BoomMicrophone/nixpkgs-suyu
1
0
Fork 0
Code Issues Pull requests Projects Releases Packages Wiki Activity
nixpkgs-suyu/lib/default.nix

151 lines
6.7 KiB
Nix
Raw Normal View History

/lib/default.nix: Add file docstring & unit test hint
2017-03-30 22:03:40 +02:00
/* Library of low-level helper functions for nix expressions.
*
* Please implement (mostly) exhaustive unit tests
* for new functions in `./tests.nix'.
*/
lib/default.nix: reorder imports for clarity Group imports according to the kind of functions they contain, in a more descriptive manner.
2016-09-15 01:21:42 +02:00
let
* Put utility functions like fold in lib/default.nix. svn path=/nixpkgs/trunk/; revision=6605
2006-09-25 12:07:59 +02:00
lib: Add composeManyExtensions
2020-11-11 03:36:19 +01:00
inherit (import ./fixed-points.nix { inherit lib; }) makeExtensible;
Convert libs to a fixed-point This does break the API of being able to import any lib file and get its libs, however I'm not sure people did this. I made this while exploring being able to swap out docFn with a stub in #2305, to avoid functor performance problems. I don't know if that is going to move forward (or if it is a problem or not,) but after doing all this work figured I'd put it up anyway :) Two notable advantages to this approach: 1. when a lib inherits another lib's functions, it doesn't automatically get put in to the scope of lib 2. when a lib implements a new obscure functions, it doesn't automatically get put in to the scope of lib Using the test script (later in this commit) I got the following diff on the API: + diff master fixed-lib 11764a11765,11766 > .types.defaultFunctor > .types.defaultTypeMerge 11774a11777,11778 > .types.isOptionType > .types.isType 11781a11786 > .types.mkOptionType 11788a11794 > .types.setType 11795a11802 > .types.types This means that this commit _adds_ to the API, however I can't find a way to fix these last remaining discrepancies. At least none are _removed_. Test script (run with nix-repl in the PATH): #!/bin/sh set -eux repl() { suff=${1:-} echo "(import ./lib)$suff" \ | nix-repl 2>&1 } attrs_to_check() { repl "${1:-}" \ | tr ';' $'\n' \ | grep "\.\.\." \ | cut -d' ' -f2 \ | sed -e "s/^/${1:-}./" \ | sort } summ() { repl "${1:-}" \ | tr ' ' $'\n' \ | sort \ | uniq } deep_summ() { suff="${1:-}" depth="${2:-4}" depth=$((depth - 1)) summ "$suff" for attr in $(attrs_to_check "$suff" | grep -v "types.types"); do if [ $depth -eq 0 ]; then summ "$attr" | sed -e "s/^/$attr./" else deep_summ "$attr" "$depth" | sed -e "s/^/$attr./" fi done } ( cd nixpkgs #git add . #git commit -m "Auto-commit, sorry" || true git checkout fixed-lib deep_summ > ../fixed-lib git checkout master deep_summ > ../master ) if diff master fixed-lib; then echo "SHALLOW MATCH!" fi ( cd nixpkgs git checkout fixed-lib repl .types )
2017-07-29 02:05:35 +02:00
lib: make extensible This allows the lib fixed point to be extended with myLib = lib.extend (self: super: { foo = "foo"; }) With this it's possible to have the new modified lib attrset available to all modules when using evalModules myLib.evalModules { modules = [ ({ lib, ... }: { options.bar = lib.mkOption { default = lib.foo; }; }) ]; } => { config = { bar = "foo"; ... }; options = ...; }
2018-04-06 18:51:10 +02:00
lib = makeExtensible (self: let
callLibs = file: import file { lib = self; };
lib: Use Nix's static scope checking, fix error message, optimize Nix can perform static scope checking, but whenever code is inside a `with` expression, the analysis breaks down, because it can't know statically what's in the attribute set whose attributes were brought into scope. In those cases, Nix has to assume that everything works out. Except it doesnt. Removing `with` from lib/ revealed an undefined variable in an error message. If that doesn't convince you that we're better off without `with`, I can tell you that this PR results in a 3% evaluation performance improvement because Nix can look up local variables by index. This adds up with applications like the module system. Furthermore, removing `with` makes the binding site of each variable obvious, which helps with comprehension.
2020-10-20 13:47:24 +02:00
in {
Convert libs to a fixed-point This does break the API of being able to import any lib file and get its libs, however I'm not sure people did this. I made this while exploring being able to swap out docFn with a stub in #2305, to avoid functor performance problems. I don't know if that is going to move forward (or if it is a problem or not,) but after doing all this work figured I'd put it up anyway :) Two notable advantages to this approach: 1. when a lib inherits another lib's functions, it doesn't automatically get put in to the scope of lib 2. when a lib implements a new obscure functions, it doesn't automatically get put in to the scope of lib Using the test script (later in this commit) I got the following diff on the API: + diff master fixed-lib 11764a11765,11766 > .types.defaultFunctor > .types.defaultTypeMerge 11774a11777,11778 > .types.isOptionType > .types.isType 11781a11786 > .types.mkOptionType 11788a11794 > .types.setType 11795a11802 > .types.types This means that this commit _adds_ to the API, however I can't find a way to fix these last remaining discrepancies. At least none are _removed_. Test script (run with nix-repl in the PATH): #!/bin/sh set -eux repl() { suff=${1:-} echo "(import ./lib)$suff" \ | nix-repl 2>&1 } attrs_to_check() { repl "${1:-}" \ | tr ';' $'\n' \ | grep "\.\.\." \ | cut -d' ' -f2 \ | sed -e "s/^/${1:-}./" \ | sort } summ() { repl "${1:-}" \ | tr ' ' $'\n' \ | sort \ | uniq } deep_summ() { suff="${1:-}" depth="${2:-4}" depth=$((depth - 1)) summ "$suff" for attr in $(attrs_to_check "$suff" | grep -v "types.types"); do if [ $depth -eq 0 ]; then summ "$attr" | sed -e "s/^/$attr./" else deep_summ "$attr" "$depth" | sed -e "s/^/$attr./" fi done } ( cd nixpkgs #git add . #git commit -m "Auto-commit, sorry" || true git checkout fixed-lib deep_summ > ../fixed-lib git checkout master deep_summ > ../master ) if diff master fixed-lib; then echo "SHALLOW MATCH!" fi ( cd nixpkgs git checkout fixed-lib repl .types )
2017-07-29 02:05:35 +02:00
# often used, or depending on very little
trivial = callLibs ./trivial.nix;
fixedPoints = callLibs ./fixed-points.nix;
# datatypes
attrsets = callLibs ./attrsets.nix;
lists = callLibs ./lists.nix;
strings = callLibs ./strings.nix;
stringsWithDeps = callLibs ./strings-with-deps.nix;
# packaging
customisation = callLibs ./customisation.nix;
rename lib/maintainers-list.nix into maintainers/maintainer-list.nix Many commits unrelated to `lib` touch that file, this will make `git log ./lib` much saner. This is what I meant in https://github.com/NixOS/nixpkgs/pull/36119#issuecomment-370184101.
2018-03-04 04:09:35 +01:00
maintainers = import ../maintainers/maintainer-list.nix;
lib: Add lib.teams Also add a freedesktop maintainer group as an example.
