NIX PILLS

by Luca Bruno


  • 1: give it a try
  • 2: installation
  • 3: the environment
  • 4: the language
  • 5: functions and imports
  • 6: our first derivation
  • 7: a working derivation

  • 1: give it a try

    the core of a nix based system is the nix store, usually installed under /nix/store , and some tools to manipulate the store

    derivations/packages are stored in the nix store as follows: /nix/store/hash-name where the hash uniquely identifies the derivation, and name is the name of the derivation

    and things in the nix store are immutable

    so where does bash find libc?

             $ ldd  `which bash`
             libc.so.6 => /nix/store/94n6....b978n-glibc-2.19/lib/libc.so.6 (0x00007f0248cce000)
    
    turns out that when bash was built, it used that specific version of glibc and at runtime it will require exactly that glibc version. the version in the derivation name is only a name for us humans. you may end up having a different hash given the same derivation name

    what does it all mean?

  • you could run mysql 5.2 with glibc-2.18, and mysql 5.5 with glibc-2.19
  • you could use your python module with python 2.7 compiled with gcc 4.6 and the same python module with python 3 compiled with gcc 4.8
  • all in the same system! in other words, no dependency hell, not even a dependency resolution algorithm - straight dependencies from derivations to other derivations


    2: installation

    right after copying the store, the installation process initializes the database with the current information. it's under /nix/var/nix/db . it is an sqlite database that keeps track of the dependencies between derivations. the schema is very simple: there's a table of valid paths, mapping from auto increment integer to store path. then there's a dependency relation from one path to other paths. never change /nix/store manually because that wouldn't be in sync with the sqlite db, unless you know what you are doing

    a profile in nix is a general and very convenient concept for realizing rollbacks. profiles are used to compose more components that are spread among multiple paths, under a new unified path. not only, profiles are made up of multiple generations: they are versioned. whenever you change a profile, a new generation is created. generations thus can be switched and rollback-ed

    let's take a closer look at our profile:

       $> ls -l ~/.nix-profile/

    the installation basically reproduced the hierarchy of the nix derivation in the profile by means of symbolic links. but that's only the contents of the latest generation of our profile. in fact, ~/.nix-profile itself is a symbolic link to /nix/var/nix/profiles/default. in turn, that's a symlink to default-1-link in the same directory. yes, that means it's the generation #1 of the default profile. finally that's a symlink to the nix store "user-environment" derivation that you saw printed during the installation process

    nixpkgs is the repository containing nix expressions:

                https://github.com/NixOS/nixpkgs

    the installer downloaded the package descriptions from commit.

    channels are a set of packages and expressions available for download.

    with profiles we're able to manage multiple generations of a composition of packages, while with channels we're able to download binaries


    3: the environment

    let's start by switching user with

            $> su - nix

    if your ~/.profile got evaluated, then you should now be able to run commands like nix-env and nix-store

    Install something

    Let's install nix repl, a simple command line tool for playing with the nix language. Yes, nix is a pure, lazy, functional language, not only a set of tools to manage derivations

    $ nix-env -i nix-repl
    installing `nix-repl-1.7-1734e8a'
    these paths will be fetched (18.61 MiB download, 69.53 MiB unpacked):
    [...]
    building path(s) `/nix/store/f01lfzbw7n0yzhsjd33xfj77li9raljv-user-environment'
    created 24 symlinks in user environment
    

    Now you can run nix repl

    Things to notice:

    We did install software as user, only for the nix user

    It created a new user environment. That's a new generation of our nix user profile

    The nix-env tool manages environments, profiles and their generations

    We installed nix repl by derivation name minus the version. I repeat: we did specify the derivation name (minus the version) to install

    We can list generations without walking through the /nix hierarchy:

    $ nix-env --list-generations
       1   2014-07-24 09:23:30
       2   2014-07-25 08:45:01   (current)
    

    List installed derivations:

    $ nix-env -q
    nix-1.7
    nix-repl-1.7-1734e8a
    

    So, where did nix repl really got installed? 'which nix repl' is ~/.nix-profile/bin/nix-repl which points to the store

    We can also list the derivation paths with

          nix-env -q --out-path
    

    So that's how those derivation paths are called: the output of a build

    rollback / switch generation

         $ nix-env -i man

    The last command installed "man". We should be at generation #3, unless you changed something in the middle. Let's say we want to rollback to the old generation:

            $ nix-env --rollback
    

    switching from generation 3 to 2

    Now nix-env -q does not list "man" anymore. ls -l `which man` should now be your system installed one.

    Enough with the joke, let's go back to the last generation:

             $ nix-env -G 3
    

    switching from generation 2 to 3

    I invite you to read the manpage of nix-env. nix-env requires an operation to perform, then there are common options for all operations, and there are options specific to an operation

    You can of course also delete and upgrade packages

    querying the store

    So far we learned how to query and manipulate the environment. But all of the environment components point to the store.

    To query and manipulate the store, there's the nix-store command. We can do neat things, but we'll only see some queries for now.

    Show direct runtime dependencies of nix repl:

    $ nix-store -q --references `which nix-repl`
    /nix/store/94n64qy99ja0vgbkf675nyk39g9b978n-glibc-2.19
    /nix/store/8jm0wksask7cpf85miyakihyfch1y21q-gcc-4.8.3
    /nix/store/25lg5iqy68k25hdv17yac72kcnwlh4lm-boehm-gc-7.2d
    /nix/store/g21di262aql6xskx15z3qiw3zh3wmjlb-nix-1.7
    /nix/store/jxs2k83npdin18ysnd104xm4sy29m8xp-readline-6.2
    

    The argument to nix-store can be anything as long as it points to the nix store. It will follow symlinks.

    It may not make sense for you right now, but let's print reverse dependencies of nix repl:

    $ nix-store -q --referrers `which nix-repl`
    /nix/store/8rj57vahlndqwg4q095x5qvfa1g4rmvr-env-manifest.nix
    /nix/store/9c8ak2h7c6vbr9kqk8i2n96bwg55br7y-env-manifest.nix
    /nix/store/dr1y2saask2k4y2wrmxw443302pp02z2-user-environment
    nix/store/f01lfzbw7n0yzhsjd33xfj77li9raljv-user-environment
    

    Did you expect it? Our environments depend upon nix repl. Yes, the environments are in the store, and since there are symlinks to nix repl, therefore the environment depends upon nix repl. It lists two environments, generation 2 and generation 3. The manifest.nix file contains metadata about the environment, such as which derivations are installed. So that nix-env can list them, upgrade or remove them. Guess what, the current manifest.nix can be found in ~/.nix-profile/manifest.nix.

    closures

    The closure of a derivation is the list of all dependencies, recursively, down to the bare minimum necessary to use that derivation.

    $ nix-store -qR `which man`
    [...]
    

    Copying all those derivations to the nix store of another machine makes you able to run "man" out of the box on that other machine

    That's the base of nix deployment, you can already foresee the potential when deploying software in the cloud (hint: nix-copy-closure and nix-store --export)

    A nicer view of the closure:

    $ nix-store -q --tree `which man`
    [...]
    

    With the above command, you can know exactly why a runtime dependency, being it direct or indirect, has been picked for a given derivation.

    Same applies to environments of course. As an exercise run

      $  nix-store -q --tree ~/.nix-profile
      [...]
    

    see that the first children are direct dependencies of the user environment: the installed derivations, and the manifest.nix

    dependency resolution

    There isn't anything like apt which solves a SAT problem in order to satisfy dependencies with lower and upper bounds on versions. Because there's no need. A derivation X depends on derivation Y, always

    Fancy disrupt

    $ nix-env -e '*'
    uninstalling `man-1.6g'
    uninstalling `nix-repl-1.7-1734e8a'
    uninstalling `nix-1.7'
    [...]
    

    Ops, that uninstalled all derivations from the environment, including nix. We are not able to run nix-env, what now?

    Environments are a convenience for the user, but nix is still there, in the store!

    First pick one nix-1.7 derivation:

      ls /nix/store/*nix-1.7,
    say /nix/store/g21di262aql6xskx15z3qiw3zh3wmjlb-nix-1.7.

    The first possibility is to rollback:

    $ /nix/store/g21di262aql6xskx15z3qiw3zh3wmjlb-nix-1.7/bin/nix-env --rollback

    The second possibility is to install nix, thus creating a new generation:

    $ /nix/store/g21di262aql6xskx15z3qiw3zh3wmjlb-nix-1.7/bin/nix-env -i
         /nix/store/g21di262aql6xskx15z3qiw3zh3wmjlb-nix-1.7
    

    channels

    So where are we getting packages from? There's a list of channels from which we get packages, usually we use a single channel. The tool to manage channels is nix-channel.

           $ nix-channel --list
            nixpkgs http://nixos.org/channels/nixpkgs-unstable
    

    That's basically the contents of ~/.nix-channels. Note: ~/.nix-channels is not a symlink to the nix store!

    To update the channel run

        $ nix-channel --update
    

    It will download the new nix expressions (descriptions of the packages), create a new generation of the channels profile and unpack under ~/.nix-defexpr/channels

    We learned how to query the user environment and to manipulate it by installing and uninstalling software. Upgrading software is as straight as it gets by reading the manual (nix-env -u '*' will upgrade all packages in the environment)

    Everytime we change the environment, a new generation gets created. Switching between generations is easy and immediate

    Then we queried the store. We inspected the dependencies and reverse dependencies of store paths.

    We still see symlinks to compose paths from the nix store, our lovely trick

    Quick analogy with programming languages:

    you have the heap with all the objects, that's the nix store

    you have objects that point to other objects, those are the derivations


    4: the language

    the nix language is used to write derivations. it's only about writing utility functions for making things convenient

    in nix, everything is an expression, there are no statements (this is common to many fp languages)

    values in nix are immutable

    value types

    We've installed nix repl in the previous pill. If you didn't,

         nix-env -i nix-repl
    

    nix supports basic arithmetic operations: +, -, /, and *

    nix-repl> 1 + 3
    4
    nix-repl> 6 / 3
    2
    

    other operators are ||, &&, ! for booleans

    relational operators such as !=, ==, <, >, <=, >=

    nix has integer (not floating point), string, path, boolean and null simple types. there are lists, sets and functions. these types are enough to build an operating system

    nix is strongly typed, but it's not statically typed. that is, you cannot mix strings and integers, you must first do the conversion

    try to use / closly between two numbers:

    nix-repl> 2/3
    /home/nix/2/3
    

    nix parsed 2/3 as a relative path to the current directory. Paths are parsed as long as there's a slash. Therefore to specify the current directory, use ./

    In addition, nix also parses urls

    Not all urls or paths can be parsed this way. If a syntax error occurs, it's still possible to fallback to plain strings. Parsing urls and paths are convenient for additional safety

    identifiers

    Not much to say, except that dash (-) is allowed in identifiers. That's convenient since many packages use dash in its name. In fact:

    nix-repl> a-b
    error: undefined variable `a-b' at (string):1:1
    nix-repl> a - b
    error: undefined variable `a' at (string):1:1
    

    As you can see, a-b is parsed as identifier, not as operation between a and b.

    strings

    It's important to understand the syntax for strings. When reading nix expressions at the beginning, you may find dollars ($) ambiguous in their usage.

    Strings are enclosed by double quotes ("), or two single quotes ('')

    nix-repl> "foo"
    "foo"
    nix-repl> ''foo''
    "foo"
    

    It's possible to interpolate whole nix expressions inside strings with ${...} and only with ${...}, not $foo or {$foo} or anything else

    nix-repl> foo = "strval"
    nix-repl> "$foo"
    "$foo"
    
    nix-repl> "${foo}"
    "strval"
    
    nix-repl> "${2+3}"
    error: cannot coerce an integer to a string, at (string):1:2
    

    Using the syntax with two single quotes, it's useful for writing double quotes inside strings instead of escaping:

    nix-repl> ''test " test''
    "test \\" test"
    
    nix-repl> ''${foo}''
    "strval"
    

    Escaping ${...} within double quoted strings is done with the backslash. Within two single quotes, it's done with '':

    nix-repl> "\\${foo}"
    "${foo}"
    
    nix-repl> ''test ''${foo} test''
    "test ${foo} test"
    

    No other magic about strings for now

    lists

    Lists are a sequence of expressions delimited by space (not comma):

    nix-repl> [ 2 "foo" true (2+3) ]
    [ 2 "foo" true 5 ]
    
    Lists, like anything else in nix, are immutable. Adding or removing elements from a list is possible, but will return a new list.

    sets

    Sets are an association between a string key and a nix expression. Keys can only be strings. When writing sets you can also use identifiers as keys.

    nix-repl> s = { foo = "bar"; a-b = "baz"; "123" = "num"; }
    
    nix-repl> s
    { 123 = "num"; a-b = "baz"; foo = "bar"; }
    
    you need semicomma (;) after every key-value assignment. do not confuse sets with argument sets used in functions

    To access elements in the set:

    nix-repl> s.a-b
    "baz"
    
    nix-repl> s."123"
    "num"
    
    Yes, you can use strings for non-identifiers to address keys in the set

    You cannot refer inside a set to elements of the same set:

    nix-repl> { a = 3; b = a+4; }
    error: undefined variable `a' at (string):1:10
    
    To do so, use recursive sets:
    nix-repl> rec { a = 3; b = a+4; }
    { a = 3; b = 7; }
    
    the // operator is an operator between two sets. the result is the union of the two sets. in case of conflicts between attribute names, the value on the right set is preferred:
    nix-repl> x = {a = 1; b = 2;}
    nix-repl> x
    { a = 1; b = 2; }
    
    nix-repl> y = {c = 3; d = 4;}
    nix-repl> z = x // y
    nix-repl> z
    { a = 1; b = 2; c = 3; d = 4; }
    

    if

    Expressions, not statements.

    nix-repl> a = 3
    nix-repl> b = 4
    nix-repl> if a > b then "yes" else "no"
    "no"
    
    You can't have only the "then" branch, you must specify also the "else" branch, because an expression must have a value in all cases

    let

    This kind of expression is used to define local variables to inner expressions.

    nix-repl> let a = "foo"; in a
    "foo"
    
    The syntax is: first assign variables, then "in" expression. The overall result will be the final expression after "in".
    nix-repl> let a = 3; b = 4; in a + b
    7
    nix-repl> let a = 3; in let b = 4; in a + b
    7
    
    With let you cannot assign twice to the same variable. You can however shadow outer variables:
    nix-repl> let a = 3; a = 8; in a
    error: attribute `a' at (string):1:12 already defined at (string):1:5
    
    nix-repl> let a = 3; in let a = 8; in a
    8
    
    You cannot refer to variables in a let expression outside of it:
    nix-repl> let a = (let b = 3; in b); in b
    error: undefined variable `b' at (string):1:31
    
    You can refer to variables in the let expression when assigning variables like with recursive sets:
    nix-repl> let a = 4; b = a + 5; in b
    9
    
    So beware when you want to refer to a variable from the outer scope, but it's being defined in the current let expression. Same applies to recursive sets.

    with

    You decide per-expression when to include symbols into the scope.

    nix-repl> longName = { a = 3; b = 4; }
    
    nix-repl> longName.a + longName.b
    7
    nix-repl> with longName; a + b
    7
    
    That's it, it takes a set and includes symbols in the scope of the inner expression

    If a symbol exists in the outer scope and also in the "with" scope, it will not be shadowed. You can however still refer to the set:

    nix-repl> let a = 10; in with longName; a + b
    14
    
    nix-repl> let a = 10; in with longName; longName.a + b
    7
    

    laziness

    nix evaluates expression only when needed. This is a great feature when working with packages

    nix-repl> let a = builtins.div 4 0; b = 6; in b
    6
    
    Since "a" is not needed, there's no error about division by zero, because the expression is not in need to be evaluated. That's why we can have all the packages defined here, yet access to specific packages very fast

    5: functions and imports

    nameless and single parameter

    Functions are anonymous (lambdas), and only have a single parameter. The syntax is extremely simple. Type the parameter name, then ":", then the body of the function.

    nix-repl> x: x*2
    «lambda»

    So here we defined a function that takes a parameter x, and returns x*2. The problem is that we cannot use it in any way, because it's unnamed... joke!

