASDF 3 this way: ABCL, Allegro CL, CLASP, Clozure CL, CMUCL, ECL, GNU CLISP, LispWorks, MKCL, SBCL. The following implementations only provide ASDF 2: MOCL, XCL. The following implementations don't provide ASDF: Corman CL, GCL, Genera, MCL, SCL. The latter implementations are not actively maintained (except maybe GCL); if some of them are ever released again, they probably will include ASDF 3.

For maximum convenience you might want to have ASDF loaded whenever you start your Lisp implementation, for example by loading it from the startup script or dumping a custom core — check your Lisp implementation's manual for details. SLIME notably sports a slime-asdf contrib that makes life easier with ASDF.

Next: , Previous: Loading a pre-installed ASDF, Up: Loading ASDF

3.2 Checking whether ASDF is loaded

To check that ASDF is properly loaded, you can run this form: 

     (asdf:asdf-version)

If it returns a string, that is the version of ASDF that is currently installed. If that version is suitably recent (say, 3.1.2 or later), then you can skip directly to next chapter: See Configuring ASDF.

If it raises an error, then either ASDF is not loaded, or you are using a very old version of ASDF, and need to install ASDF 3.

For more precision in detecting versions old and new, see How do I detect the ASDF version?.

If you are experiencing problems with ASDF, please try upgrading to the latest released version, using the method below, before you contact us and raise an issue.

Next: , Previous: Checking whether ASDF is loaded, Up: Loading ASDF

3.3 Upgrading ASDF

If your implementation already provides ASDF 3 or later (and it should), but you want a more recent ASDF version than your implementation provides, then you just need to ensure the more recent ASDF is installed in a configured path, like any other system. We recommend you download an official tarball or checkout a release from git into ~/common-lisp/asdf/. (see Configuring ASDF to find your systems).

Once the source code for ASDF is installed, you don't need any extra step to load it beyond the usual (require "asdf"): ASDF 3 will automatically look whether an updated version of itself is available amongst the regularly configured systems, before it compiles anything else.

If your implementation fails to provide ASDF 3 or later, see Replacing your implementation's ASDF.

Next: , Previous: Upgrading ASDF, Up: Loading ASDF

3.4 Replacing your implementation's ASDF

All maintained implementations now provide ASDF 3 in their latest release. If your doesn't, we recommend you upgrade it.

Now, if you insist on using an old implementation that didn't provide ASDF or provided an old version, we recommend installing a recent ASDF, as explained below, into your implementation's installation directory. Thus your modified implementation will now provide ASDF 3. This requires proper write permissions and may necessitate execution as a system administrator.

The ASDF source repository contains a tool to help you upgrade your implementation's ASDF. You can invoke it from the shell command-line as tools/asdf-tools install-asdf lispworks (where you can replace lispworks by the name of the relevant implementation), or you can (load "tools/install-asdf.lisp") from your Lisp REPL.

This script works on Allegro CL, Clozure CL, CMU CL, ECL, GCL, GNU CLISP, LispWorks, MKCL, SBCL, SCL, XCL. It doesn't work on ABCL, Corman CL, Genera, MCL, MOCL. Happily, ABCL is usually pretty up to date and shouldn't need that script. GCL requires a very recent version, and hasn't been tested much. Corman CL, Genera, MCL are obsolete anyway. MOCL is incomplete.

Previous: Replacing your implementation's ASDF, Up: Loading ASDF

3.5 Loading ASDF from source

If you write build scripts that must remain portable to old machines with old implementations that you cannot ensure have been upgraded or modified to provide a recent ASDF, you may have to install the file asdf.lisp somewhere and load it with: 

     (load "/path/to/your/installed/asdf.lisp")

The single file asdf.lisp is all you normally need to use ASDF.

You can extract this file from latest release tarball on the ASDF website. If you are daring and willing to report bugs, you can get the latest and greatest version of ASDF from its git repository. See Getting the latest version.

For scripts that try to use ASDF simply via require at first, and make heroic attempts to load it the hard way if at first they don't succeed, see tools/load-asdf.lisp distributed with the ASDF source repository, or the code of cl-launch.

Next: , Previous: Loading ASDF, Up: Top

4 Configuring ASDF

For standard use cases, ASDF should work pretty much out of the box. We recommend you skim the sections on configuring ASDF to find your systems and choose the method of installing Lisp software that works best for you. Then skip directly to See Using ASDF. That will probably be enough. You are unlikely to have to worry about the way ASDF stores object files, and resetting the ASDF configuration is usually only needed in corner cases.

Next: , Previous: Configuring ASDF, Up: Configuring ASDF

4.1 Configuring ASDF to find your systems

In order to compile and load your systems, ASDF must be configured to find the .asd files that contain system definitions.

There are a number of different techniques for setting yourself up with ASDF, starting from easiest to the most complex:

Note that your Operating System distribution or your system administrator may already have configured system-managed libraries for you.

Next: , Previous: Configuring ASDF to find your systems, Up: Configuring ASDF

4.2 Configuring ASDF to find your systems — old style

The old way to configure ASDF to find your systems is by pushing directory pathnames onto the variable asdf:*central-registry*.

You must configure this variable between the time you load ASDF and the time you first try to use it. Loading and configuring ASDF presumably happen as part of some initialization script that builds or starts your Common Lisp software system. (For instance, some SBCL users used to put it in their ~/.sbclrc.)

The asdf:*central-registry* is empty by default in ASDF 2 or ASDF 3, but is still supported for compatibility with ASDF 1. When used, it takes precedence over the above source-registry.5

For example, let's say you want ASDF to find the .asd file /home/me/src/foo/foo.asd. In your lisp initialization file, you could have the following: 

     (require "asdf")
     (push "/home/me/src/foo/" asdf:*central-registry*)

Note the trailing slash: when searching for a system, ASDF will evaluate each entry of the central registry and coerce the result to a pathname.6 The trailing directory name separator is necessary to tell Lisp that you're discussing a directory rather than a file. If you leave it out, ASDF is likely to look in /home/me/src/ instead of /home/me/src/foo/ as you intended, and fail to find your system definition.

Typically there are a lot of .asd files, and a common idiom was to put symbolic links to all of one's .asd files in a common directory and push that directory (the “link farm”) onto asdf:*central-registry*, instead of pushing each individual system directory.

ASDF knows to follow symlinks to the actual location of the systems.7

For example, if #p"/home/me/cl/systems/" is an element of *central-registry*, you could set up the system foo as follows: 

     $ cd /home/me/cl/systems/
     $ ln -s ~/src/foo/foo.asd .

This old style for configuring ASDF is not recommended for new users, but it is supported for old users, and for users who want a simple way to programmatically control what directories are added to the ASDF search path.

Next: , Previous: Configuring ASDF to find your systems --- old style, Up: Configuring ASDF

4.3 Configuring where ASDF stores object files

ASDF lets you configure where object files will be stored. Sensible defaults are provided and you shouldn't normally have to worry about it.

This allows the same source code repository to be shared between several versions of several Common Lisp implementations, between several users using different compilation options, with users who lack write privileges on shared source directories, etc. This also keeps source directories from being cluttered with object/fasl files.

Starting with ASDF 2, the asdf-output-translations facility was added to ASDF itself. This facility controls where object files will be stored. This facility is fully described in a chapter of this manual, Controlling where ASDF saves compiled files.

Note that before ASDF 2, other ASDF add-ons offered the same functionality, each in subtly different and incompatible ways: ASDF-Binary-Locations, cl-launch, common-lisp-controller. ASDF-Binary-Locations is now not needed anymore and should not be used. cl-launch 3.000 and common-lisp-controller 7.2 have been updated to delegate object file placement to ASDF.

Previous: Configuring where ASDF stores object files, Up: Configuring ASDF

4.4 Resetting the ASDF configuration

When you dump and restore an image, or when you tweak your configuration, you may want to reset the ASDF configuration. For that you may use the following function:

— Function: clear-configuration

Undoes any ASDF configuration regarding source-registry or output-translations.

This function is pushed onto the uiop:*image-dump-hook* by default, which means that if you save an image using uiop:dump-image, or via asdf:image-op and asdf:program-op, it will be automatically called to clear your configuration. If for some reason you prefer to call your implementation's underlying functionality, be sure to call clear-configuration manually, or push it into your implementation's equivalent of uiop:*image-dump-hook*, e.g. sb-ext:*save-hooks* on SBCL, or ext:*before-save-initializations* on CMUCL and SCL, etc.

Next: , Previous: Configuring ASDF, Up: Top

5 Using ASDF

Next: , Previous: Using ASDF, Up: Using ASDF

5.1 Loading a system

The system foo is loaded (and compiled, if necessary) by evaluating the following Lisp form: 

     (asdf:load-system :foo)

On some implementations (see Convenience Functions), ASDF hooks into the cl:require facility and you can just use: 

     (require :foo)

Note that the canonical name of a system is a string, conventionally lowercase. A system name can also be specified as a symbol (including a keyword), in which case its symbol-name is taken and lowercased. The name must be a suitable value for the :name initarg to make-pathname in whatever filesystem the system is to be found.

The lower-casing-symbols behaviour is unconventional, but was selected after some consideration. The type of systems we want to support either have lowercase as customary case (Unix, Mac, Windows) or silently convert lowercase to uppercase (lpns).

Next: , Previous: Loading a system, Up: Using ASDF

5.2 Convenience Functions

ASDF provides three commands for the most common system operations: load-system, compile-system, and test-system. It also provides require-system, a version of load-system that skips trying to update systems that are already loaded. And it provides make, a function that uses whichever operation was specified by the author of the target system, which by default behaves like load-system.

Because ASDF is an extensible system for defining operations on components, it also provides a generic function operate, so you may arbitrarily operate on your systems beyond the default operations. (At the interactive REPL, users often use its shorter alias oos, which stands for operate-on-system, a name inherited from mk-defsystem.) You'll use operate whenever you want to do something beyond compiling, loading and testing.

Note that output from ASDF and ASDF extensions are sent to the CL stream *standard-output*, so rebinding that stream around calls to asdf:operate should redirect all output from ASDF operations.

load-system applies operate with the operation from *load-system-operation* the system, and any provided keyword arguments. *load-system-operation* by default is load-op; it would be load-bundle-op by default on ECL, if only an implementation bug were fixed.

require-system skips any update to systems that have already been loaded, in the spirit of cl:require. It does it by calling load-system with a keyword option excluding already loaded systems.8. On actively maintained free software implementations (namely recent versions of ABCL, Clozure CL, CMUCL, ECL, GNU CLISP, MKCL and SBCL), once ASDF itself is loaded, cl:require too can load ASDF systems, by falling back on require-system for module names not recognized by the implementation.

Note that cl:require and require-system are appropriate to load code that is not being modified during the current programming session. This notably includes the implementation-provided extension modules that cl:require can load. This also includes any number of ASDF systems that the user isn't either developing or debugging, for which a previously installed version is deemed to be satisfactory. require-system and through it cl:require can load these systems without any problem. But for code that you are actively developing, debugging, or otherwise modifying, you should use load-system, so ASDF will pick on your modifications and transitively re-build the modified files and everything that depends on them.

Finally, starting with ASDF 3.1, a function make is also available, that does “The Right Thing” with your system. The default behaviour is to load the system as if by load-system; but instead of this default, system authors can specify the intended use of their system by specifying their desired operation with a :build-operation argument in the system definition.

Previous: Convenience Functions, Up: Using ASDF

5.3 Moving on

That's all you need to know to use ASDF to load systems written by others. The rest of this manual deals with writing system definitions for Common Lisp software you write yourself, including how to extend ASDF to define new operation and component types.

Next: , Previous: Using ASDF, Up: Top

6 Defining systems with defsystem

This chapter describes how to use ASDF to define systems and develop software.

Next: , Previous: Defining systems with defsystem, Up: Defining systems with defsystem

6.1 The defsystem form

This section begins with an example of a system definition, then gives the full grammar of defsystem.

Let's look at a simple system. This is a complete file that should be saved as hello-lisp.asd (in order that ASDF can find it when ordered to operate on the system named "hello-lisp"). 

     ;; Usual Lisp comments are allowed here
     
     (defsystem "hello-lisp"
       :description "hello-lisp: a sample Lisp system."
       :version "0.0.1"
       :author "Joe User <joe@example.com>"
       :licence "Public Domain"
       :depends-on ("optima.ppcre" "command-line-arguments")
       :components ((:file "packages")
                    (:file "macros" :depends-on ("packages"))
                    (:file "hello" :depends-on ("macros"))))

Some notes about this example:

This is all you need to know to define simple systems. The next example is much more involved, to give you a glimpse of how you can do more complex things. However, since it's ultimately arbitrary Lisp code, there is no bottom to the rabbit hole.

Next: , Previous: The defsystem form, Up: Defining systems with defsystem

6.2 A more involved example

Let's illustrate some more involved uses of defsystem via a slightly convoluted example: 

     (in-package :asdf-user)
     
     (defsystem "foo"
       :version (:read-file-form "variables" :at (3 2))
       :components
       ((:file "package")
        (:file "variables" :depends-on ("package"))
        (:module "mod"
          :depends-on ("package")
          :serial t
          :components ((:file "utils")
                       (:file "reader")
                       (:file "cooker")
                       (:static-file "data.raw"))
          :output-files (compile-op (o c) (list "data.cooked"))
          :perform (compile-op :after (o c)
             (cook-data
              :in (component-pathname (find-component c "data.raw"))
              :out (first (output-files o c)))))
        (:file "foo" :depends-on ("mod"))))
     
     (defmethod action-description
         ((o compile-op) (c (eql (find-component "foo" "mod"))))
       "cooking data")

Here are some notes about this example:

Next: , Previous: A more involved example, Up: Defining systems with defsystem

6.3 The defsystem grammar

     system-definition := ( defsystem system-designator system-option* )
     
     system-option := :defsystem-depends-on system-list
                      | :weakly-depends-on system-list
                      | :class class-name (see discussion below)
                      | system-option
                      | module-option
                      | option
     
     # These are only available since ASDF 3 (actually its alpha release 2.27)
     system-option := :homepage string
                      | :bug-tracker string
                      | :mailto string
                      | :long-name string
                      | :source-control source-control
                      | :version version-specifier
     
     source-control := (keyword string)
     
     module-option := :components component-list
                      | :serial [ t | nil ]
     
     option :=
             | :pathname pathname-specifier
             | :default-component-class class-name
             | :perform method-form
             | :explain method-form
             | :output-files method-form
             | :operation-done-p method-form
             | :if-feature feature-expression
             | :depends-on ( dependency-def* )
             | :in-order-to ( dependency+ )
     
     
     system-list := ( simple-component-name* )
     
     component-list := ( component-def* )
     
     component-def  := ( component-type simple-component-name option* )
     
     component-type := :module | :file | :static-file | other-component-type
     
     other-component-type := symbol-by-name
                             (see Component types)
     
     # This is used in :depends-on, as opposed to ``dependency,''
     # which is used in :in-order-to
     dependency-def := simple-component-name
                    | ( :feature feature-expression dependency-def )
                    | ( :version simple-component-name version-specifier )
                    | ( :require module-name )
     
     # ``dependency'' is used in :in-order-to, as opposed to
     # ``dependency-def''
     dependency := (dependent-op requirement+)
     requirement := (required-op required-component+)
     dependent-op := operation-name
     required-op := operation-name
     
     simple-component-name := string
                           |  symbol
     
     pathname-specifier := pathname | string | symbol
     
     method-form := (operation-name qual lambda-list &rest body)
     qual := method qualifier
     
     component-dep-fail-option := :fail | :try-next | :ignore
     
     feature-expression := keyword
                           | (:and feature-expression*)
                           | (:or feature-expression*)
                           | (:not feature-expression)

6.3.1 Component names

Component names (simple-component-name) may be either strings or symbols.

6.3.2 Component types

Component type names, even if expressed as keywords, will be looked up by name in the current package and in the asdf package, if not found in the current package. So a component type my-component-type, in the current package my-system-asd can be specified as :my-component-type, or my-component-type.

system and its subclasses are not allowed as component types for such children components.

6.3.3 System class names

A system class name will be looked up in the same way as a Component type (see above), except that only system and its subclasses are allowed. Typically, one will not need to specify a system class name, unless using a non-standard system class defined in some ASDF extension, typically loaded through DEFSYSTEM-DEPENDS-ON, see below. For such class names in the ASDF package, we recommend that the :class option be specified using a keyword symbol, such as 

     :class :MY-NEW-SYSTEM-SUBCLASS

This practice will ensure that package name conflicts are avoided. Otherwise, the symbol MY-NEW-SYSTEM-SUBCLASS will be read into the current package before it has been exported from the ASDF extension loaded by :defsystem-depends-on, causing a name conflict in the current package.

6.3.4 Defsystem depends on

The :defsystem-depends-on option to defsystem allows the programmer to specify another ASDF-defined system or set of systems that must be loaded before the system definition is processed. Typically this is used to load an ASDF extension that is used in the system definition.

6.3.5 Weakly depends on

We do NOT recommend you use this feature. If you are tempted to write a system foo that weakly-depends-on a system bar, we recommend that you should instead write system foo in a parametric way, and offer some special variable and/or some hook to specialize its behaviour; then you should write a system foo+bar that does the hooking of things together.