2019-10-27 22:44:24 +01:00
teams = callLibs ../maintainers/team-list.nix;
Convert libs to a fixed-point This does break the API of being able to import any lib file and get its libs, however I'm not sure people did this. I made this while exploring being able to swap out docFn with a stub in #2305, to avoid functor performance problems. I don't know if that is going to move forward (or if it is a problem or not,) but after doing all this work figured I'd put it up anyway :) Two notable advantages to this approach: 1. when a lib inherits another lib's functions, it doesn't automatically get put in to the scope of lib 2. when a lib implements a new obscure functions, it doesn't automatically get put in to the scope of lib Using the test script (later in this commit) I got the following diff on the API: + diff master fixed-lib 11764a11765,11766 > .types.defaultFunctor > .types.defaultTypeMerge 11774a11777,11778 > .types.isOptionType > .types.isType 11781a11786 > .types.mkOptionType 11788a11794 > .types.setType 11795a11802 > .types.types This means that this commit _adds_ to the API, however I can't find a way to fix these last remaining discrepancies. At least none are _removed_. Test script (run with nix-repl in the PATH): #!/bin/sh set -eux repl() { suff=${1:-} echo "(import ./lib)$suff" \ | nix-repl 2>&1 } attrs_to_check() { repl "${1:-}" \ | tr ';' $'\n' \ | grep "\.\.\." \ | cut -d' ' -f2 \ | sed -e "s/^/${1:-}./" \ | sort } summ() { repl "${1:-}" \ | tr ' ' $'\n' \ | sort \ | uniq } deep_summ() { suff="${1:-}" depth="${2:-4}" depth=$((depth - 1)) summ "$suff" for attr in $(attrs_to_check "$suff" | grep -v "types.types"); do if [ $depth -eq 0 ]; then summ "$attr" | sed -e "s/^/$attr./" else deep_summ "$attr" "$depth" | sed -e "s/^/$attr./" fi done } ( cd nixpkgs #git add . #git commit -m "Auto-commit, sorry" || true git checkout fixed-lib deep_summ > ../fixed-lib git checkout master deep_summ > ../master ) if diff master fixed-lib; then echo "SHALLOW MATCH!" fi ( cd nixpkgs git checkout fixed-lib repl .types )
2017-07-29 02:05:35 +02:00
meta = callLibs ./meta.nix;
sources = callLibs ./sources.nix;
lib: add versions library
2018-03-03 05:26:32 +01:00
versions = callLibs ./versions.nix;
Convert libs to a fixed-point This does break the API of being able to import any lib file and get its libs, however I'm not sure people did this. I made this while exploring being able to swap out docFn with a stub in #2305, to avoid functor performance problems. I don't know if that is going to move forward (or if it is a problem or not,) but after doing all this work figured I'd put it up anyway :) Two notable advantages to this approach: 1. when a lib inherits another lib's functions, it doesn't automatically get put in to the scope of lib 2. when a lib implements a new obscure functions, it doesn't automatically get put in to the scope of lib Using the test script (later in this commit) I got the following diff on the API: + diff master fixed-lib 11764a11765,11766 > .types.defaultFunctor > .types.defaultTypeMerge 11774a11777,11778 > .types.isOptionType > .types.isType 11781a11786 > .types.mkOptionType 11788a11794 > .types.setType 11795a11802 > .types.types This means that this commit _adds_ to the API, however I can't find a way to fix these last remaining discrepancies. At least none are _removed_. Test script (run with nix-repl in the PATH): #!/bin/sh set -eux repl() { suff=${1:-} echo "(import ./lib)$suff" \ | nix-repl 2>&1 } attrs_to_check() { repl "${1:-}" \ | tr ';' $'\n' \ | grep "\.\.\." \ | cut -d' ' -f2 \ | sed -e "s/^/${1:-}./" \ | sort } summ() { repl "${1:-}" \ | tr ' ' $'\n' \ | sort \ | uniq } deep_summ() { suff="${1:-}" depth="${2:-4}" depth=$((depth - 1)) summ "$suff" for attr in $(attrs_to_check "$suff" | grep -v "types.types"); do if [ $depth -eq 0 ]; then summ "$attr" | sed -e "s/^/$attr./" else deep_summ "$attr" "$depth" | sed -e "s/^/$attr./" fi done } ( cd nixpkgs #git add . #git commit -m "Auto-commit, sorry" || true git checkout fixed-lib deep_summ > ../fixed-lib git checkout master deep_summ > ../master ) if diff master fixed-lib; then echo "SHALLOW MATCH!" fi ( cd nixpkgs git checkout fixed-lib repl .types )
2017-07-29 02:05:35 +02:00
# module system
modules = callLibs ./modules.nix;
options = callLibs ./options.nix;
types = callLibs ./types.nix;
# constants
licenses = callLibs ./licenses.nix;
systems = callLibs ./systems;
lib/cli: unexport symbols & sort with generators lib/cli is very similar to generators, so it should follow largely the same interface. Similar to how generators isn’t exported, we should also namespace cli by default (plus “cli” is only three characters to type).
2020-01-22 23:22:23 +01:00
# serialization
cli = callLibs ./cli.nix;
generators = callLibs ./generators.nix;
Convert libs to a fixed-point This does break the API of being able to import any lib file and get its libs, however I'm not sure people did this. I made this while exploring being able to swap out docFn with a stub in #2305, to avoid functor performance problems. I don't know if that is going to move forward (or if it is a problem or not,) but after doing all this work figured I'd put it up anyway :) Two notable advantages to this approach: 1. when a lib inherits another lib's functions, it doesn't automatically get put in to the scope of lib 2. when a lib implements a new obscure functions, it doesn't automatically get put in to the scope of lib Using the test script (later in this commit) I got the following diff on the API: + diff master fixed-lib 11764a11765,11766 > .types.defaultFunctor > .types.defaultTypeMerge 11774a11777,11778 > .types.isOptionType > .types.isType 11781a11786 > .types.mkOptionType 11788a11794 > .types.setType 11795a11802 > .types.types This means that this commit _adds_ to the API, however I can't find a way to fix these last remaining discrepancies. At least none are _removed_. Test script (run with nix-repl in the PATH): #!/bin/sh set -eux repl() { suff=${1:-} echo "(import ./lib)$suff" \ | nix-repl 2>&1 } attrs_to_check() { repl "${1:-}" \ | tr ';' $'\n' \ | grep "\.\.\." \ | cut -d' ' -f2 \ | sed -e "s/^/${1:-}./" \ | sort } summ() { repl "${1:-}" \ | tr ' ' $'\n' \ | sort \ | uniq } deep_summ() { suff="${1:-}" depth="${2:-4}" depth=$((depth - 1)) summ "$suff" for attr in $(attrs_to_check "$suff" | grep -v "types.types"); do if [ $depth -eq 0 ]; then summ "$attr" | sed -e "s/^/$attr./" else deep_summ "$attr" "$depth" | sed -e "s/^/$attr./" fi done } ( cd nixpkgs #git add . #git commit -m "Auto-commit, sorry" || true git checkout fixed-lib deep_summ > ../fixed-lib git checkout master deep_summ > ../master ) if diff master fixed-lib; then echo "SHALLOW MATCH!" fi ( cd nixpkgs git checkout fixed-lib repl .types )
2017-07-29 02:05:35 +02:00
# misc
lib: move assertMsg and assertOneOf to their own library file Since the `assertOneOf` uses `lib.generators`, they are not really trivial anymore and should go into their own library file.
2018-08-08 19:26:52 +02:00
asserts = callLibs ./asserts.nix;
Convert libs to a fixed-point This does break the API of being able to import any lib file and get its libs, however I'm not sure people did this. I made this while exploring being able to swap out docFn with a stub in #2305, to avoid functor performance problems. I don't know if that is going to move forward (or if it is a problem or not,) but after doing all this work figured I'd put it up anyway :) Two notable advantages to this approach: 1. when a lib inherits another lib's functions, it doesn't automatically get put in to the scope of lib 2. when a lib implements a new obscure functions, it doesn't automatically get put in to the scope of lib Using the test script (later in this commit) I got the following diff on the API: + diff master fixed-lib 11764a11765,11766 > .types.defaultFunctor > .types.defaultTypeMerge 11774a11777,11778 > .types.isOptionType > .types.isType 11781a11786 > .types.mkOptionType 11788a11794 > .types.setType 11795a11802 > .types.types This means that this commit _adds_ to the API, however I can't find a way to fix these last remaining discrepancies. At least none are _removed_. Test script (run with nix-repl in the PATH): #!/bin/sh set -eux repl() { suff=${1:-} echo "(import ./lib)$suff" \ | nix-repl 2>&1 } attrs_to_check() { repl "${1:-}" \ | tr ';' $'\n' \ | grep "\.\.\." \ | cut -d' ' -f2 \ | sed -e "s/^/${1:-}./" \ | sort } summ() { repl "${1:-}" \ | tr ' ' $'\n' \ | sort \ | uniq } deep_summ() { suff="${1:-}" depth="${2:-4}" depth=$((depth - 1)) summ "$suff" for attr in $(attrs_to_check "$suff" | grep -v "types.types"); do if [ $depth -eq 0 ]; then summ "$attr" | sed -e "s/^/$attr./" else deep_summ "$attr" "$depth" | sed -e "s/^/$attr./" fi done } ( cd nixpkgs #git add . #git commit -m "Auto-commit, sorry" || true git checkout fixed-lib deep_summ > ../fixed-lib git checkout master deep_summ > ../master ) if diff master fixed-lib; then echo "SHALLOW MATCH!" fi ( cd nixpkgs git checkout fixed-lib repl .types )
2017-07-29 02:05:35 +02:00
debug = callLibs ./debug.nix;
misc = callLibs ./deprecated.nix;
lib: move assertMsg and assertOneOf to their own library file Since the `assertOneOf` uses `lib.generators`, they are not really trivial anymore and should go into their own library file.
2018-08-08 19:26:52 +02:00
Convert libs to a fixed-point This does break the API of being able to import any lib file and get its libs, however I'm not sure people did this. I made this while exploring being able to swap out docFn with a stub in #2305, to avoid functor performance problems. I don't know if that is going to move forward (or if it is a problem or not,) but after doing all this work figured I'd put it up anyway :) Two notable advantages to this approach: 1. when a lib inherits another lib's functions, it doesn't automatically get put in to the scope of lib 2. when a lib implements a new obscure functions, it doesn't automatically get put in to the scope of lib Using the test script (later in this commit) I got the following diff on the API: + diff master fixed-lib 11764a11765,11766 > .types.defaultFunctor > .types.defaultTypeMerge 11774a11777,11778 > .types.isOptionType > .types.isType 11781a11786 > .types.mkOptionType 11788a11794 > .types.setType 11795a11802 > .types.types This means that this commit _adds_ to the API, however I can't find a way to fix these last remaining discrepancies. At least none are _removed_. Test script (run with nix-repl in the PATH): #!/bin/sh set -eux repl() { suff=${1:-} echo "(import ./lib)$suff" \ | nix-repl 2>&1 } attrs_to_check() { repl "${1:-}" \ | tr ';' $'\n' \ | grep "\.\.\." \ | cut -d' ' -f2 \ | sed -e "s/^/${1:-}./" \ | sort } summ() { repl "${1:-}" \ | tr ' ' $'\n' \ | sort \ | uniq } deep_summ() { suff="${1:-}" depth="${2:-4}" depth=$((depth - 1)) summ "$suff" for attr in $(attrs_to_check "$suff" | grep -v "types.types"); do if [ $depth -eq 0 ]; then summ "$attr" | sed -e "s/^/$attr./" else deep_summ "$attr" "$depth" | sed -e "s/^/$attr./" fi done } ( cd nixpkgs #git add . #git commit -m "Auto-commit, sorry" || true git checkout fixed-lib deep_summ > ../fixed-lib git checkout master deep_summ > ../master ) if diff master fixed-lib; then echo "SHALLOW MATCH!" fi ( cd nixpkgs git checkout fixed-lib repl .types )
2017-07-29 02:05:35 +02:00
# domain-specific
fetchers = callLibs ./fetchers.nix;
# Eval-time filesystem handling
filesystem = callLibs ./filesystem.nix;
lib: Collect system/platform related files Previously, platforms was a random thing in top-level
2017-02-09 03:27:22 +01:00
# back-compat aliases
lib: Use Nix's static scope checking, fix error message, optimize Nix can perform static scope checking, but whenever code is inside a `with` expression, the analysis breaks down, because it can't know statically what's in the attribute set whose attributes were brought into scope. In those cases, Nix has to assume that everything works out. Except it doesnt. Removing `with` from lib/ revealed an undefined variable in an error message. If that doesn't convince you that we're better off without `with`, I can tell you that this PR results in a 3% evaluation performance improvement because Nix can look up local variables by index. This adds up with applications like the module system. Furthermore, removing `with` makes the binding site of each variable obvious, which helps with comprehension.