    We can store functions in variables.

    nix-repl> double = x: x * 2
    nix-repl> double
    «lambda»
    nix-repl> double 3
    6

    So, we defined a function x: x*2 that takes one parameter x, and returns x*2. This function is then assigned to the variable double

    Finally we did our first function call: double 3

    In summary: to call a function, name the variable, then space, then the argument. Nothing else to say, it's as easy as that.

    More than one parameter

    How do we create a function that accepts more than one parameter? For people not used to functional programming, this may take a while to grasp. Let's do it step by step.

    nix-repl> mul = a: (b: a * b)

    nix-repl> mul

    «lambda»

    nix-repl> mul 3

    «lambda»

    nix-repl> (mul 3) 4

    12

    We defined a function that takes the parameter "a", the body returns another function. This other function takes a parameter "b" and returns a*b

    Therefore, calling "mul 3" returns this kind of function: b: 3*b. In turn, we call the returned function with 4, and get the expected result

    You don't have to use parenthesis at all, nix has sane priorities when parsing the code:

    nix-repl> mul = a: b: a * b

    nix-repl> mul

    «lambda»

    nix-repl> mul 3

    «lambda»

    nix-repl> mul 3 4

    12

    nix-repl> mul (6 + 7) (8 + 9)

    221

    Since the argument is separated by a space, to pass more complex expressions you need parenthesis. In other common languages you would write mul(6 + 7, 8 + 9)

    Given that functions have only one parameter, it is straightforward to use partial application:

    nix-repl> foo = mul 3

    nix-repl> foo 4

    12

    nix-repl> foo 5

    15

    We stored the function returned by mul 3 into a variable foo, then reused it

    Arguments set

    Now this is a very cool feature of nix. It is possible to pattern match over a set in the parameter. We write an alternative version of mul = a: b: a*b first by using a set as argument, then using pattern matching.

    nix-repl> mul = s: s.a * s.b

    nix-repl> mul { a = 3; b = 4; }

    12

    nix-repl> mul = { a, b }: a * b

    nix-repl> mul { a = 3; b = 4; }

    12

    In the second case we defined an arguments set. It's like defining a set, except without values. We require that the passed set contains the keys "a" and "b". Then we can use those "a" and "b" in the function body directly

    nix-repl> mul { a = 3; b = 4; c = 6; }

    error: anonymous function at (string):1:2 called with unexpected argument `c', at (string):1:1

    nix-repl> mul { a = 3; }

    error: anonymous function at (string):1:2 called without required argument `b', at (string):1:1

    Only a set with exactly the attributes required by the function is accepted, nothing more, nothing less

    Default and variadic attributes

    It is possible to specify default values of attributes in the arguments set:

    nix-repl> mul = { a, b ? 2 }: a * b

    nix-repl> mul { a = 3; }

    6

    nix-repl> mul { a = 3; b = 4; }

    12

    Also you can allow passing more attributes (variadic) than the expected ones:

    nix-repl> mul = { a, b, ... }: a * b

    nix-repl> mul { a = 3; b = 4; c = 2; }

    12

    However, in the function body you cannot access the "c" attribute. The solution is to give a name to the given set with the @-pattern:

    nix-repl> mul = s@{ a, b, ... }: a * b * s.c

    nix-repl> mul { a = 3; b = 4; c = 2; }

    24

    That's it, you give a name to the whole parameter with name@ before the set pattern.

    Advantages of using argument sets:

    Named unordered arguments: you don't have to remember the order of the arguments.

    You can pass sets, that adds a whole new layer of flexibility and convenience.

    Disadvantages:

    Partial application does not work with argument sets.

    Imports

    The "import" function is built-in and provides a way to parse a .nix file. The natural approach is to define each component in a .nix file, then compose by importing these files.

    Let's start with the bare metal.

    a.nix:

    3

    b.nix:

    4

    mul.nix:

    a: b: a * b

    nix-repl> a = import ./a.nix

    nix-repl> b = import ./b.nix

    nix-repl> mul = import ./mul.nix

    nix-repl> mul a b

    12

    Note that the scope of the imported file does not inherit the scope of the importer

    test.nix:

    x

    nix-repl> let x = 5; in import ./test.nix

    error: undefined variable `x' at /home/lethal/test.nix:1:1

    So how do we pass information to the module? Use functions, like we did with mul.nix

    A more complex example:

    test.nix:

    { a, b ? 3, trueMsg ? "yes", falseMsg ? "no" }:

    if a > b

    then builtins.trace trueMsg true

    else builtins.trace falseMsg false

    nix-repl> import ./test.nix { a = 5; trueMsg = "ok"; }

    trace: ok

    true

    In test.nix we return a function. It accepts a set, with default attributes b, trueMsg and falseMsg

    builtins.trace is a built-in function that takes two arguments. The first is the message to display, the second is the value to return. It's usually used for debugging purposes

    Then we import test.nix, and call the function with that set. So when is the message shown? Only when it's in need to be evaluated.


    6: our first derivation

    Derivations are the building blocks of a nix system, from a file system view point. The nix language is used to describe such derivations.

    the derivation function

    The derivation built-in function is used to create derivations.

    A derivation from a nix language view point is simply a set, with some attributes. Therefore you can pass the derivation around with variables like anything else. That's where the real power comes in

    The derivation function receives a set as first argument. This set requires at least the following three attributes:

    name: the name of the derivation.

    In the nix store the format is hash-name, that's the name

    system: is the name of the system in which the derivation can be built

    For example, x86_64-linux

    builder: it is the binary program that builds the derivation

    First of all, what's the name of our system as seen by nix?

    nix-repl> builtins.currentSystem

    "x86_64-linux"

    Let's try to fake the name of the system:

    nix-repl> d = derivation { name = "myname"; builder = "mybuilder"; system = "mysystem"; }

    nix-repl> d

    «derivation /nix/store/z3hhlxbckx4g3n9sw91nnvlkjvyw754p-myname.drv»

    Oh oh, what's that? Did it build the derivation?

    No it didn't, but it did create the .drv file

    nix repl does not build derivations unless you tell to do so

    digression about .drv files

    What's that .drv file? It is the specification of how to build the derivation

    Before continuing, some analogies with the C language:

    .nix files are like .c files

    .drv files are like .o files

    the .drv describes how to build a derivation, it's the bare minimum information. out paths are then the product of the build. both drv paths and out paths are stored in the nix store as you can see

    what's in that .drv file? You can read it:

    $ cat /nix/store/*-myname.drv

    Derive(

    [("out", "/nix/store/40s0qmrfb45vlh6610rk29ym318dswdr-myname", "", "")]

    , []

    , []

    , "mysystem"

    , "mybuilder"

    , []

    , [ ("builder", "mybuilder")

    , ("name", "myname")

    , ("out", "/nix/store/40s0qmrfb45vlh6610rk29ym318dswdr-myname")

    , ("system", "mysystem")

    ]

    Ok we can see there's an out path, but it does not exist yet. We never told nix to build it, but we know beforehand where the build output will be. Why? nix let us know the path beforehand and keep evaluating the nix expressions, but it's still empty because no build was ever made

    Important: the hash of the out path is based solely on the input derivations in the current version of nix, not on the contents of the build product (It's possible however to have content-addressable derivations for e.g. tarballs)

    Many things are empty in that .drv, however I write a summary of the .drv format for you:

    - the output paths (they can be multiple ones). By default nix creates one out path called "out"

    - the list of input derivations. It's empty because we are not referring to any other derivation. Otherwise, there would a list of other .drv files

    - the system and the builder executable (yes, it's a fake one)

    - then a list of environment variables passed to the builder

    That's it, the minimum necessary information to build our derivation.

    Important note: the environment variables passed to the builder are just those you see in the .drv plus some other nix related configuration (number of cores, temp dir, ...). The builder will not inherit any variable from your running shell, otherwise builds would suffer from non-determinism

    Back to our fake derivation. Let's build our really fake derivation:

    nix-repl> d = derivation { name = "myname"; builder = "mybuilder"; system = "mysystem"; }

    nix-repl> :b d

    [...]

    these derivations will be built:

    /nix/store/z3hhlxbckx4g3n9sw91nnvlkjvyw754p-myname.drv

    building path(s) `/nix/store/40s0qmrfb45vlh6610rk29ym318dswdr-myname'

    error: a `mysystem' is required to build `/nix/store/z3hhlxbckx4g3n9sw91nnvlkjvyw754p-myname.drv', but I am a `x86_64-linux'

    the :b is a nix repl specific command to build a derivation. You can see more commands with :?

    So in the output you can see that it takes the .drv as information on how to build the derivation. Then it says it's trying to produce our out path. Finally the error: that derivation can't be built on our system

    We're doing the build inside nix repl, but what if we don't want to use nix repl? You can realise a .drv with:

    $ nix-store -r /nix/store/z3hhlxbckx4g3n9sw91nnvlkjvyw754p-myname.drv

    You will get the same output as before

    Let's fix the system attribute:

    nix-repl> d = derivation { name = "myname"; builder = "mybuilder"; system = builtins.currentSystem; }

    nix-repl> :b d

    [...]

    build error: invalid file name `mybuilder'

    A step forward: of course, that "mybuilder" executable does not really exist

    What's in a derivation set

    First of all, the returned value is a plain set:

    nix-repl> d = derivation { name = "myname"; builder = "mybuilder"; system = "mysystem"; }

    nix-repl> builtins.isAttrs d

    true

    You can guess what builtins.isAttrs does, it returns true if the argument is a set

    nix-repl> builtins.attrNames d

    [ "all" "builder" "drvAttrs" "drvPath" "name" "out" "outPath" "outputName" "system" "type" ]

    builtins.attrNames returns a list of keys of the given set

    Start from drvAttrs:

    nix-repl> d.drvAttrs

    { builder = "mybuilder"; name = "myname"; system = "mysystem"; }

    That's basically the input we gave to the derivation function. Also d.name, d.system and d.builder attributes are straight the ones we gave as input

    nix-repl> (d == d.out)

    true

    So out is just the derivation itself, it seems weird but the reason is that we only have one output from the derivation. That's also the reason why d.all is a singleton

    The d.drvPath is the path of the .drv file: /nix/store/z3hhlxbckx4g3n9sw91nnvlkjvyw754p-myname.drv .

    Something interesting is the type attribute: "derivation". nix does add a little of magic to sets with type 'derivation', but not that much. To let you understand, you can create yourself a set with that type, it's a simple set:

    nix-repl> { type = "derivation"; }

    «derivation ???»

    Of course it has no other information, so nix doesn't know what to say :-) But you get it, the type = "derivation" is just a convention for nix and for us to understand the set is a derivation

    When writing packages, we are interested in the outputs. The other metadata is needed for nix to know how to create the drv path and the out path. The outPath attribute is the build path in the nix store:

    /nix/store/40s0qmrfb45vlh6610rk29ym318dswdr-myname

    Referring to other derivations

    How do we refer to other packages? How do we refer to other derivations in terms of files on the disk?

    We use the 'outPath'. The 'outPath' tells where the files are of that derivation. To make it more convenient, nix is able to do a conversion from a derivation set to a string

    nix-repl> d.outPath

    "/nix/store/40s0qmrfb45vlh6610rk29ym318dswdr-myname"

    nix-repl> builtins.toString d

    "/nix/store/40s0qmrfb45vlh6610rk29ym318dswdr-myname"

    nix does the "set to string conversion" as long as there is the 'outPath' attribute (much like a toString method in other languages):

    nix-repl> builtins.toString { outPath = "foo"; }

    "foo"

    nix-repl> builtins.toString { a = "b"; }

    error: cannot coerce a set to a string, at (string):1:1

    Say we want to use binaries from coreutils (ignore the nixpkgs etc.):

    nix-repl> :l

    Added 3950 variables.

    nix-repl> coreutils

    «derivation /nix/store/1zcs1y4n27lqs0gw4v038i303pb89rw6-coreutils-8.21.drv»

    nix-repl> builtins.toString coreutils

    "/nix/store/8w4cbiy7wqvaqsnsnb3zvabq1cp2zhyz-coreutils-8.21"

    Apart the nixpkgs stuff, just think we added to the scope a series of variables. One of them is 'coreutils'. It is the derivation of the coreutils package you all know of from other Linux distributions. It contains basic binaries for GNU/Linux systems:

    $ ls /nix/store/*coreutils*/bin

    [......]

