Nix 2.93.3
Lix: A modern, delicious implementation of the Nix package manager; unstable internal interfaces
Loading...
Searching...
No Matches
nix::UnkeyedValidPathInfo Struct Reference
Inheritance diagram for nix::UnkeyedValidPathInfo:
nix::ValidPathInfo nix::NarInfo

Public Member Functions

 UnkeyedValidPathInfo (const UnkeyedValidPathInfo &other)=default
 UnkeyedValidPathInfo (Hash narHash)
 DECLARE_CMP (UnkeyedValidPathInfo)

Public Attributes

std::optional< StorePathderiver
Hash narHash
StorePathSet references
time_t registrationTime = 0
uint64_t narSize = 0
uint64_t id = 0
bool ultimate = false
StringSet sigs
std::optional< ContentAddressca

Member Data Documentation

◆ ca

std::optional<ContentAddress> nix::UnkeyedValidPathInfo::ca

If non-empty, an assertion that the path is content-addressed, i.e., that the store path is computed from a cryptographic hash of the contents of the path, plus some other bits of data like the "name" part of the path. Such a path doesn't need signatures, since we don't have to trust anybody's claim that the path is the output of a particular derivation. (In the extensional store model, we have to trust that the contents of an output path of a derivation were actually produced by that derivation. In the intensional model, we have to trust that a particular output path was produced by a derivation; the path then implies the contents.)

Ideally, the content-addressability assertion would just be a Boolean, and the store path would be computed from the name component, ‘narHash’ and ‘references’. However, we support many types of content addresses.

◆ narHash

Hash nix::UnkeyedValidPathInfo::narHash
Todo
document this

◆ ultimate

bool nix::UnkeyedValidPathInfo::ultimate = false

Whether the path is ultimately trusted, that is, it's a derivation output that was built locally.


The documentation for this struct was generated from the following file: