1
0
Fork 1
mirror of https://github.com/NixOS/nixpkgs.git synced 2024-11-30 01:20:40 +00:00
nixpkgs/lib/strings.nix

778 lines
22 KiB
Nix
Raw Normal View History

/* String manipulation functions. */
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 01:05:35 +01:00
{ lib }:
let
inherit (builtins) length;
in
rec {
2013-11-12 12:48:19 +00:00
inherit (builtins)
compareVersions
elem
elemAt
filter
fromJSON
head
isInt
isList
isString
match
parseDrvName
readFile
replaceStrings
split
storeDir
stringLength
substring
tail
toJSON
typeOf
unsafeDiscardStringContext
;
/* Concatenate a list of strings.
Type: concatStrings :: [string] -> string
Example:
concatStrings ["foo" "bar"]
=> "foobar"
*/
concatStrings = builtins.concatStringsSep "";
/* Map a function over a list and concatenate the resulting strings.
Type: concatMapStrings :: (a -> string) -> [a] -> string
Example:
concatMapStrings (x: "a" + x) ["foo" "bar"]
=> "afooabar"
*/
concatMapStrings = f: list: concatStrings (map f list);
/* Like `concatMapStrings` except that the f functions also gets the
position as a parameter.
Type: concatImapStrings :: (int -> a -> string) -> [a] -> string
Example:
concatImapStrings (pos: x: "${toString pos}-${x}") ["foo" "bar"]
=> "1-foo2-bar"
*/
concatImapStrings = f: list: concatStrings (lib.imap1 f list);
/* Place an element between each element of a list
Type: intersperse :: a -> [a] -> [a]
Example:
intersperse "/" ["usr" "local" "bin"]
=> ["usr" "/" "local" "/" "bin"].
*/
intersperse =
# Separator to add between elements
separator:
# Input list
list:
2012-08-13 19:19:31 +01:00
if list == [] || length list == 1
then list
2015-07-24 14:55:39 +01:00
else tail (lib.concatMap (x: [separator x]) list);
/* Concatenate a list of strings with a separator between each element
Type: concatStringsSep :: string -> [string] -> string
Example:
concatStringsSep "/" ["usr" "local" "bin"]
=> "usr/local/bin"
*/
2015-07-24 01:38:46 +01:00
concatStringsSep = builtins.concatStringsSep or (separator: list:
lib.foldl' (x: y: x + y) "" (intersperse separator list));
/* Maps a function over a list of strings and then concatenates the
result with the specified separator interspersed between
elements.
Type: concatMapStringsSep :: string -> (a -> string) -> [a] -> string
Example:
concatMapStringsSep "-" (x: toUpper x) ["foo" "bar" "baz"]
=> "FOO-BAR-BAZ"
*/
concatMapStringsSep =
# Separator to add between elements
sep:
# Function to map over the list
f:
# List of input strings
list: concatStringsSep sep (map f list);
/* Same as `concatMapStringsSep`, but the mapping function
additionally receives the position of its argument.
Type: concatIMapStringsSep :: string -> (int -> a -> string) -> [a] -> string
Example:
concatImapStringsSep "-" (pos: x: toString (x / pos)) [ 6 6 6 ]
=> "6-3-2"
*/
concatImapStringsSep =
# Separator to add between elements
sep:
# Function that receives elements and their positions
f:
# List of input strings
list: concatStringsSep sep (lib.imap1 f list);
/* Construct a Unix-style, colon-separated search path consisting of
the given `subDir` appended to each of the given paths.
Type: makeSearchPath :: string -> [string] -> string
Example:
makeSearchPath "bin" ["/root" "/usr" "/usr/local"]
=> "/root/bin:/usr/bin:/usr/local/bin"
makeSearchPath "bin" [""]
=> "/bin"
*/
makeSearchPath =
# Directory name to append
subDir:
# List of base paths
paths:
concatStringsSep ":" (map (path: path + "/" + subDir) (filter (x: x != null) paths));
/* Construct a Unix-style search path by appending the given
`subDir` to the specified `output` of each of the packages. If no
output by the given name is found, fallback to `.out` and then to
the default.