The (deprecated) :weakly-depends-on option to defsystem allows the programmer to specify another ASDF-defined system or set of systems that ASDF should try to load, but need not load in order to be successful. Typically this is used if there are a number of systems that, if present, could provide additional functionality, but which are not necessary for basic function.

Currently, although it is specified to be an option only to defsystem, this option is accepted at any component, but it probably only makes sense at the defsystem level. Programmers are cautioned not to use this component option except at the defsystem level, as this anomalous behaviour may be removed without warning.

6.3.6 Pathname specifiers

A pathname specifier (pathname-specifier) may be a pathname, a string or a symbol. When no pathname specifier is given for a component, which is the usual case, the component name itself is used.

If a string is given, which is the usual case, the string will be interpreted as a Unix-style pathname where / characters will be interpreted as directory separators. Usually, Unix-style relative pathnames are used (i.e. not starting with /, as opposed to absolute pathnames); they are relative to the path of the parent component. Finally, depending on the component-type, the pathname may be interpreted as either a file or a directory, and if it's a file, a file type may be added corresponding to the component-type, or else it will be extracted from the string itself (if applicable).

For instance, the component-type :module wants a directory pathname, and so a string "foo/bar" will be interpreted as the pathname #p"foo/bar/". On the other hand, the component-type :file wants a file of type lisp, and so a string "foo/bar" will be interpreted as the pathname #p"foo/bar.lisp", and a string "foo/bar.quux" will be interpreted as the pathname #p"foo/bar.quux.lisp". Finally, the component-type :static-file wants a file without specifying a type, and so a string "foo/bar" will be interpreted as the pathname #p"foo/bar", and a string "foo/bar.quux" will be interpreted as the pathname #p"foo/bar.quux".

ASDF interprets the string ".." as the pathname directory component word :back, which when merged, goes back one level in the directory hierarchy.

If a symbol is given, it will be translated into a string, and downcased in the process. The downcasing of symbols is unconventional, but was selected after some consideration. Observations suggest that the type of systems we want to support either have lowercase as customary case (Unix, Mac, windows) or silently convert lowercase to uppercase (lpns), so this makes more sense than attempting to use :case :common as argument to make-pathname, which is reported not to work on some implementations.

Pathname objects may be given to override the path for a component. Such objects are typically specified using reader macros such as #p or #.(make-pathname ...). Note however, that #p... is a shorthand for #.(parse-namestring ...) and that the behaviour of parse-namestring is completely non-portable, unless you are using Common Lisp logical-pathnames, which themselves involve other non-portable behaviour (see Using logical pathnames, below). Pathnames made with #.(make-pathname ...) can usually be done more easily with the string syntax above. The only case that you really need a pathname object is to override the component-type default file type for a given component. Therefore, pathname objects should only rarely be used. Unhappily, ASDF 1 used not to properly support parsing component names as strings specifying paths with directories, and the cumbersome #.(make-pathname ...) syntax had to be used. An alternative to #. read-time evaluation is to use (eval `(defsystem ... ,pathname ...)).

Note that when specifying pathname objects, ASDF does not do any special interpretation of the pathname influenced by the component type, unlike the procedure for pathname-specifying strings. On the one hand, you have to be careful to provide a pathname that correctly fulfills whatever constraints are required from that component type (e.g. naming a directory or a file with appropriate type); on the other hand, you can circumvent the file type that would otherwise be forced upon you if you were specifying a string.

6.3.7 Version specifiers

Version specifiers are strings to be parsed as period-separated lists of integers. I.e., in the example, "0.2.1" is to be interpreted, roughly speaking, as (0 2 1). In particular, version "0.2.1" is interpreted the same as "0.0002.1", though the latter is not canonical and may lead to a warning being issued. Also, "1.3" and "1.4" are both strictly uiop:version< to "1.30", quite unlike what would have happened had the version strings been interpreted as decimal fractions.

Instead of a string representing the version, the :version argument can be an expression that is resolved to such a string using the following trivial domain-specific language: in addition to being a literal string, it can be an expression of the form (:read-file-form <pathname-or-string> :at <access-at-specifier>), which will be resolved by reading a form in the specified pathname (read as a subpathname of the current system if relative or a unix-namestring). You may use a uiop:access-at specifier with the (optional) :at keyword, by default the specifier is 0, meaning the first form is returned; subforms can also be specified, with e.g. (1 2 2) specifying “the third subform (index 2) of the third subform (index 2) of the second form (index 1)” in the file (mind the off-by-one error in the English language).

System definers are encouraged to use version identifiers of the form x.y.z for major version, minor version and patch level, where significant API incompatibilities are signaled by an increased major number.

See Common attributes of components.

6.3.8 Require

Use the implementation's own require to load the module-name.

It is good taste to use :if-feature :implementation-name rather than #+implementation-name to only depend on the specified module on the specific implementation that provides it. See if-feature-option.

6.3.9 Using logical pathnames

We do not generally recommend the use of logical pathnames, especially not so to newcomers to Common Lisp. However, we do support the use of logical pathnames by old timers, when such is their preference.

To use logical pathnames, you will have to provide a pathname object as a :pathname specifier to components that use it, using such syntax as #p"LOGICAL-HOST:absolute;path;to;component.lisp".

You only have to specify such logical pathname for your system or some top-level component. Sub-components' relative pathnames, specified using the string syntax for names, will be properly merged with the pathnames of their parents. The specification of a logical pathname host however is not otherwise directly supported in the ASDF syntax for pathname specifiers as strings.

The asdf-output-translation layer will avoid trying to resolve and translate logical pathnames. The advantage of this is that you can define yourself what translations you want to use with the logical pathname facility. The disadvantage is that if you do not define such translations, any system that uses logical pathnames will behave differently under asdf-output-translations than other systems you use.

If you wish to use logical pathnames you will have to configure the translations yourself before they may be used. ASDF currently provides no specific support for defining logical pathname translations.

Note that the reasons we do not recommend logical pathnames are that (1) there is no portable way to set up logical pathnames before they are used, (2) logical pathnames are limited to only portably use a single character case, digits and hyphens. While you can solve the first issue on your own, describing how to do it on each of fifteen implementations supported by ASDF is more than we can document. As for the second issue, mind that the limitation is notably enforced on SBCL, and that you therefore can't portably violate the limitations but must instead define some encoding of your own and add individual mappings to name physical pathnames that do not fit the restrictions. This can notably be a problem when your Lisp files are part of a larger project in which it is common to name files or directories in a way that includes the version numbers of supported protocols, or in which files are shared with software written in different programming languages where conventions include the use of underscores, dots or CamelCase in pathnames.

6.3.10 Serial dependencies

If the :serial t option is specified for a module, ASDF will add dependencies for each child component, on all the children textually preceding it. This is done as if by :depends-on. 

     :serial t
     :components ((:file "a") (:file "b") (:file "c"))

is equivalent to 

     :components ((:file "a")
                  (:file "b" :depends-on ("a"))
                  (:file "c" :depends-on ("a" "b")))

6.3.11 Source location (:pathname)

The :pathname option is optional in all cases for systems defined via defsystem, and generally is unnecessary. In the simple case, source files will be found in the same directory as the system or, in the case of modules, in a subdirectory with the same name as the module.

More specifically, ASDF follows a hairy set of rules that are designed so that

  1. find-system will load a system from disk and have its pathname default to the right place.
  2. This pathname information will not be overwritten with *default-pathname-defaults* (which could be somewhere else altogether) if the user loads up the .asd file into his editor and interactively re-evaluates that form.

If a system is being loaded for the first time, its top-level pathname will be set to:

If a system is being redefined, the top-level pathname will be

6.3.12 if-feature option

This option allows you to specify a feature expression to be evaluated as if by #+ to conditionally include a component in your build. If the expression is false, the component is dropped as well as any dependency pointing to it. As compared to using #+ which is expanded at read-time, this allows you to have an object in your component hierarchy that can be used for manipulations beside building your project, and that is accessible to outside code that wishes to reason about system structure.

Programmers should be careful to consider when the :if-feature is evaluated. Recall that ASDF first computes a build plan, and then executes that plan. ASDF will check to see whether or not a feature is present at planning time, not during the build. It follows that one cannot use :if-feature to check features that are set during the course of the build. It can only be used to check the state of features before any build operations have been performed.

This option was added in ASDF 3. For more information, See Required features.

6.3.13 if-component-dep-fails option

This option was removed in ASDF 3. Its semantics was limited in purpose and dubious to explain, and its implementation was breaking a hole into the ASDF object model. Please use the if-feature option instead.

6.3.14 feature requirement

This requirement was removed in ASDF 3.1. Please do not use it. In most cases, :if-feature (see if-feature-option) will provide an adequate substitute.

The feature requirement used to ensure that a chain of component dependencies would fail when a key feature was absent. Used in conjunction with :if-component-dep-fails this provided a roundabout way to express conditional compilation.

Next: , Previous: The defsystem grammar, Up: Defining systems with defsystem

6.4 Other code in .asd files

Files containing defsystem forms are regular Lisp files that are executed by load. Consequently, you can put whatever Lisp code you like into these files. However, it is recommended to keep such forms to a minimal, and to instead define defsystem extensions that you use with :defsystem-depends-on.

If however, you might insist on including code in the .asd file itself, e.g., to examine and adjust the compile-time environment, possibly adding appropriate features to *features*. If so, here are some conventions we recommend you follow, so that users can control certain details of execution of the Lisp in .asd files:

Previous: Other code in .asd files, Up: Defining systems with defsystem

6.5 The package-inferred-system extension

Starting with release 3.1.2, ASDF supports a one-package-per-file style of programming, whereby each file is its own system, and dependencies are deduced from the defpackage form (or its variant uiop:define-package).

In this style, packages refer to a system with the same name (downcased); and if a system is defined with :class package-inferred-system, then system names that start with that name (using the slash / separator) refer to files under the filesystem hierarchy where the system is defined. For instance, if system my-lib is defined in /foo/bar/my-lib/my-lib.asd, then system my-lib/src/utility will be found in file /foo/bar/my-lib/src/utility.lisp.

This style was made popular by faslpath and quick-build before, and at the cost of a stricter package discipline, seems to make for more maintainable code. It is used by ASDF itself (starting with ASDF 3), by lisp-interface-library, and a few other libraries.

To use this style, choose a toplevel system name, e.g. my-lib, and create a file my-lib.asd with the :class :package-inferred-system option in its defsystem. For instance: 

     #-asdf3.1 (error "my-lib requires ASDF 3.1")
     (defsystem "my-lib"
       :class :package-inferred-system
       :depends-on ("my-lib/interface/all"
                    "my-lib/src/all"
                    "my-lib/extras/all")
       :in-order-to ((test-op (load-op "my-lib/test/all")))
       :perform (test-op (o c) (symbol-call :my-lib/test/all :test-suite)))
     
     (defsystem "my-lib/test" :depends-on ("my-lib/test/all"))
     
     (register-system-packages "my-lib/interface/all" '(:my-lib-interface))
     (register-system-packages "my-lib/src/all" '(:my-lib-implementation))
     (register-system-packages "my-lib/test/all" '(:my-lib-test))
     
     (register-system-packages
      "closer-mop"
      '(:c2mop :closer-common-lisp :c2cl :closer-common-lisp-user :c2cl-user))

In the code above, the first line checks that we are using ASDF 3.1, which provides package-inferred-system.

The function register-system-packages has to be called to register packages used or provided by your system and its components where the name of the system that provides the package is not the downcase of the package name.

Then, file interface/order.lisp under the lil hierarchy, that defines abstract interfaces for order comparisons, starts with the following form, dependencies being trivially computed from the :use and :mix clauses: 

     (uiop:define-package :lil/interface/order
       (:use :closer-common-lisp
        :lil/interface/definition
        :lil/interface/base
        :lil/interface/eq :lil/interface/group)
       (:mix :fare-utils :uiop :alexandria)
       (:export ...))

ASDF can tell that this file depends on system closer-mop (registered above), lil/interface/definition, lil/interface/base, lil/interface/eq, and lil/interface/group (package and system names match, and they will be looked up hierarchically).

ASDF also detects dependencies from :import-from clauses. You may thus import a well-defined set of symbols from an existing package as loaded from suitably named system; or if you prefer to use any such symbol fully qualified by a package prefix, you may declare a dependency on such a package and its corresponding system via an :import-from clause with an empty list of symbols, as in: 

     (defpackage :foo/bar
       (:use :cl)
       (:import-from :foo/baz #:sym1 #:sym2)
       (:import-from :foo/quux)
       (:export ...))

The form uiop:define-package is supported as well as defpackage, and has many options that prove useful in this context, such as :use-reexport and :mix-reexport that allow for “inheritance” of symbols being exported.

Next: , Previous: Defining systems with defsystem, Up: Top

7 The Object model of ASDF

ASDF is designed in an object-oriented way from the ground up. Both a system's structure and the operations that can be performed on systems follow a extensible protocol, allowing programmers to add new behaviours to ASDF. For example, cffi adds support for special FFI description files that interface with C libraries and for wrapper files that embed C code in Lisp. abcl-jar supports creating Java JAR archives in ABCL. poiu supports compiling code in parallel using background processes.

The key classes in ASDF are component and operation. A component represents an individual source file or a group of source files, and the products (e.g., fasl files) produced from it. An operation represents a transformation that can be performed on a component, turning them from source files to intermediate results to final outputs. Components are related by dependencies, specified in system definitions.

When ordered to operate with some operation on a component (usually a system), ASDF will first compute a plan by traversing the dependency graph using function make-plan.10 The resulting plan object contains an ordered list of actions. An action is a pair of an operation and a component, representing a particular build step to be performed. The ordering of the plan ensures that no action is performed before all its dependencies have been fulfilled.11

In this chapter, we describe ASDF's object-oriented protocol, the classes that make it up, and the generic functions on those classes. These generic functions often take both an operation and a component as arguments: much of the power and configurability of ASDF is provided by this use of CLOS's multiple dispatch. We will describe the built-in component and operation classes, and explain how to extend the ASDF protocol by defining new classes and methods for ASDF's generic functions. We will also describe the many hooks that can be configured to customize the behaviour of existing functions.

Next: , Previous: The object model of ASDF, Up: The object model of ASDF

7.1 Operations

An operation object of the appropriate type is instantiated whenever the user wants to do something with a system like

Operations can be invoked directly, or examined to see what their effects would be without performing them. There are a bunch of methods specialised on operation and component type that actually do the grunt work. Operations are invoked on systems via operate (see operate).

ASDF contains a number of pre-defined operation classes for common, and even fairly uncommon tasks that you might want to do with it. In addition, ASDF contains “abstract” operation classes that programmers can use as building blocks to define ASDF extensions. We discuss these in turn below.

Operations are invoked on systems via operate.

— Generic function: operate operation component &rest initargs &key force force-not verbose &allow-other-keys
— Generic function: oos operation component &rest initargs &key &allow-other-keys

operate invokes operation on system. oos is a synonym for operate (it stands for operate-on-system).

operation is a symbol that is passed, along with the supplied initargs, to make-operation (which will call make-instance) to create the operation object. component is a component designator, usually a string or symbol that designates a system, sometimes a list of strings or symbols that designate a subcomponent of a system.

The initargs are passed to the make-instance call when creating the operation object. Note that dependencies may cause the operation to invoke other operations on the system or its components: the new operations will be created with the same initargs as the original one.

If force is :all, then all systems are forced to be recompiled even if not modified since last compilation. If force is t, then only the system being loaded is forced to be recompiled even if not modified since last compilation, but other systems are not affected. If force is a list, then it specifies a list of systems that are forced to be recompiled even if not modified since last compilation. If force-not is :all, then all systems are forced not to be recompiled even if modified since last compilation. If force-not is t, then all systems but the system being loaded are forced not to be recompiled even if modified since last compilation (note: this was changed in ASDF 3.1.2). If force-not is a list, then it specifies a list of systems that are forced not to be recompiled even if modified since last compilation.

Both force and force-not apply to systems that are dependencies and were already compiled. force-not takes precedences over force, as it should, really, but unhappily only since ASDF 3.1.2. Moreover, systems which have been registered as immutable by register-immutable-system (since ASDF 3.1.5) are always considered forced-not, and even their .asd are not refreshed from the filesystem. See Miscellaneous Functions.

To see what operate would do, you can use: 

          (asdf:traverse operation-class system-name)

Next: , Previous: Operations, Up: Operations

7.1.1 Predefined operations of ASDF

All the operations described in this section are in the asdf package. They are invoked via the operate generic function. 

     (asdf:operate 'asdf:operation-name :system-name {operation-options ...})
— Operation: compile-op

This operation compiles the specified component. A cl-source-file will be compile-file'd. All the children and dependencies of a system or module will be recursively compiled by compile-op.

compile-op depends on prepare-op which itself depends on a load-op of all of a component's dependencies, as well as of its parent's dependencies. When operate is called on compile-op, all these dependencies will be loaded as well as compiled; yet, some parts of the system main remain unloaded, because nothing depends on them. Use load-op to load a system.

— Operation: load-op

This operation loads the compiled code for a specified component. A cl-source-file will have its compiled fasl loaded, which fasl is the output of compile-op that load-op depends on.

load-op will recursively load all the children of a system or module.

load-op also depends on prepare-op which itself depends on a load-op of all of a component's dependencies, as well as of its parent's dependencies.

