nix-super/src/libexpr/flake/flake.hh

146 lines
4.3 KiB
C++
Raw Normal View History

2019-06-04 20:45:16 +03:00
#pragma once
2019-02-12 19:23:11 +02:00
#include "types.hh"
#include "flakeref.hh"
#include "lockfile.hh"
2020-04-16 17:54:34 +03:00
#include "value.hh"
2019-02-12 19:23:11 +02:00
namespace nix {
class EvalState;
namespace fetchers { struct Tree; }
namespace flake {
2019-03-10 08:05:05 +02:00
Respect lock files of inputs + fine-grained lock file control When computing a lock file, we now respect the lock files of flake inputs. This is important for usability / reproducibility. For example, the 'nixops' flake depends on the 'nixops-aws' and 'nixops-hetzner' repositories. So when the 'nixops' flake is used in another flake, we want the versions of 'nixops-aws' and 'nixops-hetzner' locked by the the 'nixops' flake because those presumably have been tested. This can lead to a proliferation of versions of flakes like 'nixpkgs' (since every flake's lock file could depend on a different version of 'nixpkgs'). This is not a major issue when using Nixpkgs overlays or NixOS modules, since then the top-level flake composes those overlays/modules into *its* version of Nixpkgs and all other versions are ignored. Lock file computation has been made a bit more lazy so it won't try to fetch all those versions of 'nixpkgs'. However, in case it's necessary to minimize flake versions, there now are two input attributes that allow this. First, you can copy an input from another flake, as follows: inputs.nixpkgs.follows = "dwarffs/nixpkgs"; This states that the calling flake's 'nixpkgs' input shall be the same as the 'nixpkgs' input of the 'dwarffs' input. Second, you can override inputs of inputs: inputs.nixpkgs.url = github:edolstra/nixpkgs/<hash>; inputs.nixops.inputs.nixpkgs.url = github:edolstra/nixpkgs/<hash>; or equivalently, using 'follows': inputs.nixpkgs.url = github:edolstra/nixpkgs/<hash>; inputs.nixops.inputs.nixpkgs.follows = "nixpkgs"; This states that the 'nixpkgs' input of the 'nixops' input shall be the same as the calling flake's 'nixpkgs' input. Finally, at '-v' Nix now prints the changes to the lock file, e.g. $ nix flake update ~/Misc/eelco-configurations/hagbard inputs of flake 'git+file:///home/eelco/Misc/eelco-configurations?subdir=hagbard' changed: updated 'nixpkgs': 'github:edolstra/nixpkgs/7845bf5f4b3013df1cf036e9c9c3a55a30331db9' -> 'github:edolstra/nixpkgs/03f3def66a104a221aac8b751eeb7075374848fd' removed 'nixops' removed 'nixops/nixops-aws' removed 'nixops/nixops-hetzner' removed 'nixops/nixpkgs'
2020-01-24 23:05:11 +02:00
struct FlakeInput;
typedef std::map<FlakeId, FlakeInput> FlakeInputs;
/* FlakeInput is the 'Flake'-level parsed form of the "input" entries
* in the flake file.
*
* A FlakeInput is normally constructed by the 'parseFlakeInput'
* function which parses the input specification in the '.flake' file
* to create a 'FlakeRef' (a fetcher, the fetcher-specific
* representation of the input specification, and possibly the fetched
* local store path result) and then creating this FlakeInput to hold
* that FlakeRef, along with anything that might override that
* FlakeRef (like command-line overrides or "follows" specifications).
*
* A FlakeInput is also sometimes constructed directly from a FlakeRef
* instead of starting at the flake-file input specification
* (e.g. overrides, follows, and implicit inputs).
*
* A FlakeInput will usually have one of either "ref" or "follows"
* set. If not otherwise specified, a "ref" will be generated to a
* 'type="indirect"' flake, which is treated as simply the name of a
* flake to be resolved in the registry.