2020-10-20 13:47:24 +02:00
platforms = self.systems.doubles;
Convert libs to a fixed-point This does break the API of being able to import any lib file and get its libs, however I'm not sure people did this. I made this while exploring being able to swap out docFn with a stub in #2305, to avoid functor performance problems. I don't know if that is going to move forward (or if it is a problem or not,) but after doing all this work figured I'd put it up anyway :) Two notable advantages to this approach: 1. when a lib inherits another lib's functions, it doesn't automatically get put in to the scope of lib 2. when a lib implements a new obscure functions, it doesn't automatically get put in to the scope of lib Using the test script (later in this commit) I got the following diff on the API: + diff master fixed-lib 11764a11765,11766 > .types.defaultFunctor > .types.defaultTypeMerge 11774a11777,11778 > .types.isOptionType > .types.isType 11781a11786 > .types.mkOptionType 11788a11794 > .types.setType 11795a11802 > .types.types This means that this commit _adds_ to the API, however I can't find a way to fix these last remaining discrepancies. At least none are _removed_. Test script (run with nix-repl in the PATH): #!/bin/sh set -eux repl() { suff=${1:-} echo "(import ./lib)$suff" \ | nix-repl 2>&1 } attrs_to_check() { repl "${1:-}" \ | tr ';' $'\n' \ | grep "\.\.\." \ | cut -d' ' -f2 \ | sed -e "s/^/${1:-}./" \ | sort } summ() { repl "${1:-}" \ | tr ' ' $'\n' \ | sort \ | uniq } deep_summ() { suff="${1:-}" depth="${2:-4}" depth=$((depth - 1)) summ "$suff" for attr in $(attrs_to_check "$suff" | grep -v "types.types"); do if [ $depth -eq 0 ]; then summ "$attr" | sed -e "s/^/$attr./" else deep_summ "$attr" "$depth" | sed -e "s/^/$attr./" fi done } ( cd nixpkgs #git add . #git commit -m "Auto-commit, sorry" || true git checkout fixed-lib deep_summ > ../fixed-lib git checkout master deep_summ > ../master ) if diff master fixed-lib; then echo "SHALLOW MATCH!" fi ( cd nixpkgs git checkout fixed-lib repl .types )
2017-07-29 02:05:35 +02:00
lib.kernel: make public Remove the "version" parameter in order to make it more widely available. Starts making some kernel configuration helpers available. The intent is to be able to better build and check the linux kernel configuration.
2019-09-18 10:34:02 +02:00
# linux kernel configuration
kernel = callLibs ./kernel.nix;
lib: Add more configure flag helpers Add with/without to match enable/disable, and add `--{enable,with}-key=value` versions of both.
2018-05-11 17:12:15 +02:00
inherit (builtins) add addErrorContext attrNames concatLists
deepSeq elem elemAt filter genericClosure genList getAttr
hasAttr head isAttrs isBool isInt isList isString length
lessThan listToAttrs pathExists readFile replaceStrings seq
lib: Add lib.trace for consistency This puts it among the trace* family of functions derived from it.
2020-10-20 16:34:15 +02:00
stringLength sub substring tail trace;
lib: Use Nix's static scope checking, fix error message, optimize Nix can perform static scope checking, but whenever code is inside a `with` expression, the analysis breaks down, because it can't know statically what's in the attribute set whose attributes were brought into scope. In those cases, Nix has to assume that everything works out. Except it doesnt. Removing `with` from lib/ revealed an undefined variable in an error message. If that doesn't convince you that we're better off without `with`, I can tell you that this PR results in a 3% evaluation performance improvement because Nix can look up local variables by index. This adds up with applications like the module system. Furthermore, removing `with` makes the binding site of each variable obvious, which helps with comprehension.
2020-10-20 13:47:24 +02:00
inherit (self.trivial) id const pipe concat or and bitAnd bitOr bitXor
lib: Add lib.isFloat for consistency Unlike the other three is* functions in lib.trivial, it was only available as lib.trivial.isFloat
2020-10-20 15:24:59 +02:00
bitNot boolToString mergeAttrs flip mapNullable inNixShell isFloat min max
lib.trivial.warnIf: init It's a common pattern in Nixpkgs to want to emit a warning in certain cases, but not actually change behaviours. This is often expressed as either if cond then lib.warn "Don't do that thing" x else x Or (if cond then lib.warn "Don't do that thing" else lib.id) x Neither of which really expresses the intent here, because it looks like 'x' is being chosen conditionally. To make this clearer, I introduce a "warnIf" function, which makes it clear that the only thing being affected by the condition is whether the warning is generated, not the value being returned.
2021-04-27 12:56:51 +02:00
importJSON importTOML warn warnIf info showWarnings nixpkgsVersion version
mod compare splitByAndCompare functionArgs setFunctionArgs isFunction
toHexString toBaseDigits;
lib: Use Nix's static scope checking, fix error message, optimize Nix can perform static scope checking, but whenever code is inside a `with` expression, the analysis breaks down, because it can't know statically what's in the attribute set whose attributes were brought into scope. In those cases, Nix has to assume that everything works out. Except it doesnt. Removing `with` from lib/ revealed an undefined variable in an error message. If that doesn't convince you that we're better off without `with`, I can tell you that this PR results in a 3% evaluation performance improvement because Nix can look up local variables by index. This adds up with applications like the module system. Furthermore, removing `with` makes the binding site of each variable obvious, which helps with comprehension.
2020-10-20 13:47:24 +02:00
inherit (self.fixedPoints) fix fix' converge extends composeExtensions
lib: Add composeManyExtensions
2020-11-11 03:36:19 +01:00
composeManyExtensions makeExtensible makeExtensibleWithCustomName;
lib: Use Nix's static scope checking, fix error message, optimize Nix can perform static scope checking, but whenever code is inside a `with` expression, the analysis breaks down, because it can't know statically what's in the attribute set whose attributes were brought into scope. In those cases, Nix has to assume that everything works out. Except it doesnt. Removing `with` from lib/ revealed an undefined variable in an error message. If that doesn't convince you that we're better off without `with`, I can tell you that this PR results in a 3% evaluation performance improvement because Nix can look up local variables by index. This adds up with applications like the module system. Furthermore, removing `with` makes the binding site of each variable obvious, which helps with comprehension.
2020-10-20 13:47:24 +02:00
inherit (self.attrsets) attrByPath hasAttrByPath setAttrByPath
bundlerEnv: ensure dependencies always included Suppose I have a Gemfile like this: source "https://rubygems.org" gem "actioncable" gem "websocket-driver", group: :test The gemset.nix generated by Bundix 2.4.1 will set ActionCable's groups to [ "default" ], and websocket-driver's to [ "test" ]. This means that the generated bundlerEnv wouldn't include websocket-driver unless the test group was included, even though it's required by the default group. This is arguably a bug in Bundix (websocket-driver's groups should probably be [ "default" "test" ] or just [ "default" ]), but there's no reason bundlerEnv should omit dependencies even given such an input -- it won't necessarily come from Bundix, and it would be good for bundlerEnv to do the right thing. To fix this, filterGemset is now a recursive function, that adds dependencies of gems in the group to the filtered gemset until it stabilises on the gems that match the required groups, and all of their recursive dependencies.