Type: string -> string -> [package] -> string
2016-04-13 13:33:21 +01:00
Example:
makeSearchPathOutput "dev" "bin" [ pkgs.openssl pkgs.zlib ]
=> "/nix/store/9rz8gxhzf8sw4kf2j2f1grr49w8zx5vj-openssl-1.0.1r-dev/bin:/nix/store/wwh7mhwh269sfjkm6k5665b5kgp7jrk2-zlib-1.2.8/bin"
2016-04-13 13:33:21 +01:00
*/
makeSearchPathOutput =
# Package output to use
output:
# Directory name to append
subDir:
# List of packages
pkgs: makeSearchPath subDir (map (lib.getOutput output) pkgs);
2016-04-13 13:33:21 +01:00
/* Construct a library search path (such as RPATH) containing the
libraries for a set of packages
Example:
makeLibraryPath [ "/usr" "/usr/local" ]
=> "/usr/lib:/usr/local/lib"
pkgs = import <nixpkgs> { }
makeLibraryPath [ pkgs.openssl pkgs.zlib ]
=> "/nix/store/9rz8gxhzf8sw4kf2j2f1grr49w8zx5vj-openssl-1.0.1r/lib:/nix/store/wwh7mhwh269sfjkm6k5665b5kgp7jrk2-zlib-1.2.8/lib"
*/
makeLibraryPath = makeSearchPathOutput "lib" "lib";
/* Construct a binary search path (such as $PATH) containing the
binaries for a set of packages.
Example:
makeBinPath ["/root" "/usr" "/usr/local"]
=> "/root/bin:/usr/bin:/usr/local/bin"
*/
makeBinPath = makeSearchPathOutput "bin" "bin";
2017-04-19 20:41:28 +01:00
/* Depending on the boolean `cond', return either the given string
or the empty string. Useful to concatenate against a bigger string.
Type: optionalString :: bool -> string -> string
Example:
optionalString true "some-string"
=> "some-string"
optionalString false "some-string"
=> ""
*/
optionalString =
# Condition
cond:
# String to return if condition is true
string: if cond then string else "";
/* Determine whether a string has given prefix.
Type: hasPrefix :: string -> string -> bool
Example:
hasPrefix "foo" "foobar"
=> true
hasPrefix "foo" "barfoo"
=> false
*/
hasPrefix =
# Prefix to check for
pref:
# Input string
str: substring 0 (stringLength pref) str == pref;
/* Determine whether a string has given suffix.
Type: hasSuffix :: string -> string -> bool
Example:
hasSuffix "foo" "foobar"
=> false
hasSuffix "foo" "barfoo"
=> true
*/
hasSuffix =
# Suffix to check for
suffix:
# Input string
content:
let
2016-08-10 19:06:28 +01:00
lenContent = stringLength content;
lenSuffix = stringLength suffix;
in lenContent >= lenSuffix &&
substring (lenContent - lenSuffix) lenContent content == suffix;
/* Determine whether a string contains the given infix
Type: hasInfix :: string -> string -> bool
Example:
hasInfix "bc" "abcd"
=> true
hasInfix "ab" "abcd"
=> true
hasInfix "cd" "abcd"
=> true
hasInfix "foo" "abcd"
=> false
*/
hasInfix = infix: content:
let
drop = x: substring 1 (stringLength x) x;
in hasPrefix infix content
|| content != "" && hasInfix infix (drop content);
/* Convert a string to a list of characters (i.e. singleton strings).
This allows you to, e.g., map a function over each character. However,
note that this will likely be horribly inefficient; Nix is not a
general purpose programming language. Complex string manipulations
should, if appropriate, be done in a derivation.
Also note that Nix treats strings as a list of bytes and thus doesn't
handle unicode.
Type: stringToCharacters :: string -> [string]
Example:
stringToCharacters ""
=> [ ]
stringToCharacters "abc"
=> [ "a" "b" "c" ]
stringToCharacters "💩"
=> [ "<EFBFBD>" "<EFBFBD>" "<EFBFBD>" "<EFBFBD>" ]
*/
2015-07-24 14:45:41 +01:00
stringToCharacters = s:
map (p: substring p 1 s) (lib.range 0 (stringLength s - 1));
/* Manipulate a string character by character and replace them by
strings before concatenating the results.