— Operation: prepare-op

This operation ensures that the dependencies of a component and its recursive parents are loaded (as per load-op), as a prerequisite before compile-op and load-op operations may be performed on a given component.

— Operation: load-source-op, prepare-source-op

load-source-op will load the source for the files in a module rather than the compiled fasl output. It has a prepare-source-op analog to prepare-op, that ensures the dependencies are themselves loaded via load-source-op.

— Operation: test-op

This operation will perform some tests on the module. The default method will do nothing. The default dependency is to require load-op to be performed on the module first. Its operation-done-p method returns nil, which means that the operation is never done – we assume that if you invoke the test-op, you want to test the system, even if you have already done so.

The results of this operation are not defined by ASDF. It has proven difficult to define how the test operation should signal its results to the user in a way that is compatible with all of the various test libraries and test techniques in use in the community, and given the fact that ASDF operations do not return a value indicating success or failure. For those willing to go to the effort, we suggest defining conditions to signal when a test-op fails, and storing in those conditions information that describes which tests fail.

People typically define a separate test system to hold the tests. Doing this avoids unnecessarily adding a test framework as a dependency on a library. For example, one might have 

          (defsystem "foo"
            :in-order-to ((test-op (test-op "foo/test")))
           ...)
          
          (defsystem "foo/test"
            :depends-on ("foo" "fiveam") ; fiveam is a test framework library
            ...)

Then one defines perform methods on test-op such as the following: 

          (defsystem "foo/test"
            :depends-on ("foo" "fiveam") ; fiveam is a test framework library
            :perform (test-op (o s)
                              (uiop:symbol-call :fiveam '#:run!
                                 (uiop:find-symbol* '#:foo-test-suite
                                                      :foo-tests)))
            ...)
— Operation: compile-bundle-op, monolithic-compile-bundle-op, load-bundle-op, monolithic-load-bundle-op, deliver-asd-op, monolithic-deliver-asd-op, lib-op, monolithic-lib-op, dll-op, monolithic-dll-op, image-op, program-op

These are “bundle” operations, that can create a single-file “bundle” for all the contents of each system in an application, or for the entire application.

compile-bundle-op will create a single fasl file for each of the systems needed, grouping all its many fasls in one, so you can deliver each system as a single fasl monolithic-compile-bundle-op will create a single fasl file for the target system and all its dependencies, so you can deliver your entire application as a single fasl. load-bundle-op will load the output of compile-bundle-op. Note that if it the output is not up-to-date, compile-bundle-op may load the intermediate fasls as a side-effect. Bundling fasls together matters a lot on ECL, where the dynamic linking involved in loading tens of individual fasls can be noticeably more expensive than loading a single one.

NB: compile-bundle-op, monolithic-compile-bundle-op, load-bundle-op, monolithic-load-bundle-op, deliver-asd-op, monolithic-deliver-asd-op were respectively called fasl-op, monolithic-fasl-op, load-fasl-op, monolithic-load-fasl-op, binary-op, monolithic-binary-op before ASDF 3.1. The old names still exist for backward compatibility, though they poorly label what is going on.

Once you have created a fasl with compile-bundle-op, you can use precompiled-system to deliver it in a way that is compatible with clients having dependencies on your system, whether it is distributed as source or as a single binary; the .asd file to be delivered with the fasl will look like this: 

          (defsystem :mysystem :class :precompiled-system
            :fasl (some expression that will evaluate to a pathname))

Or you can use deliver-asd-op to let ASDF create such a system for you as well as the compile-bundle-op output, or monolithic-deliver-asd-op. This allows you to deliver code for your systems or applications as a single file. Of course, if you want to test the result in the current image, before you try to use any newly created .asd files, you should not forget to (asdf:clear-configuration) or at least (asdf:clear-source-registry), so it re-populates the source-registry from the filesystem.

The program-op operation will create an executable program from the specified system and its dependencies. You can use UIOP for its pre-image-dump hooks, its post-image-restore hooks, and its access to command-line arguments. And you can specify an entry point my-app:main by specifying in your defsystem the option :entry-point "my-app:main". Depending on your implementation, running (asdf:operate 'asdf:program-op :my-app) may quit the current Lisp image upon completion. See the example in test/hello-world-example.asd and test/hello.lisp, as built and tested by test/test-program.script and test/make-hello-world.lisp. image-op will dump an image that may not be standalone and does not start its own function, but follows the usual execution convention of the underlying Lisp, just with more code pre-loaded, for use as an intermediate build result or with a wrapper invocation script.

There is also lib-op for building a linkable .a file (Windows: .lib) from all linkable object dependencies (FFI files, and on ECL, Lisp files too), and its monolithic equivalent monolithic-lib-op. And there is also dll-op (respectively its monolithic equivalent monolithic-lib-op) for building a linkable .so file (Windows: .dll, MacOS X: .dynlib) to create a single dynamic library for all the extra FFI code to be linked into each of your systems (respectively your entire application).

All these “bundle” operations are available since ASDF 3 on all actively supported Lisp implementations, but may be unavailable on unmaintained legacy implementations. This functionality was previously available for select implementations, as part of a separate system asdf-bundle, itself descended from the ECL-only asdf-ecl.

The pathname of the output of bundle operations is subject to output-translation as usual, unless the operation is equal to the :build-operation argument to defsystem. This behaviour is not very satisfactory and may change in the future. Maybe you have suggestions on how to better configure it?

— Operation: concatenate-source-op, monolithic-concatenate-source-op, load-concatenated-source-op, compile-concatenated-source-op, load-compiled-concatenated-source-op, monolithic-load-concatenated-source-op, monolithic-compile-concatenated-source-op, monolithic-load-compiled-concatenated-source-op

These operations, as their respective names indicate, will concatenate all the cl-source-file source files in a system (or in a system and all its dependencies, if monolithic), in the order defined by dependencies, then load the result, or compile and then load the result.

These operations are useful to deliver a system or application as a single source file, and for testing that said file loads properly, or compiles and then loads properly.

ASDF itself is delivered as a single source file this way, using monolithic-concatenate-source-op, prepending a prelude and the uiop library before the asdf/defsystem system itself.

Previous: Predefined operations of ASDF, Up: Operations

7.1.2 Creating new operations

ASDF was designed to be extensible in an object-oriented fashion. To teach ASDF new tricks, a programmer can implement the behaviour he wants by creating a subclass of operation.

ASDF's pre-defined operations are in no way “privileged”, but it is requested that developers never use the asdf package for operations they develop themselves. The rationale for this rule is that we don't want to establish a “global asdf operation name registry”, but also want to avoid name clashes.

Your operation must usually provide methods for one or more of the following generic functions:

An operation may provide methods for the following generic functions:

Operations that print output should send that output to the standard CL stream *standard-output*, as the Lisp compiler and loader do.

Next: , Previous: Operations, Up: The object model of ASDF

7.2 Components

A component represents an individual source file or a group of source files, and the things that get transformed into. A system is a component at the top level of the component hierarchy, that can be found via find-system. A source-file is a component representing a single source-file and the successive output files into which it is transformed. A module is an intermediate component itself grouping several other components, themselves source-files or further modules.

A system designator is a system itself, or a string or symbol that behaves just like any other component name (including with regard to the case conversion rules for component names).

A component designator, relative to a base component, is either a component itself, or a string or symbol, or a list of designators.

— Function: find-system system-designator &optional (error-p t)

Given a system designator, find-system finds and returns a system. If no system is found, an error of type missing-component is thrown, or nil is returned if error-p is false.

To find and update systems, find-system funcalls each element in the *system-definition-search-functions* list, expecting a pathname to be returned, or a system object, from which a pathname may be extracted, and that will be registered. The resulting pathname (if any) is loaded if one of the following conditions is true:

When system definitions are loaded from .asd files, they are implicitly loaded into the ASDF-USER package, which uses ASDF, UIOP and UIOP/COMMON-LISP12 Programmers who do anything non-trivial in a .asd file, such as defining new variables, functions or classes, should include defpackage and in-package forms in this file, so they will not overwrite each others' extensions. Such forms might also help the files behave identically if loaded manually with cl:load for development or debugging, though we recommend you use the function asdf::load-asd instead, which the slime-asdf contrib knows about.

The default value of *system-definition-search-functions* is a list of three functions. The first function looks in each of the directories given by evaluating members of *central-registry* for a file whose name is the name of the system and whose type is asd; the first such file is returned, whether or not it turns out to actually define the appropriate system. The second function does something similar, for the directories specified in the source-registry, but searches the filesystem only once and caches its results. The third function makes the package-inferred-system extension work, see The package-inferred-system extension.

Because of the way these search functions are defined, you should put the definition for a system foo in a file named foo.asd, in a directory that is in the central registry or which can be found using the source registry configuration.

It is often useful to define multiple systems in a same file, but ASDF can only locate a system's definition file based on the system name. For this reason, ASDF 3's system search algorithm has been extended to allow a file foo.asd to contain secondary systems named foo/bar, foo/baz, foo/quux, etc., in addition to the primary system named foo. The first component of a system name, separated by the slash character, /, is called the primary name of a system. The primary name may be extracted by function asdf::primary-system-name; when ASDF 3 is told to find a system whose name has a slash, it will first attempt to load the corresponding primary system, and will thus see any such definitions, and/or any definition of a package-inferred-system.13 If your file foo.asd also defines systems that do not follow this convention, e.g., a system named foo-test, ASDF will not be able to automatically locate a definition for these systems, and will only see their definition if you explicitly find or load the primary system using e.g. (asdf:find-system "foo") before you try to use them. We strongly recommend against this practice, though it is currently supported for backward compatibility.

— Function: primary-system-name name

Internal (not exported) function, asdf::primary-system-name. Returns the primary system name (the portion before the slash, /, in a secondary system name) from name.

— Function: locate-system name

This function should typically not be invoked directly. It is exported as part of the API only for programmers who wish to provide their own *system-definition-search-functions*.

Given a system name designator, try to locate where to load the system definition from. Returns five values: foundp, found-system, pathname, previous, and previous-time. foundp is true when a system was found, either a new as yet unregistered one, or a previously registered one. The found-system return value will be a system object, if a system definition is found in your source registry. The system definition will not be loaded if it hasn't been loaded already. pathname when not null is a path from which to load the system, either associated with found-system, or with the previous system. If previous is not null, it will be a previously loaded system object of the same name (note that the system definition is previously-loaded: the system itself may or may not be). previous-time when not null is the timestamp of the previous system definition file, at the time when the previous system definition was loaded.

For example, if your current registry has foo.asd in /current/path/to/foo.asd, but system foo was previously loaded from /previous/path/to/foo.asd then locate-system will return the following values:

  1. foundp will be T,
  2. found-system will be NIL,
  3. pathname will be #p"/current/path/to/foo.asd",
  4. previous will be an object of type SYSTEM with system-source-file slot value of #p"/previous/path/to/foo.asd"
  5. previous-time will be the timestamp of #p"/previous/path/to/foo.asd" at the time it was loaded.

— Function: find-component base path

Given a base component (or designator for such), and a path, find the component designated by the path starting from the base.

If path is a component object, it designates itself, independently from the base.

If path is a string, or symbol denoting a string via coerce-name, then base is resolved to a component object, which must be a system or module, and the designated component is the child named by the path.

If path is a cons cell, find-component with the base and the car of the path, and the resulting object is used as the base for a tail call to find-component with the car of the path.

If base is a component object, it designates itself.

If base is null, then path is used as the base, with nil as the path.

If base is a string, or symbol denoting a string via coerce-name, it designates a system as per find-system.

If base is a cons cell, it designates the component found by find-component with its car as base and cdr as path.

Next: , Previous: Components, Up: Components

7.2.1 Common attributes of components

All components, regardless of type, have the following attributes. All attributes except name are optional.

7.2.1.1 Name

A component name is a string or a symbol. If a symbol, its name is taken and lowercased. This translation is performed by the exported function coerce-name.

Unless overridden by a :pathname attribute, the name will be interpreted as a pathname specifier according to a Unix-style syntax. See Pathname specifiers.

7.2.1.2 Version identifier

This optional attribute specifies a version for the current component. The version should typically be a string of integers separated by dots, for example ‘1.0.11’. For more information on version specifiers, see The defsystem grammar.

A version may then be queried by the generic function version-satisfies, to see if :version dependencies are satisfied, and when specifying dependencies, a constraint of minimal version to satisfy can be specified using e.g. (:version "mydepname" "1.0.11").

Note that in the wild, we typically see version numbering only on components of type system. Presumably it is much less useful within a given system, wherein the library author is responsible to keep the various files in synch.

7.2.1.3 Required features

Traditionally defsystem users have used #+ reader conditionals to include or exclude specific per-implementation files. For example, CFFI, the portable C foreign function interface contained lines like: 

          #+sbcl       (:file "cffi-sbcl")

An unfortunate side effect of this approach is that no single implementation can read the entire system. This causes problems if, for example, one wished to design an archive-op that would create an archive file containing all the sources, since for example the file cffi-sbcl.lisp above would be invisible when running the archive-op on any implementation other than SBCL.

Starting with ASDF 3, components may therefore have an :if-feature option. The value of this option should be a feature expression using the same syntax as #+ does. If that feature expression evaluates to false, any reference to the component will be ignored during compilation, loading and/or linking. Since the expression is read by the normal reader, you must explicitly prefix your symbols with : so they be read as keywords; this is as contrasted with the #+ syntax that implicitly reads symbols in the keyword package by default.

For instance, :if-feature (:and :x86 (:or :sbcl :cmu :scl)) specifies that the given component is only to be compiled and loaded when the implementation is SBCL, CMUCL or Scieneer CL on an x86 machine. You cannot write it as :if-feature (and x86 (or sbcl cmu scl)) since the symbols would not be read as keywords.

See if-feature-option.

7.2.1.4 Dependencies

This attribute specifies dependencies of the component on its siblings. It is optional but often necessary.

There is an excitingly complicated relationship between the initarg and the method that you use to ask about dependencies

Dependencies are between (operation component) pairs. In your initargs for the component, you can say 

     :in-order-to ((compile-op (load-op "a" "b") (compile-op "c"))
                   (load-op (load-op "foo")))

This means the following things:

The syntax is approximately 

(this-op @{(other-op required-components)@}+)

simple-component-name := string
                      |  symbol

required-components := simple-component-name
                     | (required-components required-components)

component-name := simple-component-name
                | (:version simple-component-name minimum-version-object)

Side note:

This is on a par with what ACL defsystem does. mk-defsystem is less general: it has an implied dependency 

  for all source file x, (load x) depends on (compile x)

and using a :depends-on argument to say that b depends on a actually means that 

  (compile b) depends on (load a)

This is insufficient for e.g. the McCLIM system, which requires that all the files are loaded before any of them can be compiled ]

End side note

In ASDF, the dependency information for a given component and operation can be queried using (component-depends-on operation component), which returns a list 

     ((load-op "a") (load-op "b") (compile-op "c") ...)

component-depends-on can be subclassed for more specific component/operation types: these need to (call-next-method) and append the answer to their dependency, unless they have a good reason for completely overriding the default dependencies.

If it weren't for CLISP, we'd be using LIST method combination to do this transparently. But, we need to support CLISP. If you have the time for some CLISP hacking, I'm sure they'd welcome your fixes.

A minimal version can be specified for a component you depend on (typically another system), by specifying (:version "other-system" "1.2.3") instead of simply "other-system" as the dependency. See the discussion of the semantics of :version in the defsystem grammar.

7.2.1.5 pathname

This attribute is optional and if absent (which is the usual case), the component name will be used.

See Pathname specifiers, for an explanation of how this attribute is interpreted.

Note that the defsystem macro (used to create a “top-level” system) does additional processing to set the filesystem location of the top component in that system. This is detailed elsewhere. See Defining systems with defsystem.

To find the CL pathname corresponding to a component, use

— Function: component-pathname component

Returns the pathname corresponding to component. For components such as source files, this will be a filename pathname. For example: 

          CL-USER> (asdf:component-pathname (asdf:find-system "xmls"))
          #P"/Users/rpg/lisp/xmls/"

and 

          CL-USER> (asdf:component-pathname
                     (asdf:find-component
                        (asdf:find-system "xmls") "xmls"))
          #P"/Users/rpg/lisp/xmls/xmls.lisp"
7.2.1.6 properties

This attribute is optional.

Packaging systems often require information about files or systems in addition to that specified by ASDF's pre-defined component attributes. Programs that create vendor packages out of ASDF systems therefore have to create “placeholder” information to satisfy these systems. Sometimes the creator of an ASDF system may know the additional information and wish to provide it directly.

(component-property component property-name) and associated setf method will allow the programmatic update of this information. Property names are compared as if by EQL, so use symbols or keywords or something.

Next: , Previous: Common attributes of components, Up: Components

7.2.2 Pre-defined subclasses of component

— Component: source-file

A source file is any file that the system does not know how to generate from other components of the system.

Note that this is not necessarily the same thing as “a file containing data that is typically fed to a compiler”. If a file is generated by some pre-processor stage (e.g. a .h file from .h.in by autoconf) then it is not, by this definition, a source file. Conversely, we might have a graphic file that cannot be automatically regenerated, or a proprietary shared library that we received as a binary: these do count as source files for our purposes.

Subclasses of source-file exist for various languages. FIXME: describe these.