    I remind you, inside strings it's possible to interpolate nix expressions with ${...}:

    nix-repl> "${d}"

    "/nix/store/40s0qmrfb45vlh6610rk29ym318dswdr-myname"

    nix-repl> "${coreutils}"

    "/nix/store/8w4cbiy7wqvaqsnsnb3zvabq1cp2zhyz-coreutils-8.21"

    That's very convenient, because then we could refer to e.g. the bin/true binary like this:

    nix-repl> "${coreutils}/bin/true"

    "/nix/store/8w4cbiy7wqvaqsnsnb3zvabq1cp2zhyz-coreutils-8.21/bin/true"

    In the previous attempt we used a fake builder, "mybuilder" which obviously does not exist. But we can use for example bin/true, which always exits with 0 (success)

    nix-repl> :l

    nix-repl> d = derivation { name = "myname"; builder = "${coreutils}/bin/true"; system = builtins.currentSystem; }

    nix-repl> :b d

    ...

    builder for `/nix/store/d4xczdij7xazjfm5kn4nmphx63mpv676-myname.drv' failed to produce output path `/nix/store/fy5lyr5iysn4ayyxvpnsya8r5y5bwjnl-myname'

    Another step forward, it executed the builder (bin/true), but the builder did not create the out path of course, it just exited with 0

    Obvious note: everytime we change the derivation, a new hash is created

    Let's examine the new .drv now that we referred to another derivation:

    $ cat /nix/store/d4xczdij7xazjfm5kn4nmphx63mpv676-myname.drv

    Derive(

    [("out", "/nix/store/fy5lyr5iysn4ayyxvpnsya8r5y5bwjnl-myname", "", "")]

    , [("/nix/store/1zcs1y4n27lqs0gw4v038i303pb89rw6-coreutils-8.21.drv", ["out"])]

    , []

    , "x86_64-linux"

    , "/nix/store/8w4cbiy7wqvaqsnsnb3zvabq1cp2zhyz-coreutils-8.21/bin/true"

    , []

    , [ ("builder", "/nix/store/8w4cbiy7wqvaqsnsnb3zvabq1cp2zhyz-coreutils-8.21/bin/true")

    , ("name", "myname")

    , ("out", "/nix/store/fy5lyr5iysn4ayyxvpnsya8r5y5bwjnl-myname")

    , ("system", "x86_64-linux")

    ]

    Aha! nix added a dependency to our myname.drv, it's the coreutils.drv. Before doing our build, nix should build the coreutils.drv. But since coreutils is already in our nix store, no build is needed, it's already there with out path /nix/store/8w4cbiy7wqvaqsnsnb3zvabq1cp2zhyz-coreutils-8.21

    When is the derivation built

    nix does not build derivations during evaluation of nix expressions. In fact, that's why we have to do

    :b drv

    in nix repl, or use nix-store -r in the first place. An important separation is made in nix:

    Instantiate/Evaluation time:

    the nix expression is parsed, interpreted and finally returns a derivation set

    during evaluation, you can refer to other derivations because nix will create .drv files and

    we will know out paths beforehand. This is achieved with nix-instantiate

    Realise/Build time:

    the .drv from the derivation set is built,

    first building .drv inputs (build dependencies)

    this is achieved with nix-store -r

    Think of it as of compile time and link time like with C/C++ projects. You first compile all source files to object files. Then link object files in a single executable

    In nix, first the nix expression (usually in a .nix file) is compiled to .drv, then each .drv is built and the product is installed in the relative out paths

    With the derivation function we provide a set of information on how to build a package, and we get back the information about where the package was built

    nix converts a set to a string when there's an outPath, that's very convenient. With that, it's easy to refer to other derivations

    When nix builds a derivation, it first creates a .drv file from a derivation expression, and uses it to build the output. It does so recursively for all the dependencies (inputs). It "executes" the .drv files like a machine. Not much magic after all


    7: a working derivation

    --- make all exercises as root ------

    using a script as builder

    What's the easiest way to run a sequence of commands for building something? A bash script. We write a custom bash script, and we want it to be our builder

    Given a builder.sh, we want the derivation to run 'bash builder.sh'

    We don't use hash bangs in builder.sh, because at the time we are writing builder.sh we do not know the path to bash in the nix store

    We don't even use /usr/bin/env, because then we lose the cool stateless property of nix. Not to say PATH gets cleared when building therefore it wouldn't work anyway.

    In summary: we want the builder to be bash, and pass it an argument, builder.sh

    Turns out the derivation function accepts an optional args attribute that is exactly used to pass arguments to the builder executable

    First of all, let's write our 'builder.sh' in the current directory:

    declare -xp

    echo foo > $out

    nix creates the out path of the derivation. In the .drv there's a list of environment variables passed to the builder. One of them is $out

    What we have to do is to create something in $out, be it a file or a directory (we are creating a file)

    In addition, we also debug ('declare -xp') the environment variables during the build process. We cannot use 'env', because env is part of coreutils and we don't have a dependency to it. Not yet. It's plain bash, only bash. we get a blessed bash for free from our magic nixpkgs stuff:

    nix-repl> :l

    Added 3950 variables.

    nix-repl> "${bash}"

    "/nix/store/ihmkc7z2wqk3bbipfnlh0yjrlfkkgnv6-bash-4.2-p45"

    Great, with the usual trick we can then refer to bin/bash and create our derivation:

    nix-repl> d = derivation { name = "foo"; builder = "${bash}/bin/bash"; args = [ ./builder.sh ]; system = builtins.currentSystem; }

    nix-repl> :b d

    these derivations will be built:

    /nix/store/ybnysdh5k6cjznhg4afjgbhr6czbwb4s-foo.drv

    building path(s) `/nix/store/72v14vk4li47n8sx3z2ibd802ihpqyvx-foo'

    these derivations will be built:

    /nix/store/ibwr68l3rjlx02kgz61dkkkrlpgljfxd-simple.drv

    [...]

    this derivation produced the following outputs:

    out -> /nix/store/w024zci0x1hh1wj6gjq0jagkc1sgrf5r-foo

    What? We did it! The contents of /nix/store/w024zci0x1hh1wj6gjq0jagkc1sgrf5r-foo is really 'foo'. We built our first derivation

    Note: we used ./builder.sh, not "./builder.sh". This way it gets parsed as path. Try using the string version, it will say it cannot find ./builder.sh , because that would be relative to the temporary build directory

    the builder environment

    Let's inspect those debugged environment variables during the build process.

    $HOME is not your home, and /homeless-shelter doesn't exist at all. We force packages to not depend upon $HOME during the build process

    $PATH plays the same game of $HOME

    $NIX_BUILD_CORES and $NIX_STORE are nix configurations

    $PWD and $TMP clearly shows nix created a temporary build directory

    Then builder, name, out and system are variables set due to the .drv contents

    And that's how we used the $out variable in our derivation, put stuff inside it. It's like nix reserved a slot in the nix store for us, and we must fill it. In terms of autotools, that will be the --prefix path. That's a big difference between nix and other package managers. That's the essence of stateless packaging. You don't install the package in a global common path under /, you install it in a local isolated path under your nix store slot

    the .drv contents

    We added something else this time to the derivation. The args attribute. Let's see how this changed the .drv compared to the previous pill:

    $ cat /nix/store/g6jj1mjzq68i66rbqyb3gpx3k0x606af-foo.drv

    Derive(

    [("out", "/nix/store/w024zci0x1hh1wj6gjq0jagkc1sgrf5r-foo", "", "")]

    , [("/nix/store/jdggv3q1sb15140qdx0apvyrps41m4lr-bash-4.2-p45.drv", ["out"])]

    , ["/nix/store/5d1i99yd1fy4wkyx85iz5bvh78j2j96r-builder.sh"]

    , "x86_64-linux"

    , "/nix/store/ihmkc7z2wqk3bbipfnlh0yjrlfkkgnv6-bash-4.2-p45/bin/bash"

    , ["/nix/store/5d1i99yd1fy4wkyx85iz5bvh78j2j96r-builder.sh"]

    , [ ("builder", "/nix/store/ihmkc7z2wqk3bbipfnlh0yjrlfkkgnv6-bash-4.2-p45/bin/bash")

    , ("name", "foo")

    , ("out", "/nix/store/w024zci0x1hh1wj6gjq0jagkc1sgrf5r-foo")

    , ("system", "x86_64-linux")

    ]

    )

    Perfect, much like the usual .drv, except there's a list of arguments in there passed to the builder (bash), with the builder.sh... builder.sh... what? It's not pointing to my home's builder.sh

    nix automatically copies files or directories needed for the build in the nix store, to ensure, for example, that they do not get changed during the build process. Also to ensure the deployment to be stateless and independent of the building machine

    Not only builder.sh is in the arguments passed to the builder, it's also in the input derivations

    Being builder.sh a plain file, it has no .drv associated with it. The store path will be computed based on the hash of its contents, and the name itself

    Packaging a simple C executable

    Start off writing a simple.c file:

    void main () {

    puts ("Simple!");

    }

    And its simple_builder.sh:

    export PATH="$coreutils/bin:$gcc/bin"

    mkdir $out

    gcc -o $out/simple $src

    Don't spend time understanding where those variables come from. Let's write the derivation and build it:

    nix-repl> :l

    nix-repl> simple = derivation { name = "simple"; builder = "${bash}/bin/bash"; args = [ ./simple_builder.sh ]; gcc = gcc; coreutils = coreutils; src = ./simple.c; system = builtins.currentSystem; }

    nix-repl> :b simple

    this derivation produced the following outputs:

    out -> /nix/store/ni66p4jfqksbmsl616llx3fbs1d232d4-simple

    Perfect, now you can run /nix/store/ni66p4jfqksbmsl616llx3fbs1d232d4-simple/simple in your shell

    We added two new attributes to the derivation call, "gcc" and "coreutils" - please, don't get an headache by reading "gcc = gcc". on the left, it's the attribute name of the set, on the right, there's an expression, it's the gcc derivation. same goes for coreutils

    We also added the "src" attribute, nothing magic it's just a name with the ./simple.c path. Like for simple_builder.sh, simple.c will be added to the store

    The trick: every attribute in the set

    - will be converted to a string and

    - PASSED AS ENIRONMENT VARIABLE to the builder

    Now it's all clear. $coreutils and $gcc are then the out paths of the derivations, and of course appending "/bin" will point to their binaries. Same goes for the src variable, $src is the path to simple.c in the nix store

    In simple_builder.sh we set the PATH for gcc and coreutils binaries, so that gcc can find the necessary binaries like "cat", "readlink", etc

    Then we create $out as a directory and inside it we put the binary

    Note: instead of running plain gcc (or mkdir), it would have been equivalent to run $gcc/bin/gcc (or $coreutils/bin/mkdir)

    Enough with nix repl

    Drop out of nix repl, write a simple.nix file:

    with (import {});

    derivation {

    name = "simple";

    builder = "${bash}/bin/bash";

    args = [ ./simple_builder.sh ];

    inherit gcc coreutils;

    src = ./simple.c;

    system = builtins.currentSystem;

    }

    Now you can build it with

    $> nix-build simple.nix

    It will create a symlink "result" in the current directory, pointing to the out path of the derivation

    The nix-build tool does two main jobs:

    nix-instantiate: parse simple.nix and return the .drv file

    nix-store -r: realise the .drv, which actually builds the derivation

    Finally creates the symlink

    Look at the first line of the .nix file. We have an "import" function call nested in a 'with' expression. I recall import accepts one argument, a nix file to parse. In this case it parsed a function out of the file.

    Afterwards we call the parsed function with the empty set

    Let me underline it: "import {}" are two function calls, not one. Read it like "(import ) {}"

    The final returned value of that import is a set. To simplify it: it's a set of derivations. Using the "with" expression we drop them into the scope. We basically simulated what :l does in nix repl, so we can easily access derivations such as bash, gcc and coreutils

    Then we meet the 'inherit' keyword. Doing 'inherit foo', is the same as doing foo = foo. Doing 'inherit foo bar', is the same as doing foo = foo; bar = bar. Literally. This syntax only makes sense inside sets. Don't think it's black magic, it's just a convenience to avoid repeating the same name twice, once for the attribute name, once for the variable in the scope

    ===== 8: generic builders

    In the previous pill we packaged a simple .c file, which was being compiled with a raw gcc call. That's not a good example of project. Many use autotools, and since we're going to generalize our builder, better do it with the most used build system.

    GNU hello world, despite its name, is a simple yet complete project using autotools. Fetch the latest tarball here:

    http://ftp.gnu.org/gnu/hello/hello-2.9.tar.gz

    Let's create a builder script for GNU hello world, 'hello_builder.sh':

    export PATH="$gnutar/bin:$gcc/bin:$gnumake/bin:$coreutils/bin:$gawk/bin:$gzip/bin:$gnugrep/bin:$gnused/bin:$binutils/bin"

    tar -xzf $src

    cd hello-2.9

    ./configure --prefix=$out

    make

    make install

    and the derivation 'hello.nix':

    with (import {});

    derivation {

    name = "hello";

    builder = "${bash}/bin/bash";

    args = [ ./hello_builder.sh ];

    inherit gnutar gzip gnumake gcc binutils coreutils gawk gnused gnugrep;

    src = ./hello-2.9.tar.gz;

    system = builtins.currentSystem;

    }

    Now build it with

    $> nix-build hello.nix

    and you can launch

    $> result/bin/hello

    Nothing easier, but do we have to create a builder.sh for each package? Do we always have to pass the dependencies to the derivation function?

    Please note the --prefix=$out

    a generic builder

    let's a create a generic builder.sh for autotools projects:

    set -e

    unset PATH

    for p in $buildInputs; do

    export PATH=$p/bin${PATH:+:}$PATH

    done

    tar -xf $src

    for d in *; do

    if [ -d "$d" ]; then

    cd "$d"

    break

    fi

    done

    ./configure --prefix=$out

    make

    make install

    What do we do here?

    -- exit the build on any error with set -e

    -- unset PATH, because it's initially set to a non-existant path

    -- for each path in $buildInputs, we append bin to PATH

    -- unpack the source.

    -- find a directory where the source has been unpacked and cd into it

    -- compile and install

    As you can see, there's no reference to "hello" in the builder anymore. It still does several assumptions, but it's certainly more generic

    Now let's rewrite hello.nix:

    with (import {});

    derivation {

    name = "hello";

    builder = "${bash}/bin/bash";

    args = [ ./builder.sh ];

    buildInputs = [ gnutar gzip gnumake gcc binutils coreutils gawk gnused gnugrep ];

    src = ./hello-2.9.tar.gz;

    system = builtins.currentSystem;

    }

    All clear, except that buildInputs. nix is able to convert a list to a string. It first converts the elements to strings, and then concatenates them separated by a space:

    nix-repl> builtins.toString 123

    "123"

    nix-repl> builtins.toString [ 123 456 ]

    "123 456"

    Recall that derivations can be converted to a string, hence:

    nix-repl> :l

    Added 3950 variables.

    nix-repl> builtins.toString gnugrep

    "/nix/store/g5gdylclfh6d224kqh9sja290pk186xd-gnugrep-2.14"

    nix-repl> builtins.toString [ gnugrep gnused ]

    "/nix/store/g5gdylclfh6d224kqh9sja290pk186xd-gnugrep-2.14 /nix/store/krgdc4sknzpw8iyk9p20lhqfd52kjmg0-gnused-4.2.2"

    simple! the 'buildInputs' variable is a string with out paths separated by space, perfect for bash usage in a for loop

    a more convenient derivation function

    We managed to write a builder that can be used for multiple autotools projects. But in the hello.nix expression we are specifying tools that are common to more projects; we don't want to pass them everytime.

