From e3a6e380337820d17c154c54eaf50ab95bba5c0d Mon Sep 17 00:00:00 2001
From: Silvan Mosberger <silvan.mosberger@tweag.io>
Date: Tue, 13 Feb 2024 22:46:11 +0100
Subject: [PATCH] lib.fileset.toList: init

---
 lib/fileset/default.nix  | 37 +++++++++++++++++++++++++++++++++++++
 lib/fileset/internal.nix | 23 +++++++++++++++++++++++
 lib/fileset/tests.sh     | 24 ++++++++++++++++++++++--
 3 files changed, 82 insertions(+), 2 deletions(-)

diff --git a/lib/fileset/default.nix b/lib/fileset/default.nix
index c007b60def0a..ce9afc796a3f 100644
--- a/lib/fileset/default.nix
+++ b/lib/fileset/default.nix
@@ -23,6 +23,10 @@
 
     Add files in file sets to the store to use as derivation sources.
 
+  - [`lib.fileset.toList`](#function-library-lib.fileset.toList):
+
+    The list of files contained in a file set.
+
   Combinators:
   - [`lib.fileset.union`](#function-library-lib.fileset.union)/[`lib.fileset.unions`](#function-library-lib.fileset.unions):
 
@@ -102,6 +106,7 @@ let
     _coerceMany
     _toSourceFilter
     _fromSourceFilter
+    _toList
     _unionMany
     _fileFilter
     _printFileset
@@ -412,6 +417,38 @@ in {
         filter = sourceFilter;
       };
 
+
+  /*
+    The list of file paths contained in the given file set.
+
+    :::{.note}
+    This function is strict in the entire file set.
+    This is in contrast with combinators [`lib.fileset.union`](#function-library-lib.fileset.union),
+    [`lib.fileset.intersection`](#function-library-lib.fileset.intersection) and [`lib.fileset.difference`](#function-library-lib.fileset.difference).
+
+    Thus it is recommended to call `toList` on file sets created using the combinators,
+    instead of doing list processing on the result of `toList`.
+    :::
+
+    The resulting list of files can be turned back into a file set using [`lib.fileset.unions`](#function-library-lib.fileset.unions).
+
+    Type:
+      toList :: FileSet -> [ Path ]
+
+    Example:
+      toList ./.
+      [ ./README.md ./Makefile ./src/main.c ./src/main.h ]
+
+      toList (difference ./. ./src)
+      [ ./README.md ./Makefile ]
+  */
+  toList =
+    # The file set whose file paths to return.
+    # This argument can also be a path,
+    # which gets [implicitly coerced to a file set](#sec-fileset-path-coercion).
+    fileset:
+    _toList (_coerce "lib.fileset.toList: Argument" fileset);
+
   /*
     The file set containing all files that are in either of two given file sets.
     This is the same as [`unions`](#function-library-lib.fileset.unions),
diff --git a/lib/fileset/internal.nix b/lib/fileset/internal.nix
index f4fcc83e1012..0d97ef174568 100644
--- a/lib/fileset/internal.nix
+++ b/lib/fileset/internal.nix
@@ -18,6 +18,7 @@ let
     attrNames
     attrValues
     mapAttrs
+    mapAttrsToList
     optionalAttrs
     zipAttrsWith
     ;
@@ -29,6 +30,7 @@ let
   inherit (lib.lists)
     all
     commonPrefix
+    concatLists
     elemAt
     filter
     findFirst
@@ -539,6 +541,27 @@ rec {
           ${baseNameOf root} = rootPathType;
         };
 
+  # Turns a file set into the list of file paths it includes.
+  # Type: fileset -> [ Path ]
+  _toList = fileset:
+    let
+      recurse = path: tree:
+        if isAttrs tree then
+          concatLists (mapAttrsToList (name: value:
+            recurse (path + "/${name}") value
+          ) tree)
+        else if tree == "directory" then
+          recurse path (readDir path)
+        else if tree == null then
+          [ ]
+        else
+          [ path ];
+    in
+    if fileset._internalIsEmptyWithoutBase then
+      [ ]
+    else
+      recurse fileset._internalBase fileset._internalTree;
+
   # Transforms the filesetTree of a file set to a shorter base path, e.g.
   # _shortenTreeBase [ "foo" ] (_create /foo/bar null)
   # => { bar = null; }
diff --git a/lib/fileset/tests.sh b/lib/fileset/tests.sh
index af8338eb7855..654a7cefc9f1 100755
--- a/lib/fileset/tests.sh
+++ b/lib/fileset/tests.sh
@@ -275,7 +275,6 @@ createTree() {
 # )
 # checkFileset './a' # Pass the fileset as the argument
 checkFileset() {
-    # New subshell so that we can have a separate trap handler, see `trap` below
     local fileset=$1
 
     # Create the tree
@@ -283,16 +282,20 @@ checkFileset() {
 
     # Process the tree into separate arrays for included paths, excluded paths and excluded files.
     local -a included=()
+    local -a includedFiles=()
     local -a excluded=()
     local -a excludedFiles=()
     for p in "${!tree[@]}"; do
         case "${tree[$p]}" in
             1)
                 included+=("$p")
+                # If keys end with a `/` we treat them as directories, otherwise files
+                if [[ ! "$p" =~ /$ ]]; then
+                    includedFiles+=("$p")
+                fi
                 ;;
             0)
                 excluded+=("$p")
-                # If keys end with a `/` we treat them as directories, otherwise files
                 if [[ ! "$p" =~ /$ ]]; then
                     excludedFiles+=("$p")
                 fi
@@ -302,6 +305,10 @@ checkFileset() {
         esac
     done
 
+    # Test that lib.fileset.toList contains exactly the included files.
+    # The /#/./ part prefixes each element with `./`
+    expectEqual "toList ($fileset)" "sort lessThan [ ${includedFiles[*]/#/./} ]"
+
     expression="toSource { root = ./.; fileset = $fileset; }"
 
     # We don't have lambda's in bash unfortunately,
@@ -511,6 +518,19 @@ expectEqual '_toSourceFilter (_create /. { foo = "regular"; }) "/foo" ""' 'true'
 expectEqual '_toSourceFilter (_create /. { foo = null; }) "/foo" ""' 'false'
 
 
+## lib.fileset.toList
+# This function is mainly tested in checkFileset
+
+# The error context for an invalid argument must be correct
+expectFailure 'toList null' 'lib.fileset.toList: Argument is of type null, but it should be a file set or a path instead.'
+
+# Works for the empty fileset
+expectEqual 'toList _emptyWithoutBase' '[ ]'
+
+# Works on empty paths
+expectEqual 'toList ./.' '[ ]'
+
+
 ## lib.fileset.union, lib.fileset.unions