Type: stringAsChars :: (string -> string) -> string -> string
Example:
stringAsChars (x: if x == "a" then "i" else x) "nax"
=> "nix"
*/
stringAsChars =
# Function to map over each individual character
f:
# Input string
s: concatStrings (
map f (stringToCharacters s)
);
/* Escape occurrence of the elements of `list` in `string` by
prefixing it with a backslash.
Type: escape :: [string] -> string -> string
Example:
escape ["(" ")"] "(foo)"
=> "\\(foo\\)"
*/
2015-07-24 14:45:41 +01:00
escape = list: replaceChars list (map (c: "\\${c}") list);
lib: Make escapeShellArg more robust Quoting various characters that the shell *may* interpret specially is a very fragile thing to do. I've used something more robust all over the place in various Nix expression I've written just because I didn't trust escapeShellArg. Here is a proof of concept showing that I was indeed right in distrusting escapeShellArg: with import <nixpkgs> {}; let payload = runCommand "payload" {} '' # \x00 is not allowed for Nix strings, so let's begin at 1 for i in $(seq 1 255); do echo -en "\\x$(printf %02x $i)" done > "$out" ''; escapers = with lib; { current = escapeShellArg; better = arg: let backslashEscapes = stringToCharacters "\"\\ ';$`()|<>\r\t*[]&!~#"; search = backslashEscapes ++ [ "\n" ]; replace = map (c: "\\${c}") backslashEscapes ++ [ "'\n'" ]; in replaceStrings search replace (toString arg); best = arg: "'${replaceStrings ["'"] ["'\\''"] (toString arg)}'"; }; testWith = escaper: let escaped = escaper (builtins.readFile payload); in runCommand "test" {} '' if ! r="$(bash -c ${escapers.best "echo -nE ${escaped}"} 2> /dev/null)" then echo bash eval error > "$out" exit 0 fi if echo -n "$r" | cmp -s "${payload}"; then echo success > "$out" else echo failed > "$out" fi ''; in runCommand "results" {} '' echo "Test results:" ${lib.concatStrings (lib.mapAttrsToList (name: impl: '' echo " ${name}: $(< "${testWith impl}")" '') escapers)} exit 1 '' The resulting output is the following: Test results: best: success better: success current: bash eval error I did the "better" implementation just to illustrate that the method of quoting only "harmful" characters results in madness in terms of implementation and performance. Signed-off-by: aszlig <aszlig@redmoonstudios.org> Cc: @edolstra, @zimbatm
2016-06-20 17:31:49 +01:00
/* Quote string to be used safely within the Bourne shell.
Type: escapeShellArg :: string -> string
Example:
lib: Make escapeShellArg more robust Quoting various characters that the shell *may* interpret specially is a very fragile thing to do. I've used something more robust all over the place in various Nix expression I've written just because I didn't trust escapeShellArg. Here is a proof of concept showing that I was indeed right in distrusting escapeShellArg: with import <nixpkgs> {}; let payload = runCommand "payload" {} '' # \x00 is not allowed for Nix strings, so let's begin at 1 for i in $(seq 1 255); do echo -en "\\x$(printf %02x $i)" done > "$out" ''; escapers = with lib; { current = escapeShellArg; better = arg: let backslashEscapes = stringToCharacters "\"\\ ';$`()|<>\r\t*[]&!~#"; search = backslashEscapes ++ [ "\n" ]; replace = map (c: "\\${c}") backslashEscapes ++ [ "'\n'" ]; in replaceStrings search replace (toString arg); best = arg: "'${replaceStrings ["'"] ["'\\''"] (toString arg)}'"; }; testWith = escaper: let escaped = escaper (builtins.readFile payload); in runCommand "test" {} '' if ! r="$(bash -c ${escapers.best "echo -nE ${escaped}"} 2> /dev/null)" then echo bash eval error > "$out" exit 0 fi if echo -n "$r" | cmp -s "${payload}"; then echo success > "$out" else echo failed > "$out" fi ''; in runCommand "results" {} '' echo "Test results:" ${lib.concatStrings (lib.mapAttrsToList (name: impl: '' echo " ${name}: $(< "${testWith impl}")" '') escapers)} exit 1 '' The resulting output is the following: Test results: best: success better: success current: bash eval error I did the "better" implementation just to illustrate that the method of quoting only "harmful" characters results in madness in terms of implementation and performance. Signed-off-by: aszlig <aszlig@redmoonstudios.org> Cc: @edolstra, @zimbatm