— Component: module

A module is a collection of sub-components.

A module component has the following extra initargs:

The default operation knows how to traverse a module, so most operations will not need to provide methods specialised on modules.

module may be subclassed to represent components such as foreign-language linked libraries or archive files.

— Component: system

system is a subclass of module.

A system is a module with a few extra attributes for documentation purposes; these are given elsewhere. See The defsystem grammar.

Users can create new classes for their systems: the default defsystem macro takes a :class keyword argument.

Previous: Pre-defined subclasses of component, Up: Components

7.2.3 Creating new component types

New component types are defined by subclassing one of the existing component classes and specializing methods on the new component class.

As an example, suppose we have some implementation-dependent functionality that we want to isolate in one subdirectory per Lisp implementation our system supports. We create a subclass of cl-source-file: 

     (defclass unportable-cl-source-file (cl-source-file)
       ())

Function asdf:implementation-type (exported since 2.014.14) gives us the name of the subdirectory. All that's left is to define how to calculate the pathname of an unportable-cl-source-file. 

     (defmethod component-pathname ((component unportable-cl-source-file))
       (merge-pathnames*
        (parse-unix-namestring (format nil "~(~A~)/" (asdf:implementation-type)))
        (call-next-method)))

The new component type is used in a defsystem form in this way: 

     (defsystem :foo
         :components
         ((:file "packages")
          ...
          (:unportable-cl-source-file "threads"
           :depends-on ("packages" ...))
          ...
         )

Next: , Previous: Components, Up: The object model of ASDF

7.3 Dependencies

To be successfully build-able, this graph of actions must be acyclic. If, as a user, extender or implementer of ASDF, you introduce a cycle into the dependency graph, ASDF will fail loudly. To clearly distinguish the direction of dependencies, ASDF 3 uses the words requiring and required as applied to an action depending on the other: the requiring action depends-on the completion of all required actions before it may itself be performed.

Using the defsystem syntax, users may easily express direct dependencies along the graph of the object hierarchy: between a component and its parent, its children, and its siblings. By defining custom CLOS methods, you can express more elaborate dependencies as you wish. Most common operations, such as load-op, compile-op or load-source-op are automatically propagate “downward” the component hierarchy and are “covariant” with it: to act the operation on the parent module, you must first act it on all the children components, with the action on the parent being parent of the action on each child. Other operations, such as prepare-op and prepare-source-op (introduced in ASDF 3) are automatically propagated “upward” the component hierarchy and are “contravariant” with it: to perform the operation of preparing for compilation of a child component, you must perform the operation of preparing for compilation of its parent component, and so on, ensuring that all the parent's dependencies are (compiled and) loaded before the child component may be compiled and loaded. Yet other operations, such as test-op or load-bundle-op remain at the system level, and are not propagated along the hierarchy, but instead do something global on the system.

Previous: Dependencies, Up: The object model of ASDF

7.4 Functions

— Function: version-satisfies version version-spec

Does version satisfy the version-spec. A generic function. ASDF provides built-in methods for version being a component or string. version-spec should be a string. If it's a component, its version is extracted as a string before further processing.

A version string satisfies the version-spec if after parsing, the former is no older than the latter. Therefore "1.9.1", "1.9.2" and "1.10" all satisfy "1.9.1", but "1.8.4" or "1.9" do not. For more information about how version-satisfies parses and interprets version strings and specifications, see The defsystem grammar (version specifiers) and Common attributes of components.

Note that in versions of ASDF prior to 3.0.1, including the entire ASDF 1 and ASDF 2 series, version-satisfies would also require that the version and the version-spec have the same major version number (the first integer in the list); if the major version differed, the version would be considered as not matching the spec. But that feature was not documented, therefore presumably not relied upon, whereas it was a nuisance to several users. Starting with ASDF 3.0.1, version-satisfies does not treat the major version number specially, and returns T simply if the first argument designates a version that isn't older than the one specified as a second argument. If needs be, the (:version ...) syntax for specifying dependencies could be in the future extended to specify an exclusive upper bound for compatible versions as well as an inclusive lower bound.

Next: , Previous: The object model of ASDF, Up: Top

8 Controlling where ASDF searches for systems

Next: , Previous: Controlling where ASDF searches for systems, Up: Controlling where ASDF searches for systems

8.1 Configurations

Configurations specify paths where to find system files.

  1. The search registry may use some hardcoded wrapping registry specification. This allows some implementations (notably SBCL) to specify where to find some special implementation-provided systems that need to precisely match the version of the implementation itself.
  2. An application may explicitly initialize the source-registry configuration using the configuration API (see Configuration API, below) in which case this takes precedence. It may itself compute this configuration from the command-line, from a script, from its own configuration file, etc.
  3. The source registry will be configured from the environment variable CL_SOURCE_REGISTRY if it exists.
  4. The source registry will be configured from user configuration file $XDG_CONFIG_DIRS/common-lisp/source-registry.conf (which defaults to ~/.config/common-lisp/source-registry.conf) if it exists.
  5. The source registry will be configured from user configuration directory $XDG_CONFIG_DIRS/common-lisp/source-registry.conf.d/ (which defaults to ~/.config/common-lisp/source-registry.conf.d/) if it exists.
  6. The source registry will be configured from default user configuration trees ~/common-lisp/ (since ASDF 3.1.2 only), ~/.sbcl/systems/ (on SBCL only), $XDG_DATA_HOME/common-lisp/systems/ (no recursion, link farm) $XDG_DATA_HOME/common-lisp/source/. The XDG_DATA_HOME directory defaults to ~/.local/share/. On Windows, the local-appdata and appdata directories are used instead.
  7. The source registry will be configured from system configuration file /etc/common-lisp/source-registry.conf if it exists.
  8. The source registry will be configured from system configuration directory /etc/common-lisp/source-registry.conf.d/ if it exists.
  9. The source registry will be configured from a default configuration. This configuration may allow for implementation-specific systems to be found, for systems to be found the current directory (at the time that the configuration is initialized) as well as :directory entries for $XDG_DATA_DIRS/common-lisp/systems/ and :tree entries for $XDG_DATA_DIRS/common-lisp/source/, where XDG_DATA_DIRS defaults to /usr/local/share and /usr/share on Unix, and the common-appdata directory on Windows.
  10. The source registry may include implementation-dependent directories that correspond to implementation-provided extensions.

Each of these configurations is specified as an s-expression in a trivial domain-specific language (defined below). Additionally, a more shell-friendly syntax is available for the environment variable (defined yet below).

Each of these configurations is only used if the previous configuration explicitly or implicitly specifies that it includes its inherited configuration.

Additionally, some implementation-specific directories may be automatically prepended to whatever directories are specified in configuration files, no matter if the last one inherits or not.

Next: , Previous: Configurations, Up: Controlling where ASDF searches for systems

8.2 Truenames and other dangers

One great innovation of the original ASDF was its ability to leverage CL:TRUENAME to locate where your source code was and where to build it, allowing for symlink farms as a simple but effective configuration mechanism that is easy to control programmatically. ASDF 3 still supports this configuration style, and it is enabled by default; however we recommend you instead use our source-registry configuration mechanism described below, because it is easier to setup in a portable way across users and implementations.

Additionally, some people dislike truename, either because it is very slow on their system, or because they are using content-addressed storage where the truename of a file is related to a digest of its individual contents, and not to other files in the same intended project. For these people, ASDF 3 allows to eschew the TRUENAME mechanism, by setting the variable asdf:*resolve-symlinks* to nil.

PS: Yes, if you haven't read Vernor Vinge's short but great classic “True Names... and Other Dangers” then you're in for a treat.

Next: , Previous: Truenames and other dangers, Up: Controlling where ASDF searches for systems

8.3 XDG base directory

Note that we purport to respect the XDG base directory specification as to where configuration files are located, where data files are located, where output file caches are located. Mentions of XDG variables refer to that document.

http://standards.freedesktop.org/basedir-spec/basedir-spec-latest.html

This specification allows the user to specify some environment variables to customize how applications behave to his preferences.

On Windows platforms, even when not using Cygwin, and starting with ASDF 3.1.5, we still do a best effort at following the XDG base directory specification, even though it doesn't exactly fit common practice for Windows applications. However, we replace the fixed Unix paths ~/.local, /usr/local and /usr with their rough Windows equivalent Local AppData, AppData, Common AppData, etc. Since support for querying the Windows registry is not possible to do in reasonable amounts of portable Common Lisp code, ASDF 3 relies on the environment variables that Windows usually exports, and are hopefully in synch with the Windows registry. If you care about the details, see uiop/configuration.lisp and don't hesitate to suggest improvements.

Next: , Previous: XDG base directory, Up: Controlling where ASDF searches for systems

8.4 Backward Compatibility

For backward compatibility as well as to provide a practical backdoor for hackers, ASDF will first search for .asd files in the directories specified in asdf:*central-registry* before it searches in the source registry above.

See Configuring ASDF to find your systems — old style.

By default, asdf:*central-registry* will be empty.

This old mechanism will therefore not affect you if you don't use it, but will take precedence over the new mechanism if you do use it.

Next: , Previous: Backward Compatibility, Up: Controlling where ASDF searches for systems

8.5 Configuration DSL

Here is the grammar of the s-expression (SEXP) DSL for source-registry configuration: 

     ;; A configuration is a single SEXP starting with the keyword
     ;; :source-registry followed by a list of directives.
     CONFIGURATION := (:source-registry DIRECTIVE ...)
     
     ;; A directive is one of the following:
     DIRECTIVE :=
         ;; INHERITANCE DIRECTIVE:
         ;; Your configuration expression MUST contain
         ;; exactly one of the following:
         :inherit-configuration |
         ;; splices inherited configuration (often specified last) or
         :ignore-inherited-configuration |
         ;; drop inherited configuration (specified anywhere)
     
         ;; forward compatibility directive (since ASDF 2.011.4), useful when
         ;; you want to use new configuration features but have to bootstrap
         ;; the newer required ASDF from an older release that doesn't
         ;; support said features:
         :ignore-invalid-entries |
     
         ;; add a single directory to be scanned (no recursion)
         (:directory DIRECTORY-PATHNAME-DESIGNATOR) |
     
         ;; add a directory hierarchy, recursing but
         ;; excluding specified patterns
         (:tree DIRECTORY-PATHNAME-DESIGNATOR) |
     
         ;; override the defaults for exclusion patterns
         (:exclude EXCLUSION-PATTERN ...) |
         ;; augment the defaults for exclusion patterns
         (:also-exclude EXCLUSION-PATTERN ...) |
         ;; Note that the scope of a an exclude pattern specification is
         ;; the rest of the current configuration expression or file.
     
         ;; splice the parsed contents of another config file
         (:include REGULAR-FILE-PATHNAME-DESIGNATOR) |
     
         ;; This directive specifies that some default must be spliced.
         :default-registry
     
     REGULAR-FILE-PATHNAME-DESIGNATOR
         := PATHNAME-DESIGNATOR ; interpreted as a file
     DIRECTORY-PATHNAME-DESIGNATOR
         := PATHNAME-DESIGNATOR ; interpreted as a directory
     
     PATHNAME-DESIGNATOR :=
         NIL | ;; Special: skip this entry.
         ABSOLUTE-COMPONENT-DESIGNATOR ;; see pathname DSL
     
     EXCLUSION-PATTERN := a string without wildcards, that will be matched
         exactly against the name of a any subdirectory in the directory
         component of a path. e.g. "_darcs" will match
         #p"/foo/bar/_darcs/src/bar.asd"

Pathnames are designated using another DSL, shared with the output-translations configuration DSL below. The DSL is resolved by the function asdf::resolve-location, to be documented and exported at some point in the future. 

     ABSOLUTE-COMPONENT-DESIGNATOR :=
         (ABSOLUTE-COMPONENT-DESIGNATOR RELATIVE-COMPONENT-DESIGNATOR ...) |
         STRING |
         ;; namestring (better be absolute or bust, directory assumed where
         ;; applicable).  In output-translations, directory is assumed and
         ;; **/*.*.* added if it's last.  On MCL, a MacOSX-style POSIX
         ;; namestring (for MacOS9 style, use #p"..."); Note that none of the
         ;; above applies to strings used in *central-registry*, which
         ;; doesn't use this DSL: they are processed as normal namestrings.
         ;; however, you can compute what you put in the *central-registry*
         ;; based on the results of say
         ;; (asdf::resolve-location "/Users/fare/cl/cl-foo/")
         PATHNAME |
         ;; pathname (better be an absolute path, or bust)
         ;; In output-translations, unless followed by relative components,
         ;; it better have appropriate wildcards, as in **/*.*.*
         :HOME | ; designates the user-homedir-pathname ~/
         :USER-CACHE | ; designates the default location for the user cache
         :HERE |
         ;; designates the location of the configuration file
         ;; (or *default-pathname-defaults*, if invoked interactively)
         :ROOT
         ;; magic, for output-translations source only: paths that are relative
         ;; to the root of the source host and device
     
     They keyword :SYSTEM-CACHE is not accepted in ASDF 3.1 and beyond: it
     was a security hazard.
     
     RELATIVE-COMPONENT-DESIGNATOR :=
         (RELATIVE-COMPONENT-DESIGNATOR RELATIVE-COMPONENT-DESIGNATOR ...) |
         STRING |
           ;; relative directory pathname as interpreted by
           ;; parse-unix-namestring.
           ;; In output translations, if last component, **/*.*.* is added
         PATHNAME | ; pathname; unless last component, directory is assumed.
         :IMPLEMENTATION |
            ;; directory based on implementation, e.g. sbcl-1.0.45-linux-x64
         :IMPLEMENTATION-TYPE |
            ;; a directory based on lisp-implementation-type only, e.g. sbcl
         :DEFAULT-DIRECTORY |
            ;; a relativized version of the default directory
         :*/ | ;; any direct subdirectory (since ASDF 2.011.4)
         :**/ | ;; any recursively inferior subdirectory (since ASDF 2.011.4)
         :*.*.* | ;; any file (since ASDF 2.011.4)
     
     The keywords :UID and :USERNAME are no longer supported.

For instance, as a simple case, my ~/.config/common-lisp/source-registry.conf, which is the default place ASDF looks for this configuration, once contained: 

     (:source-registry
       (:tree (:home "cl")) ;; will expand to e.g. "/home/joeluser/cl/"
       :inherit-configuration)

Next: , Previous: Configuration DSL, Up: Controlling where ASDF searches for systems

8.6 Configuration Directories

Configuration directories consist in files each containing a list of directives without any enclosing (:source-registry ...) form. The files will be sorted by namestring as if by string< and the lists of directives of these files with be concatenated in order. An implicit :inherit-configuration will be included at the end of the list.

System-wide or per-user Common Lisp software distributions such as Debian packages or some future version of clbuild may then include files such as /etc/common-lisp/source-registry.conf.d/10-foo.conf or ~/.config/common-lisp/source-registry.conf.d/10-foo.conf to easily and modularly register configuration information about software being distributed.

The convention is that, for sorting purposes, the names of files in such a directory begin with two digits that determine the order in which these entries will be read. Also, the type of these files must be .conf, which not only simplifies the implementation by allowing for more portable techniques in finding those files, but also makes it trivial to disable a file, by renaming it to a different file type.

Directories may be included by specifying a directory pathname or namestring in an :include directive, e.g.: 

             (:include "/foo/bar/")

Hence, to achieve the same effect as my example ~/.config/common-lisp/source-registry.conf above, I could simply create a file ~/.config/common-lisp/source-registry.conf.d/33-home-fare-cl.conf alone in its directory with the following contents: 

     (:tree "/home/fare/cl/")

Previous: Configuration Directories, Up: Configuration Directories

8.6.1 The :here directive

The :here directive is an absolute pathname designator that refers to the directory containing the configuration file currently being processed.

The :here directive is intended to simplify the delivery of complex CL systems, and for easy configuration of projects shared through revision control systems, in accordance with our design principle that each participant should be able to provide all and only the information available to him or her.

Consider a person X who has set up the source code repository for a complex project with a master directory dir/. Ordinarily, one might simply have the user add a directive that would look something like this: 

        (:tree "path/to/dir")

But what if X knows that there are very large subtrees under dir that are filled with, e.g., Java source code, image files for icons, etc.? All of the asdf system definitions are contained in the subdirectories dir/src/lisp/ and dir/extlib/lisp/, and these are the only directories that should be searched.

In this case, X can put into dir/ a file asdf.conf that contains the following: 

     (:source-registry
        (:tree (:here "src/lisp/"))
        (:tree (:here "extlib/lisp"))
        (:directory (:here "outlier/")))

Then when someone else (call her Y) checks out a copy of this repository, she need only add 

     (:include "/path/to/my/checkout/directory/asdf.conf")

to one of her previously-existing asdf source location configuration files, or invoke initialize-source-registry with a configuration form containing that s-expression. ASDF will find the .conf file that X has provided, and then set up source locations within the working directory according to X's (relative) instructions.

Next: , Previous: Configuration Directories, Up: Controlling where ASDF searches for systems

8.7 Shell-friendly syntax for configuration

When considering environment variable CL_SOURCE_REGISTRY ASDF will skip to next configuration if it's an empty string. It will READ the string as a SEXP in the DSL if it begins with a paren (, otherwise it will be interpreted much like TEXINPUTS, as a list of paths, where

* paths are separated by a : (colon) on Unix platforms (including cygwin), by a ; (semicolon) on other platforms (mainly, Windows).