*/
struct FlakeInput
{
std::optional<FlakeRef> ref;
bool isFlake = true; // true = process flake to get outputs, false = (fetched) static source path
Respect lock files of inputs + fine-grained lock file control When computing a lock file, we now respect the lock files of flake inputs. This is important for usability / reproducibility. For example, the 'nixops' flake depends on the 'nixops-aws' and 'nixops-hetzner' repositories. So when the 'nixops' flake is used in another flake, we want the versions of 'nixops-aws' and 'nixops-hetzner' locked by the the 'nixops' flake because those presumably have been tested. This can lead to a proliferation of versions of flakes like 'nixpkgs' (since every flake's lock file could depend on a different version of 'nixpkgs'). This is not a major issue when using Nixpkgs overlays or NixOS modules, since then the top-level flake composes those overlays/modules into *its* version of Nixpkgs and all other versions are ignored. Lock file computation has been made a bit more lazy so it won't try to fetch all those versions of 'nixpkgs'. However, in case it's necessary to minimize flake versions, there now are two input attributes that allow this. First, you can copy an input from another flake, as follows: inputs.nixpkgs.follows = "dwarffs/nixpkgs"; This states that the calling flake's 'nixpkgs' input shall be the same as the 'nixpkgs' input of the 'dwarffs' input. Second, you can override inputs of inputs: inputs.nixpkgs.url = github:edolstra/nixpkgs/<hash>; inputs.nixops.inputs.nixpkgs.url = github:edolstra/nixpkgs/<hash>; or equivalently, using 'follows': inputs.nixpkgs.url = github:edolstra/nixpkgs/<hash>; inputs.nixops.inputs.nixpkgs.follows = "nixpkgs"; This states that the 'nixpkgs' input of the 'nixops' input shall be the same as the calling flake's 'nixpkgs' input. Finally, at '-v' Nix now prints the changes to the lock file, e.g. $ nix flake update ~/Misc/eelco-configurations/hagbard inputs of flake 'git+file:///home/eelco/Misc/eelco-configurations?subdir=hagbard' changed: updated 'nixpkgs': 'github:edolstra/nixpkgs/7845bf5f4b3013df1cf036e9c9c3a55a30331db9' -> 'github:edolstra/nixpkgs/03f3def66a104a221aac8b751eeb7075374848fd' removed 'nixops' removed 'nixops/nixops-aws' removed 'nixops/nixops-hetzner' removed 'nixops/nixpkgs'
2020-01-24 23:05:11 +02:00
std::optional<InputPath> follows;
FlakeInputs overrides;
};
struct ConfigFile
{
using ConfigValue = std::variant<std::string, int64_t, Explicit<bool>, std::vector<std::string>>;
std::map<std::string, ConfigValue> settings;
void apply();
};
/* The contents of a flake.nix file. */
2019-02-21 07:53:01 +02:00
struct Flake
{
FlakeRef originalRef; // the original flake specification (by the user)
FlakeRef resolvedRef; // registry references and caching resolved to the specific underlying flake
FlakeRef lockedRef; // the specific local store result of invoking the fetcher
std::optional<std::string> description;
std::shared_ptr<const fetchers::Tree> sourceInfo;
Respect lock files of inputs + fine-grained lock file control When computing a lock file, we now respect the lock files of flake inputs. This is important for usability / reproducibility. For example, the 'nixops' flake depends on the 'nixops-aws' and 'nixops-hetzner' repositories. So when the 'nixops' flake is used in another flake, we want the versions of 'nixops-aws' and 'nixops-hetzner' locked by the the 'nixops' flake because those presumably have been tested. This can lead to a proliferation of versions of flakes like 'nixpkgs' (since every flake's lock file could depend on a different version of 'nixpkgs'). This is not a major issue when using Nixpkgs overlays or NixOS modules, since