2018-12-11 22:18:22 +01:00
getAttrFromPath attrVals attrValues getAttrs catAttrs filterAttrs
Convert libs to a fixed-point This does break the API of being able to import any lib file and get its libs, however I'm not sure people did this. I made this while exploring being able to swap out docFn with a stub in #2305, to avoid functor performance problems. I don't know if that is going to move forward (or if it is a problem or not,) but after doing all this work figured I'd put it up anyway :) Two notable advantages to this approach: 1. when a lib inherits another lib's functions, it doesn't automatically get put in to the scope of lib 2. when a lib implements a new obscure functions, it doesn't automatically get put in to the scope of lib Using the test script (later in this commit) I got the following diff on the API: + diff master fixed-lib 11764a11765,11766 > .types.defaultFunctor > .types.defaultTypeMerge 11774a11777,11778 > .types.isOptionType > .types.isType 11781a11786 > .types.mkOptionType 11788a11794 > .types.setType 11795a11802 > .types.types This means that this commit _adds_ to the API, however I can't find a way to fix these last remaining discrepancies. At least none are _removed_. Test script (run with nix-repl in the PATH): #!/bin/sh set -eux repl() { suff=${1:-} echo "(import ./lib)$suff" \ | nix-repl 2>&1 } attrs_to_check() { repl "${1:-}" \ | tr ';' $'\n' \ | grep "\.\.\." \ | cut -d' ' -f2 \ | sed -e "s/^/${1:-}./" \ | sort } summ() { repl "${1:-}" \ | tr ' ' $'\n' \ | sort \ | uniq } deep_summ() { suff="${1:-}" depth="${2:-4}" depth=$((depth - 1)) summ "$suff" for attr in $(attrs_to_check "$suff" | grep -v "types.types"); do if [ $depth -eq 0 ]; then summ "$attr" | sed -e "s/^/$attr./" else deep_summ "$attr" "$depth" | sed -e "s/^/$attr./" fi done } ( cd nixpkgs #git add . #git commit -m "Auto-commit, sorry" || true git checkout fixed-lib deep_summ > ../fixed-lib git checkout master deep_summ > ../master ) if diff master fixed-lib; then echo "SHALLOW MATCH!" fi ( cd nixpkgs git checkout fixed-lib repl .types )
2017-07-29 02:05:35 +02:00
filterAttrsRecursive foldAttrs collect nameValuePair mapAttrs
mapAttrs' mapAttrsToList mapAttrsRecursive mapAttrsRecursiveCond
genAttrs isDerivation toDerivation optionalAttrs
zipAttrsWithNames zipAttrsWith zipAttrs recursiveUpdateUntil
recursiveUpdate matchAttrs overrideExisting getOutput getBin
lib/attrsets: add getMan function
2020-05-11 11:08:29 +02:00
getLib getDev getMan chooseDevOutputs zipWithNames zip
lib/attrsets: add cartesianProductOfSets function
2021-01-25 16:59:46 +01:00
recurseIntoAttrs dontRecurseIntoAttrs cartesianProductOfSets;
lib: Use Nix's static scope checking, fix error message, optimize Nix can perform static scope checking, but whenever code is inside a `with` expression, the analysis breaks down, because it can't know statically what's in the attribute set whose attributes were brought into scope. In those cases, Nix has to assume that everything works out. Except it doesnt. Removing `with` from lib/ revealed an undefined variable in an error message. If that doesn't convince you that we're better off without `with`, I can tell you that this PR results in a 3% evaluation performance improvement because Nix can look up local variables by index. This adds up with applications like the module system. Furthermore, removing `with` makes the binding site of each variable obvious, which helps with comprehension.
2020-10-20 13:47:24 +02:00
inherit (self.lists) singleton forEach foldr fold foldl foldl' imap0 imap1
Convert libs to a fixed-point This does break the API of being able to import any lib file and get its libs, however I'm not sure people did this. I made this while exploring being able to swap out docFn with a stub in #2305, to avoid functor performance problems. I don't know if that is going to move forward (or if it is a problem or not,) but after doing all this work figured I'd put it up anyway :) Two notable advantages to this approach: 1. when a lib inherits another lib's functions, it doesn't automatically get put in to the scope of lib 2. when a lib implements a new obscure functions, it doesn't automatically get put in to the scope of lib Using the test script (later in this commit) I got the following diff on the API: + diff master fixed-lib 11764a11765,11766 > .types.defaultFunctor > .types.defaultTypeMerge 11774a11777,11778 > .types.isOptionType > .types.isType 11781a11786 > .types.mkOptionType 11788a11794 > .types.setType 11795a11802 > .types.types This means that this commit _adds_ to the API, however I can't find a way to fix these last remaining discrepancies. At least none are _removed_. Test script (run with nix-repl in the PATH): #!/bin/sh set -eux repl() { suff=${1:-} echo "(import ./lib)$suff" \ | nix-repl 2>&1 } attrs_to_check() { repl "${1:-}" \ | tr ';' $'\n' \ | grep "\.\.\." \ | cut -d' ' -f2 \ | sed -e "s/^/${1:-}./" \ | sort } summ() { repl "${1:-}" \ | tr ' ' $'\n' \ | sort \ | uniq } deep_summ() { suff="${1:-}" depth="${2:-4}" depth=$((depth - 1)) summ "$suff" for attr in $(attrs_to_check "$suff" | grep -v "types.types"); do if [ $depth -eq 0 ]; then summ "$attr" | sed -e "s/^/$attr./" else deep_summ "$attr" "$depth" | sed -e "s/^/$attr./" fi done } ( cd nixpkgs #git add . #git commit -m "Auto-commit, sorry" || true git checkout fixed-lib deep_summ > ../fixed-lib git checkout master deep_summ > ../master ) if diff master fixed-lib; then echo "SHALLOW MATCH!" fi ( cd nixpkgs git checkout fixed-lib repl .types )
2017-07-29 02:05:35 +02:00
concatMap flatten remove findSingle findFirst any all count
optional optionals toList range partition zipListsWith zipLists
lib: Add more configure flag helpers Add with/without to match enable/disable, and add `--{enable,with}-key=value` versions of both.
2018-05-11 17:12:15 +02:00
reverseList listDfs toposort sort naturalSort compareLists take
drop sublist last init crossLists unique intersectLists
lib: add groupBy (#38612)
2018-06-10 19:31:09 +02:00
subtractLists mutuallyExclusive groupBy groupBy';
lib: Use Nix's static scope checking, fix error message, optimize Nix can perform static scope checking, but whenever code is inside a `with` expression, the analysis breaks down, because it can't know statically what's in the attribute set whose attributes were brought into scope. In those cases, Nix has to assume that everything works out. Except it doesnt. Removing `with` from lib/ revealed an undefined variable in an error message. If that doesn't convince you that we're better off without `with`, I can tell you that this PR results in a 3% evaluation performance improvement because Nix can look up local variables by index. This adds up with applications like the module system. Furthermore, removing `with` makes the binding site of each variable obvious, which helps with comprehension.
2020-10-20 13:47:24 +02:00
inherit (self.strings) concatStrings concatMapStrings concatImapStrings
Convert libs to a fixed-point This does break the API of being able to import any lib file and get its libs, however I'm not sure people did this. I made this while exploring being able to swap out docFn with a stub in #2305, to avoid functor performance problems. I don't know if that is going to move forward (or if it is a problem or not,) but after doing all this work figured I'd put it up anyway :) Two notable advantages to this approach: 1. when a lib inherits another lib's functions, it doesn't automatically get put in to the scope of lib 2. when a lib implements a new obscure functions, it doesn't automatically get put in to the scope of lib Using the test script (later in this commit) I got the following diff on the API: + diff master fixed-lib 11764a11765,11766 > .types.defaultFunctor > .types.defaultTypeMerge 11774a11777,11778 > .types.isOptionType > .types.isType 11781a11786 > .types.mkOptionType 11788a11794 > .types.setType 11795a11802 > .types.types This means that this commit _adds_ to the API, however I can't find a way to fix these last remaining discrepancies. At least none are _removed_. Test script (run with nix-repl in the PATH): #!/bin/sh set -eux repl() { suff=${1:-} echo "(import ./lib)$suff" \ | nix-repl 2>&1 } attrs_to_check() { repl "${1:-}" \ | tr ';' $'\n' \ | grep "\.\.\." \ | cut -d' ' -f2 \ | sed -e "s/^/${1:-}./" \ | sort } summ() { repl "${1:-}" \ | tr ' ' $'\n' \ | sort \ | uniq } deep_summ() { suff="${1:-}" depth="${2:-4}" depth=$((depth - 1)) summ "$suff" for attr in $(attrs_to_check "$suff" | grep -v "types.types"); do if [ $depth -eq 0 ]; then summ "$attr" | sed -e "s/^/$attr./" else deep_summ "$attr" "$depth" | sed -e "s/^/$attr./" fi done } ( cd nixpkgs #git add . #git commit -m "Auto-commit, sorry" || true git checkout fixed-lib deep_summ > ../fixed-lib git checkout master deep_summ > ../master ) if diff master fixed-lib; then echo "SHALLOW MATCH!" fi ( cd nixpkgs git checkout fixed-lib repl .types )
2017-07-29 02:05:35 +02:00
intersperse concatStringsSep concatMapStringsSep
concatImapStringsSep makeSearchPath makeSearchPathOutput
lib.makePerlPath -> perlPackages.makePerlPath
2018-12-15 04:50:31 +01:00
makeLibraryPath makeBinPath optionalString
python: Make .isPyPy flag more accurate. nix-repl> map (s: s.isPyPy) [ python python3 pypy pypy3 ] [ false false true true ]
2019-02-04 09:09:38 +01:00
hasInfix hasPrefix hasSuffix stringToCharacters stringAsChars escape
lib: Add more configure flag helpers Add with/without to match enable/disable, and add `--{enable,with}-key=value` versions of both.
2018-05-11 17:12:15 +02:00
escapeShellArg escapeShellArgs replaceChars lowerChars
upperChars toLower toUpper addContextFrom splitString
lib: Add getName to mirror getVersion
2019-11-24 17:19:32 +01:00
removePrefix removeSuffix versionOlder versionAtLeast
getName getVersion
lib: Add more configure flag helpers Add with/without to match enable/disable, and add `--{enable,with}-key=value` versions of both.