2016-06-20 17:31:49 +01:00
escapeShellArg "esc'ape\nme"
=> "'esc'\\''ape\nme'"
*/
lib: Make escapeShellArg more robust Quoting various characters that the shell *may* interpret specially is a very fragile thing to do. I've used something more robust all over the place in various Nix expression I've written just because I didn't trust escapeShellArg. Here is a proof of concept showing that I was indeed right in distrusting escapeShellArg: with import <nixpkgs> {}; let payload = runCommand "payload" {} '' # \x00 is not allowed for Nix strings, so let's begin at 1 for i in $(seq 1 255); do echo -en "\\x$(printf %02x $i)" done > "$out" ''; escapers = with lib; { current = escapeShellArg; better = arg: let backslashEscapes = stringToCharacters "\"\\ ';$`()|<>\r\t*[]&!~#"; search = backslashEscapes ++ [ "\n" ]; replace = map (c: "\\${c}") backslashEscapes ++ [ "'\n'" ]; in replaceStrings search replace (toString arg); best = arg: "'${replaceStrings ["'"] ["'\\''"] (toString arg)}'"; }; testWith = escaper: let escaped = escaper (builtins.readFile payload); in runCommand "test" {} '' if ! r="$(bash -c ${escapers.best "echo -nE ${escaped}"} 2> /dev/null)" then echo bash eval error > "$out" exit 0 fi if echo -n "$r" | cmp -s "${payload}"; then echo success > "$out" else echo failed > "$out" fi ''; in runCommand "results" {} '' echo "Test results:" ${lib.concatStrings (lib.mapAttrsToList (name: impl: '' echo " ${name}: $(< "${testWith impl}")" '') escapers)} exit 1 '' The resulting output is the following: Test results: best: success better: success current: bash eval error I did the "better" implementation just to illustrate that the method of quoting only "harmful" characters results in madness in terms of implementation and performance. Signed-off-by: aszlig <aszlig@redmoonstudios.org> Cc: @edolstra, @zimbatm
2016-06-20 17:31:49 +01:00
escapeShellArg = arg: "'${replaceStrings ["'"] ["'\\''"] (toString arg)}'";
2016-06-12 18:11:37 +01:00
lib: Make escapeShellArg more robust Quoting various characters that the shell *may* interpret specially is a very fragile thing to do. I've used something more robust all over the place in various Nix expression I've written just because I didn't trust escapeShellArg. Here is a proof of concept showing that I was indeed right in distrusting escapeShellArg: with import <nixpkgs> {}; let payload = runCommand "payload" {} '' # \x00 is not allowed for Nix strings, so let's begin at 1 for i in $(seq 1 255); do echo -en "\\x$(printf %02x $i)" done > "$out" ''; escapers = with lib; { current = escapeShellArg; better = arg: let backslashEscapes = stringToCharacters "\"\\ ';$`()|<>\r\t*[]&!~#"; search = backslashEscapes ++ [ "\n" ]; replace = map (c: "\\${c}") backslashEscapes ++ [ "'\n'" ]; in replaceStrings search replace (toString arg); best = arg: "'${replaceStrings ["'"] ["'\\''"] (toString arg)}'"; }; testWith = escaper: let escaped = escaper (builtins.readFile payload); in runCommand "test" {} '' if ! r="$(bash -c ${escapers.best "echo -nE ${escaped}"} 2> /dev/null)" then echo bash eval error > "$out" exit 0 fi if echo -n "$r" | cmp -s "${payload}"; then echo success > "$out" else echo failed > "$out" fi ''; in runCommand "results" {} '' echo "Test results:" ${lib.concatStrings (lib.mapAttrsToList (name: impl: '' echo " ${name}: $(< "${testWith impl}")" '') escapers)} exit 1 '' The resulting output is the following: Test results: best: success better: success current: bash eval error I did the "better" implementation just to illustrate that the method of quoting only "harmful" characters results in madness in terms of implementation and performance. Signed-off-by: aszlig <aszlig@redmoonstudios.org> Cc: @edolstra, @zimbatm
2016-06-20 17:31:49 +01:00
/* Quote all arguments to be safely passed to the Bourne shell.