* each entry is a directory to add to the search path.

* if the entry ends with a double slash // then it instead indicates a tree in the subdirectories of which to recurse.

* if the entry is the empty string (which may only appear once), then it indicates that the inherited configuration should be spliced there.

Next: , Previous: Shell-friendly syntax for configuration, Up: Controlling where ASDF searches for systems

8.8 Search Algorithm

In case that isn't clear, the semantics of the configuration is that when searching for a system of a given name, directives are processed in order.

When looking in a directory, if the system is found, the search succeeds, otherwise it continues.

When looking in a tree, if one system is found, the search succeeds. If multiple systems are found, the consequences are unspecified: the search may succeed with any of the found systems, or an error may be raised. ASDF currently returns the first system found, XCVB currently raised an error. If none is found, the search continues.

Exclude statements specify patterns of subdirectories the systems from which to ignore. Typically you don't want to use copies of files kept by such version control systems as Darcs. Exclude statements are not propagated to further included or inherited configuration files or expressions; instead the defaults are reset around every configuration statement to the default defaults from asdf::*default-source-registry-exclusions*.

Include statements cause the search to recurse with the path specifications from the file specified.

An inherit-configuration statement cause the search to recurse with the path specifications from the next configuration (see Configurations above).

Next: , Previous: Search Algorithm, Up: Controlling where ASDF searches for systems

8.9 Caching Results

The implementation is allowed to either eagerly compute the information from the configurations and file system, or to lazily re-compute it every time, or to cache any part of it as it goes. In practice, the recommended source-registry eagerly collects and caches results and you need to explicitly flush the cache for change to be taken into account, whereas the old-style *central-registry* mechanism queries the filesystem every time.

To explicitly flush any information cached by the system after a change was made in the filesystem, See Configuration API, and e.g. call asdf:clear-source-registry.

Starting with ASDF 3.1.4, you can also explicitly build a persistent cache of the .asd files found under a tree: when recursing into a directory declared by :tree and its transitive subdirectories, if a file .cl-source-registry.cache exists containing a form that is a list starting with :source-registry-cache followed by a list of strings, as in (:source-registry-cache "foo/bar.asd" "path/to/more.asd" ...), then the strings are assumed to be unix-namestrings designating the available asd files under that tree, and the recursion otherwise stops. The list can also be empty, allowing to stop a costly recursion in a huge directory tree.

To update such a cache after you install, update or remove source repositories, you can run a script distributed with ASDF: tools/cl-source-registry-cache.lisp /path/to/directory. To wholly invalidate the cache, you can delete the file .cl-source-registry.cache in that directory. In either case, for an existing Lisp process to see this change, it needs to clear its own cache with e.g. (asdf:clear-source-registry).

Developers may safely create a cache in their development tree, and we recommend they do it at the top of their source tree if it contains more than a small number of files and directories; they only need update it when they create, remove or move .asd files. Software distribution managers may also safely create such a cache, but they must be careful to update it every time they install, update or remove a software source repository or installation package. Finally, advanced developers who juggle with a lot of code in their source-registry may manually manage such a cache, to allow for faster startup of Lisp programs.

This persistence cache can help you reduce startup latency. For instance, on one machine with hundreds of source repositories, such a cache shaves half a second at the startup of every #!/usr/bin/cl script using SBCL, more on other implementations; this makes a notable difference as to their subjective interactivity and usability. The speedup will only happen if the implementation-provided ASDF is recent enough (3.1.3.7 or later); it is not enough for a recent ASDF upgrade to be present, since the upgrade will itself be found but after the old version has scanned the directories without heeding such a cache. To upgrade the implementation-provided ASDF, use our script tools/install-asdf.lisp.

Next: , Previous: Caching Results, Up: Controlling where ASDF searches for systems

8.10 Configuration API

The specified functions are exported from your build system's package. Thus for ASDF the corresponding functions are in package ASDF, and for XCVB the corresponding functions are in package XCVB.

— Function: initialize-source-registry &optional PARAMETER

will read the configuration and initialize all internal variables. You may extend or override configuration from the environment and configuration files with the given PARAMETER, which can be nil (no configuration override), or a SEXP (in the SEXP DSL), a string (as in the string DSL), a pathname (of a file or directory with configuration), or a symbol (fbound to function that when called returns one of the above).

— Function: clear-source-registry

undoes any source registry configuration and clears any cache for the search algorithm. You might want to call this function (or better, clear-configuration) before you dump an image that would be resumed with a different configuration, and return an empty configuration. Note that this does not include clearing information about systems defined in the current image, only about where to look for systems not yet defined.

— Function: ensure-source-registry &optional PARAMETER

checks whether a source registry has been initialized. If not, initialize it with the given PARAMETER.

Every time you use ASDF's find-system, or anything that uses it (such as operate, load-system, etc.), ensure-source-registry is called with parameter nil, which the first time around causes your configuration to be read. If you change a configuration file, you need to explicitly initialize-source-registry again, or maybe simply to clear-source-registry (or clear-configuration) which will cause the initialization to happen next time around.

Next: , Previous: Configuration API, Up: Controlling where ASDF searches for systems

8.11 Introspection

Next: , Previous: Introspection, Up: Introspection

8.11.1 *source-registry-parameter* variable

We have made available the variable *source-registry-parameter* that can be used by code that wishes to introspect about the (past) configuration of ASDF's source registry. This variable should never be set! It will be set as a side-effect of calling initialize-source-registry; user code should treat it as read-only.

Previous: *source-registry-parameter* variable, Up: Introspection

8.11.2 Information about system dependencies

ASDF makes available three functions to read system interdependencies. These are intended to aid programmers who wish to perform dependency analyses.

— Function: system-defsystem-depends-on system
— Function: system-depends-on system
— Function: system-weakly-depends-on system

Returns a list of names of systems that are weakly depended on by system. Weakly depended on systems are optionally loaded only if ASDF can find them; failure to find such systems does not cause an error in loading.

Note that the return value for system-weakly-depends-on is simpler than the return values of the other two system dependency introspection functions.

Next: , Previous: Introspection, Up: Controlling where ASDF searches for systems

8.12 Status

This mechanism is vastly successful, and we have declared that asdf:*central-registry* is not recommended anymore, though we will continue to support it. All hooks into implementation-specific search mechanisms have been integrated in the wrapping-source-registry that everyone uses implicitly.

Next: , Previous: Status, Up: Controlling where ASDF searches for systems

8.13 Rejected ideas

Alternatives I (FRR) considered and rejected while developing ASDF 2 included:

  1. Keep asdf:*central-registry* as the master with its current semantics, and somehow the configuration parser expands the new configuration language into a expanded series of directories of subdirectories to lookup, pre-recursing through specified hierarchies. This is kludgy, and leaves little space of future cleanups and extensions.
  2. Keep asdf:*central-registry* as the master but extend its semantics in completely new ways, so that new kinds of entries may be implemented as a recursive search, etc. This seems somewhat backwards.
  3. Completely remove asdf:*central-registry* and break backwards compatibility. Hopefully this will happen in a few years after everyone migrate to a better ASDF and/or to XCVB, but it would be very bad to do it now.
  4. Replace asdf:*central-registry* by a symbol-macro with appropriate magic when you dereference it or setf it. Only the new variable with new semantics is handled by the new search procedure. Complex and still introduces subtle semantic issues.

I've been suggested the below features, but have rejected them, for the sake of keeping ASDF no more complex than strictly necessary.

Next: , Previous: Rejected ideas, Up: Controlling where ASDF searches for systems

8.14 TODO

Previous: TODO, Up: Controlling where ASDF searches for systems

8.15 Credits for the source-registry

Thanks a lot to Stelian Ionescu for the initial idea.

Thanks to Rommel Martinez for the initial implementation attempt.

All bad design ideas and implementation bugs are mine, not theirs. But so are good design ideas and elegant implementation tricks.

— Francois-Rene Rideau fare@tunes.org, Mon, 22 Feb 2010 00:07:33 -0500

Next: , Previous: Controlling where ASDF searches for systems, Up: Top

9 Controlling where ASDF saves compiled files

Each Common Lisp implementation has its own format for compiled files or fasls.14 If you use multiple implementations (or multiple versions of the same implementation), you'll soon find your source directories littered with various fasls, dfsls, cfsls and so on. Worse yet, multiple implementations use the same file extension and some implementations maintain the same file extension while changing formats from version to version (or platform to platform). This can lead to many errors and much confusion as you switch from one implementation to the next. Finally, this requires write access to the source directory, and therefore precludes sharing of a same source code directory between multiple users.

Since ASDF 2, ASDF includes the asdf-output-translations facility to mitigate the problem.

Next: , Previous: Controlling where ASDF saves compiled files, Up: Controlling where ASDF saves compiled files

9.1 Configurations

Configurations specify mappings from input locations to output locations. Once again we rely on the XDG base directory specification for configuration. See XDG base directory.

  1. Some hardcoded wrapping output translations configuration may be used. This allows special output translations (or usually, invariant directories) to be specified corresponding to the similar special entries in the source registry.
  2. An application may explicitly initialize the output-translations configuration using the Configuration API in which case this takes precedence. (see Configuration API.) It may itself compute this configuration from the command-line, from a script, from its own configuration file, etc.
  3. The source registry will be configured from the environment variable ASDF_OUTPUT_TRANSLATIONS if it exists.
  4. The source registry will be configured from user configuration file $XDG_CONFIG_DIRS/common-lisp/asdf-output-translations.conf (which defaults to ~/.config/common-lisp/asdf-output-translations.conf) if it exists.
  5. The source registry will be configured from user configuration directory $XDG_CONFIG_DIRS/common-lisp/asdf-output-translations.conf.d/ (which defaults to ~/.config/common-lisp/asdf-output-translations.conf.d/) if it exists.
  6. The source registry will be configured from system configuration file /etc/common-lisp/asdf-output-translations.conf if it exists.
  7. The source registry will be configured from system configuration directory /etc/common-lisp/asdf-output-translations.conf.d/ if it exists.

Each of these configurations is specified as a SEXP in a trivial domain-specific language (see Configuration DSL). Additionally, a more shell-friendly syntax is available for the environment variable (see Shell-friendly syntax for configuration).

When processing an entry in the above list of configuration methods, ASDF will stop unless that entry explicitly or implicitly specifies that it includes its inherited configuration.

Note that by default, a per-user cache is used for output files. This allows the seamless use of shared installations of software between several users, and takes files out of the way of the developers when they browse source code, at the expense of taking a small toll when developers have to clean up output files and find they need to get familiar with output-translations first.15

Next: , Previous: Output Configurations, Up: Controlling where ASDF saves compiled files

9.2 Backward Compatibility

We purposely do not provide backward compatibility with earlier versions of ASDF-Binary-Locations (8 Sept 2009), common-lisp-controller (7.0) or cl-launch (2.35), each of which had similar general capabilities. The APIs of these programs were not designed for easy user configuration through configuration files. Recent versions of common-lisp-controller (7.2) and cl-launch (3.000) use the new asdf-output-translations API as defined below. ASDF-Binary-Locations is fully superseded and not to be used anymore.

This incompatibility shouldn't inconvenience many people. Indeed, few people use and customize these packages; these few people are experts who can trivially adapt to the new configuration. Most people are not experts, could not properly configure these features (except inasmuch as the default configuration of common-lisp-controller and/or cl-launch might have been doing the right thing for some users), and yet will experience software that “just works”, as configured by the system distributor, or by default.

Nevertheless, if you are a fan of ASDF-Binary-Locations, we provide a limited emulation mode:

— Function: enable-asdf-binary-locations-compatibility &key centralize-lisp-binaries default-toplevel-directory include-per-user-information map-all-source-files source-to-target-mappings

This function will initialize the new asdf-output-translations facility in a way that emulates the behaviour of the old ASDF-Binary-Locations facility. Where you would previously set global variables *centralize-lisp-binaries*, *default-toplevel-directory*, *include-per-user-information*, *map-all-source-files* or *source-to-target-mappings* you will now have to pass the same values as keyword arguments to this function. Note however that as an extension the :source-to-target-mappings keyword argument will accept any valid pathname designator for asdf-output-translations instead of just strings and pathnames.

If you insist, you can also keep using the old ASDF-Binary-Locations (the one available as an extension to load of top of ASDF, not the one built into a few old versions of ASDF), but first you must disable asdf-output-translations with (asdf:disable-output-translations), or you might experience “interesting” issues.

Also, note that output translation is enabled by default. To disable it, use (asdf:disable-output-translations).

Next: , Previous: Output Backward Compatibility, Up: Controlling where ASDF saves compiled files

9.3 Configuration DSL

Here is the grammar of the SEXP DSL for asdf-output-translations configuration: 

;; A configuration is single SEXP starting with keyword :source-registry
;; followed by a list of directives.
CONFIGURATION := (:output-translations DIRECTIVE ...)

;; A directive is one of the following:
DIRECTIVE :=
    ;; INHERITANCE DIRECTIVE:
    ;; Your configuration expression MUST contain
    ;; exactly one of either of these:
    :inherit-configuration |
      ;; splices inherited configuration (often specified last)
    :ignore-inherited-configuration |
      ;; drop inherited configuration (specified anywhere)

    ;; forward compatibility directive (since ASDF 2.011.4), useful when
    ;; you want to use new configuration features but have to bootstrap a
    ;; the newer required ASDF from an older release that doesn't have
    ;; said features:
    :ignore-invalid-entries |

    ;; include a configuration file or directory
    (:include PATHNAME-DESIGNATOR) |

    ;; enable global cache in ~/.common-lisp/cache/sbcl-1.0.45-linux-amd64/
    ;; or something.
    :enable-user-cache |
    ;; Disable global cache. Map / to /
    :disable-cache |

    ;; add a single directory to be scanned (no recursion)
    (DIRECTORY-DESIGNATOR DIRECTORY-DESIGNATOR)

    ;; use a function to return the translation of a directory designator
    (DIRECTORY-DESIGNATOR (:function TRANSLATION-FUNCTION))

DIRECTORY-DESIGNATOR :=
    NIL | ; As source: skip this entry. As destination: same as source
    T | ; as source matches anything, as destination
        ; maps pathname to itself.
    ABSOLUTE-COMPONENT-DESIGNATOR ; same as in the source-registry language

TRANSLATION-FUNCTION :=
    SYMBOL | ;; symbol naming a function that takes two arguments:
             ;; the pathname to be translated and the matching
             ;; DIRECTORY-DESIGNATOR
    LAMBDA   ;; A form which evaluates to a function taking two arguments:
             ;; the pathname to be translated and the matching
             ;; DIRECTORY-DESIGNATOR

Relative components better be either relative or subdirectories of the path before them, or bust.

The last component, if not a pathname, is notionally completed by /**/*.*. You can specify more fine-grained patterns by using a pathname object as the last component e.g. #p"some/path/**/foo*/bar-*.fasl"

You may use #+features to customize the configuration file.

The second designator of a mapping may be nil, indicating that files are not mapped to anything but themselves (same as if the second designator was the same as the first).

When the first designator is t, the mapping always matches. When the first designator starts with :root, the mapping matches any host and device. In either of these cases, if the second designator isn't t and doesn't start with :root, then strings indicating the host and pathname are somehow copied in the beginning of the directory component of the source pathname before it is translated.

When the second designator is t, the mapping is the identity. When the second designator starts with :root, the mapping preserves the host and device of the original pathname. Notably, this allows you to map files to a subdirectory of the whichever directory the file is in. Though the syntax is not quite as easy to use as we'd like, you can have an (source destination) mapping entry such as follows in your configuration file, or you may use enable-asdf-binary-locations-compatibility with :centralize-lisp-binaries nil which will do the same thing internally for you: 

     #.(let ((wild-subdir
               (make-pathname :directory '(:relative :wild-inferiors)))
             (wild-file
               (make-pathname :name :wild :version :wild :type :wild)))
        `((:root ,wild-subdir ,wild-file)
          (:root ,wild-subdir :implementation ,wild-file)))

Starting with ASDF 2.011.4, you can use the simpler: `(:root (:root :**/ :implementation :*.*.*))

:include statements cause the search to recurse with the path specifications from the file specified.

If the translate-pathname mechanism cannot achieve a desired translation, the user may provide a function which provides the required algorithm. Such a translation function is specified by supplying a list as the second directory-designator the first element of which is the keyword :function, and the second element of which is either a symbol which designates a function or a lambda expression. The function designated by the second argument must take two arguments, the first being the pathname of the source file, the second being the wildcard that was matched. When invoked, the function should return the translated pathname.

An :inherit-configuration statement causes the search to recurse with the path specifications from the next configuration in the bulleted list. See Configurations, above.

Next: , Previous: Output Configuration DSL, Up: Controlling where ASDF saves compiled files

9.4 Configuration Directories

Configuration directories consist of files, each of which contains a list of directives without any enclosing (:output-translations ...) form. The files will be sorted by namestring as if by string< and the lists of directives of these files with be concatenated in order. An implicit :inherit-configuration will be included at the end of the list.

System-wide or per-user Common Lisp software distributions such as Debian packages or some future version of clbuild may then include files such as /etc/common-lisp/asdf-output-translations.conf.d/10-foo.conf or ~/.config/common-lisp/asdf-output-translations.conf.d/10-foo.conf to easily and modularly register configuration information about software being distributed.