    A natural approach would be to create a function that accepts an attribute set and merge it with another attribute set containing values common to many projects

    Create autotools.nix:

    pkgs: attrs:

    with pkgs;

    let defaultAttrs = {

    builder = "${bash}/bin/bash";

    args = [ ./builder.sh ];

    baseInputs = [ gnutar gzip gnumake gcc binutils coreutils gawk gnused gnugrep ];

    buildInputs = [];

    system = builtins.currentSystem;

    };

    in

    derivation (defaultAttrs // attrs)

    The whole nix expression of this autotools.nix file will evaluate to a function. This function accepts a parameter pkgs, then returns a function which accepts a parameter attrs

    The body of the function is simple:

    -- drop in the scope the magic pkgs attribute set

    -- Within a 'let' expression we define an helper variable, 'defaultAttrs',

    which serves as a set of common attributes used in derivations

    -- we create the derivation with that strange expression

    The // operator is an operator between two sets. The result is the union of the two sets. In case of conflicts between attribute names, the value on the right set is preferred. So we use 'defaultAttrs' as base set, and add (or override) the attributes from 'attrs'

    A couple of examples ought to be enough to clear out the behavior of the operator:

    nix-repl> { a = "b"; } // { c = "d"; }

    { a = "b"; c = "d"; }

    nix-repl> { a = "b"; } // { a = "c"; }

    { a = "c"; }

    Complete the new builder.sh by adding $baseInputs in the for loop together with $buildInputs. As you noticed, we passed that new variable in the derivation. Instead of merging buildInputs with the base ones, we prefer to preserve buildInputs as seen by the caller, so we keep them separated. Just a matter of choice

    Then we rewrite hello.nix as follows:

    let

    pkgs = import {};

    mkDerivation = import ./autotools.nix pkgs;

    in mkDerivation {

    name = "hello";

    src = ./hello-2.9.tar.gz;

    }

    Finally! We got a very simple description of a package!

    We assigned to pkgs the import that we did in the previous expressions in the "with"

    The 'mkDerivation' variable is a nice example of partial application, look at it as (import ./autotools.nix) pkgs. First we import the expression, then we apply the pkgs parameter. That will give us a function that accepts the attribute set attrs

    We create the derivation specifying only name and src. If the project eventually needed other dependencies to be in PATH, then we would simply add those to buildInputs (not specified in hello.nix because empty)

    Note: we didn't use any other library - special C flags may be needed to find include files of other libraries at compile time, and ld flags at link time

    Conclusion

    We are feeling the way a nix system grows up: it's about creating and composing derivations with the nix language.

    Analogy:

    in C you create objects in the heap, and then you compose them inside new objects. Pointers are used to refer to other objects.

    In nix you create derivations stored in the nix store, and then you compose them by creating new derivations. Store paths are used to refer to other derivations. ===== 9: automatic runtime dependencies

    Build dependencies

    Let's start analyzing build dependencies for our GNU hello world package:

    $ nix-instantiate hello.nix

    /nix/store/z77vn965a59irqnrrjvbspiyl2rph0jp-hello.drv

    $ nix-store -q --references /nix/store/z77vn965a59irqnrrjvbspiyl2rph0jp-hello.drv

    /nix/store/0q6pfasdma4as22kyaknk4kwx4h58480-hello-2.9.tar.gz

    /nix/store/1zcs1y4n27lqs0gw4v038i303pb89rw6-coreutils-8.21.drv

    /nix/store/2h4b30hlfw4fhqx10wwi71mpim4wr877-gnused-4.2.2.drv

    /nix/store/39bgdjissw9gyi4y5j9wanf4dbjpbl07-gnutar-1.27.1.drv

    /nix/store/7qa70nay0if4x291rsjr7h9lfl6pl7b1-builder.sh

    /nix/store/g6a0shr58qvx2vi6815acgp9lnfh9yy8-gnugrep-2.14.drv

    /nix/store/jdggv3q1sb15140qdx0apvyrps41m4lr-bash-4.2-p45.drv

    /nix/store/pglhiyp1zdbmax4cglkpz98nspfgbnwr-gnumake-3.82.drv

    /nix/store/q9l257jn9lndbi3r9ksnvf4dr8cwxzk7-gawk-4.1.0.drv

    /nix/store/rgyrqxz1ilv90r01zxl0sq5nq0cq7v3v-binutils-2.23.1.drv

    /nix/store/qzxhby795niy6wlagfpbja27dgsz43xk-gcc-wrapper-4.8.3.drv

    /nix/store/sk590g7fv53m3zp0ycnxsc41snc2kdhp-gzip-1.6.drv

    It has exactly the derivations referenced in the derivation function, nothing more, nothing less. Some of them might not be used at all, however given that our generic 'mkDerivation' function always pulls such dependencies for every package you build from now on, you will have these packages in the nix store

    Why are we looking at .drv files? Because the hello.drv file is the representation of the build action to perform in order to build the 'hello' out path, and as such it also contains the input derivations needed to be built before building 'hello'

    about NAR files

    NAR is the nix ARchive. First question: why not tar? Because commonly used archivers are not deterministic. They add padding, they do not sort files, they add timestamps, etc..

    Hence NAR, a very simple deterministic archive format being used by nix for deployment. NARs are also used extensively within nix itself

    To create NAR archives, it's possible to use

    $> nix-store --dump

    and

    $> nix-store --restore

    Those two commands work regardless of /nix/store

    Runtime dependencies

    Something is different for runtime dependencies however. Build dependencies are automatically recognized by nix once they are used in any derivation call, but we never specify what are the runtime dependencies for a derivation

    There's really black magic involved. It's something that at first glance makes you think "no, this can't work in the long term", but at the same it works so well that a whole operating system is built on top of this magic.

    In other words, nix automatically computes all the runtime dependencies of a derivation, and it's possible thanks to the hash of the store paths.

    Steps:

    Dump the derivation as NAR, a serialization of the derivation output. Works fine whether it's a single file or a directory.

    For each build dependency .drv and its relative out path, search the contents of the NAR for this out path.

    If found, then it's a runtime dependency.

    You get really all the runtime dependencies, and that's why nix deployments are so easy

    $ nix-instantiate hello.nix

    /nix/store/z77vn965a59irqnrrjvbspiyl2rph0jp-hello.drv

    $ nix-store -r /nix/store/z77vn965a59irqnrrjvbspiyl2rph0jp-hello.drv

    /nix/store/a42k52zwv6idmf50r9lps1nzwq9khvpf-hello

    $ nix-store -q --references /nix/store/a42k52zwv6idmf50r9lps1nzwq9khvpf-hello

    /nix/store/94n64qy99ja0vgbkf675nyk39g9b978n-glibc-2.19

    /nix/store/8jm0wksask7cpf85miyakihyfch1y21q-gcc-4.8.3

    /nix/store/a42k52zwv6idmf50r9lps1nzwq9khvpf-hello

    Ok glibc and gcc. Well, gcc really should not be a runtime dependency!

    $ strings result/bin/hello | grep gcc

    /nix/store/94n64qy99ja0vgbkf675nyk39g9b978n-glibc-2.19/lib:/nix/store/8jm0wksask7cpf85miyakihyfch1y21q-gcc-4.8.3/lib64

    Oh nix added gcc because its out path is mentioned in the "hello" binary. Why is that? That's the ld rpath. It's the list of directories where libraries can be found at runtime. In other distributions, this is usually not abused. But in nix, we have to refer to particular versions of libraries, thus the rpath has an important role

    The build process adds that gcc lib path thinking it may be useful at runtime, but really it's not. How do we get rid of it? nix authors have written another magical tool called patchelf, which is able to reduce the rpath to the paths that are really used by the binary

    Not only, even after reducing the rpath the hello binary would still depend upon gcc. Because of debugging information. For that, the well known strip can be used

    Another phase in the builder

    The builder has these phases already:

    -- First the environment is set up

    -- Unpack phase: we unpack the sources in the current directory

    (remember, nix changes dir to a temporary directory first)

    -- Change source root to the directory that has been unpacked

    -- Configure phase: ./configure

    -- Build phase: make

    -- Install phase: make install

    We add a new phase after the installation phase, which we call fixup phase

    At the end of the builder.sh follows:

    find $out -type f -exec patchelf --shrink-rpath '{}' \\; -exec strip '{}' \\; 2>/dev/null

    That is, for each file we run patchelf --shrink-rpath and strip. Note that we used two new commands here, find and patchelf. These two deserve a place in baseInputs of autotools.nix as findutils and patchelf.

    Rebuild hello.nix and...:

    $ nix-build hello.nix

    [...]

    $ nix-store -q --references result

    /nix/store/94n64qy99ja0vgbkf675nyk39g9b978n-glibc-2.19

    /nix/store/md4a3zv0ipqzsybhjb8ndjhhga1dj88x-hello

    ...only glibc is the runtime dependency. Exactly what we wanted.

    The package is self-contained, copy its closure on another machine and you will be able to run it. I remind you the very few components under the /nix/store necessary to run nix when we installed it. The hello binary will use that exact version of glibc library and interpreter, not the system one:

    $ ldd result/bin/hello

    linux-vdso.so.1 (0x00007fff11294000)

    libc.so.6 => /nix/store/94n64qy99ja0vgbkf675nyk39g9b978n-glibc-2.19/lib/libc.so.6 (0x00007f7ab7362000)

    /nix/store/94n64qy99ja0vgbkf675nyk39g9b978n-glibc-2.19/lib/ld-linux-x86-64.so.2 (0x00007f7ab770f000)

    Of course, the executable runs fine as long as everything is under the /nix/store path

    Conclusion

    nix is able to compute all runtime dependencies automatically for us. Not only shared libraries, but also referenced executables, scripts, Python libraries, etc

    This makes packages self-contained, because we're sure (apart data and configuration) that copying the runtime closure on another machine is sufficient to run the program

    ===== 10: developing with nix-shell

    the nix-shell tool drops us in a shell by setting up the necessary environment variables to hack a derivation

    it does not build the derivation, it only serves as a preparation

    I remind you, in a nix environment you don't have access to libraries and programs unless you install them with nix-env. however installing libraries with nix-env is not good practice. we prefer to have isolated environments for development

    $ nix-shell hello.nix

    [nix-shell]$ make

    bash: make: command not found

    [nix-shell]$ echo $baseInputs

    /nix/store/jff4a6zqi0yrladx3kwy4v6844s3swpc-gnutar-1.27.1 [...]

    we call nix-shell on a nix expression which returns a derivation. we then enter a new bash shell, but it's really useless: we expected to have the GNU hello world build inputs available in PATH, including GNU make, but it's not the case

    but, we have the environment variables that we set in the derivation, like $baseInputs, $buildInputs, $src and so on

    that means we can source our builder.sh, and it will build the derivation. you may get an error in the installation phase, because the user may not have the permission to write to /nix/store:

    [nix-shell]$ source builder.sh

    ...

    it didn't install, but it built. things to notice:

    we sourced builder.sh, therefore it ran all the steps including setting up the PATH for us

    the working directory is no more a temp directory created by nix-build, but the current directory. therefore, hello-2.9 has been unpacked there

    we're able to cd into hello-2.9 and type make, because now it's available

    in other words, nix-shell drops us in a shell with the same (or almost) environment used to run the builder!

    a builder for nix-shell

    we can improve our builder to be more nix-shell friendly

    we were able to source builder.sh because it was in our current directory, but that's not nice. we want the builder.sh that is stored in the nix store, the one that would be used by nix-build. to do so, the right way is to pass the usual environment variable through the derivation

    Note: $builder is already defined, but it's the bash executable, not our builder.sh, our builder.sh is an argument to bash

    we don't want to run the whole builder, we only want it to setup the necessary environment for manually building the project. So we'll write two files, one for setting up the environment, and the real builder.sh that runs with nix-build

    Additionally, we'll wrap the phases in functions, it may be useful, and move the set -e to the builder instead of the setup. The set -e is annoying in nix-shell

    The codebase is becoming a little long

    Noteworthy is the setup = ./setup.sh; attribute in the derivation, which adds setup.sh to the nix store and as usual, adds a $setup environment variable in the builder. Thanks to that, we can split builder.sh into setup.sh and builder.sh. what builder.sh does is sourcing $setup and calling the genericBuild function. everything else is just some bash changes

    Now back to nix-shell:

    $ nix-shell hello.nix

    [nix-shell]$ source $setup

    [nix-shell]$

    Now you can run, for example, unpackPhase which unpacks $src and enters the directory. And you can run commands like ./configure, make etc. manually, or run phases with their respective functions.

    It's all that straight, nix-shell builds the .drv file and its input dependencies, then drops into a shell by setting up the environment variables necessary to build the .drv, in particular those passed to the derivation function.

    Conclusion

    with nix-shell we're able to drop into an isolated environment for developing a project, with the necessary dependencies just like nix-build does, except we can build and debug the project manually, step by step like you would do in any other operating system. note that we did never install gcc, make, etc. system-wide. these tools and libraries are available per-build

    ===== 11: the garbage collector

    When using nix tools, often derivations are built. This include both .drv files and out paths. These artifacts go in the nix store, and we never cared about deleting them until now.

    How does it work

    Other package managers, like dpkg, have somehow a way to remove unused software. However, nix is much more precise compared to other systems.

    How do we determine whether a store path is still needed? The same way programming languages with a garbage collector decide whether an object is still alive. Programming languages with a garbage collector have an important concept in order to keep track of live objects: GC roots. A GC root is an object that is always alive. All objects recursively referred to by a GC root are live. Therefore, the garbage collection process starts from GC roots, and recursively mark referenced objects as live. All other objects can be collected and deleted.

    In nix there's this same concept. Instead of being objects, of course, GC roots are store paths. The implementation is very simple and transparent to the user. GC roots are stored under /nix/var/nix/gcroots. If there's a symlink to a store path, then that store path is a GC root. nix allows this directory to have subdirectories: it will simply recurse directories in search of symlinks to store paths.