then the top-level flake composes those overlays/modules into *its* version of Nixpkgs and all other versions are ignored. Lock file computation has been made a bit more lazy so it won't try to fetch all those versions of 'nixpkgs'. However, in case it's necessary to minimize flake versions, there now are two input attributes that allow this. First, you can copy an input from another flake, as follows: inputs.nixpkgs.follows = "dwarffs/nixpkgs"; This states that the calling flake's 'nixpkgs' input shall be the same as the 'nixpkgs' input of the 'dwarffs' input. Second, you can override inputs of inputs: inputs.nixpkgs.url = github:edolstra/nixpkgs/<hash>; inputs.nixops.inputs.nixpkgs.url = github:edolstra/nixpkgs/<hash>; or equivalently, using 'follows': inputs.nixpkgs.url = github:edolstra/nixpkgs/<hash>; inputs.nixops.inputs.nixpkgs.follows = "nixpkgs"; This states that the 'nixpkgs' input of the 'nixops' input shall be the same as the calling flake's 'nixpkgs' input. Finally, at '-v' Nix now prints the changes to the lock file, e.g. $ nix flake update ~/Misc/eelco-configurations/hagbard inputs of flake 'git+file:///home/eelco/Misc/eelco-configurations?subdir=hagbard' changed: updated 'nixpkgs': 'github:edolstra/nixpkgs/7845bf5f4b3013df1cf036e9c9c3a55a30331db9' -> 'github:edolstra/nixpkgs/03f3def66a104a221aac8b751eeb7075374848fd' removed 'nixops' removed 'nixops/nixops-aws' removed 'nixops/nixops-hetzner' removed 'nixops/nixpkgs'
2020-01-24 23:05:11 +02:00
FlakeInputs inputs;
ConfigFile config; // 'nixConfig' attribute
~Flake();
};
Flake getFlake(EvalState & state, const FlakeRef & flakeRef, bool allowLookup);
/* Fingerprint of a locked flake; used as a cache key. */
typedef Hash Fingerprint;
struct LockedFlake
{
2019-03-29 17:18:25 +02:00
Flake flake;
LockFile lockFile;
Fingerprint getFingerprint() const;
};
struct LockFlags
{
/* Whether to ignore the existing lock file, creating a new one
from scratch. */
bool recreateLockFile = false;
/* Whether to update the lock file at all. If set to false, if any
change to the lock file is needed (e.g. when an input has been
added to flake.nix), you get a fatal error. */
bool updateLockFile = true;
/* Whether to write the lock file to disk. If set to true, if the
any changes to the lock file are needed and the flake is not
writable (i.e. is not a local Git working tree or similar), you
get a fatal error. If set to false, Nix will use the modified
lock file in memory only, without writing it to disk. */
bool writeLockFile = true;
/* Whether to use the registries to lookup indirect flake
references like 'nixpkgs'. */
std::optional<bool> useRegistries = std::nullopt;
/* Whether to apply flake's nixConfig attribute to the configuration */
bool applyNixConfig = false;
/* Whether mutable flake references (i.e. those without a Git
revision or similar) without a corresponding lock are
allowed. Mutable flake references with a lock are always
allowed. */
bool allowMutable = true;
2020-02-05 15:48:49 +02:00
/* Whether to commit changes to flake.lock. */
bool commitLockFile = false;
2021-03-26 17:14:38 +02:00
/* Flake inputs to be overridden. */
std::map<InputPath, FlakeRef> inputOverrides;
/* Flake inputs to be updated. This means that any existing lock
for those inputs will be ignored. */
std::set<InputPath> inputUpdates;
};
LockedFlake lockFlake(
EvalState & state,
const FlakeRef & flakeRef,
const LockFlags & lockFlags);
2019-04-16 15:27:54 +03:00
void callFlake(
EvalState & state,
const LockedFlake & lockedFlake,
Value & v);
2019-02-12 19:23:11 +02:00
}
void emitTreeAttrs(
EvalState & state,
const fetchers::Tree & tree,
const fetchers::Input & input,
Value & v, bool emptyRevFallback = false);
}