The convention is that, for sorting purposes, the names of files in such a directory begin with two digits that determine the order in which these entries will be read. Also, the type of these files must be .conf, which not only simplifies the implementation by allowing for more portable techniques in finding those files, but also makes it trivial to disable a file, by renaming it to a different file type.

Directories may be included by specifying a directory pathname or namestring in an :include directive, e.g.: 

        (:include "/foo/bar/")

Next: , Previous: Output Configuration Directories, Up: Controlling where ASDF saves compiled files

9.5 Shell-friendly syntax for configuration

When considering environment variable ASDF_OUTPUT_TRANSLATIONS ASDF will skip to the next configuration if it's an empty string. It will READ the string as an SEXP in the DSL if it begins with a paren ( and it will be interpreted as a list of directories. Directories should come by pairs, indicating a mapping directive. Entries are separated by a : (colon) on Unix platforms (including cygwin), by a ; (semicolon) on other platforms (mainly, Windows).

The magic empty entry, if it comes in what would otherwise be the first entry in a pair, indicates the splicing of inherited configuration. If it comes as the second entry in a pair, it indicates that the directory specified first is to be left untranslated (which has the same effect as if the directory had been repeated). Thus "/foo:/bar::/baz:" means that things under directory /foo/ are translated to be under /bar/, then include the inherited configuration, then specify that things under directory /baz/ are not translated.

Next: , Previous: Output Shell-friendly syntax for configuration, Up: Controlling where ASDF saves compiled files

9.6 Semantics of Output Translations

From the specified configuration, a list of mappings is extracted in a straightforward way: mappings are collected in order, recursing through included or inherited configuration as specified. To this list is prepended some implementation-specific mappings, and is appended a global default.

The list is then compiled to a mapping table as follows: for each entry, in order, resolve the first designated directory into an actual directory pathname for source locations. If no mapping was specified yet for that location, resolve the second designated directory to an output location directory add a mapping to the table mapping the source location to the output location, and add another mapping from the output location to itself (unless a mapping already exists for the output location).

Based on the table, a mapping function is defined, mapping source pathnames to output pathnames: given a source pathname, locate the longest matching prefix in the source column of the mapping table. Replace that prefix by the corresponding output column in the same row of the table, and return the result. If no match is found, return the source pathname. (A global default mapping the filesystem root to itself may ensure that there will always be a match, with same fall-through semantics).

Next: , Previous: Semantics of Output Translations, Up: Controlling where ASDF saves compiled files

9.7 Caching Results

The implementation is allowed to either eagerly compute the information from the configurations and file system, or to lazily re-compute it every time, or to cache any part of it as it goes. To explicitly flush any information cached by the system, use the API below.

Next: , Previous: Output Caching Results, Up: Controlling where ASDF saves compiled files

9.8 Output location API

The specified functions are exported from package ASDF.

— Function: initialize-output-translations &optional PARAMETER

will read the configuration and initialize all internal variables. You may extend or override configuration from the environment and configuration files with the given PARAMETER, which can be nil (no configuration override), or a SEXP (in the SEXP DSL), a string (as in the string DSL), a pathname (of a file or directory with configuration), or a symbol (fbound to function that when called returns one of the above).

— Function: disable-output-translations

will initialize output translations in a way that maps every pathname to itself, effectively disabling the output translation facility.

— Function: clear-output-translations

undoes any output translation configuration and clears any cache for the mapping algorithm. You might want to call this function (or better, clear-configuration) before you dump an image that would be resumed with a different configuration, and return an empty configuration. Note that this does not include clearing information about systems defined in the current image, only about where to look for systems not yet defined.

— Function: ensure-output-translations &optional PARAMETER

checks whether output translations have been initialized. If not, initialize them with the given PARAMETER. This function will be called before any attempt to operate on a system.

— Function: apply-output-translations PATHNAME

Applies the configured output location translations to PATHNAME (calls ensure-output-translations for the translations).

Every time you use ASDF's output-files, or anything that uses it (that may compile, such as operate, perform, etc.), ensure-output-translations is called with parameter nil, which the first time around causes your configuration to be read. If you change a configuration file, you need to explicitly initialize-output-translations again, or maybe clear-output-translations (or clear-configuration), which will cause the initialization to happen next time around.

Previous: Output location API, Up: Controlling where ASDF saves compiled files

9.9 Credits for output translations

Thanks a lot to Peter van Eynde for Common Lisp Controller and to Bjorn Lindberg and Gary King for ASDF-Binary-Locations.

All bad design ideas and implementation bugs are to mine, not theirs. But so are good design ideas and elegant implementation tricks.

— Francois-Rene Rideau fare@tunes.org

Next: , Previous: Controlling where ASDF saves compiled files, Up: Top

10 Error handling

10.1 ASDF errors

If ASDF detects an incorrect system definition, it will signal a generalised instance of SYSTEM-DEFINITION-ERROR.

Operations may go wrong (for example when source files contain errors). These are signalled using generalised instances of OPERATION-ERROR.

10.2 Compilation error and warning handling

ASDF checks for warnings and errors when a file is compiled. The variables *compile-file-warnings-behaviour* and *compile-file-failure-behaviour* control the handling of any such events. The valid values for these variables are :error, :warn, and :ignore.

Next: , Previous: Error handling, Up: Top

11 Miscellaneous additional functionality

ASDF includes several additional features that are generally useful for system definition and development.

Next: , Previous: Miscellaneous additional functionality, Up: Miscellaneous additional functionality

11.1 Controlling file compilation

When declaring a component (system, module, file), you can specify a keyword argument :around-compile function. If left unspecified (and therefore unbound), the value will be inherited from the parent component if any, or with a default of nil if no value is specified in any transitive parent.

The argument must be either nil, an fbound symbol, a lambda-expression (e.g. (lambda (thunk) ...(funcall thunk ...) ...)) a function object (e.g. using #.#' but that's discouraged because it prevents the introspection done by e.g. asdf-dependency-grovel), or a string that when read yields a symbol or a lambda-expression. nil means the normal compile-file function will be called. A non-nil value designates a function of one argument that will be called with a function that will invoke compile-file* with various arguments; the around-compile hook may supply additional keyword arguments to pass to that call to compile-file*.

One notable argument that is heeded by compile-file* is :compile-check, a function called when the compilation was otherwise a success, with the same arguments as compile-file; the function shall return true if the compilation and its resulting compiled file respected all system-specific invariants, and false (nil) if it broke any of those invariants; it may issue warnings or errors before it returns nil. (NB: The ability to pass such extra flags is only available starting with ASDF 2.22.3.) This feature is notably exercised by asdf-finalizers.

By using a string, you may reference a function, symbol and/or package that will only be created later during the build, but isn't yet present at the time the defsystem form is evaluated. However, if your entire system is using such a hook, you may have to explicitly override the hook with nil for all the modules and files that are compiled before the hook is defined.

Using this hook, you may achieve such effects as: locally renaming packages, binding *readtables* and other syntax-controlling variables, handling warnings and other conditions, proclaiming consistent optimization settings, saving code coverage information, maintaining meta-data about compilation timings, setting gensym counters and PRNG seeds and other sources of non-determinism, overriding the source-location and/or timestamping systems, checking that some compile-time side-effects were properly balanced, etc.

Note that there is no around-load hook. This is on purpose. Some implementations such as ECL, GCL or MKCL link object files, which allows for no such hook. Other implementations allow for concatenating FASL files, which doesn't allow for such a hook either. We aim to discourage something that's not portable, and has some dubious impact on performance and semantics even when it is possible. Things you might want to do with an around-load hook are better done around-compile, though it may at times require some creativity (see e.g. the package-renaming system).

Next: , Previous: Controlling file compilation, Up: Miscellaneous additional functionality

11.2 Controlling source file character encoding

Starting with ASDF 2.21, components accept a :encoding option so authors may specify which character encoding should be used to read and evaluate their source code. When left unspecified, the encoding is inherited from the parent module or system; if no encoding is specified at any point, or if nil is explicitly specified, an extensible protocol described below is followed, that ultimately defaults to :utf-8 since ASDF 3.

The protocol to determine the encoding is to call the function detect-encoding, which itself, if provided a valid file, calls the function specified by *encoding-detection-hook*, or else defaults to the *default-encoding*. The *encoding-detection-hook* is by default bound to function always-default-encoding, that always returns the contents of *default-encoding*. *default-encoding* is bound to :utf-8 by default (before ASDF 3, the default was :default).

Whichever encoding is returned must be a portable keyword, that will be translated to an implementation-specific external-format designator by function encoding-external-format, which itself simply calls the function specified *encoding-external-format-hook*; that function by default is default-encoding-external-format, that only recognizes :utf-8 and :default, and translates the former to the implementation-dependent *utf-8-external-format*, and the latter to itself (that itself is portable but has an implementation-dependent meaning).

In other words, there now are plenty of extension hooks, but by default ASDF enforces the previous de facto standard behaviour of using :utf-8, independently from whatever configuration the user may be using. Thus, system authors can now rely on :utf-8 being used while compiling their files, even if the user is currently using :koi8-r or :euc-jp as their interactive encoding. (Before ASDF 3, there was no such guarantee, :default was used, and only plain ASCII was safe to include in source code.)

Some legacy implementations only support 8-bit characters, and some implementations provide 8-bit only variants. On these implementations, the *utf-8-external-format* gracefully falls back to :default, and Unicode characters will be read as multi-character mojibake. To detect such situations, UIOP will push the :asdf-unicode feature on implementations that support Unicode, and you can use reader-conditionalization to protect any :encoding encoding statement, as in #+asdf-unicode :encoding #+asdf-unicode :utf-8. We recommend that you avoid using unprotected :encoding specifications until after ASDF 2.21 or later becomes widespread. As of May 2015, all maintained implementations provide ASDF 3, so you may prudently start using this and other features without such protection.

While it offers plenty of hooks for extension, and one such extension is available (see asdf-encodings below), ASDF itself only recognizes one encoding beside :default, and that is :utf-8, which is the de facto standard, already used by the vast majority of libraries that use more than ASCII. On implementations that do not support unicode, the feature :asdf-unicode is absent, and the :default external-format is used to read even source files declared as :utf-8. On these implementations, non-ASCII characters intended to be read as one CL character may thus end up being read as multiple CL characters. In most cases, this shouldn't affect the software's semantics: comments will be skipped just the same, strings with be read and printed with slightly different lengths, symbol names will be accordingly longer, but none of it should matter. But a few systems that actually depend on unicode characters may fail to work properly, or may work in a subtly different way. See for instance lambda-reader.

We invite you to embrace UTF-8 as the encoding for non-ASCII characters starting today, even without any explicit specification in your .asd files. Indeed, on some implementations and configurations, UTF-8 is already the :default, and loading your code may cause errors if it is encoded in anything but UTF-8. Therefore, even with the legacy behaviour, non-UTF-8 is guaranteed to break for some users, whereas UTF-8 is pretty much guaranteed not to break anywhere (provided you do not use a BOM), although it might be read incorrectly on some implementations. :utf-8 has been the default value of *default-encoding* since ASDF 3.

If you need non-standard character encodings for your source code, use the extension system asdf-encodings, by specifying :defsystem-depends-on ("asdf-encodings") in your defsystem. This extension system will register support for more encodings using the *encoding-external-format-hook* facility, so you can explicitly specify :encoding :latin1 in your .asd file. Using the *encoding-detection-hook* it will also eventually implement some autodetection of a file's encoding from an emacs-style -*- mode: lisp ; coding: latin1 -*- declaration, or otherwise based on an analysis of octet patterns in the file. At this point, asdf-encoding only supports the encodings that are supported as part of your implementation. Since the list varies depending on implementations, we still recommend you use :utf-8 everywhere, which is the most portable (next to it is :latin1).

Recent versions of Quicklisp include asdf-encodings; if you're not using it, you may get this extension using git: git clone https://gitlab.common-lisp.net/asdf/asdf-encodings.git or git clone git@gitlab.common-lisp.net:asdf/asdf-encodings.git. You can also browse the repository on https://gitlab.common-lisp.net/asdf/asdf-encodings.

When you use asdf-encodings, any .asd file loaded will use the autodetection algorithm to determine its encoding. If you depend on this detection happening, you should explicitly load asdf-encodings early in your build. Note that :defsystem-depends-on cannot be used here: by the time the :defsystem-depends-on is loaded, the enclosing defsystem form has already been read.

In practice, this means that the *default-encoding* is usually used for .asd files. Currently, this defaults to :utf-8, and you should be safe using Unicode characters in those files. This might matter, for instance, in meta-data about author's names. Otherwise, the main data in these files is component (path)names, and we don't recommend using non-ASCII characters for these, for the result probably isn't very portable.

Next: , Previous: Controlling source file character encoding, Up: Miscellaneous additional functionality

11.3 Miscellaneous Functions

These functions are exported by ASDF for your convenience.

— Function: system-relative-pathname system name &key type

It's often handy to locate a file relative to some system. The system-relative-pathname function meets this need.

It takes two mandatory arguments system and name and a keyword argument type: system is name of a system, whereas name and optionally type specify a relative pathname, interpreted like a component pathname specifier by coerce-pathname. See Pathname specifiers.

It returns a pathname built from the location of the system's source directory and the relative pathname. For example: 

          > (asdf:system-relative-pathname 'cl-ppcre "regex.data")
          #P"/repository/other/cl-ppcre/regex.data"
— Function: system-source-directory system-designator

ASDF does not provide a turnkey solution for locating data (or other miscellaneous) files that are distributed together with the source code of a system. Programmers can use system-source-directory to find such files. Returns a pathname object. The system-designator may be a string, symbol, or ASDF system object.

— Function: clear-system system-designator

It is sometimes useful to force recompilation of a previously loaded system. For these cases, (asdf:clear-system :foo) will remove the system from the table of currently loaded systems: the next time the system foo or one that depends on it is re-loaded, foo will be loaded again.16

Note that this does not and cannot undo the previous loading of the system. Common Lisp has no provision for such an operation, and its reliance on irreversible side-effects to global data structures makes such a thing impossible in the general case. If the software being re-loaded is not conceived with hot upgrade in mind, re-loading may cause many errors, warnings or subtle silent problems, as packages, generic function signatures, structures, types, macros, constants, etc. are being redefined incompatibly. It is up to the user to make sure that reloading is possible and has the desired effect. In some cases, extreme measures such as recursively deleting packages, unregistering symbols, defining methods on update-instance-for-redefined-class and much more are necessary for reloading to happen smoothly. ASDF itself goes to extensive effort to make a hot upgrade possible with respect to its own code. If you want, you can reuse some of its utilities such as uiop:define-package and uiop:with-upgradability, and get inspiration (or disinspiration) from what it does in header.lisp and upgrade.lisp.

— Function: register-preloaded-system name &rest keys

A system with name name, created by make-instance with extra keys keys (e.g. :version), is registered as preloaded. That is, its code has already been loaded into the current image, and if at some point some other system :depends-on it yet no source code is found, it is considered as already provided, and ASDF will not raise a missing-component error.

This function is particularly useful if you distribute your code as fasls with either compile-bundle-op or monolithic-compile-bundle-op, and want to register systems so that dependencies will work uniformly whether you're using your software from source or from fasl.

— Function: register-immutable-system name &rest keys

A system with name name, created by make-instance with extra keys keys (e.g. :version), is registered as immutable. That is, its code has already been loaded into the current image, and if at some point some other system :depends-on it, it is considered as already provided, and no attempt will be made to search for an updated version from the source-registry or any other method. There will be no search for an .asd file, and no missing-component error.

This function (available since ASDF 3.1.5) is particularly useful if you distribute a large body of code as a precompiled image, and want to allow users to extend the image with further extension systems, but without making thousands of filesystem requests looking for inexistent (or worse, out of date) source code for all the systems that came bundled with the image but aren't distributed as source code to regular users.

— Function: run-shell-command control-string &rest args

This function is obsolete and present only for the sake of backwards-compatibility: “If it's not backwards, it's not compatible”. We strongly discourage its use. Its current behaviour is only well-defined on Unix platforms (which include MacOS X and cygwin). On Windows, anything goes. The following documentation is only for the purpose of your migrating away from it in a way that preserves semantics.

Instead we recommend the use run-program, described in the next section, and available as part of ASDF since ASDF 3.

run-shell-command takes as arguments a format control-string and arguments to be passed to format after this control-string to produce a string. This string is a command that will be evaluated with a POSIX shell if possible; yet, on Windows, some implementations will use CMD.EXE, while others (like SBCL) will make an attempt at invoking a POSIX shell (and fail if it is not present).

Previous: Miscellaneous Functions, Up: Miscellaneous additional functionality

11.4 Some Utility Functions

The below functions are not exported by ASDF itself, but by UIOP, available since ASDF 3. Some of them have precursors in ASDF 2, but we recommend you rely on ASDF 3 for active developments. UIOP provides many, many more utility functions, and we recommend you read its README and sources for more information.

— Function: parse-unix-namestring name &key type defaults dot-dot ensure-directory &allow-other-keys

Coerce NAME into a PATHNAME using standard Unix syntax.