    So we have a list of GC roots. At this point, deleting dead store paths is as easy as you can imagine. We have the list of all live store paths, hence the rest of the store paths are dead. In particular, nix first moves dead store paths to /nix/store/trash which is an atomic operation. Afterwards, the trash is emptied.

    Playing with the GC

    Before playing with the GC, first run the nix garbage collector once, so that we have a cleaned up playground for our experiments:

    $ nix-collect-garbage

    finding garbage collector roots...

    [...]

    deleting unused links...

    note: currently hard linking saves -0.00 MiB

    1169 store paths deleted, 228.43 MiB freed

    Perfect, if you run it again it won't find anything new to delete, as expected. What's left in the nix store is everything being referenced from the GC roots. Let's install for a moment bsd-games:

    $ nix-env -iA nixpkgs.bsdgames

    $ readlink -f `which fortune`

    /nix/store/b3lxx3d3ggxcggvjw5n0m1ya1gcrmbyn-bsd-games-2.17/bin/fortune

    $ nix-store -q --roots `which fortune`

    /nix/var/nix/profiles/default-9-link

    $ nix-env --list-generations

    [...]

    9 2014-08-20 12:44:14 (current)

    The nix-store command can be used to query the GC roots that refer to a given derivation. In this case, our current user environment does refer to bsd-games. Now remove it, collect garbage and note that bsd-games is still in the nix store:

    $ nix-env -e bsd-games

    uninstalling `bsd-games-2.17'

    $ nix-collect-garbage

    [...]

    $ ls /nix/store/b3lxx3d3ggxcggvjw5n0m1ya1gcrmbyn-bsd-games-2.17

    bin share

    That's because the old generation is still in the nix store because it's a GC root. As we'll see below, all profiles and their generations are GC roots. Removing a GC root is simple. Let's try deleting the generation that refers to bsd-games, collect garbage, and note that now bsd-games is no longer in the nix store:

    $ rm /nix/var/nix/profiles/default-9-link

    $ nix-env --list-generations

    [...]

    8 2014-07-28 10:23:24

    10 2014-08-20 12:47:16 (current)

    $ nix-collect-garbage

    [...]

    $ ls /nix/store/b3lxx3d3ggxcggvjw5n0m1ya1gcrmbyn-bsd-games-2.17

    ls: cannot access /nix/store/b3lxx3d3ggxcggvjw5n0m1ya1gcrmbyn-bsd-games-2.17: No such file or directory

    Note: nix-env --list-generations does not rely on any particular metadata. It is able to list generations based solely on the file names under the profiles directory. However we removed the link from /nix/var/nix/profiles, not from /nix/var/nix/gcroots. Turns out, that /nix/var/nix/gcroots/profiles is a symlink to /nix/var/nix/profiles. That is very handy. It means any profile and its generations are GC roots.

    It's as simple as that, anything under /nix/var/nix/gcroots is a GC root. And anything not being garbage collected is because it's referred from one of the GC roots.

    Indirect roots

    I remind you that building the GNU hello world package with nix-build produces a result symlink in the current directory. Despite the collected garbage done above, the hello program is still working: therefore it has not been garbage collected. Clearly, since there's no other derivation that depends upon the GNU hello world package, it must be a GC root.

    In fact, nix-build automatically adds the result symlink as a GC root. Yes, not the built derivation, but the symlink. These GC roots are added under /nix/var/nix/gcroots/auto .

    $ ls -l /nix/var/nix/gcroots/auto/

    total 8

    drwxr-xr-x 2 nix nix 4096 Aug 20 10:24 ./

    drwxr-xr-x 3 nix nix 4096 Jul 24 10:38 ../

    lrwxrwxrwx 1 nix nix 16 Jul 31 10:51 xlgz5x2ppa0m72z5qfc78b8wlciwvgiz -> /home/nix/result/

    Don't care about the name of the symlink. What's important is that a symlink exists that point to /home/nix/result. This is called an indirect GC root. That is, the GC root is effectively specified outside of /nix/var/nix/gcroots. Whatever result points to, it will not be garbage collected.

    How do we remove the derivation then? There are two possibilities:

    - remove the indirect GC root from /nix/var/nix/gcroots/auto

    - remove the result symlink

    In the first case, the derivation will be deleted from the nix store, and result becomes a dangling symlink. In the second case, the derivation is removed as well as the indirect root in /nix/var/nix/gcroots/auto .

    Running nix-collect-garbage after deleting the GC root or the indirect GC root, will remove the derivation from the store.

    Cleanup everything

    What's the main source of software duplication in the nix store? Clearly, GC roots due to nix-build and profile generations. Doing a nix-build results in a GC root for a build that somehow will refer to a specific version of glibc, and other libraries. After an upgrade, if that build is not deleted by the user, it will not be garbage collected. Thus the old dependencies referred to by the build will not be deleted either.

    Same goes for profiles. Manipulating the nix-env profile will create further generations. Old generations refer to old software, thus increasing duplication in the nix store after an upgrade.

    What are the basic steps for upgrading and removing everything old, including old generations? In other words, do an upgrade similar to other systems, where they forget everything about the older state:

    $ nix-channel --update

    $ nix-env -u --always

    $ rm /nix/var/nix/gcroots/auto/*

    $ nix-collect-garbage -d

    First, we download a new version of the nixpkgs channel, which holds the description of all the software. Then we upgrade our installed packages with nix-env -u. That will bring us into a fresh new generation with all updated software.

    Then we remove all the indirect roots generated by nix-build: beware, this will result in dangling symlinks. You may be smarter and also remove the target of those symlinks.

    Finally, the -d option of nix-collect-garbage is used to delete old generations of all profiles, then collect garbage. After this, you lose the ability to rollback to any previous generation. So make sure the new generation is working well before running the command.

    Cleaning up everything down to the oldest bit of software after an upgrade seems a bit contrived, but that's the price of having multiple generations, multiple profiles, multiple versions of software, thus rollbacks etc.. The price of having many possibilities.

    ===== 12: the inputs design pattern

    Repositories in nix

    nix is a tool for build and deployment, it does not enforce any particular repository format. A repository of packages is the main usage for nix, but not the only possibility. See it more like a consequence due to the need of organizing packages. nix is a language, and it is powerful enough to let you choose the format of your own repository. In this sense, it is not declarative, but functional. There is no preset directory structure or preset packaging policy. It's all about you and nix.

    The single repository pattern

    Systems like Debian scatter packages in several small repositories. Personally, this makes it hard to track interdependent changes and to contribute to new packages. Systems like Gentoo instead, put package descriptions all in a single repository. The nix reference for packages is nixpkgs, a single repository of all descriptions of all packages. I find this approach very natural and attractive for new contributions.

    The natural implementation in nix is to create a top-level nix expression, and one expression for each package. The top-level expression imports and combines all expressions in a giant attribute set with name -> package pairs.

    But isn't that heavy? It isn't, because nix is a lazy language. And that's why nixpkgs is able to maintain such a big software repository in a giant attribute set.

    Packaging graphviz

    We have packaged GNU hello world, I guess you would like to package something else for creating at least a repository of two projects :-) . I chose graphviz, which uses the standard autotools build system, requires no patching and dependencies are optional.

    Download graphviz. The graphviz.nix expression is straightforward:

    let

    pkgs = import {};

    mkDerivation = import ./autotools.nix pkgs;

    in mkDerivation {

    name = "graphviz";

    src = ./graphviz-2.38.0.tar.gz;

    }

    Build with nix-build graphviz.nix and you will get runnable binaries under result/bin. Notice how we did reuse the same autotools.nix of hello.nix. Let's create a simple png:

    $ echo 'graph test { a -- b }'|result/bin/dot -Tpng -o test.png

    Format: "png" not recognized. Use one of: canon cmap [...]

    Oh of course... graphviz can't know about png. It built only the output formats it supports natively, without using any extra library.

    I remind you, in autotools.nix there's a buildInputs variablewhich gets concatenated to baseInputs. That would be the perfect place to add a build dependency. We created that variable exactly for this reason to be overridable from package expressions.

    This 2.38 version of graphviz has several plugins to output png. For simplicity, we will use libgd.

    Digression about gcc and ld wrappers

    The gd, jpeg, fontconfig and bzip2 libraries (dependencies of gd) don't use pkg-config to specify which flags to pass to the compiler. Since there's no global location for libraries, we need to tell gcc and ld where to find includes and libs.

    The nixpkgs provides gcc and binutils, and we are using them for our packaging. Not only, it also provides wrappers for them which allow passing extra arguments to gcc and ld, bypassing the project build systems:

    NIX_CFLAGS_COMPILE: extra flags to gcc at compile time

    NIX_LDFLAGS: extra flags to ld

    What can we do about it? We can employ the same trick we did for PATH: automatically filling the variables from buildInputs. This is the relevant snippet of setup.sh:

    for p in $baseInputs $buildInputs; do

    if [ -d $p/bin ]; then

    export PATH="$p/bin${PATH:+:}$PATH"

    fi

    if [ -d $p/include ]; then

    export NIX_CFLAGS_COMPILE="-I $p/include${NIX_CFLAGS_COMPILE:+ }$NIX_CFLAGS_COMPILE"

    fi

    if [ -d $p/lib ]; then

    export NIX_LDFLAGS="-rpath $p/lib -L $p/lib${NIX_LDFLAGS:+ }$NIX_LDFLAGS"

    fi

    done

    Now by adding derivations to buildInputs, will add the lib, include and bin paths automatically in setup.sh.

    The -rpath flag in ld is needed because at runtime, the executable must use exactly that version of the library. If unneeded paths are specified, the fixup phase will shrink the rpath for us!

    Completing graphviz with gd

    Finish the expression for graphviz with gd support (note the use of the with expression in buildInputs to avoid repeating pkgs):

    let

    pkgs = import {};

    mkDerivation = import ./autotools.nix pkgs;

    in mkDerivation {

    name = "graphviz";

    src = ./graphviz-2.38.0.tar.gz;

    buildInputs = with pkgs; [ gd fontconfig libjpeg bzip2 ];

    }

    Now you can create the png! Ignore any error from fontconfig, especially if you are in a chroot.

    The repository expression

    Now that we have two packages, what's a good way to put them together in a single repository? We do something like nixpkgs does. With nixpkgs, we import it and then we peek derivations by accessing the giant attribute set. For us nixers, this a good technique, because it abstracts from the file names. We don't refer to a package by REPO/some/sub/dir/package.nix but by importedRepo.package (or pkgs.package in our examples).

    Create a default.nix in the current directory:

    {

    hello = import ./hello.nix;

    graphviz = import ./graphviz.nix;

    }

    Ready to use! Try it with nix-repl:

    $ nix-repl

    nix-repl> :l default.nix

    Added 2 variables.

    nix-repl> hello

    «derivation /nix/store/dkib02g54fpdqgpskswgp6m7bd7mgx89-hello.drv»

    nix-repl> graphviz

    «derivation /nix/store/zqv520v9mk13is0w980c91z7q1vkhhil-graphviz.drv»

    With nix-build:

    $ nix-build default.nix -A hello

    [...]

    $ result/bin/hello

    Hello, world!

    The -A argument is used to access an attribute of the set from the given .nix expression.

    Important: why did we choose the default.nix? Because when a directory (by default the current directory) has a default.nix, that default.nix will be used (see import here). In fact you can run nix-build -A hello without specifying default.nix. For pythoners, it is similar to __init__.py. With nix-env, to install the package in your user environment:

    $ nix-env -f . -iA graphviz

    [...]

    $ dot -V

    The -f option is used to specify the expression to use, in this case the current directory, therefore ./default.nix. The -i stands for installation. The -A is the same as above for nix-build. We reproduced the very basic behavior of nixpkgs.

    The inputs pattern

    After a long preparation, we finally arrived. I know you have a big doubt in this moment. It's about the hello.nix and graphviz.nix. They are very, very dependent on nixpkgs:

    First big problem: they import nixpkgs directly. In autotools.nix instead we pass nixpkgs as an argument. That's a much better approach.

    Second problem: what if we want a variant of graphviz without libgd support?

    Third problem: what if we want to test graphviz with a particular libgd version?

    The current answer to the above questions is: change the expression to match your needs (or change the callee to match your needs).

    With the inputs pattern, we choose to give another answer: let the user change the inputs of the expression (or change the caller to pass different inputs). By inputs of an expression, we refer to the set of derivations needed to build that expression. In this case: mkDerivation from autotools. Recall that mkDerivation has an implicit dependency on the toolchain. libgd and its dependencies

    The src is also an input but it's pointless to change the source from the caller. For version bumps, in nixpkgs we prefer to write another expression (e.g. because patches are needed or different inputs are needed).

    Goal: make package expressions independent of the repository

    How do we achieve that? The answer is simple: use functions to declare inputs for a derivation. Doing it for graphviz.nix, will make the derivation independent of the repository and customizable:

    { mkDerivation, gdSupport ? true, gd, fontconfig, libjpeg, bzip2 }:

    mkDerivation {

    name = "graphviz";

    src = ./graphviz-2.38.0.tar.gz;

    buildInputs = if gdSupport then [ gd fontconfig libjpeg bzip2 ] else [];

    }

    I recall that "{...}: ..." is the syntax for defining functions accepting an attribute set as argument.

    We made gd and its dependencies optional. If gdSupport is true (by default), we will fill buildInputs and thus graphviz will be built with gd support, otherwise it won't.

    Now back to default.nix:

    let

    pkgs = import {};

    mkDerivation = import ./autotools.nix pkgs;

    in with pkgs; {

    hello = import ./hello.nix { inherit mkDerivation; };

    graphviz = import ./graphviz.nix { inherit mkDerivation gd fontconfig libjpeg bzip2; };

    graphvizCore = import ./graphviz.nix {

    inherit mkDerivation gd fontconfig libjpeg bzip2;

    gdSupport = false;

    };

    }

    So we factorized the import of nixpkgs and mkDerivation, and also added a variant of graphviz with gd support disabled. The result is that both hello.nix (exercise for the reader) and graphviz.nix are independent of the repository and customizable by passing specific inputs.

    If you wanted to build graphviz with a specific version of gd, it would suffice to pass gd = ...;.

    If you wanted to change the toolchain, you may pass a different mkDerivation function.

    Clearing up the syntax:

    In the end we return an attribute set from default.nix. With "let" we define some local variables. We bring pkgs into the scope when defining the packages set, which is very convenient instead of typing everytime "pkgs". We import hello.nix and graphviz.nix, which will return a function, and call it with a set of inputs to get back the derivation.