2018-05-11 17:12:15 +02:00
nameFromURL enableFeature enableFeatureAs withFeature
withFeatureAs fixedWidthString fixedWidthNumber isStorePath
Convert libs to a fixed-point This does break the API of being able to import any lib file and get its libs, however I'm not sure people did this. I made this while exploring being able to swap out docFn with a stub in #2305, to avoid functor performance problems. I don't know if that is going to move forward (or if it is a problem or not,) but after doing all this work figured I'd put it up anyway :) Two notable advantages to this approach: 1. when a lib inherits another lib's functions, it doesn't automatically get put in to the scope of lib 2. when a lib implements a new obscure functions, it doesn't automatically get put in to the scope of lib Using the test script (later in this commit) I got the following diff on the API: + diff master fixed-lib 11764a11765,11766 > .types.defaultFunctor > .types.defaultTypeMerge 11774a11777,11778 > .types.isOptionType > .types.isType 11781a11786 > .types.mkOptionType 11788a11794 > .types.setType 11795a11802 > .types.types This means that this commit _adds_ to the API, however I can't find a way to fix these last remaining discrepancies. At least none are _removed_. Test script (run with nix-repl in the PATH): #!/bin/sh set -eux repl() { suff=${1:-} echo "(import ./lib)$suff" \ | nix-repl 2>&1 } attrs_to_check() { repl "${1:-}" \ | tr ';' $'\n' \ | grep "\.\.\." \ | cut -d' ' -f2 \ | sed -e "s/^/${1:-}./" \ | sort } summ() { repl "${1:-}" \ | tr ' ' $'\n' \ | sort \ | uniq } deep_summ() { suff="${1:-}" depth="${2:-4}" depth=$((depth - 1)) summ "$suff" for attr in $(attrs_to_check "$suff" | grep -v "types.types"); do if [ $depth -eq 0 ]; then summ "$attr" | sed -e "s/^/$attr./" else deep_summ "$attr" "$depth" | sed -e "s/^/$attr./" fi done } ( cd nixpkgs #git add . #git commit -m "Auto-commit, sorry" || true git checkout fixed-lib deep_summ > ../fixed-lib git checkout master deep_summ > ../master ) if diff master fixed-lib; then echo "SHALLOW MATCH!" fi ( cd nixpkgs git checkout fixed-lib repl .types )
2017-07-29 02:05:35 +02:00
toInt readPathsFromFile fileContents;
lib: Use Nix's static scope checking, fix error message, optimize Nix can perform static scope checking, but whenever code is inside a `with` expression, the analysis breaks down, because it can't know statically what's in the attribute set whose attributes were brought into scope. In those cases, Nix has to assume that everything works out. Except it doesnt. Removing `with` from lib/ revealed an undefined variable in an error message. If that doesn't convince you that we're better off without `with`, I can tell you that this PR results in a 3% evaluation performance improvement because Nix can look up local variables by index. This adds up with applications like the module system. Furthermore, removing `with` makes the binding site of each variable obvious, which helps with comprehension.
2020-10-20 13:47:24 +02:00
inherit (self.stringsWithDeps) textClosureList textClosureMap
Convert libs to a fixed-point This does break the API of being able to import any lib file and get its libs, however I'm not sure people did this. I made this while exploring being able to swap out docFn with a stub in #2305, to avoid functor performance problems. I don't know if that is going to move forward (or if it is a problem or not,) but after doing all this work figured I'd put it up anyway :) Two notable advantages to this approach: 1. when a lib inherits another lib's functions, it doesn't automatically get put in to the scope of lib 2. when a lib implements a new obscure functions, it doesn't automatically get put in to the scope of lib Using the test script (later in this commit) I got the following diff on the API: + diff master fixed-lib 11764a11765,11766 > .types.defaultFunctor > .types.defaultTypeMerge 11774a11777,11778 > .types.isOptionType > .types.isType 11781a11786 > .types.mkOptionType 11788a11794 > .types.setType 11795a11802 > .types.types This means that this commit _adds_ to the API, however I can't find a way to fix these last remaining discrepancies. At least none are _removed_. Test script (run with nix-repl in the PATH): #!/bin/sh set -eux repl() { suff=${1:-} echo "(import ./lib)$suff" \ | nix-repl 2>&1 } attrs_to_check() { repl "${1:-}" \ | tr ';' $'\n' \ | grep "\.\.\." \ | cut -d' ' -f2 \ | sed -e "s/^/${1:-}./" \ | sort } summ() { repl "${1:-}" \ | tr ' ' $'\n' \ | sort \ | uniq } deep_summ() { suff="${1:-}" depth="${2:-4}" depth=$((depth - 1)) summ "$suff" for attr in $(attrs_to_check "$suff" | grep -v "types.types"); do if [ $depth -eq 0 ]; then summ "$attr" | sed -e "s/^/$attr./" else deep_summ "$attr" "$depth" | sed -e "s/^/$attr./" fi done } ( cd nixpkgs #git add . #git commit -m "Auto-commit, sorry" || true git checkout fixed-lib deep_summ > ../fixed-lib git checkout master deep_summ > ../master ) if diff master fixed-lib; then echo "SHALLOW MATCH!" fi ( cd nixpkgs git checkout fixed-lib repl .types )
2017-07-29 02:05:35 +02:00
noDepEntry fullDepEntry packEntry stringAfter;
lib: Use Nix's static scope checking, fix error message, optimize Nix can perform static scope checking, but whenever code is inside a `with` expression, the analysis breaks down, because it can't know statically what's in the attribute set whose attributes were brought into scope. In those cases, Nix has to assume that everything works out. Except it doesnt. Removing `with` from lib/ revealed an undefined variable in an error message. If that doesn't convince you that we're better off without `with`, I can tell you that this PR results in a 3% evaluation performance improvement because Nix can look up local variables by index. This adds up with applications like the module system. Furthermore, removing `with` makes the binding site of each variable obvious, which helps with comprehension.
2020-10-20 13:47:24 +02:00
inherit (self.customisation) overrideDerivation makeOverridable
lib: Add more configure flag helpers Add with/without to match enable/disable, and add `--{enable,with}-key=value` versions of both.
2018-05-11 17:12:15 +02:00
callPackageWith callPackagesWith extendDerivation hydraJob
lib: Create `makeScopeWithSplicing` It's ugly as hell, but I suppose it is needed to codify how to make spliced package sets.
2020-11-18 18:12:34 +01:00
makeScope makeScopeWithSplicing;
lib: Use Nix's static scope checking, fix error message, optimize Nix can perform static scope checking, but whenever code is inside a `with` expression, the analysis breaks down, because it can't know statically what's in the attribute set whose attributes were brought into scope. In those cases, Nix has to assume that everything works out. Except it doesnt. Removing `with` from lib/ revealed an undefined variable in an error message. If that doesn't convince you that we're better off without `with`, I can tell you that this PR results in a 3% evaluation performance improvement because Nix can look up local variables by index. This adds up with applications like the module system. Furthermore, removing `with` makes the binding site of each variable obvious, which helps with comprehension.
2020-10-20 13:47:24 +02:00
inherit (self.meta) addMetaAttrs dontDistribute setName updateName
lib: implement `setPrio` For when `hiPrio` and `lowPrio` are not enough.
2018-11-18 09:26:13 +01:00
appendToName mapDerivationAttrset setPrio lowPrio lowPrioSet hiPrio
Convert libs to a fixed-point This does break the API of being able to import any lib file and get its libs, however I'm not sure people did this. I made this while exploring being able to swap out docFn with a stub in #2305, to avoid functor performance problems. I don't know if that is going to move forward (or if it is a problem or not,) but after doing all this work figured I'd put it up anyway :) Two notable advantages to this approach: 1. when a lib inherits another lib's functions, it doesn't automatically get put in to the scope of lib 2. when a lib implements a new obscure functions, it doesn't automatically get put in to the scope of lib Using the test script (later in this commit) I got the following diff on the API: + diff master fixed-lib 11764a11765,11766 > .types.defaultFunctor > .types.defaultTypeMerge 11774a11777,11778 > .types.isOptionType > .types.isType 11781a11786 > .types.mkOptionType 11788a11794 > .types.setType 11795a11802 > .types.types This means that this commit _adds_ to the API, however I can't find a way to fix these last remaining discrepancies. At least none are _removed_. Test script (run with nix-repl in the PATH): #!/bin/sh set -eux repl() { suff=${1:-} echo "(import ./lib)$suff" \ | nix-repl 2>&1 } attrs_to_check() { repl "${1:-}" \ | tr ';' $'\n' \ | grep "\.\.\." \ | cut -d' ' -f2 \ | sed -e "s/^/${1:-}./" \ | sort } summ() { repl "${1:-}" \ | tr ' ' $'\n' \ | sort \ | uniq } deep_summ() { suff="${1:-}" depth="${2:-4}" depth=$((depth - 1)) summ "$suff" for attr in $(attrs_to_check "$suff" | grep -v "types.types"); do if [ $depth -eq 0 ]; then summ "$attr" | sed -e "s/^/$attr./" else deep_summ "$attr" "$depth" | sed -e "s/^/$attr./" fi done } ( cd nixpkgs #git add . #git commit -m "Auto-commit, sorry" || true git checkout fixed-lib deep_summ > ../fixed-lib git checkout master deep_summ > ../master ) if diff master fixed-lib; then echo "SHALLOW MATCH!" fi ( cd nixpkgs git checkout fixed-lib repl .types )
2017-07-29 02:05:35 +02:00
hiPrioSet;
lib: Use Nix's static scope checking, fix error message, optimize Nix can perform static scope checking, but whenever code is inside a `with` expression, the analysis breaks down, because it can't know statically what's in the attribute set whose attributes were brought into scope. In those cases, Nix has to assume that everything works out. Except it doesnt. Removing `with` from lib/ revealed an undefined variable in an error message. If that doesn't convince you that we're better off without `with`, I can tell you that this PR results in a 3% evaluation performance improvement because Nix can look up local variables by index. This adds up with applications like the module system. Furthermore, removing `with` makes the binding site of each variable obvious, which helps with comprehension.