Unix syntax is used whether or not the underlying system is Unix; on non-Unix systems it is only usable for relative pathnames. In order to manipulate relative pathnames portably, it is crucial to possess a portable pathname syntax independent of the underlying OS. This is what parse-unix-namestring provides, and why we use it in ASDF.

When given a pathname object, just return it untouched. When given nil, just return nil. When given a non-null symbol, first downcase its name and treat it as a string. When given a string, portably decompose it into a pathname as below.

#\/ separates directory components.

The last #\/-separated substring is interpreted as follows: 1- If type is :directory or ensure-directory is true, the string is made the last directory component, and its name and type are nil. if the string is empty, it's the empty pathname with all slots nil. 2- If type is nil, the substring is a file-namestring, and its name and type are separated by split-name-type. 3- If type is a string, it is the given type, and the whole string is the name.

Directory components with an empty name the name . are removed. Any directory named .. is read as dot-dot, which must be one of :back or :up and defaults to :back.

host, device and version components are taken from defaults, which itself defaults to *nil-pathname*. *nil-pathname* is also used if defaults is nil. No host or device can be specified in the string itself, which makes it unsuitable for absolute pathnames outside Unix.

For relative pathnames, these components (and hence the defaults) won't matter if you use merge-pathnames* but will matter if you use merge-pathnames, which is an important reason to always use merge-pathnames*.

Arbitrary keys are accepted, and the parse result is passed to ensure-pathname with those keys, removing type, defaults and dot-dot. When you're manipulating pathnames that are supposed to make sense portably even though the OS may not be Unixish, we recommend you use :want-relative t so that parse-unix-namestring will throw an error if the pathname is absolute.

— Function: merge-pathnames* specified &optional defaults

This function is a replacement for merge-pathnames that uses the host and device from the defaults rather than the specified pathname when the latter is a relative pathname. This allows ASDF and its users to create and use relative pathnames without having to know beforehand what are the host and device of the absolute pathnames they are relative to.

— Function: subpathname pathname subpath &key type

This function takes a pathname and a subpath and a type. If subpath is already a pathname object (not namestring), and is an absolute pathname at that, it is returned unchanged; otherwise, subpath is turned into a relative pathname with given type as per parse-unix-namestring with :want-relative t :type type, then it is merged with the pathname-directory-pathname of pathname, as per merge-pathnames*.

We strongly encourage the use of this function for portably resolving relative pathnames in your code base.

— Function: subpathname* pathname subpath &key type

This function returns nil if the base pathname is nil, otherwise acts like subpathname.

— Function: run-program command &key ignore-error-status force-shell input output error-output if-input-does-not-exist if-output-exists if-error-output-exists element-type external-format &allow-other-keys

run-program takes a command argument that is either a list of a program name or path and its arguments, or a string to be executed by a shell. It spawns the command, waits for it to return, verifies that it exited cleanly (unless told not too below), and optionally captures and processes its output. It accepts many keyword arguments to configure its behaviour.

run-program returns three values: the first for the output, the second for the error-output, and the third for the return value. (Beware that before ASDF 3.0.2.11, it didn't handle input or error-output, and returned only one value, the one for the output if any handler was specified, or else the exit code; please upgrade ASDF, or at least UIOP, to rely on the new enhanced behaviour.)

output is its most important argument; it specifies how the output is captured and processed. If it is nil, then the output is redirected to the null device, that will discard it. If it is :interactive, then it is inherited from the current process (beware: this may be different from your *standard-output*, and under SLIME will be on your *inferior-lisp* buffer). If it is t, output goes to your current *standard-output* stream. Otherwise, output should be a value that is a suitable first argument to slurp-input-stream (see below), or a list of such a value and keyword arguments. In this case, run-program will create a temporary stream for the program output; the program output, in that stream, will be processed by a call to slurp-input-stream, using output as the first argument (or if it's a list the first element of output and the rest as keywords). The primary value resulting from that call (or nil if no call was needed) will be the first value returned by run-program. E.g., using :output :string will have it return the entire output stream as a string. And using :output '(:string :stripped t) will have it return the same string stripped of any ending newline.

error-output is similar to output, except that the resulting value is returned as the second value of run-program. t designates the *error-output*. Also :output means redirecting the error output to the output stream, in which case nil is returned.

input is similar to output, except that vomit-output-stream is used, no value is returned, and t designates the *standard-input*.

element-type and external-format are passed on to your Lisp implementation, when applicable, for creation of the output stream.

One and only one of the stream slurping or vomiting may or may not happen in parallel in parallel with the subprocess, depending on options and implementation, and with priority being given to output processing. Other streams are completely produced or consumed before or after the subprocess is spawned, using temporary files.

force-shell forces evaluation of the command through a shell, even if it was passed as a list rather than a string. If a shell is used, it is /bin/sh on Unix or CMD.EXE on Windows, except on implementations that (erroneously, IMNSHO) insist on consulting $SHELL like clisp.

ignore-error-status causes run-program to not raise an error if the spawned program exits in error. Following POSIX convention, an error is anything but a normal exit with status code zero. By default, an error of type subprocess-error is raised in this case.

run-program works on all platforms supported by ASDF, except Genera. See the source code for more documentation.

— Function: slurp-input-stream processor input-stream &key

slurp-input-stream is a generic function of two arguments, a target object and an input stream, and accepting keyword arguments. Predefined methods based on the target object are as follows:

Next: , Previous: Miscellaneous additional functionality, Up: Top

12 Getting the latest version

Decide which version you want. The master branch is where development happens; its HEAD is usually OK, including the latest fixes and portability tweaks, but an occasional regression may happen despite our (limited) test suite.

The release branch is what cautious people should be using; it has usually been tested more, and releases are cut at a point where there isn't any known unresolved issue.

You may get the ASDF source repository using git: git clone https://gitlab.common-lisp.net/asdf/asdf.git

You will find the above referenced tags in this repository. You can also browse the repository on https://gitlab.common-lisp.net/asdf/asdf.

Discussion of ASDF development is conducted on the mailing list (see Mailing list).

Next: , Previous: Getting the latest version, Up: Top

13 FAQ

Next: , Previous: FAQ, Up: FAQ

13.1 “Where do I report a bug?”

ASDF bugs are tracked on launchpad: https://launchpad.net/asdf.

If you're unsure about whether something is a bug, or for general discussion, use the asdf-devel mailing list (see Mailing list).

Next: , Previous: Where do I report a bug?, Up: FAQ

13.2 Mailing list

Discussion of ASDF development is conducted on the mailing list asdf-devel@common-lisp.net. http://common-lisp.net/cgi-bin/mailman/listinfo/asdf-devel

Next: , Previous: Mailing list, Up: FAQ

13.3 “What has changed between ASDF 1, ASDF 2, and ASDF 3?”

We released ASDF 2.000 on May 31st 2010, ASDF 3.0.0 on May 15th 2013, ASDF 3.1.2 on May 6th 2014. Releases of ASDF 2 and now ASDF 3 have since then been included in all actively maintained CL implementations that used to bundle ASDF 1, plus many implementations that previously did not. ASDF has been made to work with all actively maintained CL implementations and even a few implementations that are not actively maintained. See “What has changed between ASDF 1 ASDF 2 and ASDF 3?”.

Furthermore, it is possible to upgrade from ASDF 1 to ASDF 2 or ASDF 3 on the fly (though we recommend instead upgrading your implementation or replacing its ASDF module). For this reason, we have stopped supporting ASDF 1 and ASDF 2. If you are using ASDF 1 or ASDF 2 and are experiencing any kind of issues or limitations, we recommend you upgrade to ASDF 3 — and we explain how to do that. See Loading ASDF.

Note that in the context of compatibility requirements, ASDF 2.27, released on Feb 1st 2013, and further releases up to 2.33, count as pre-releases of ASDF 3, and define the :asdf3 feature, though the first stable release of ASDF 3 was release 3.0.1. Significant new or improved functionality were added in ASDF 3.1; the :asdf3.1 feature is present in recent enough versions to detect this functionality; the first stable release since then was ASDF 3.1.2. New *features* are only added at major milestones, and the next one will probably be :asdf3.2.

Next: , Previous: What has changed between ASDF 1 ASDF 2 and ASDF 3?, Up: What has changed between ASDF 1 ASDF 2 and ASDF 3?

13.3.1 What are ASDF 1, ASDF 2, and ASDF 3?

ASDF 1 refers to any release earlier than 1.369 or so (from August 2001 to October 2009), and to any development revision earlier than 2.000 (May 2010). If your copy of ASDF doesn't even contain version information, it's an old ASDF 1. Revisions between 1.656 and 1.728 may count as development releases for ASDF 2.

ASDF 2 refers to releases from 2.000 (May 31st 2010) to 2.26 (Oct 30th 2012), and any development revision newer than ASDF 1 and older than 2.27 (Feb 1st 2013).

ASDF 3 refers to releases from 2.27 (Feb 1st 2013) to 2.33 and 3.0.0 onward (May 15th 2013). 2.27 to 2.33 count as pre-releases to ASDF 3.

ASDF 3.1 refers to releases from 3.1.2 (May 6th 2014) onward. These releases are also considered part of ASDF 3.

Next: , Previous: What are ASDF 1 2 3?, Up: What has changed between ASDF 1 ASDF 2 and ASDF 3?

13.3.2 How do I detect the ASDF version?

All releases of ASDF push :asdf onto *features*. Releases starting with ASDF 2 push :asdf2 onto *features*. Releases starting with ASDF 3 (including 2.27 and later pre-releases) push :asdf3 onto *features*. Furthermore, releases starting with ASDF 3.1.2 (May 2014), though they count as ASDF 3, include enough progress that they also push :asdf3.1 onto *features*. You may depend on the presence or absence of these features to write code that takes advantage of recent ASDF functionality but still works on older versions, or at least detects the old version and signals an error.

Additionally, all releases starting with ASDF 2 define a function (asdf:asdf-version) you may use to query the version. All releases starting with 2.013 display the version number prominently on the second line of the asdf.lisp source file.

If you are experiencing problems or limitations of any sort with ASDF 1 or ASDF 2, we recommend that you should upgrade to the latest release, be it ASDF 3 or other.

Finally, here is a code snippet to programmatically determine what version of ASDF is loaded, if any, that works on all versions including very old ones: 

     (when (find-package :asdf)
       (let ((ver (symbol-value
                        (or (find-symbol (string :*asdf-version*) :asdf)
                            (find-symbol (string :*asdf-revision*) :asdf)))))
         (etypecase ver
           (string ver)
           (cons (with-output-to-string (s)
                   (loop for (n . m) on ver
                         do (princ n s)
                            (when m (princ "." s)))))
           (null "1.0"))))

If it returns nil then ASDF is not installed. Otherwise it should return a string. If it returns "1.0", then it can actually be any version before 1.77 or so, or some buggy variant of 1.x. If it returns anything older than "3.0.1", you really need to upgrade your implementation or at least upgrade its ASDF. See Replacing your implementation's ASDF.

Next: , Previous: How do I detect the ASDF version?, Up: What has changed between ASDF 1 ASDF 2 and ASDF 3?

13.3.3 ASDF can portably name files in subdirectories

Common Lisp namestrings are not portable, except maybe for logical pathname namestrings, that themselves have various limitations and require a lot of setup that is itself ultimately non-portable.

In ASDF 1, the only portable ways to refer to pathnames inside systems and components were very awkward, using #.(make-pathname ...) and #.(merge-pathnames ...). Even the above were themselves were inadequate in the general case due to host and device issues, unless horribly complex patterns were used. Plenty of simple cases that looked portable actually weren't, leading to much confusion and greavance.

ASDF 2 implements its own portable syntax for strings as pathname specifiers. Naming files within a system definition becomes easy and portable again. See system-relative-pathname, merge-pathnames*, coerce-pathname.

On the other hand, there are places where systems used to accept namestrings where you must now use an explicit pathname object: (defsystem ... :pathname "LOGICAL-HOST:PATH;TO;SYSTEM;" ...) must now be written with the #p syntax: (defsystem ... :pathname #p"LOGICAL-HOST:PATH;TO;SYSTEM;" ...)

See Pathname specifiers.

Next: , Previous: ASDF can portably name files in subdirectories, Up: What has changed between ASDF 1 ASDF 2 and ASDF 3?

13.3.4 Output translations

A popular feature added to ASDF was output pathname translation: asdf-binary-locations, common-lisp-controller, cl-launch and other hacks were all implementing it in ways both mutually incompatible and difficult to configure.

Output pathname translation is essential to share source directories of portable systems across multiple implementations or variants thereof, or source directories of shared installations of systems across multiple users, or combinations of the above.

In ASDF 2, a standard mechanism is provided for that, asdf-output-translations, with sensible defaults, adequate configuration languages, a coherent set of configuration files and hooks, and support for non-Unix platforms.

See Controlling where ASDF saves compiled files.

Next: , Previous: Output translations, Up: What has changed between ASDF 1 ASDF 2 and ASDF 3?

13.3.5 Source Registry Configuration

Configuring ASDF used to require special magic to be applied just at the right moment, between the moment ASDF is loaded and the moment it is used, in a way that is specific to the user, the implementation he is using and the application he is building.

This made for awkward configuration files and startup scripts that could not be shared between users, managed by administrators or packaged by distributions.

ASDF 2 provides a well-documented way to configure ASDF, with sensible defaults, adequate configuration languages, and a coherent set of configuration files and hooks.

We believe it's a vast improvement because it decouples application distribution from library distribution. The application writer can avoid thinking where the libraries are, and the library distributor (dpkg, clbuild, advanced user, etc.) can configure them once and for every application. Yet settings can be easily overridden where needed, so whoever needs control has exactly as much as required.

At the same time, ASDF 2 remains compatible with the old magic you may have in your build scripts (using *central-registry* and *system-definition-search-functions*) to tailor the ASDF configuration to your build automation needs, and also allows for new magic, simpler and more powerful magic.

See Controlling where ASDF searches for systems.

Next: , Previous: Source Registry Configuration, Up: What has changed between ASDF 1 ASDF 2 and ASDF 3?

13.3.6 Usual operations are made easier to the user

In ASDF 1, you had to use the awkward syntax (asdf:oos 'asdf:load-op :foo) to load a system, and similarly for compile-op, test-op.

In ASDF 2, you can use shortcuts for the usual operations: (asdf:load-system :foo), and similarly for compile-system, test-system.

Next: , Previous: Usual operations are made easier to the user, Up: What has changed between ASDF 1 ASDF 2 and ASDF 3?

13.3.7 Many bugs have been fixed

The following issues and many others have been fixed:

Next: , Previous: Many bugs have been fixed, Up: What has changed between ASDF 1 ASDF 2 and ASDF 3?

13.3.8 ASDF itself is versioned

Between new features, old bugs fixed, and new bugs introduced, there were various releases of ASDF in the wild, and no simple way to check which release had which feature set. People using or writing systems had to either make worst-case assumptions as to what features were available and worked, or take great pains to have the correct version of ASDF installed.

With ASDF 2, we provide a new stable set of working features that everyone can rely on from now on. Use #+asdf2 to detect presence of ASDF 2, (asdf:version-satisfies (asdf:asdf-version) "2.345.67") to check the availability of a version no earlier than required.

Next: , Previous: ASDF itself is versioned, Up: What has changed between ASDF 1 ASDF 2 and ASDF 3?

13.3.9 ASDF can be upgraded

When an old version of ASDF was loaded, it was very hard to upgrade ASDF in your current image without breaking everything. Instead you had to exit the Lisp process and somehow arrange to start a new one from a simpler image. Something that can't be done from within Lisp, making automation of it difficult, which compounded with difficulty in configuration, made the task quite hard. Yet as we saw before, the task would have been required to not have to live with the worst case or non-portable subset of ASDF features.

With ASDF 2, it is easy to upgrade from ASDF 2 to later versions from within Lisp, and not too hard to upgrade from ASDF 1 to ASDF 2 from within Lisp. We support hot upgrade of ASDF and any breakage is a bug that we will do our best to fix. There are still limitations on upgrade, though, most notably the fact that after you upgrade ASDF, you must also reload or upgrade all ASDF extensions.

Next: , Previous: ASDF can be upgraded, Up: What has changed between ASDF 1 ASDF 2 and ASDF 3?

13.3.10 Decoupled release cycle

When vendors were releasing their Lisp implementations with ASDF, they had to basically never change version because neither upgrade nor downgrade was possible without breaking something for someone, and no obvious upgrade path was visible and recommendable.

With ASDF 2, upgrade is possible, easy and can be recommended. This means that vendors can safely ship a recent version of ASDF, confident that if a user isn't fully satisfied, he can easily upgrade ASDF and deal with a supported recent version of it. This means that release cycles will be causally decoupled, the practical consequence of which will mean faster convergence towards the latest version for everyone.

Next: , Previous: Decoupled release cycle, Up: What has changed between ASDF 1 ASDF 2 and ASDF 3?

13.3.11 Pitfalls of the transition to ASDF 2

The main pitfalls in upgrading to ASDF 2 seem to be related to the output translation mechanism.

Other issues include the following:

Next: , Previous: Pitfalls of the transition to ASDF 2, Up: What has changed between ASDF 1 ASDF 2 and ASDF 3?

13.3.12 Pitfalls of the upgrade to ASDF 3

While ASDF 3 is largely compatible with ASDF 2, there are a few pitfalls when upgrading from ASDF 2, due to limitations in ASDF 2.