    The "inherit x" syntax is equivalent to "x = x". So "inherit gd" here, combined to the above "with pkgs;" is equivalent to "x = pkgs.gd"

    ===== 13: the callPackage design pattern

    The next design pattern worth noting is what I'd like to call the callPackage pattern. This technique is extensively used in nixpkgs, it's the current standard for importing packages in a repository.

    The callPackage convenience

    In the previous pill, we underlined the fact that the inputs pattern is great to decouple packages from the repository, in that we can pass manually the inputs to the derivation. The derivation declares its inputs, and the caller passes the arguments. However as with usual programming languages, we declare parameter names, and then we have to pass arguments. We do the job twice. With package management, we often see common patterns. In the case of nixpkgs it's the following.

    Some package derivation:

    { input1, input2, ... }:

    ...

    Repository derivation:

    rec {

    lib1 = import package1.nix { inherit input1 input2 ...; };

    program2 = import package1.nix { inherit inputX inputY lib1 ...; };

    }

    Where inputs may even be packages in the repository itself (note the rec keyword). The pattern here is clear, often inputs have the same name of the attributes in the repository itself. Our desire is to pass those inputs from the repository automatically, and in case be able to specify a particular argument (that is, override the automatically passed default argument).

    To achieve this, we will define a callPackage function with the following synopsis:

    {

    lib1 = callPackage package1.nix { };

    program2 = callPackage package2.nix { someoverride = overriddenDerivation; };

    }

    What should it do? Import the given expression, which in turn returns a function. Determine the name of its arguments. Pass default arguments from the repository set, and let us override those arguments.

    Implementing callPackage

    First of all, we need a way to introspect (reflection or whatever) at runtime the argument names of a function. That's because we want to automatically pass such arguments. Then callPackage requires access to the whole packages set, because it needs to find the packages to pass automatically.

    We start off simple with nix repl:

    nix-repl> add = { a ? 3, b }: a+b

    nix-repl> builtins.functionArgs add

    { a = true; b = false; }

    nix provides a builtin function to introspect the names of the arguments of a function. In addition, for each argument, it tells whether the argument has a default value or not. We don't really care about default values in our case. We are only interested in the argument names.

    Now we need a set with all the values, let's call it values. And a way to intersect the attributes of values with the function arguments:

    nix-repl> values = { a = 3; b = 5; c = 10; }

    nix-repl> builtins.intersectAttrs values (builtins.functionArgs add)

    { a = true; b = false; }

    nix-repl> builtins.intersectAttrs (builtins.functionArgs add) values

    { a = 3; b = 5; }

    Perfect, note from the example above that the intersectAttrs returns a set whose names are the intersection, and the attribute values are taken from the second set.

    We're done, we have a way to get argument names from a function, and match with an existing set of attributes. This is our simple implementation of callPackage:

    nix-repl> callPackage = set: f: f (builtins.intersectAttrs (builtins.functionArgs f) set)

    nix-repl> callPackage values add

    8

    nix-repl> with values; add { inherit a b; }

    8

    Clearing up the syntax:

    We define a callPackage variable which is a function.

    First it accepts a set, and it returns another function accepting another parameter. In other words, let's simplify by saying it accepts two parameters.

    The second parameter is the function to "autocall".

    We take the argument names of the function and intersect with the set of all values.

    Finally we call the passed function f with the resulting intersection.

    In the code above, I've also shown that the callPackage call is equivalent to directly calling add a b.

    We achieved what we wanted. Automatically call functions given a set of possible arguments. If an argument is not found in the set, that's nothing special. It's a function call with a missing parameter, and that's an error (unless the function has varargs ... as explained in the 5th pill).

    Or not. We missed something. Being able to override some of the parameters. We may not want to always call functions with values taken from the big set. Then we add a further parameter, which takes a set of overrides:

    nix-repl> callPackage = set: f: overrides: f ((builtins.intersectAttrs (builtins.functionArgs f) set) // overrides)

    nix-repl> callPackage values add { }

    8

    nix-repl> callPackage values add { b = 12; }

    15

    Apart from the increasing number of parenthesis, it should be clear that we simply do a set union between the default arguments, and the overriding set.

    Use callPackage to simplify the repository

    Given our brand new tool, we can simplify the repository expression (default.nix). Let me write it down first:

    let

    nixpkgs = import {};

    allPkgs = nixpkgs // pkgs;

    callPackage = path: overrides:

    let f = import path;

    in f ((builtins.intersectAttrs (builtins.functionArgs f) allPkgs) // overrides);

    pkgs = with nixpkgs; {

    mkDerivation = import ./autotools.nix nixpkgs;

    hello = callPackage ./hello.nix { };

    graphviz = callPackage ./graphviz.nix { };

    graphvizCore = callPackage ./graphviz.nix { gdSupport = false; };

    };

    in pkgs

    Wow, there's a lot to say here:

    We renamed the old pkgs of the previous pill to nixpkgs. Our package set is now instead named pkgs. Sorry for the confusion.

    We needed a way to pass pkgs to callPackage somehow. Instead of returning the set of packages directly from default.nix, we first assign it to a let variable and reuse it in callPackage.

    For convenience, in callPackage we first import the file, instead of calling it directly. Otherwise for each package we would have to write the import.

    Since our expressions use packages from nixpkgs, in callPackage we use allPkgs, which is the union of nixpkgs and our packages.

    We moved mkDerivation in pkgs itself, so that it gets also passed automatically.

    Note how easy is to override arguments in the case of graphviz without gd. But most importantly, how easy it was to merge two repositories: nixpkgs and our pkgs!

    The reader should notice a magic thing happening. We're defining pkgs in terms of callPackage, and callPackage in terms of pkgs. That magic is possible thanks to lazy evaluation.

    Conclusion

    The "callPackage" pattern has simplified a lot our repository. We're able to import packages that require some named arguments and call them automatically, given the set of all packages.

    We've also introduced some useful builtin functions that allows us to introspect nix functions and manipulate attributes. These builtin functions are not usually used when packaging software, rather to provide tools for packaging. That's why they are not documented in the nix manual.

    Writing a repository in nix is an evolution of writing convenient functions for combining the packages. This demonstrates even more how nix is a generic tool to build and deploy something, and how suitable it is to create software repositories with your own conventions. ===== 14: the override design pattern

    About composability

    Functional languages are known for being able to compose functions.

    In particular, you gain a lot from functions that are able to manipulate the original value into a new value having the same structure. So that in the end we're able to call multiple functions to have the desired modifications.

    In nix we mostly talk about functions that accept inputs in order to return derivations. In our world we want nice utility functions that are able to manipulate those structures. These utilities add some useful properties to the original value, and we must be able to apply more utilities on top of it.

    For example let's say we have an initial derivation "drv" and we want it to be a "drv with debugging information" and also to apply some custom patches:

    debugVersion (applyPatches [ ./patch1.patch ./patch2.patch ] drv)

    The final result will be still the original derivation plus some changes. That's both interesting and very different from other packaging approaches, which is a consequence of using a functional language to describe packages.

    Designing such utilities is not trivial in a functional language that is not statically typed, because understanding what can or cannot be composed is difficult. But we try to do the best.

    The override pattern

    In the pill 12 we introduced the inputs design pattern. We do not return a derivation picking dependencies directly from the repository, rather we declare the inputs and let the callers pass the necessary arguments.

    In our repository we have a set of attributes that import the expressions of the packages and pass these arguments, getting back a derivation. Let's take for example the graphviz attribute:

    graphviz = import ./graphviz.nix { inherit mkDerivation gd fontconfig libjpeg bzip2; };

    If we wanted to produce a derivation of graphviz with a customized gd version, we would have to repeat most of the above plus specifying an alternative gd:

    mygraphviz = import ./graphviz.nix {

    inherit mkDerivation fontconfig libjpeg bzip2;

    gd = customgd;

    };

    That's hard to maintain. Using callPackage it would be easier:

    mygraphviz = callPackage ./graphviz.nix { gd = customgd; };

    But we may still be diverging from the original graphviz in the repository.

    We would like to avoid specifying the nix expression again, instead reuse the original graphviz attribute in the repository and add our overrides like this:

    mygraphviz = graphviz.override { gd = customgd; };

    The difference is obvious, as well as the advantages of this approach.

    Note: that .override is not a "method" in the OO sense as you may think. nix is a functional language. That .override is simply an attribute of a set.

    The override implementation

    I remind you, the graphviz attribute in the repository is the derivation returned by the function imported from graphviz.nix. We would like to add a further attribute named "override" to the returned set.

    Let's start simple by first creating a function "makeOverridable" that takes a function and a set of original arguments to be passed to the function.

    Contract: the wrapped function must return a set.

    Let's write a lib.nix:

    {

    makeOverridable = f: origArgs:

    let

    origRes = f origArgs;

    in

    origRes // { override = newArgs: f (origArgs // newArgs); };

    }

    So makeOverridable takes a function and a set of original arguments. It returns the original returned set, plus a new override attribute.

    This override attribute is a function taking a set of new arguments, and returns the result of the original function called with the original arguments unified with the new arguments. What a mess.

    Let's try it with nix repl:

    $ nix-repl

    nix-repl> :l lib.nix

    Added 1 variables.

    nix-repl> f = { a, b }: { result = a+b; }

    nix-repl> f { a = 3; b = 5; }

    { result = 8; }

    nix-repl> res = makeOverridable f { a = 3; b = 5; }

    nix-repl> res

    { override = «lambda»; result = 8; }

    nix-repl> res.override { a = 10; }

    { result = 15; }

    Note that the function f does not return the plain sum but a set, because of the contract. You didn't forget already, did you? :-)

    The variable res is the result of the function call without any override. It's easy to see in the definition of makeOverridable. In addition you can see the new override attribute being a function.

    Calling that .override with a set will invoke the original function with the overrides, as expected.

    But: we can't override again! Because the returned set with result 15 does not have an override attribute!

    That's bad, it breaks further compositions.

    The solution is simple, the .override function should make the result overridable again:

    rec {

    makeOverridable = f: origArgs:

    let

    origRes = f origArgs;

    in

    origRes // { override = newArgs: makeOverridable f (origArgs // newArgs); };

    }

    Please note the rec keyword. It's necessary so that we can refer to makeOverridable from makeOverridable itself.

    Now let's try overriding twice:

    nix-repl> :l lib.nix

    Added 1 variables.

    nix-repl> f = { a, b }: { result = a+b; }

    nix-repl> res = makeOverridable f { a = 3; b = 5; }

    nix-repl> res2 = res.override { a = 10; }

    nix-repl> res2

    { override = «lambda»; result = 15; }

    nix-repl> res2.override { b = 20; }

    { override = «lambda»; result = 30; }

    Success! The result is 30, as expected because a is overridden to 10 in the first override, and b to 20.

    Now it would be nice if callPackage made our derivations overridable. That was the goal of this pill after all. This is an exercise for the reader.

    Conclusion

    The "override" pattern simplifies the way we customize packages starting from an existing set of packages. This opens a world of possibilities about using a central repository like nixpkgs, and defining overrides on our local machine without even modifying the original package.

    Dream of a custom isolated nix-shell environment for testing graphviz with a custom gd:

    debugVersion (graphviz.override { gd = customgd; })

    Once a new version of the overridden package comes out in the repository, the customized package will make use of it automatically.

    The key in nix is to find powerful yet simple abstractions in order to let the user customize his environment with highest consistency and lowest maintenance time, by using predefined composable components. ===== 15: nix search paths

    Welcome to the 15th nix pill. In the previous 14th pill we have introduced the "override" pattern, useful for writing variants of derivations by passing different inputs.

    Assuming you followed the previous posts, I hope you are now ready to understand nixpkgs. But we have to find nixpkgs in our system first! So this is the step: introducing some options and environment variables used by nix tools.

    The NIX_PATH

    The NIX_PATH environment variable is very important. It's very similar to the PATH environment variable. The syntax is similar, several paths are separated by a colon ":". nix will then search for something in those paths from left to right.

    Who uses NIX_PATH? The nix expressions! Yes, NIX_PATH is not of much use by the nix tools themselves, rather it's used when writing nix expressions.

    In the shell for example, when you execute the command "ping", it's being searched in the PATH directories. The first one found is the one being used.

    In nix it's exactly the same, however the syntax is different. Instead of just typing "ping" you have to type . Yes, I know... you are already thinking of .

    However don't stop reading here, let's keep going.

    What's NIX_PATH good for? nix expressions may refer to an "abstract" path such as , and it's possible to override it from the command line.

    For ease we will use nix-instantiate --eval to do our tests. I remind you, nix-instantiate is used to evaluate nix expressions and generate the .drv files. Here we are not interested in building derivations, so evaluation is enough. It can be used for one-shot expressions.

    Fake it a little

    It's useless from a nix view point, but I think it's useful for your own understanding. Let's use PATH itself as NIX_PATH, and try to locate ping (or another binary if you don't have it).

    $ nix-instantiate --eval -E ''

    error: file `ping' was not found in the nix search path (add it using $NIX_PATH or -I)

    $ NIX_PATH=$PATH nix-instantiate --eval -E ''

    /bin/ping

    $ nix-instantiate -I /bin --eval -E ''

    /bin/ping

    Great. At first attempt nix obviously said could not be found anywhere in the search path. Note that the -I option accepts a single directory. Paths added with -I take precedence over NIX_PATH.

    The NIX_PATH also accepts a different yet very handy syntax: "somename=somepath". That is, instead of searching inside a directory for a name, we specify exactly the value of that name.

    $ NIX_PATH="ping=/bin/ping" nix-instantiate --eval -E ''

    /bin/ping

    $ NIX_PATH="ping=/bin/foo" nix-instantiate --eval -E ''

    error: file `ping' was not found in the nix search path (add it using $NIX_PATH or -I)

    Note in the second case how nix checks whether the path exists or not.

    The path to repository

    You are out of curiosity, right?

    $ nix-instantiate --eval -E ''

    /home/nix/.nix-defexpr/channels/nixpkgs

    $ echo $NIX_PATH

    nixpkgs=/home/nix/.nix-defexpr/channels/nixpkgs

    You may have a different path, depending on how you added channels etc.. Anyway that's the whole point. The stranger that we used in our nix expressions, is referring to a path in the filesystem specified by NIX_PATH.

    You can list that directory and realize it's simply a checkout of the nixpkgs repository at a specific commit (hint: .version-suffix).

    The NIX_PATH variable is exported by nix.sh, and that's the reason why I always asked you to source nix.sh at the beginning of my posts.