2016-06-12 18:11:37 +01:00
Type: escapeShellArgs :: [string] -> string
2016-06-12 18:11:37 +01:00
Example:
lib: Make escapeShellArg more robust Quoting various characters that the shell *may* interpret specially is a very fragile thing to do. I've used something more robust all over the place in various Nix expression I've written just because I didn't trust escapeShellArg. Here is a proof of concept showing that I was indeed right in distrusting escapeShellArg: with import <nixpkgs> {}; let payload = runCommand "payload" {} '' # \x00 is not allowed for Nix strings, so let's begin at 1 for i in $(seq 1 255); do echo -en "\\x$(printf %02x $i)" done > "$out" ''; escapers = with lib; { current = escapeShellArg; better = arg: let backslashEscapes = stringToCharacters "\"\\ ';$`()|<>\r\t*[]&!~#"; search = backslashEscapes ++ [ "\n" ]; replace = map (c: "\\${c}") backslashEscapes ++ [ "'\n'" ]; in replaceStrings search replace (toString arg); best = arg: "'${replaceStrings ["'"] ["'\\''"] (toString arg)}'"; }; testWith = escaper: let escaped = escaper (builtins.readFile payload); in runCommand "test" {} '' if ! r="$(bash -c ${escapers.best "echo -nE ${escaped}"} 2> /dev/null)" then echo bash eval error > "$out" exit 0 fi if echo -n "$r" | cmp -s "${payload}"; then echo success > "$out" else echo failed > "$out" fi ''; in runCommand "results" {} '' echo "Test results:" ${lib.concatStrings (lib.mapAttrsToList (name: impl: '' echo " ${name}: $(< "${testWith impl}")" '') escapers)} exit 1 '' The resulting output is the following: Test results: best: success better: success current: bash eval error I did the "better" implementation just to illustrate that the method of quoting only "harmful" characters results in madness in terms of implementation and performance. Signed-off-by: aszlig <aszlig@redmoonstudios.org> Cc: @edolstra, @zimbatm
2016-06-20 17:31:49 +01:00
escapeShellArgs ["one" "two three" "four'five"]
=> "'one' 'two three' 'four'\\''five'"
2016-06-12 18:11:37 +01:00
*/
escapeShellArgs = concatMapStringsSep " " escapeShellArg;
2017-12-02 18:03:40 +00:00
/* Turn a string into a Nix expression representing that string
Type: string -> string
2017-12-02 18:03:40 +00:00
Example:
escapeNixString "hello\${}\n"
=> "\"hello\\\${}\\n\""
*/
escapeNixString = s: escape ["$"] (toJSON s);
2017-12-02 18:03:40 +00:00
2020-10-18 11:15:40 +01:00
/* Turn a string into an exact regular expression
Type: string -> string
Example:
escapeRegex "[^a-z]*"
=> "\\[\\^a-z]\\*"
*/
escapeRegex = escape (stringToCharacters "\\[{()^$?*+|.");
/* Quotes a string if it can't be used as an identifier directly.
Type: string -> string
Example:
escapeNixIdentifier "hello"
=> "hello"
escapeNixIdentifier "0abc"
=> "\"0abc\""
*/
escapeNixIdentifier = s:
# Regex from https://github.com/NixOS/nix/blob/d048577909e383439c2549e849c5c2f2016c997e/src/libexpr/lexer.l#L91
if match "[a-zA-Z_][a-zA-Z0-9_'-]*" s != null
then s else escapeNixString s;
/* Escapes a string such that it is safe to include verbatim in an XML
document.