2020-10-20 13:47:24 +02:00
inherit (self.sources) pathType pathIsDirectory cleanSourceFilter
Convert libs to a fixed-point This does break the API of being able to import any lib file and get its libs, however I'm not sure people did this. I made this while exploring being able to swap out docFn with a stub in #2305, to avoid functor performance problems. I don't know if that is going to move forward (or if it is a problem or not,) but after doing all this work figured I'd put it up anyway :) Two notable advantages to this approach: 1. when a lib inherits another lib's functions, it doesn't automatically get put in to the scope of lib 2. when a lib implements a new obscure functions, it doesn't automatically get put in to the scope of lib Using the test script (later in this commit) I got the following diff on the API: + diff master fixed-lib 11764a11765,11766 > .types.defaultFunctor > .types.defaultTypeMerge 11774a11777,11778 > .types.isOptionType > .types.isType 11781a11786 > .types.mkOptionType 11788a11794 > .types.setType 11795a11802 > .types.types This means that this commit _adds_ to the API, however I can't find a way to fix these last remaining discrepancies. At least none are _removed_. Test script (run with nix-repl in the PATH): #!/bin/sh set -eux repl() { suff=${1:-} echo "(import ./lib)$suff" \ | nix-repl 2>&1 } attrs_to_check() { repl "${1:-}" \ | tr ';' $'\n' \ | grep "\.\.\." \ | cut -d' ' -f2 \ | sed -e "s/^/${1:-}./" \ | sort } summ() { repl "${1:-}" \ | tr ' ' $'\n' \ | sort \ | uniq } deep_summ() { suff="${1:-}" depth="${2:-4}" depth=$((depth - 1)) summ "$suff" for attr in $(attrs_to_check "$suff" | grep -v "types.types"); do if [ $depth -eq 0 ]; then summ "$attr" | sed -e "s/^/$attr./" else deep_summ "$attr" "$depth" | sed -e "s/^/$attr./" fi done } ( cd nixpkgs #git add . #git commit -m "Auto-commit, sorry" || true git checkout fixed-lib deep_summ > ../fixed-lib git checkout master deep_summ > ../master ) if diff master fixed-lib; then echo "SHALLOW MATCH!" fi ( cd nixpkgs git checkout fixed-lib repl .types )
2017-07-29 02:05:35 +02:00
cleanSource sourceByRegex sourceFilesBySuffices
lib: Add more configure flag helpers Add with/without to match enable/disable, and add `--{enable,with}-key=value` versions of both.
2018-05-11 17:12:15 +02:00
commitIdFromGitRepo cleanSourceWith pathHasContext
nixos/version: fix case where .git is a symlink Before c9214c394b248e1f26e45dbe1be2bd82363af3a6 and 9d396d2e426b9a765de0b8999aa214f1259633e6 if .git is symlink the version would gracefully default to no git revision. With those changes an exception is thrown instead. This introduces a new function `pathIsGitRepo` that checks if `commitIdFromGitRepo` fails without error so we don't have to reimplement this logic again and can fail gracefully.
2020-01-20 00:44:07 +01:00
canCleanSource pathIsRegularFile pathIsGitRepo;
lib: Use Nix's static scope checking, fix error message, optimize Nix can perform static scope checking, but whenever code is inside a `with` expression, the analysis breaks down, because it can't know statically what's in the attribute set whose attributes were brought into scope. In those cases, Nix has to assume that everything works out. Except it doesnt. Removing `with` from lib/ revealed an undefined variable in an error message. If that doesn't convince you that we're better off without `with`, I can tell you that this PR results in a 3% evaluation performance improvement because Nix can look up local variables by index. This adds up with applications like the module system. Furthermore, removing `with` makes the binding site of each variable obvious, which helps with comprehension.
2020-10-20 13:47:24 +02:00
inherit (self.modules) evalModules unifyModuleSyntax
lib/modules: Don't pack submodules specially This has the beneficial side effect of allowing paths to be used as modules in types.{submodule,submoduleWith}
2019-12-05 03:29:51 +01:00
applyIfFunction mergeModules
Convert libs to a fixed-point This does break the API of being able to import any lib file and get its libs, however I'm not sure people did this. I made this while exploring being able to swap out docFn with a stub in #2305, to avoid functor performance problems. I don't know if that is going to move forward (or if it is a problem or not,) but after doing all this work figured I'd put it up anyway :) Two notable advantages to this approach: 1. when a lib inherits another lib's functions, it doesn't automatically get put in to the scope of lib 2. when a lib implements a new obscure functions, it doesn't automatically get put in to the scope of lib Using the test script (later in this commit) I got the following diff on the API: + diff master fixed-lib 11764a11765,11766 > .types.defaultFunctor > .types.defaultTypeMerge 11774a11777,11778 > .types.isOptionType > .types.isType 11781a11786 > .types.mkOptionType 11788a11794 > .types.setType 11795a11802 > .types.types This means that this commit _adds_ to the API, however I can't find a way to fix these last remaining discrepancies. At least none are _removed_. Test script (run with nix-repl in the PATH): #!/bin/sh set -eux repl() { suff=${1:-} echo "(import ./lib)$suff" \ | nix-repl 2>&1 } attrs_to_check() { repl "${1:-}" \ | tr ';' $'\n' \ | grep "\.\.\." \ | cut -d' ' -f2 \ | sed -e "s/^/${1:-}./" \ | sort } summ() { repl "${1:-}" \ | tr ' ' $'\n' \ | sort \ | uniq } deep_summ() { suff="${1:-}" depth="${2:-4}" depth=$((depth - 1)) summ "$suff" for attr in $(attrs_to_check "$suff" | grep -v "types.types"); do if [ $depth -eq 0 ]; then summ "$attr" | sed -e "s/^/$attr./" else deep_summ "$attr" "$depth" | sed -e "s/^/$attr./" fi done } ( cd nixpkgs #git add . #git commit -m "Auto-commit, sorry" || true git checkout fixed-lib deep_summ > ../fixed-lib git checkout master deep_summ > ../master ) if diff master fixed-lib; then echo "SHALLOW MATCH!" fi ( cd nixpkgs git checkout fixed-lib repl .types )
2017-07-29 02:05:35 +02:00
mergeModules' mergeOptionDecls evalOptionValue mergeDefinitions
pushDownProperties dischargeProperties filterOverrides
sortProperties fixupOptionType mkIf mkAssert mkMerge mkOverride
lib: inherit mkImageMediaOverride
2021-08-18 14:20:41 +02:00
mkOptionDefault mkDefault mkImageMediaOverride mkForce mkVMOverride
Partially revert "lib/modules: Drop mkStrict and mkFixStrictness" mkFixStrictness was never properly deprecated and should only be removed after warning for some time. This partially reverts commit 8fb9984690c878fcd768e967190957441de05d11.
2021-07-12 07:23:45 +02:00
mkFixStrictness mkOrder mkBefore mkAfter mkAliasDefinitions
Convert libs to a fixed-point This does break the API of being able to import any lib file and get its libs, however I'm not sure people did this. I made this while exploring being able to swap out docFn with a stub in #2305, to avoid functor performance problems. I don't know if that is going to move forward (or if it is a problem or not,) but after doing all this work figured I'd put it up anyway :) Two notable advantages to this approach: 1. when a lib inherits another lib's functions, it doesn't automatically get put in to the scope of lib 2. when a lib implements a new obscure functions, it doesn't automatically get put in to the scope of lib Using the test script (later in this commit) I got the following diff on the API: + diff master fixed-lib 11764a11765,11766 > .types.defaultFunctor > .types.defaultTypeMerge 11774a11777,11778 > .types.isOptionType > .types.isType 11781a11786 > .types.mkOptionType 11788a11794 > .types.setType 11795a11802 > .types.types This means that this commit _adds_ to the API, however I can't find a way to fix these last remaining discrepancies. At least none are _removed_. Test script (run with nix-repl in the PATH): #!/bin/sh set -eux repl() { suff=${1:-} echo "(import ./lib)$suff" \ | nix-repl 2>&1 } attrs_to_check() { repl "${1:-}" \ | tr ';' $'\n' \ | grep "\.\.\." \ | cut -d' ' -f2 \ | sed -e "s/^/${1:-}./" \ | sort } summ() { repl "${1:-}" \ | tr ' ' $'\n' \ | sort \ | uniq } deep_summ() { suff="${1:-}" depth="${2:-4}" depth=$((depth - 1)) summ "$suff" for attr in $(attrs_to_check "$suff" | grep -v "types.types"); do if [ $depth -eq 0 ]; then summ "$attr" | sed -e "s/^/$attr./" else deep_summ "$attr" "$depth" | sed -e "s/^/$attr./" fi done } ( cd nixpkgs #git add . #git commit -m "Auto-commit, sorry" || true git checkout fixed-lib deep_summ > ../fixed-lib git checkout master deep_summ > ../master ) if diff master fixed-lib; then echo "SHALLOW MATCH!" fi ( cd nixpkgs git checkout fixed-lib repl .types )
2017-07-29 02:05:35 +02:00
mkAliasAndWrapDefinitions fixMergeModules mkRemovedOptionModule
mkRenamedOptionModule mkMergedOptionModule mkChangedOptionModule
lib/modules: Recursive disabledModules With this change, disabledModules applies recursively, meaning if you have a module "foo.nix" with imports = [ ./bar.nix ]; then setting disabledModules = [ "foo.nix" ]; will disable both "foo.nix" and "bar.nix", whereas previously only "foo.nix" would be disabled. This change along with https://github.com/NixOS/nixpkgs/pull/61570 allows modules to be fully disabled even when they have some `mkRenamedOption` imports.