    You may wonder: then I can also specify a different nixpkgs path to, e.g., a git checkout of nixpkgs? Yes, you can and I encourage doing that. We'll talk about this in the next pill.

    Let's define a path for our repository, then! Let's say all the default.nix, graphviz.nix etc. are under /home/nix/mypkgs:

    $ export NIX_PATH=mypkgs=/home/nix/mypkgs:$NIX_PATH

    $ nix-instantiate --eval ''

    { graphviz = ; graphvizCore = ; hello = ; mkDerivation = ; }

    As expected we got the set of our packages (well except the mkDerivation utility), that's our repository.

    Until now we used nix-build directly in the directory of default.nix. However nix-build generally needs a .nix to be specified to the command line:

    $ nix-build /home/nix/mypkgs -A graphviz

    /nix/store/dhps2c1fnsfxssdri3yxbdiqmnhzg6yr-graphviz

    $ nix-build '' -A graphviz

    /nix/store/dhps2c1fnsfxssdri3yxbdiqmnhzg6yr-graphviz

    Yes, nix-build also accepts paths with angular brackets. We first evaluate the whole repository (default.nix) and then peek the graphviz attribute.

    A big word about nix-env

    The nix-env command is a little different than nix-instantiate and nix-build. Whereas nix-instantiate and nix-build require a starting nix expression, nix-env does not.

    You may be crippled by this concept at the beginning, you may think nix-env uses NIX_PATH to find the nixpkgs repository. But that's not it.

    The nix-env command uses ~/.nix-defexpr, which is also part of NIX_PATH by default, but that's only a coincidence. If you empty NIX_PATH, nix-env will still be able to find derivations because of ~/.nix-defexpr.

    So if you run nix-env -i graphviz inside your repository, it will install the nixpkgs one. Same if you set NIX_PATH to point to your repository.

    In order to specify an alternative to ~/.nix-defexpr it's possible to use the -f option:

    $ nix-env -f '' -i graphviz

    warning: there are multiple derivations named `graphviz'; using the first one

    replacing old `graphviz'

    installing `graphviz'

    Oh why did it say there's another derivation named graphviz? Because both graphviz and graphvizCore attributes in our repository have the name "graphviz" for the derivation:

    $ nix-env -f '' -qaP

    graphviz graphviz

    graphvizCore graphviz

    hello hello

    By default nix-env parses all derivations and use the derivation names to interpret the command line. So in this case "graphviz" matched two derivations. Alternatively, like for nix-build, one can use -A to specify an attribute name instead of a derivation name:

    $ nix-env -f '' -i -A graphviz

    replacing old `graphviz'

    installing `graphviz'

    This form, other than being more precise, it's also faster because nix-env does not have to parse all the derivations.

    For completeness: you may install graphvizCore with -A, since without the -A switch it's ambiguous.

    In summary, it may happen when playing with nix that nix-env peeks a different derivation than nix-build. In such case you probably specified NIX_PATH, but nix-env is instead looking into ~/.nix-defexpr.

    Why is nix-env having this different behavior? I don't know specifically by myself either, but the answers could be:

    nix-env tries to be generic, thus it does not look for "nixpkgs" in NIX_PATH, rather it looks in ~/.nix-defexpr.

    nix-env is able to merge multiple trees in ~/.nix-defexpr by looking at all the possible derivations

    It may also happen to you that you cannot match a derivation name when installing, because of the derivation name vs -A switch described above. Maybe nix-env wanted to be more friendly in this case for default user setups.

    It may or may not make sense for you, or it's like that for historical reasons, but that's how it works currently, unless somebody comes up with a better idea.

    Conclusion

    The NIX_PATH variable is the search path used by nix when using the angular brackets syntax. It's possible to refer to "abstract" paths inside nix expressions and define the "concrete" path by means of NIX_PATH, or the usual -I flag in nix tools.

    We've also explained some of the uncommon nix-env behaviors for newcomers. The nix-env tool does not use NIX_PATH to search for packages, but rather for ~/.nix-defexpr. Beware of that!

    In general do not abuse NIX_PATH, when possible use relative paths when writing your own nix expressions. Of course, in the case of in our repository, that's a perfectly fine usage of NIX_PATH. Instead, inside our repository itself, refer to expressions with relative paths like ./hello.nix.

    ===== 16: nixpkgs, the parameters

    Welcome to the 16th nix pill. In the previous 15th pill we've realized how nix finds expressions with the angular brackets syntax, so that we finally know where is located on our system.

    We can start diving into the nixpkgs repository, through all the various tools and design patterns. Please note that also nixpkgs has its own manual, underlying the difference between the general "nix" language and the "nixpkgs" repository.

    The default.nix expression

    We will not start inspecting packages at the beginning, rather the general structure of nixpkgs.

    In our custom repository we created a default.nix which composed the expressions of the various packages.

    Also nixpkgs has its own default.nix, which is the one being loaded when referring to . It does a simple thing: check whether the nix version is at least 1.7 (at the time of writing this blog post). Then import pkgs/top-level/all-packages.nix. From now on, we will refer to this set of packages as pkgs.

    The all-packages.nix is then the file that composes all the packages. Note the pkgs/ subdirectory, while nixos is in the nixos/ subdirectory.

    The all-packages.nix is a bit contrived. First of all, it's a function. It accepts a couple of interesting parameters:

    system: defaults to the current system

    config: defaults to null

    others...

    The system parameter, as per comment in the expression, it's the system for which the packages will be built. It allows for example to install i686 packages on amd64 machines.

    The config parameter is a simple attribute set. Packages can read some of its values and change the behavior of some derivations.

    The system parameter

    You will find this parameter in many other .nix expressions (e.g. release expressions). The reason is that, given pkgs accepts a system parameter, then whenever you want to import pkgs you also want to pass through the value of system. E.g.:

    myrelease.nix:

    { system ? builtins.currentSystem }:

    let pkgs = import { inherit system; };

    ...

    Why is it useful? With this parameter it's very easy to select a set of packages for a particular system. For example:

    nix-build -A psmisc --argstr system i686-linux

    This will build the psmisc derivation for i686-linux instead of x86_64-linux. This concept is very similar to multi-arch of Debian.

    The setup for cross compiling is also in nixpkgs, however it's a little contrived to talk about it and I don't know much of it either.

    The config parameter

    I'm sure on the wiki or other manuals you've read about ~/.nixpkgs/config.nix and I'm sure you've wondered whether that's hardcoded in nix. It's not, it's in nixpkgs.

    The all-packages.nix expression accepts the config parameter. If it's null, then it reads the NIXPKGS_CONFIG environment variable. If not specified, nixpkgs will peek $HOME/.nixpkgs/config.nix .

    After determining config.nix, it will be imported as nix expression, and that will be the value of config (in case it hasn't been passed as parameter to import ).

    The config is available in the resulting repository:

    $ nix-repl

    nix-repl> pkgs = import {}

    nix-repl> pkgs.config

    { }

    nix-repl> pkgs = import { config = { foo = "bar"; }; }

    nix-repl> pkgs.config

    { foo = "bar"; }

    What attributes go in config is a matter of convenience and conventions.

    For example, config.allowUnfree is an attribute that forbids building packages that have an unfree license by default. The config.pulseaudio setting tells whether to build packages with pulseaudio support or not where applicable and when the derivation obeys to the setting.

    About .nix functions

    A .nix file contains a nix expression. Thus it can also be a function.

    I remind you that nix-build expects the expression to return a derivation. Therefore it's natural to return straight a derivation from a .nix file.

    However, it's also very natural for the .nix file to accept some parameters, in order to tweak the derivation being returned.

    In this case, nix does a trick:

    If the expression is a derivation, well build it.

    If the expression is a function, call it and build the resulting derivation.

    For example you can nix-build the .nix file below:

    { pkgs ? import {} }:

    pkgs.psmisc

    nix is able to call the function because the pkgs parameter has a default value. This allows you to pass a different value for pkgs using the --arg option.

    Does it work if you have a function returning a function that returns a derivation? No, nix only calls the function it encounters once.

    Conclusion

    We've unleashed the repository. It's a function that accepts some parameters, and returns the set of all packages. Due to laziness, only the accessed derivations will be built.

    You can use this repository to build your own packages as we've seen in the previous pill when creating our own repository.

    Lately I'm a little busy with the nixOS 14.11 release and other stuff, and I'm also looking toward migrating from blogger to a more coder-oriented blogging platform. So sorry for the delayed and shorter pills :)

    ===== 17: nixpkgs, overriding packages

    Welcome to the 17th nix pill. In the previous 16th pill we have started to dive into the nixpkgs repository. nixpkgs is a function, and we've looked at some parameters like system and config.

    Today we'll talk about a special attribute: config.packageOverrides. Overriding packages in a set with fixed point can be considered another design pattern in nixpkgs.

    Overriding a package

    I recall the override design pattern from the nix pill 14. Instad of calling a function with parameters directly, we make the call (function + parameters) overridable.

    We put the override function in the returned attribute set of the original function call.

    Take for example graphviz. It has an input parameter xlibs. If it's null, then graphviz will build without X support.

    $ nix-repl

    nix-repl> :l

    Added 4360 variables.

    nix-repl> :b graphviz.override { xlibs = null; }

    This will build graphviz without X support, it's as simple as that.

    However let's say a package P depends on graphviz, how do we make P depend on the new graphviz without X support?

    In an imperative world...

    ...you could do something like this:

    pkgs = import {};

    pkgs.graphviz = pkgs.graphviz.override { xlibs = null; };

    build(pkgs.P)

    Given pkgs.P depends on pkgs.graphviz, it's easy to build P with the replaced graphviz. On a pure functional language it's not that easy because you can assign to variables only once.

    Fixed point

    The fixed point with lazy evaluation is crippling but about necessary in a language like nix. It lets us achieve something similar to what we'd do imperatively.

    Follows the definition of fixed point in nixpkgs:

    # Take a function and evaluate it with its own returned value.

    fix = f: let result = f result; in result;

    It's a function that accepts a function f, calls f result on the result just returned by f result and returns it. In other words it's f(f(f(....

    At first sight, it's an infinite loop. With lazy evaluation it isn't, because the call is done only when needed.

    nix-repl> fix = f: let result = f result; in result

    nix-repl> pkgs = self: { a = 3; b = 4; c = self.a+self.b; }

    nix-repl> fix pkgs

    { a = 3; b = 4; c = 7; }

    Without the rec keyword, we were able to refer to a and b of the same set.

    First pkgs gets called with an unevaluated thunk (pkgs(pkgs(...)

    To set the value of c then self.a and self.b are evaluated.

    The pkgs function gets called again to get the value of a and b.

    The trick is that c is not needed to be evaluated in the inner call, thus it doesn't go in an infinite loop.

    Won't go further with the explanation here. A good post about fixed point and nix can be found here.

    Overriding a set with fixed point

    Given that self.a and self.b refer to the passed set and not to the literal set in the function, we're able to override both a and b and get a new value for c:

    nix-repl> overrides = { a = 1; b = 2; }

    nix-repl> let newpkgs = pkgs (newpkgs // overrides); in newpkgs

    { a = 3; b = 4; c = 3; }

    nix-repl> let newpkgs = pkgs (newpkgs // overrides); in newpkgs // overrides

    { a = 1; b = 2; c = 3; }

    In the first case we computed pkgs with the overrides, in the second case we also included the overriden attributes in the result.

    Overriding nixpkgs packages

    We've seen how to override attributes in a set such that they get recursively picked by dependant attributes. This approach can be used for derivations too, after all nixpkgs is a giant set of attributes that depend on each other.

    To do this, nixpkgs offers config.packageOverrides. So nixpkgs returns a fixed point of the package set, and packageOverrides is used to inject the overrides.

    Create a config.nix file like this somewhere:

    {

    packageOverrides = pkgs: {

    graphviz = pkgs.graphviz.override { xlibs = null; };

    };

    }

    Now we can build e.g. asciidocFull and it will automatically use the overridden graphviz:

    nix-repl> pkgs = import { config = import ./config.nix; }

    nix-repl> :b pkgs.asciidocFull

    Note how we pass the config with packageOverrides when importing nixpkgs. Then pkgs.asciidocFull is a derivation that has graphviz input (pkgs.asciidoc is the lighter version and doesn't use graphviz at all).

    Since there's no version of asciidoc with graphviz without X support in the binary cache, nix will recompile the needed stuff for you.

    The ~/.nixpkgs/config.nix file

    In the previous pill we already talked about this file. The above config.nix that we just wrote could be the content of ~/.nixpkgs/config.nix.

    Instead of passing it explicitly whenever we import nixpkgs, it will be automatically imported by nixpkgs.

    Conclusion

    We've learned about a new design pattern: using fixed point for overriding packages in a package set.

    Whereas in an imperative setting, like with other package managers, a library is installed replacing the old version and applications will use it, in nix it's not that straight and simple. But it's more precise.

    nix applications will depend on specific versions of libraries, hence the reason why we have to recompile asciidoc to use the new graphviz library.

    The newly built asciidoc will depend on the new graphviz, and old asciidoc will keep using the old graphviz undisturbed.

    ...we will stop diving nixpkgs for a moment and talk about store paths. How does nix compute the path in the store where to place the result of builds? How to add files to the store for which we have an integrity hash?

    ===== 18: nix store paths

    the nixpkgs repository structure is a set of packages, and it's possible to override such packages so that all other packages will use the overrides. before reading existing derivations, I'd like to talk about store paths and how they are computed. In particular we are interested in fixed store paths that depend on an integrity hash (e.g. a sha256), which is usually applied to source tarballs.

    The way store paths are computed is a little contrived, mostly due to historical reasons. Our reference will be the nix source code.

    Source paths

    nix allows relative paths to be used, such that the file or directory is stored in the nix store, that is ./myfile gets stored into /nix/store/....... We want to understand how is the store path generated for such a file:

    $ echo mycontent > myfile

    I remind you, the simplest derivation you can write has a name, a builder and the system:

    $ nix-repl

    nix-repl> derivation { system = "x86_64-linux"; builder = ./myfile; name = "foo"; }

    «derivation /nix/store/y4h73bmrc9ii5bxg6i7ck6hsf5gqv8ck-foo.drv»

    Now inspect the .drv to see where is ./myfile being stored:

    $ pp-aterm -i /nix/store/y4h73bmrc9ii5bxg6i7ck6hsf5gqv8ck-foo.drv

    Derive(

    [("out", "/nix/store/hs0yi5n5nw6micqhy8l1igkbhqdkzqa1-foo", "", "")]

    , []

    , ["/nix/store/xv2iccirbrvklck36f1g7vldn5v58vck-myfile"]

    , "x86_64-linux"

    ...