Type: string -> string
Example:
escapeXML ''"test" 'test' < & >''
=> "\\[\\^a-z]\\*"
*/
escapeXML = builtins.replaceStrings
["\"" "'" "<" ">" "&"]
["&quot;" "&apos;" "&lt;" "&gt;" "&amp;"];
# Obsolete - use replaceStrings instead.
2015-07-24 14:45:41 +01:00
replaceChars = builtins.replaceStrings or (
del: new: s:
let
substList = lib.zipLists del new;
subst = c:
let found = lib.findFirst (sub: sub.fst == c) null substList; in
if found == null then
c
else
found.snd;
in
2015-07-24 14:45:41 +01:00
stringAsChars subst s);
2015-07-24 14:45:41 +01:00
# Case conversion utilities.
lowerChars = stringToCharacters "abcdefghijklmnopqrstuvwxyz";
upperChars = stringToCharacters "ABCDEFGHIJKLMNOPQRSTUVWXYZ";
/* Converts an ASCII string to lower-case.
Type: toLower :: string -> string
Example:
toLower "HOME"
=> "home"
*/
toLower = replaceChars upperChars lowerChars;
/* Converts an ASCII string to upper-case.
Type: toUpper :: string -> string
Example:
toUpper "home"
=> "HOME"
*/
toUpper = replaceChars lowerChars upperChars;
/* Appends string context from another string. This is an implementation
detail of Nix.
Strings in Nix carry an invisible `context` which is a list of strings
representing store paths. If the string is later used in a derivation
attribute, the derivation will properly populate the inputDrvs and
inputSrcs.
Example:
pkgs = import <nixpkgs> { };
addContextFrom pkgs.coreutils "bar"
=> "bar"
*/
addContextFrom = a: b: substring 0 0 a + b;
/* Cut a string with a separator and produces a list of strings which
were separated by this separator.
Example:
splitString "." "foo.bar.baz"
=> [ "foo" "bar" "baz" ]
splitString "/" "/usr/local/bin"
=> [ "" "usr" "local" "bin" ]
*/
splitString = _sep: _s:
let
2020-10-18 11:15:40 +01:00
sep = builtins.unsafeDiscardStringContext _sep;
s = builtins.unsafeDiscardStringContext _s;
splits = builtins.filter builtins.isString (builtins.split (escapeRegex sep) s);
in
2020-10-18 11:15:40 +01:00
map (v: addContextFrom _sep (addContextFrom _s v)) splits;
/* Return a string without the specified prefix, if the prefix matches.
Type: string -> string -> string
Example:
removePrefix "foo." "foo.bar.baz"
=> "bar.baz"
removePrefix "xxx" "foo.bar.baz"
=> "foo.bar.baz"
*/
removePrefix =
# Prefix to remove if it matches
prefix:
# Input string
str:
let
preLen = stringLength prefix;
sLen = stringLength str;
in
if hasPrefix prefix str then
substring preLen (sLen - preLen) str
else
str;
/* Return a string without the specified suffix, if the suffix matches.
Type: string -> string -> string
Example:
removeSuffix "front" "homefront"
=> "home"
removeSuffix "xxx" "homefront"
=> "homefront"
*/
removeSuffix =
# Suffix to remove if it matches
suffix:
# Input string
str:
let
sufLen = stringLength suffix;
sLen = stringLength str;
in
if sufLen <= sLen && suffix == substring (sLen - sufLen) sufLen str then
substring 0 (sLen - sufLen) str
else
str;
/* Return true if string v1 denotes a version older than v2.
Example:
versionOlder "1.1" "1.2"
=> true
versionOlder "1.1" "1.1"
=> false
*/
versionOlder = v1: v2: compareVersions v2 v1 == 1;
/* Return true if string v1 denotes a version equal to or newer than v2.
Example:
versionAtLeast "1.1" "1.0"
=> true
versionAtLeast "1.1" "1.1"
=> true
versionAtLeast "1.1" "1.2"
=> false
*/
versionAtLeast = v1: v2: !versionOlder v1 v2;
2019-11-24 16:19:32 +00:00
/* This function takes an argument that's either a derivation or a
derivation's "name" attribute and extracts the name part from that
argument.