2020-01-02 22:09:31 +01:00
mkAliasOptionModule doRename;
lib: Use Nix's static scope checking, fix error message, optimize Nix can perform static scope checking, but whenever code is inside a `with` expression, the analysis breaks down, because it can't know statically what's in the attribute set whose attributes were brought into scope. In those cases, Nix has to assume that everything works out. Except it doesnt. Removing `with` from lib/ revealed an undefined variable in an error message. If that doesn't convince you that we're better off without `with`, I can tell you that this PR results in a 3% evaluation performance improvement because Nix can look up local variables by index. This adds up with applications like the module system. Furthermore, removing `with` makes the binding site of each variable obvious, which helps with comprehension.
2020-10-20 13:47:24 +02:00
inherit (self.options) isOption mkEnableOption mkSinkUndeclaredOptions
Convert libs to a fixed-point This does break the API of being able to import any lib file and get its libs, however I'm not sure people did this. I made this while exploring being able to swap out docFn with a stub in #2305, to avoid functor performance problems. I don't know if that is going to move forward (or if it is a problem or not,) but after doing all this work figured I'd put it up anyway :) Two notable advantages to this approach: 1. when a lib inherits another lib's functions, it doesn't automatically get put in to the scope of lib 2. when a lib implements a new obscure functions, it doesn't automatically get put in to the scope of lib Using the test script (later in this commit) I got the following diff on the API: + diff master fixed-lib 11764a11765,11766 > .types.defaultFunctor > .types.defaultTypeMerge 11774a11777,11778 > .types.isOptionType > .types.isType 11781a11786 > .types.mkOptionType 11788a11794 > .types.setType 11795a11802 > .types.types This means that this commit _adds_ to the API, however I can't find a way to fix these last remaining discrepancies. At least none are _removed_. Test script (run with nix-repl in the PATH): #!/bin/sh set -eux repl() { suff=${1:-} echo "(import ./lib)$suff" \ | nix-repl 2>&1 } attrs_to_check() { repl "${1:-}" \ | tr ';' $'\n' \ | grep "\.\.\." \ | cut -d' ' -f2 \ | sed -e "s/^/${1:-}./" \ | sort } summ() { repl "${1:-}" \ | tr ' ' $'\n' \ | sort \ | uniq } deep_summ() { suff="${1:-}" depth="${2:-4}" depth=$((depth - 1)) summ "$suff" for attr in $(attrs_to_check "$suff" | grep -v "types.types"); do if [ $depth -eq 0 ]; then summ "$attr" | sed -e "s/^/$attr./" else deep_summ "$attr" "$depth" | sed -e "s/^/$attr./" fi done } ( cd nixpkgs #git add . #git commit -m "Auto-commit, sorry" || true git checkout fixed-lib deep_summ > ../fixed-lib git checkout master deep_summ > ../master ) if diff master fixed-lib; then echo "SHALLOW MATCH!" fi ( cd nixpkgs git checkout fixed-lib repl .types )
2017-07-29 02:05:35 +02:00
mergeDefaultOption mergeOneOption mergeEqualOption getValues
getFiles optionAttrSetToDocList optionAttrSetToDocList'
scrubOptionValue literalExample showOption showFiles
unknownModule mkOption;
lib: Use Nix's static scope checking, fix error message, optimize Nix can perform static scope checking, but whenever code is inside a `with` expression, the analysis breaks down, because it can't know statically what's in the attribute set whose attributes were brought into scope. In those cases, Nix has to assume that everything works out. Except it doesnt. Removing `with` from lib/ revealed an undefined variable in an error message. If that doesn't convince you that we're better off without `with`, I can tell you that this PR results in a 3% evaluation performance improvement because Nix can look up local variables by index. This adds up with applications like the module system. Furthermore, removing `with` makes the binding site of each variable obvious, which helps with comprehension.
2020-10-20 13:47:24 +02:00
inherit (self.types) isType setType defaultTypeMerge defaultFunctor
Convert libs to a fixed-point This does break the API of being able to import any lib file and get its libs, however I'm not sure people did this. I made this while exploring being able to swap out docFn with a stub in #2305, to avoid functor performance problems. I don't know if that is going to move forward (or if it is a problem or not,) but after doing all this work figured I'd put it up anyway :) Two notable advantages to this approach: 1. when a lib inherits another lib's functions, it doesn't automatically get put in to the scope of lib 2. when a lib implements a new obscure functions, it doesn't automatically get put in to the scope of lib Using the test script (later in this commit) I got the following diff on the API: + diff master fixed-lib 11764a11765,11766 > .types.defaultFunctor > .types.defaultTypeMerge 11774a11777,11778 > .types.isOptionType > .types.isType 11781a11786 > .types.mkOptionType 11788a11794 > .types.setType 11795a11802 > .types.types This means that this commit _adds_ to the API, however I can't find a way to fix these last remaining discrepancies. At least none are _removed_. Test script (run with nix-repl in the PATH): #!/bin/sh set -eux repl() { suff=${1:-} echo "(import ./lib)$suff" \ | nix-repl 2>&1 } attrs_to_check() { repl "${1:-}" \ | tr ';' $'\n' \ | grep "\.\.\." \ | cut -d' ' -f2 \ | sed -e "s/^/${1:-}./" \ | sort } summ() { repl "${1:-}" \ | tr ' ' $'\n' \ | sort \ | uniq } deep_summ() { suff="${1:-}" depth="${2:-4}" depth=$((depth - 1)) summ "$suff" for attr in $(attrs_to_check "$suff" | grep -v "types.types"); do if [ $depth -eq 0 ]; then summ "$attr" | sed -e "s/^/$attr./" else deep_summ "$attr" "$depth" | sed -e "s/^/$attr./" fi done } ( cd nixpkgs #git add . #git commit -m "Auto-commit, sorry" || true git checkout fixed-lib deep_summ > ../fixed-lib git checkout master deep_summ > ../master ) if diff master fixed-lib; then echo "SHALLOW MATCH!" fi ( cd nixpkgs git checkout fixed-lib repl .types )
2017-07-29 02:05:35 +02:00
isOptionType mkOptionType;
lib: Use Nix's static scope checking, fix error message, optimize Nix can perform static scope checking, but whenever code is inside a `with` expression, the analysis breaks down, because it can't know statically what's in the attribute set whose attributes were brought into scope. In those cases, Nix has to assume that everything works out. Except it doesnt. Removing `with` from lib/ revealed an undefined variable in an error message. If that doesn't convince you that we're better off without `with`, I can tell you that this PR results in a 3% evaluation performance improvement because Nix can look up local variables by index. This adds up with applications like the module system. Furthermore, removing `with` makes the binding site of each variable obvious, which helps with comprehension.
2020-10-20 13:47:24 +02:00
inherit (self.asserts)
lib: move assertMsg and assertOneOf to their own library file Since the `assertOneOf` uses `lib.generators`, they are not really trivial anymore and should go into their own library file.
2018-08-08 19:26:52 +02:00
assertMsg assertOneOf;
lib: Use Nix's static scope checking, fix error message, optimize Nix can perform static scope checking, but whenever code is inside a `with` expression, the analysis breaks down, because it can't know statically what's in the attribute set whose attributes were brought into scope. In those cases, Nix has to assume that everything works out. Except it doesnt. Removing `with` from lib/ revealed an undefined variable in an error message. If that doesn't convince you that we're better off without `with`, I can tell you that this PR results in a 3% evaluation performance improvement because Nix can look up local variables by index. This adds up with applications like the module system. Furthermore, removing `with` makes the binding site of each variable obvious, which helps with comprehension.