    Great, how did nix decide to use xv2iccirbrvklck36f1g7vldn5v58vck ? Keep looking at the nix comments.

    Note: doing nix-store --add myfile will store the file in the same store path.

    Step 1, compute the hash of the file

    The comments tell us to first compute the sha256 of the NAR serialization of the file. Can be done in two ways:

    $ nix-hash --type sha256 myfile

    2bfef67de873c54551d884fdab3055d84d573e654efa79db3c0d7b98883f9ee3

    Or:

    $ nix-store --dump myfile|sha256sum

    2bfef67de873c54551d884fdab3055d84d573e654efa79db3c0d7b98883f9ee3 -

    In general, nix understands two contents: flat for regular files, or recursive for NAR serializations which can be anything.

    Step 2, build the string description

    Then nix uses a special string which includes the hash, the path type and the file name. We store this in another file:

    $ echo -n "source:sha256:2bfef67de873c54551d884fdab3055d84d573e654efa79db3c0d7b98883f9ee3:/nix/store:myfile" > myfile.str

    Step 3, compute the final hash

    Finally the comments tell us to compute the base-32 representation of the first 160 bits (truncation) of a sha256 of the above string:

    $ nix-hash --type sha256 --truncate --base32 --flat myfile.str

    xv2iccirbrvklck36f1g7vldn5v58vck

    Output paths

    Output paths are usually generated for derivations. We use the above example because it's simple. Even if we didn't build the derivation, nix knows the out path hs0yi5n5nw6micqhy8l1igkbhqdkzqa1. This is because the out path only depends on inputs.

    It's computed in a similar way to source paths, except that the .drv is hashed and the type of derivation is output:out. In case of multiple outputs, we may have different output:.

    At the time nix computes the out path, the .drv contains an empty string for each out path. So what we do is getting our .drv and replacing the out path with an empty string:

    $ cp -f /nix/store/y4h73bmrc9ii5bxg6i7ck6hsf5gqv8ck-foo.drv myout.drv

    $ sed -i 's,/nix/store/hs0yi5n5nw6micqhy8l1igkbhqdkzqa1-foo,,g' myout.drv

    The myout.drv is the .drv state in which nix is when computing the out path for our derivation:

    $ sha256sum myout.drv

    1bdc41b9649a0d59f270a92d69ce6b5af0bc82b46cb9d9441ebc6620665f40b5 myout.drv

    $ echo -n "output:out:sha256:1bdc41b9649a0d59f270a92d69ce6b5af0bc82b46cb9d9441ebc6620665f40b5:/nix/store:foo" > myout.str

    $ nix-hash --type sha256 --truncate --base32 --flat myout.str

    hs0yi5n5nw6micqhy8l1igkbhqdkzqa1

    Then nix puts that out path in the .drv, and that's it.

    In case the .drv has input derivations, that is it references other .drv, then such .drv paths are replaced by this same algorithm which returns an hash. In other words, you get a final .drv where every other .drv path is replaced by its hash.

    Fixed-output paths

    Finally, the other most used kind of path is when we know beforehand an integrity hash of a file. This is usual for tarballs.

    A derivation can take three special attributes: outputHashMode, outputHash and outputHashAlgo which are well documented in the nix manual.

    The builder must create the out path and make sure its hash is the same as the one declared with outputHash.

    Let's say our builder should create a file whose contents is mycontent:

    $ echo mycontent > myfile

    $ sha256sum myfile

    f3f3c4763037e059b4d834eaf68595bbc02ba19f6d2a500dce06d124e2cd99bb myfile

    nix-repl> derivation { name = "bar"; system = "x86_64-linux"; builder = "none"; outputHashMode = "flat"; outputHashAlgo = "sha256"; outputHash = "f3f3c4763037e059b4d834eaf68595bbc02ba19f6d2a500dce06d124e2cd99bb"; }

    «derivation /nix/store/ymsf5zcqr9wlkkqdjwhqllgwa97rff5i-bar.drv»

    Inspect the .drv and see that it also stored the fact that it's a fixed-output derivation with sha256 algorithm, compared to the previous examples:

    $ pp-aterm -i /nix/store/ymsf5zcqr9wlkkqdjwhqllgwa97rff5i-bar.drv

    Derive(

    [("out", "/nix/store/a00d5f71k0vp5a6klkls0mvr1f7sx6ch-bar", "sha256", "f3f3c4763037e059b4d834eaf68595bbc02ba19f6d2a500dce06d124e2cd99bb")]

    ...

    It doesn't matter which input derivations are being used, the final out path must only depend on the declared hash.

    What nix does is to create an intermediate string representation of the fixed-output content:

    $ echo -n "fixed:out:sha256:f3f3c4763037e059b4d834eaf68595bbc02ba19f6d2a500dce06d124e2cd99bb:" > mycontent.str

    $ sha256sum mycontent.str

    423e6fdef56d53251c5939359c375bf21ea07aaa8d89ca5798fb374dbcfd7639 myfile.str

    Then proceed as it was a normal derivation output path:

    $ echo -n "output:out:sha256:423e6fdef56d53251c5939359c375bf21ea07aaa8d89ca5798fb374dbcfd7639:/nix/store:bar" > myfile.str

    $ nix-hash --type sha256 --truncate --base32 --flat myfile.str

    a00d5f71k0vp5a6klkls0mvr1f7sx6ch

    Hence, the store path only depends on the declared fixed-output hash.

    There are other types of store paths, but you get the idea. nix first hashes the contents, then creates a string description, and the final store path is the hash of this string.

    nix knows beforehand the out path of a derivation since it only depends on the inputs. fixed-output derivations are especially used by the nixpkgs repository for downloading and verifying source tarballs

    ===== 19: fundamentals of stdenv

    This time we will instead look into nixpkgs, in particular one of its core derivation: stdenv .

    The stdenv is not a special derivation, but it's very important for the nixpkgs repository. It serves as base for packaging software. It is used to pull in dependencies such as the GCC toolchain, GNU make, core utilities, patch and diff utilities, and so on. Basic tools needed to compile a huge pile of software currently present in nixpkgs

    What is stdenv

    First of all stdenv is a derivation. And it's a very simple one:

    $ nix-build '' -A stdenv

    /nix/store/k4jklkcag4zq4xkqhkpy156mgfm34ipn-stdenv

    $ ls -R result/

    result/:

    nix-support/ setup

    result/nix-support:

    propagated-user-env-packages

    It has just two files: /setup and /nix-support/propagated-user-env-packages. Don't care about the latter, it's even empty. The important file is /setup

    How can this simple derivation pull in all the toolchain and basic tools needed to compile packages? Let's look at the runtime dependencies:

    $ nix-store -q --references result

    /nix/store/3a45nb37s0ndljp68228snsqr3qsyp96-bzip2-1.0.6

    /nix/store/a457ywa1haa0sgr9g7a1pgldrg3s798d-coreutils-8.24

    /nix/store/zmd4jk4db5lgxb8l93mhkvr3x92g2sx2-bash-4.3-p39

    /nix/store/47sfpm2qclpqvrzijizimk4md1739b1b-gcc-wrapper-4.9.3

    ...

    How can it be? The package must be referring to those package somehow. In fact, they are hardcoded in the /setup file:

    $ head result/setup

    export SHELL=/nix/store/zmd4jk4db5lgxb8l93mhkvr3x92g2sx2-bash-4.3-p39/bin/bash

    initialPath="/nix/store/a457ywa1haa0sgr9g7a1pgldrg3s798d-coreutils-8.24 ..."

    defaultNativeBuildInputs="/nix/store/sgwq15xg00xnm435gjicspm048rqg9y6-patchelf-0.8 ..."

    The setup file

    Remember our generic builder.sh in Pill 8? It sets up a basic PATH, unpacks the source and runs the usual autotools commands for us.

    The stdenv setup file is exactly that. It sets up several environment variables like PATH and creates some helper bash functions to build a package. I invite you to read it, it's only 860 lines at the time of this writing.

    The hardcoded toolchain and utilities are used to initially fill up the environment variables so that it's more pleasant to run common commands, similarly but not equal like we did with our builder with baseInputs and buildInputs

    The build with stdenv works in phases. Phases are like unpackPhase, configurePhase, buildPhase, checkPhase, installPhase, fixupPhase. You can see the default list in the genericBuild function

    What genericBuild does is just run these phases. Default phases are just bash functions, you can easily read them

    Every phase has hooks to run commands before and after the phase has been executed. Phases can be overwritten, reordered, whatever, it's just bash code

    How to use this file? Like our old builder. To test it, we enter a fake empty derivation, source the stdenv setup, unpack the hello sources and build it:

    $ nix-shell -E 'derivation { name = "fake"; builder = "fake"; system = "x86_64-linux"; }'

    nix-shell$ unset PATH

    nix-shell$ source /nix/store/k4jklkcag4zq4xkqhkpy156mgfm34ipn-stdenv/setup

    nix-shell$ tar -xf hello-2.9.tar.gz

    nix-shell$ cd hello-2.9

    nix-shell$ configurePhase

    ...

    nix-shell$ buildPhase

    ...

    I unset PATH to further show that the stdenv is enough self-contained to build autotools packages that have no other dependencies

    So we ran the configurePhase function and buildPhase function and they worked. These bash functions should be self-explanatory, you can read the code in the setup file

    How is the setup file built

    Very little digression for completeness. The stdenv derivation is just that setup file. That setup file is just this setup.sh in nixpkgs plus some lines on top of it, put by this simple builder:

    ...

    echo "export SHELL=$shell" > $out/setup

    echo "initialPath=\\"$initialPath\\"" >> $out/setup

    echo "defaultNativeBuildInputs=\\"$defaultNativeBuildInputs\\"" >> $out/setup

    echo "$preHook" >> $out/setup

    cat "$setup" >> $out/setup

    ...

    Nothing much to say, but you can read the nix code that pass $initialPath and $defaultNativeBuildInputs. Not much interesting to continue further in this pill.

    the stdenv.mkDerivation function

    Until now we worked with plain bash scripts. What about the nix side? The nixpkgs repository offers a useful function, like we did with our old builder. It is a wrapper around the raw derivation function which pulls in the stdenv for us, and runs genericBuild. It's 'stdenv.mkDerivation'

    Note how stdenv is a derivation but it's also an attribute set which contains some other attributes, like mkDerivation. Nothing fancy here, just convenience.

    Let's write a hello.nix expression using this new discovered stdenv:

    with import {};

    stdenv.mkDerivation {

    name = "hello";

    src = ./hello-2.9.tar.gz;

    }

    Don't be scared by the with expression. It pulls the nixpkgs repository into scope, so we can directly use stdenv. It looks very similar to the hello expression in Pill 8. It builds, and runs fine:

    $ nix-build hello.nix

    ...

    /nix/store/6flbdbpq6sc1dc79xjx01bz43zwgj3wc-hello

    $ result/bin/hello

    Hello, world!

    the stdenv.mkDerivation builder

    Let's take a look at the builder used by mkDerivation. You can read the code here in nixpkgs:

    {

    ...

    builder = attrs.realBuilder or shell;

    args = attrs.args or ["-e" (attrs.builder or ./default-builder.sh)];

    stdenv = result;

    ...

    }

    Also take a look at our old derivation wrapper in previous pills! The builder is bash (that shell variable), the argument to the builder (bash) is default-builder.sh, and then we add the environment variable $stdenv in the derivation which is the stdenv derivation.

    You can open default-builder.sh and see what it does:

    source $stdenv/setup

    genericBuild

    It's what we did in Pill 10 to make the derivations nix-shell friendly. When entering the shell, the setup file only sets up the environment without building anything. When doing nix-build, it actually runs the build process.

    To get a clear understanding of the environment variables, look at the .drv of the hello derivation:

    $ pp-aterm -i $(nix-instantiate hello.nix)

    Derive(

    [("out", "/nix/store/6flbdbpq6sc1dc79xjx01bz43zwgj3wc-hello", "", "")]

    , [("/nix/store/8z4xw8a0ax1csa0l83zflsm4jw9c94w2-bash-4.3-p39.drv", ["out"]), ("/nix/store/j0905apmxw2qb4ng5j40d4ghpiwa3mi1-stdenv.drv", ["out"])]

    , ["/nix/store/0q6pfasdma4as22kyaknk4kwx4h58480-hello-2.9.tar.gz", "/nix/store/9krlzvny65gdc8s7kpb6lkx8cd02c25b-default-builder.sh"]

    , "x86_64-linux"

    , "/nix/store/zmd4jk4db5lgxb8l93mhkvr3x92g2sx2-bash-4.3-p39/bin/bash"

    , ["-e", "/nix/store/9krlzvny65gdc8s7kpb6lkx8cd02c25b-default-builder.sh"]

    , [ ("buildInputs", "")

    , ("builder", "/nix/store/zmd4jk4db5lgxb8l93mhkvr3x92g2sx2-bash-4.3-p39/bin/bash")

    , ("name", "hello")

    , ("nativeBuildInputs", "")

    , ("out", "/nix/store/6flbdbpq6sc1dc79xjx01bz43zwgj3wc-hello")

    , ("propagatedBuildInputs", "")

    , ("propagatedNativeBuildInputs", "")

    , ("src", "/nix/store/0q6pfasdma4as22kyaknk4kwx4h58480-hello-2.9.tar.gz")

    , ("stdenv", "/nix/store/k4jklkcag4zq4xkqhkpy156mgfm34ipn-stdenv")

    , ("system", "x86_64-linux")

    ]

    )

    So short I decided to paste it entirely above. The builder is bash, with -e default-builder.sh arguments. Then you can see the src and stdenv environment variables.

    Last bit, the unpackPhase in the setup is used to unpack the sources and enter the directory, again like we did in our old builder.

    Conclusion

    The stdenv is the core of the nixpkgs repository. All packages use the stdenv.mkDerivation wrapper instead of the raw derivation. It does a bunch of operations for us and also sets up a pleasant build environment.

    The overall process is simple:

    nix-build

    bash -e default-builder.sh

    source $stdenv/setup

    genericBuild

    That's it, everything you need to know about the stdenv phases is in the setup file.

    Really, take your time to read that file. Don't forget that juicy docs are also available in the nixpkgs manual.

    Next pill...

    ...we will talk about how to add dependencies to our packages, buildInputs, propagatedBuildInputs and setup hooks. These three concepts are at the base of the current nixpkgs packages composition.