Example:
getName "youtube-dl-2016.01.01"
=> "youtube-dl"
getName pkgs.youtube-dl
=> "youtube-dl"
*/
getName = x:
let
parse = drv: (parseDrvName drv).name;
2019-11-24 16:19:32 +00:00
in if isString x
then parse x
else x.pname or (parse x.name);
/* This function takes an argument that's either a derivation or a
derivation's "name" attribute and extracts the version part from that
argument.
Example:
getVersion "youtube-dl-2016.01.01"
=> "2016.01.01"
getVersion pkgs.youtube-dl
=> "2016.01.01"
*/
getVersion = x:
let
parse = drv: (parseDrvName drv).version;
in if isString x
then parse x
else x.version or (parse x.name);
/* Extract name with version from URL. Ask for separator which is
supposed to start extension.
Example:
nameFromURL "https://nixos.org/releases/nix/nix-1.7/nix-1.7-x86_64-linux.tar.bz2" "-"
=> "nix"
nameFromURL "https://nixos.org/releases/nix/nix-1.7/nix-1.7-x86_64-linux.tar.bz2" "_"
=> "nix-1.7-x86"
*/
nameFromURL = url: sep:
let
components = splitString "/" url;
filename = lib.last components;
name = head (splitString sep filename);
in assert name != filename; name;
/* Create an --{enable,disable}-<feat> string that can be passed to
standard GNU Autoconf scripts.
Example:
enableFeature true "shared"
=> "--enable-shared"
enableFeature false "shared"
=> "--disable-shared"
*/
enableFeature = enable: feat:
assert isString feat; # e.g. passing openssl instead of "openssl"
"--${if enable then "enable" else "disable"}-${feat}";
/* Create an --{enable-<feat>=<value>,disable-<feat>} string that can be passed to
standard GNU Autoconf scripts.
Example:
enableFeatureAs true "shared" "foo"
=> "--enable-shared=foo"
enableFeatureAs false "shared" (throw "ignored")
=> "--disable-shared"
*/
enableFeatureAs = enable: feat: value: enableFeature enable feat + optionalString enable "=${value}";
/* Create an --{with,without}-<feat> string that can be passed to
standard GNU Autoconf scripts.
Example:
withFeature true "shared"
=> "--with-shared"
withFeature false "shared"
=> "--without-shared"
*/
withFeature = with_: feat:
assert isString feat; # e.g. passing openssl instead of "openssl"
"--${if with_ then "with" else "without"}-${feat}";
/* Create an --{with-<feat>=<value>,without-<feat>} string that can be passed to
standard GNU Autoconf scripts.
Example:
withFeatureAs true "shared" "foo"
=> "--with-shared=foo"
withFeatureAs false "shared" (throw "ignored")
=> "--without-shared"
*/
withFeatureAs = with_: feat: value: withFeature with_ feat + optionalString with_ "=${value}";
/* Create a fixed width string with additional prefix to match
required width.
2015-07-23 16:41:35 +01:00
This function will fail if the input string is longer than the
requested length.
Type: fixedWidthString :: int -> string -> string -> string
Example:
fixedWidthString 5 "0" (toString 15)
=> "00015"
*/
fixedWidthString = width: filler: str:
let
strw = lib.stringLength str;
reqWidth = width - (lib.stringLength filler);
in
assert lib.assertMsg (strw <= width)
"fixedWidthString: requested string length (${
toString width}) must not be shorter than actual length (${
toString strw})";
if strw == width then str else filler + fixedWidthString reqWidth filler str;
/* Format a number adding leading zeroes up to fixed width.
2015-07-23 16:41:35 +01:00
Example:
fixedWidthNumber 5 15
=> "00015"
*/
fixedWidthNumber = width: n: fixedWidthString width "0" (toString n);
2020-03-07 01:44:56 +00:00
/* Convert a float to a string, but emit a warning when precision is lost
during the conversion
Example:
floatToString 0.000001
=> "0.000001"
floatToString 0.0000001
=> trace: warning: Imprecise conversion from float to string 0.000000
"0.000000"
*/
floatToString = float: let
result = toString float;
precise = float == fromJSON result;
in lib.warnIf (!precise) "Imprecise conversion from float to string ${result}"
result;
2020-03-07 01:44:56 +00:00
/* Check whether a value can be coerced to a string */
isCoercibleToString = x:
elem (typeOf x) [ "path" "string" "null" "int" "float" "bool" ] ||
(isList x && lib.all isCoercibleToString x) ||
x ? outPath ||
x ? __toString;
/* Check whether a value is a store path.