2020-10-20 13:47:24 +02:00
inherit (self.debug) addErrorContextToAttrs traceIf traceVal traceValFn
Convert libs to a fixed-point This does break the API of being able to import any lib file and get its libs, however I'm not sure people did this. I made this while exploring being able to swap out docFn with a stub in #2305, to avoid functor performance problems. I don't know if that is going to move forward (or if it is a problem or not,) but after doing all this work figured I'd put it up anyway :) Two notable advantages to this approach: 1. when a lib inherits another lib's functions, it doesn't automatically get put in to the scope of lib 2. when a lib implements a new obscure functions, it doesn't automatically get put in to the scope of lib Using the test script (later in this commit) I got the following diff on the API: + diff master fixed-lib 11764a11765,11766 > .types.defaultFunctor > .types.defaultTypeMerge 11774a11777,11778 > .types.isOptionType > .types.isType 11781a11786 > .types.mkOptionType 11788a11794 > .types.setType 11795a11802 > .types.types This means that this commit _adds_ to the API, however I can't find a way to fix these last remaining discrepancies. At least none are _removed_. Test script (run with nix-repl in the PATH): #!/bin/sh set -eux repl() { suff=${1:-} echo "(import ./lib)$suff" \ | nix-repl 2>&1 } attrs_to_check() { repl "${1:-}" \ | tr ';' $'\n' \ | grep "\.\.\." \ | cut -d' ' -f2 \ | sed -e "s/^/${1:-}./" \ | sort } summ() { repl "${1:-}" \ | tr ' ' $'\n' \ | sort \ | uniq } deep_summ() { suff="${1:-}" depth="${2:-4}" depth=$((depth - 1)) summ "$suff" for attr in $(attrs_to_check "$suff" | grep -v "types.types"); do if [ $depth -eq 0 ]; then summ "$attr" | sed -e "s/^/$attr./" else deep_summ "$attr" "$depth" | sed -e "s/^/$attr./" fi done } ( cd nixpkgs #git add . #git commit -m "Auto-commit, sorry" || true git checkout fixed-lib deep_summ > ../fixed-lib git checkout master deep_summ > ../master ) if diff master fixed-lib; then echo "SHALLOW MATCH!" fi ( cd nixpkgs git checkout fixed-lib repl .types )
2017-07-29 02:05:35 +02:00
traceXMLVal traceXMLValMarked traceSeq traceSeqN traceValSeq
lib/debug: add traceFnSeqN Immensely helpful when you want to see the changes a function makes to its value as it passes through. Example: ``` $ nix-instantiate --strict --eval -E '(with import ./lib; traceFnSeqN 2 "id" (x: x) { a.b.c = 3; })' trace: { fn = "id"; from = { a = { b = {…}; }; }; to = { a = { b = {…}; }; }; } { a = { b = { c = 3; }; }; } ```
2021-01-24 11:33:41 +01:00
traceValSeqFn traceValSeqN traceValSeqNFn traceFnSeqN traceShowVal
lib/debug: add traceValFn, traceValSeqFn, traceValSeqNFn Being able to modify the value on-the-fly before printing is very useful in practice.
2018-04-02 17:21:35 +02:00
traceShowValMarked showVal traceCall traceCall2 traceCall3
lib: Add more configure flag helpers Add with/without to match enable/disable, and add `--{enable,with}-key=value` versions of both.
2018-05-11 17:12:15 +02:00
traceValIfNot runTests testAllTrue traceCallXml attrNamesToStr;
lib: Use Nix's static scope checking, fix error message, optimize Nix can perform static scope checking, but whenever code is inside a `with` expression, the analysis breaks down, because it can't know statically what's in the attribute set whose attributes were brought into scope. In those cases, Nix has to assume that everything works out. Except it doesnt. Removing `with` from lib/ revealed an undefined variable in an error message. If that doesn't convince you that we're better off without `with`, I can tell you that this PR results in a 3% evaluation performance improvement because Nix can look up local variables by index. This adds up with applications like the module system. Furthermore, removing `with` makes the binding site of each variable obvious, which helps with comprehension.
2020-10-20 13:47:24 +02:00
inherit (self.misc) maybeEnv defaultMergeArg defaultMerge foldArgs
Convert libs to a fixed-point This does break the API of being able to import any lib file and get its libs, however I'm not sure people did this. I made this while exploring being able to swap out docFn with a stub in #2305, to avoid functor performance problems. I don't know if that is going to move forward (or if it is a problem or not,) but after doing all this work figured I'd put it up anyway :) Two notable advantages to this approach: 1. when a lib inherits another lib's functions, it doesn't automatically get put in to the scope of lib 2. when a lib implements a new obscure functions, it doesn't automatically get put in to the scope of lib Using the test script (later in this commit) I got the following diff on the API: + diff master fixed-lib 11764a11765,11766 > .types.defaultFunctor > .types.defaultTypeMerge 11774a11777,11778 > .types.isOptionType > .types.isType 11781a11786 > .types.mkOptionType 11788a11794 > .types.setType 11795a11802 > .types.types This means that this commit _adds_ to the API, however I can't find a way to fix these last remaining discrepancies. At least none are _removed_. Test script (run with nix-repl in the PATH): #!/bin/sh set -eux repl() { suff=${1:-} echo "(import ./lib)$suff" \ | nix-repl 2>&1 } attrs_to_check() { repl "${1:-}" \ | tr ';' $'\n' \ | grep "\.\.\." \ | cut -d' ' -f2 \ | sed -e "s/^/${1:-}./" \ | sort } summ() { repl "${1:-}" \ | tr ' ' $'\n' \ | sort \ | uniq } deep_summ() { suff="${1:-}" depth="${2:-4}" depth=$((depth - 1)) summ "$suff" for attr in $(attrs_to_check "$suff" | grep -v "types.types"); do if [ $depth -eq 0 ]; then summ "$attr" | sed -e "s/^/$attr./" else deep_summ "$attr" "$depth" | sed -e "s/^/$attr./" fi done } ( cd nixpkgs #git add . #git commit -m "Auto-commit, sorry" || true git checkout fixed-lib deep_summ > ../fixed-lib git checkout master deep_summ > ../master ) if diff master fixed-lib; then echo "SHALLOW MATCH!" fi ( cd nixpkgs git checkout fixed-lib repl .types )
2017-07-29 02:05:35 +02:00
maybeAttrNullable maybeAttr ifEnable checkFlag getValue
checkReqs uniqList uniqListExt condConcat lazyGenericClosure
innerModifySumArgs modifySumArgs innerClosePropagation
closePropagation mapAttrsFlatten nvs setAttr setAttrMerge
mergeAttrsWithFunc mergeAttrsConcatenateValues
mergeAttrsNoOverride mergeAttrByFunc mergeAttrsByFuncDefaults
Remove composableDerivation, closes #18763
2018-07-26 12:05:40 +02:00
mergeAttrsByFuncDefaultsClean mergeAttrBy
lib.fake{Sri => Hash}: fix and rename The previous hash was too short and caused evaluation-time errors like: invalid SRI hash 'sha256-AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA=' Additionally, since the fact that this is broken implies that nobody could have been using it, "SRI" is a bit of a vague and obscure term, `fakeSriHash` would be somewhat of a mouthful, and the relevant fetcher parameters are just called `hash`, rename it to `fakeHash`.
2020-05-12 00:04:41 +02:00
fakeHash fakeSha256 fakeSha512
Remove composableDerivation, closes #18763
2018-07-26 12:05:40 +02:00
nixType imap;
lib: Use Nix's static scope checking, fix error message, optimize Nix can perform static scope checking, but whenever code is inside a `with` expression, the analysis breaks down, because it can't know statically what's in the attribute set whose attributes were brought into scope. In those cases, Nix has to assume that everything works out. Except it doesnt. Removing `with` from lib/ revealed an undefined variable in an error message. If that doesn't convince you that we're better off without `with`, I can tell you that this PR results in a 3% evaluation performance improvement because Nix can look up local variables by index. This adds up with applications like the module system. Furthermore, removing `with` makes the binding site of each variable obvious, which helps with comprehension.
2020-10-20 13:47:24 +02:00
inherit (self.versions)
lib/versions: expose splitVersion
2019-09-24 10:47:34 +02:00
splitVersion;
lib: make extensible This allows the lib fixed point to be extended with myLib = lib.extend (self: super: { foo = "foo"; }) With this it's possible to have the new modified lib attrset available to all modules when using evalModules myLib.evalModules { modules = [ ({ lib, ... }: { options.bar = lib.mkOption { default = lib.foo; }; }) ]; } => { config = { bar = "foo"; ... }; options = ...; }
2018-04-06 18:51:10 +02:00
});
Convert libs to a fixed-point This does break the API of being able to import any lib file and get its libs, however I'm not sure people did this. I made this while exploring being able to swap out docFn with a stub in #2305, to avoid functor performance problems. I don't know if that is going to move forward (or if it is a problem or not,) but after doing all this work figured I'd put it up anyway :) Two notable advantages to this approach: 1. when a lib inherits another lib's functions, it doesn't automatically get put in to the scope of lib 2. when a lib implements a new obscure functions, it doesn't automatically get put in to the scope of lib Using the test script (later in this commit) I got the following diff on the API: + diff master fixed-lib 11764a11765,11766 > .types.defaultFunctor > .types.defaultTypeMerge 11774a11777,11778 > .types.isOptionType > .types.isType 11781a11786 > .types.mkOptionType 11788a11794 > .types.setType 11795a11802 > .types.types This means that this commit _adds_ to the API, however I can't find a way to fix these last remaining discrepancies. At least none are _removed_. Test script (run with nix-repl in the PATH): #!/bin/sh set -eux repl() { suff=${1:-} echo "(import ./lib)$suff" \ | nix-repl 2>&1 } attrs_to_check() { repl "${1:-}" \ | tr ';' $'\n' \ | grep "\.\.\." \ | cut -d' ' -f2 \ | sed -e "s/^/${1:-}./" \ | sort } summ() { repl "${1:-}" \ | tr ' ' $'\n' \ | sort \ | uniq } deep_summ() { suff="${1:-}" depth="${2:-4}" depth=$((depth - 1)) summ "$suff" for attr in $(attrs_to_check "$suff" | grep -v "types.types"); do if [ $depth -eq 0 ]; then summ "$attr" | sed -e "s/^/$attr./" else deep_summ "$attr" "$depth" | sed -e "s/^/$attr./" fi done } ( cd nixpkgs #git add . #git commit -m "Auto-commit, sorry" || true git checkout fixed-lib deep_summ > ../fixed-lib git checkout master deep_summ > ../master ) if diff master fixed-lib; then echo "SHALLOW MATCH!" fi ( cd nixpkgs git checkout fixed-lib repl .types )
2017-07-29 02:05:35 +02:00
in lib
Reference in a new issue Copy permalink