Previous: Pitfalls of the upgrade to ASDF 3, Up: What has changed between ASDF 1 ASDF 2 and ASDF 3?

13.3.13 What happened to the bundle operations?

asdf-ecl and its short-lived successor asdf-bundle are no more, having been replaced by code now built into ASDF 3. Moreover, the name of the bundle operations has changed since ASDF 3.1.3. See What happened to the bundle operations.

And yet, the feature is not enabled to be used by load-system by default on ECL as originally intended, because of a bug in ECL itself found during testing.

Some of the bundle operations were renamed after ASDF 3.1.3, and the old names have been removed. Old bundle operations, and their modern equivalents are:

Next: , Previous: What has changed between ASDF 1 ASDF 2 and ASDF 3?, Up: FAQ

13.4 Issues with installing the proper version of ASDF

Next: , Previous: Issues with installing the proper version of ASDF, Up: Issues with installing the proper version of ASDF

13.4.1 “My Common Lisp implementation comes with an outdated version of ASDF. What to do?”

If you have a recent implementation, it should already come with ASDF 3 or later. If you need a more recent version than is provided, we recommend you simply upgrade ASDF by installing a recent version in a path configured in your source-registry. See Upgrading ASDF.

If you have an old implementation that does not provide ASDF 3, we recommend you replace your implementation's ASDF. See Replacing your implementation's ASDF.

Previous: My Common Lisp implementation comes with an outdated version of ASDF. What to do?, Up: Issues with installing the proper version of ASDF

13.4.2 “I'm a Common Lisp implementation vendor. When and how should I upgrade ASDF?”

Since ASDF 2, it should always be a good time to upgrade to a recent version of ASDF. You may consult with the maintainer for which specific version they recommend, but the latest release should be correct. Though we do try to test ASDF releases against all implementations that we can, we may not be testing against all variants of your implementation, and we may not be running enough tests; we trust you to thoroughly test it with your own implementation before you release it. If there are any issues with the current release, it's a bug that you should report upstream and that we will fix ASAP.

As to how to include ASDF, we recommend the following:

Next: , Previous: Issues with installing the proper version of ASDF, Up: FAQ

13.5 Issues with configuring ASDF

Next: , Previous: Issues with configuring ASDF, Up: Issues with configuring ASDF

13.5.1 “How can I customize where fasl files are stored?”

See Controlling where ASDF saves compiled files.

Note that in the past there was an add-on to ASDF called ASDF-binary-locations, developed by Gary King. That add-on has been merged into ASDF proper, then superseded by the asdf-output-translations facility.

Note that use of asdf-output-translations can interfere with one aspect of your systems — if your system uses *load-truename* to find files (e.g., if you have some data files stored with your program), then the relocation that this ASDF customization performs is likely to interfere. Use asdf:system-relative-pathname to locate a file in the source directory of some system, and use asdf:apply-output-translations to locate a file whose pathname has been translated by the facility.

Previous: How can I customize where fasl files are stored?, Up: Issues with configuring ASDF

13.5.2 “How can I wholly disable the compiler output cache?”

To permanently disable the compiler output cache for all future runs of ASDF, you can: 

     mkdir -p ~/.config/common-lisp/asdf-output-translations.conf.d/
     echo ':disable-cache' > \
     ~/.config/common-lisp/asdf-output-translations.conf.d/99-disable-cache.conf

This assumes that you didn't otherwise configure the ASDF files (if you did, edit them again), and don't somehow override the configuration at runtime with a shell variable (see below) or some other runtime command (e.g. some call to asdf:initialize-output-translations).

To disable the compiler output cache in Lisp processes run by your current shell, try (assuming bash or zsh) (on Unix and cygwin only): 

     export ASDF_OUTPUT_TRANSLATIONS=/:

To disable the compiler output cache just in the current Lisp process, use (after loading ASDF but before using it): 

     (asdf:disable-output-translations)

Note that this does NOT belong in a .asd file. Please do not tamper with ASDF configuration from a .asd file, and only do this from your personal configuration or build scripts.

Next: , Previous: Issues with configuring ASDF, Up: FAQ

13.6 Issues with using and extending ASDF to define systems

Next: , Previous: Issues with using and extending ASDF to define systems, Up: Issues with using and extending ASDF to define systems

13.6.1 “How can I cater for unit-testing in my system?”

ASDF provides a predefined test operation, test-op. See test-op. The test operation, however, is largely left to the system definer to specify. test-op has been a topic of considerable discussion on the asdf-devel mailing list (see Mailing list), and on the launchpad bug-tracker (see Where do I report a bug?). We provide some guidelines in the discussion of test-op.

Next: , Previous: How can I cater for unit-testing in my system?, Up: Issues with using and extending ASDF to define systems

13.6.2 “How can I cater for documentation generation in my system?”

Various ASDF extensions provide some kind of doc-op operation. See also https://bugs.launchpad.net/asdf/+bug/479470.

Next: , Previous: How can I cater for documentation generation in my system?, Up: Issues with using and extending ASDF to define systems

13.6.3 “How can I maintain non-Lisp (e.g. C) source files?”

See cffi's cffi-grovel.

Next: , Previous: How can I maintain non-Lisp (e.g. C) source files?, Up: Issues with using and extending ASDF to define systems

13.6.4 “I want to put my module's files at the top level. How do I do this?”

By default, the files contained in an asdf module go in a subdirectory with the same name as the module. However, this can be overridden by adding a :pathname "" argument to the module description. For example, here is how it could be done in the spatial-trees ASDF system definition for ASDF 2 or later: 

     (asdf:defsystem "spatial-trees"
       :components
       ((:module "base"
                 :pathname ""
                 :components
                 ((:file "package")
                  (:file "basedefs" :depends-on ("package"))
                  (:file "rectangles" :depends-on ("package"))))
        (:module tree-impls
                 :depends-on ("base")
                 :pathname ""
                 :components
                 ((:file "r-trees")
                  (:file "greene-trees" :depends-on ("r-trees"))
                  (:file "rstar-trees" :depends-on ("r-trees"))
                  (:file "rplus-trees" :depends-on ("r-trees"))
                  (:file "x-trees" :depends-on ("r-trees" "rstar-trees"))))
        (:module viz
                 :depends-on ("base")
                 :pathname ""
                 :components
                 ((:static-file "spatial-tree-viz.lisp")))
        (:module tests
                 :depends-on ("base")
                 :pathname ""
                 :components
                 ((:static-file "spatial-tree-test.lisp")))
        (:static-file "LICENCE")
        (:static-file "TODO")))

All of the files in the tree-impls module are at the top level, instead of in a tree-impls/ subdirectory.

Note that the argument to :pathname can be either a pathname object or a string. A pathname object can be constructed with the #p"foo/bar/" syntax, but this is discouraged because the results of parsing a namestring are not portable. A pathname can only be portably constructed with such syntax as #.(make-pathname :directory '(:relative "foo" "bar")), and similarly the current directory can only be portably specified as #.(make-pathname :directory '(:relative)). However, as of ASDF 2, you can portably use a string to denote a pathname. The string will be parsed as a /-separated path from the current directory, such that the empty string "" denotes the current directory, and "foo/bar" (no trailing / required in the case of modules) portably denotes the same subdirectory as above. When files are specified, the last /-separated component is interpreted either as the name component of a pathname (if the component class specifies a pathname type), or as a name component plus optional dot-separated type component (if the component class doesn't specifies a pathname type).

Next: , Previous: I want to put my module's files at the top level. How do I do this?, Up: Issues with using and extending ASDF to define systems

13.6.5 How do I create a system definition where all the source files have a .cl extension?

Starting with ASDF 2.014.14, you may just pass the builtin class cl-source-file.cl as the :default-component-class argument to defsystem: 

     (defsystem my-cl-system
       :default-component-class cl-source-file.cl
       ...)

Another builtin class cl-source-file.lsp is offered for files ending in .lsp.

If you want to use a different extension for which ASDF doesn't provide builtin support, or want to support versions of ASDF earlier than 2.014.14 (but later than 2.000), you can define a class as follows: 

     ;; Prologue: make sure we're using a sane package.
     (defpackage :my-asdf-extension
        (:use :asdf :common-lisp)
        (:export #:cl-source-file.lis))
     (in-package :my-asdf-extension)
     
     (defclass cl-source-file.lis (cl-source-file)
       ((type :initform "lis")))

Then you can use it as follows: 

     (defsystem my-cl-system
       :default-component-class my-asdf-extension:cl-source-file.lis
       ...)

Of course, if you're in the same package, e.g. in the same file, you won't need to use the package qualifier before cl-source-file.lis. Actually, if all you're doing is defining this class and using it in the same file without other fancy definitions, you might skip package complications: 

     (in-package :asdf)
     (defclass cl-source-file.lis (cl-source-file)
        ((type :initform "lis")))
     (defsystem my-cl-system
       :default-component-class cl-source-file.lis
       ...)

Next: , Previous: How do I create a system definition where all the source files have a .cl extension?, Up: Issues with using and extending ASDF to define systems

13.6.6 How do I mark a source file to be loaded only and not compiled?

There is no provision in ASDF for ensuring that some components are always loaded as source, while others are always compiled. There is load-source-op (see load-source-op), but that is an operation to be applied to a system as a whole, not to one or another specific source files. While this idea often comes up in discussions, it doesn't play well with either the linking model of ECL or with various bundle operations. In addition, the dependency model of ASDF would have to be modified incompatibly to allow for such a trick.

Previous: How do I mark a source file to be loaded only and not compiled?, Up: Issues with using and extending ASDF to define systems

13.6.7 How do I work with readtables?

It is possible to configure the lisp syntax by modifying the currently-active readtable. However, this same readtable is shared globally by all software being compiled by ASDF, especially since load and compile-file both bind *readtable*, so that its value is the same across the build at the start of every file (unless overridden by some perform :around method), even if a file locally binds it to a different readtable during the build.

Therefore, the following hygiene restrictions apply. If you don't abide by these restrictions, there will be situations where your output files will be corrupted during an incremental build. We are not trying to prescribe new restrictions for the sake of good style: these restrictions have always applied implicitly, and we are simply describing what they have always been.

If you want to use readtable modifications that cannot abide by those restrictions, you must create a different readtable object and set *readtable* to temporarily bind it to your new readtable (which will be undone after processing the file).

For that, we recommend you use system named-readtables to define or combine such readtables using named-readtables:defreadtable and use them using named-readtables:in-readtable. Equivalently, you can use system cl-syntax, that itself uses named-readtables, but may someday do more with, e.g. *print-pprint-dispatch*.

For even more advanced syntax modification beyond what a readtable can express, you may consider either:

Beware that it is unsafe to use ASDF from the REPL to compile or load systems while the readtable isn't the shared readtable previously used to build software. You must manually undo any binding of *readtable* at the REPL and restore its initial value whenever you call operate (via e.g. load-system, test-system or require) from a REPL that is using a different readtable.

13.6.7.1 How should my system use a readtable exported by another system?

Use from the named-readtables system the macro named-readtables:in-readtable.

If the other system fails to use named-readtables, fix it and send a patch upstream. In the day and age of Quicklisp and clbuild, there is little reason to eschew using such an important library anymore.

13.6.7.2 How should my library make a readtable available to other systems?

Use from the named-readtables system the macro named-readtables:defreadtable.

Previous: Issues with using and extending ASDF to define systems, Up: FAQ

13.7 ASDF development FAQs

Previous: ASDF development FAQs, Up: ASDF development FAQs

13.7.1 How do run the tests interactively in a REPL?

This not-so-frequently asked question is primarily for ASDF developers, but those who encounter an unexpected error in some test may be interested, too.

Here's the procedure for experimenting with tests in a REPL: 

     ;; BEWARE! Some tests expect you to be in the .../asdf/test directory
     ;; If your REPL is not there yet, change your current directory:
     ;; under SLIME, you may: ,change-directory ~/common-lisp/asdf/test/
     ;; otherwise you may evaluate something like:
     (require "asdf") (asdf:upgrade-asdf) ;load UIOP & update asdf.lisp
     (uiop:chdir (asdf:system-relative-pathname :asdf "test/"))
     (setf *default-pathname-defaults* (uiop:getcwd))
     
     ;; Load the test script support.
     (load "script-support.lisp")
     
     ;; Initialize the script support.
     ;; This will also change your *package* to asdf-test.
     ;; NB: this function is also available from package cl-user,
     ;; and also available with the shorter name da in both packages.
     (asdf-test::debug-asdf)
     
     ;; In case you modified ASDF since you last tested it,
     ;; you need to update asdf.lisp itself by evaluating 'make' in a shell,
     ;; or (require "asdf") (asdf:load-system :asdf) in another CL REPL,
     ;; if not done in this REPL above.
     ;; *Then*, in this REPL, you need to evaluate:
     ;(asdf-test::compile-load-asdf)
     
     ;; Now, you may experiment with test code from a .script file.
     ;; See the instructions given at the end of your failing test
     ;; to identify which form is needed, e.g.
     (frob-packages)
     (asdf::with-asdf-cache () (load "test-utilities.script"))

Next: , Previous: FAQ, Up: Top

Ongoing Work

For an active list of things to be done, see the TODO file in the source repository.

Also, bugs are currently tracked on launchpad: https://launchpad.net/asdf.

Next: , Previous: Ongoing Work, Up: Top

Bibliography

Next: , Previous: Bibliography, Up: Top

Concept Index

Next: , Previous: Concept Index, Up: Top

Function and Class Index

Previous: Function and Class Index, Up: Top

Variable Index

Footnotes

[1] NB: all implementations except GNU CLISP also accept (require "ASDF"), (require 'asdf) and (require :asdf). For portability's sake, you should use (require "asdf").

[2] ~/common-lisp/ is only included in the default configuration starting with ASDF 3.1.2 or later. If your implementation provides an earlier variant of ASDF, you may need to explicitly configure it to use this path, as further explained.

[3] For Windows users, and starting with ASDF 3.1.5, start from your %LOCALAPPDATA%, which is usually ~/AppData/Local/ (but you can ask in a CMD.EXE terminal echo %LOCALAPPDATA% to make sure) and underneath create a subpath config/common-lisp/source-registry.conf.d/.

[4] By requiring the .conf extension, and ignoring other files, ASDF allows you to have disabled files, editor backups, etc. in the same directory with your active configuration files.

ASDF will also ignore files whose names start with a . character.

It is customary to start the filename with two digits, to control the sorting of the conf files in the source registry directory, and thus the order in which the directories will be scanned.

[5] It is possible to further customize the system definition file search. That's considered advanced use, and covered later: search forward for *system-definition-search-functions*. See Defining systems with defsystem.

[6] ASDF will indeed call eval on each entry. It will skip entries that evaluate to nil.

Strings and pathname objects are self-evaluating, in which case the eval step does nothing; but you may push arbitrary s-expressions onto the central registry. These s-expressions may be evaluated to compute context-dependent entries, e.g. things that depend on the value of shell variables or the identity of the user.

The variable asdf:*central-registry* is thus a list of “system directory designators”. A system directory designator is a form which will be evaluated whenever a system is to be found, and must evaluate to a directory to look in (or NIL). By “directory”, we mean “designator for a pathname with a non-empty DIRECTORY component”.

[7] On Windows, you can use Windows shortcuts instead of POSIX symlinks. if you try aliases under MacOS, we are curious to hear about your experience.

[8] For the curious, the option is :force-not (already-loaded-systems).

[9] ASDF 1 and 2 (up until 2.26) used to dynamically create and delete temporary packages asdfN, one for each .asd file, in a misguided attempt to thereby reduce name clashes; but it failed at that goal and only made things more complex. ASDF 3 just uses a shared package asdf-user instead, and relies on the usual Common Lisp conventions to avoid clashes. As far as package oddities go, you may just notice that the asdf-user package also uses uiop/common-lisp, a variant of the common-lisp package that papers over deficiencies in more obscure Common Lisp implementations; but unless you care about Corman Lisp, GCL, Genera or MCL, you shouldn't be concerned.

[10] Historically, the function that built a plan was called traverse, and returned a list of actions; it was deprecated in favor of make-plan (that returns a plan object) when the plan objects were introduced; the old function is kept for backward compatibility and debugging purposes only.

[11] The term action was used by Kent Pitman in his article, “The Description of Large Systems,” (see Bibliography). Although the term was only used by ASDF hackers starting with ASDF 2, the concept was there since the very beginning of ASDF 1, just not clearly articulated.

[12] Note that between releases 2.27 and 3.0.3, only UIOP/PACKAGE, not all of UIOP, was used; if you want your code to work with releases earlier than 3.1.2, you may have to explicitly define a package that uses UIOP, or use proper package prefix to your symbols, as in uiop:version<.

[13] ASDF 2.26 and earlier versions do not support this primary system name convention. With these versions of ASDF you must explicitly load foo.asd before you can use system foo/bar defined therein, e.g. using (asdf:find-system "foo"). We do not support ASDF 2, and recommend that you should upgrade to ASDF 3.

[14] “FASL” is short for “FASt Loading.”

[15] A CLEAN-OP would be a partial solution to this problem.

[16] Alternatively, you could touch foo.asd or remove the corresponding fasls from the output file cache.

[17] Forward incompatibility can be determined using the variable asdf/upgrade::*oldest-forward-compatible-asdf-version*, which is 2.33 at the time of this writing.