Example:
isStorePath "/nix/store/d945ibfx9x185xf04b890y4f9g3cbb63-python-2.7.11/bin/python"
=> false
isStorePath "/nix/store/d945ibfx9x185xf04b890y4f9g3cbb63-python-2.7.11"
=> true
isStorePath pkgs.python
=> true
isStorePath [] || isStorePath 42 || isStorePath {} ||
=> false
*/
isStorePath = x:
if !(isList x) && isCoercibleToString x then
let str = toString x; in
substring 0 1 str == "/"
&& dirOf str == storeDir
else
false;
/* Parse a string as an int.
Type: string -> int
Example:
toInt "1337"
=> 1337
toInt "-4"
=> -4
toInt "3.14"
=> error: floating point JSON numbers are not supported
*/
# Obviously, it is a bit hacky to use fromJSON this way.
toInt = str:
let may_be_int = fromJSON str; in
if isInt may_be_int
then may_be_int
else throw "Could not convert ${str} to int.";
/* Read a list of paths from `file`, relative to the `rootPath`.
Lines beginning with `#` are treated as comments and ignored.
Whitespace is significant.
NOTE: This function is not performant and should be avoided.
Example:
readPathsFromFile /prefix
./pkgs/development/libraries/qt-5/5.4/qtbase/series
=> [ "/prefix/dlopen-resolv.patch" "/prefix/tzdir.patch"
"/prefix/dlopen-libXcursor.patch" "/prefix/dlopen-openssl.patch"
"/prefix/dlopen-dbus.patch" "/prefix/xdg-config-dirs.patch"
"/prefix/nix-profiles-library-paths.patch"
"/prefix/compose-search-path.patch" ]
*/
readPathsFromFile = lib.warn "lib.readPathsFromFile is deprecated, use a list instead"
(rootPath: file:
let
lines = lib.splitString "\n" (readFile file);
removeComments = lib.filter (line: line != "" && !(lib.hasPrefix "#" line));
relativePaths = removeComments lines;
absolutePaths = map (path: rootPath + "/${path}") relativePaths;
in
absolutePaths);
2016-07-29 09:15:37 +01:00
/* Read the contents of a file removing the trailing \n
Type: fileContents :: path -> string
2016-07-29 09:15:37 +01:00
Example:
$ echo "1.0" > ./version
fileContents ./version
=> "1.0"
*/
fileContents = file: removeSuffix "\n" (readFile file);
/* Creates a valid derivation name from a potentially invalid one.
Type: sanitizeDerivationName :: String -> String
Example:
sanitizeDerivationName "../hello.bar # foo"
=> "-hello.bar-foo"
sanitizeDerivationName ""
=> "unknown"
sanitizeDerivationName pkgs.hello
=> "-nix-store-2g75chlbpxlrqn15zlby2dfh8hr9qwbk-hello-2.10"
*/
sanitizeDerivationName = string: lib.pipe string [
# Get rid of string context. This is safe under the assumption that the
# resulting string is only used as a derivation name
unsafeDiscardStringContext
# Strip all leading "."
(x: elemAt (match "\\.*(.*)" x) 0)
# Split out all invalid characters
# https://github.com/NixOS/nix/blob/2.3.2/src/libstore/store-api.cc#L85-L112
# https://github.com/NixOS/nix/blob/2242be83c61788b9c0736a92bb0b5c7bbfc40803/nix-rust/src/store/path.rs#L100-L125
(split "[^[:alnum:]+._?=-]+")
# Replace invalid character ranges with a "-"
(concatMapStrings (s: if lib.isList s then "-" else s))
# Limit to 211 characters (minus 4 chars for ".drv")
(x: substring (lib.max (stringLength x - 207) 0) (-1) x)
# If the result is empty, replace it with "unknown"
(x: if stringLength x == 0 then "unknown" else x)
];
}