Reformat indented strings to simple strings when appropriate

src/Nixfmt/Parser.hs

@@ -17,7 +17,7 @@ import Data.Char (isAlpha)
 import Data.Foldable (toList)
 import Data.Functor (($>))
 import Data.Maybe (fromMaybe, mapMaybe, maybeToList)
-import Data.Text (Text, pack)
+import Data.Text (Text, elem, isPrefixOf, pack)
 import qualified Data.Text as Text
 import Data.Void (Void)
 import Nixfmt.Lexer (lexeme, takeTrivia, whole)
@@ -75,7 +75,7 @@ import Text.Megaparsec (
   (<|>),
  )
 import Text.Megaparsec.Char (char, digitChar)
-import Prelude hiding (String)
+import Prelude hiding (String, elem)
 
 -- HELPER FUNCTIONS
 
@@ -324,6 +324,39 @@ indentedString =
     *> fmap fixIndentedString (sepBy indentedLine (chunk "\n"))
     <* rawSymbol TDoubleSingleQuote
 
+-- | Parser for all string types (simple, URI, or indented)
+string :: Parser Term
+string =
+  (SimpleString <$> lexeme (simpleString <|> uri))
+    <|> (classifyString <$> lexeme indentedString)
+  where
+    -- Converts indented string syntax to appropriate string type.
+    -- If the content can be represented as a simple string (no newlines, quotes or backslashes),
+    -- it's reformatted as SimpleString to maintain a consistent style.
+    classifyString s
+      | shouldBeSimpleString (value s) = SimpleString (s{value = convertIndentedEscapes (value s)})
+      | otherwise = IndentedString s
+
+    shouldBeSimpleString parts =
+      not (containsNewlines parts) && not (any (any hasQuoteOrBackSlash) parts)
+
+    containsNewlines parts = length parts > 1
+
+    hasQuoteOrBackSlash (TextPart t) = '"' `elem` t || '\\' `elem` t
+    hasQuoteOrBackSlash (Interpolation _) = False
+
+    convertIndentedEscapes = map $ map convertPart
+
+    convertPart (TextPart t) = TextPart (convertEscapes t)
+    convertPart (Interpolation x) = Interpolation x
+
+    -- Converts indented string escapes to simple string escapes
+    convertEscapes t
+      | Text.null t = t
+      | "''$" `isPrefixOf` t = "\\$" <> convertEscapes (Text.drop 3 t) -- ''$ -> \$
+      | "'''" `isPrefixOf` t = "''" <> convertEscapes (Text.drop 3 t) -- ''' -> ''
+      | otherwise = Text.take 1 t <> convertEscapes (Text.drop 1 t)
+
 -- TERMS
 
 parens :: Parser Term
@@ -358,8 +391,7 @@ selectorPath' = many $ try $ selector $ Just $ symbol TDot
 -- Everything but selection
 simpleTerm :: Parser Term
 simpleTerm =
-  (SimpleString <$> lexeme (simpleString <|> uri))
-    <|> (IndentedString <$> lexeme indentedString)
+  string
     <|> (Path <$> path)
     <|> (Token <$> (envPath <|> float <|> integer <|> identifier))
     <|> parens

standard.md

@@ -327,6 +327,10 @@ For list elements, attributes, and function arguments, the following applies:
 ### Strings
 
 - The kind of quotes used in strings (`"` vs `''`) must be preserved from the input.
+  - Exception: Indented strings (`''...''`) that contain no newlines, double quote characters or backslashes are automatically reformatted as simple strings (`"..."`).
+  - When converting indented strings to simple strings, escape sequences are rewritten to maintain semantic equivalence:
+    - `''$` becomes `\$` (escaped dollar sign)
+    - `'''` becomes `''` (quotes)
 - The non-interpolated string parts must be preserved from the input
   - E.g. changing `\t` to a tab character must not be done automatically
 
@@ -343,6 +347,22 @@ For list elements, attributes, and function arguments, the following applies:
 ''
   This is a really long string that would not fit within the line length limit
 ''
+
+# Indented strings with simple content get reformatted as simple strings
+''hello''        # becomes "hello"
+''hello world''  # becomes "hello world"
+
+# Escape sequences are rewritten when converting to simple strings
+''''${pkgs.ghostscript}/bin/ps2pdf''  # becomes "\${pkgs.ghostscript}/bin/ps2pdf"
+'''test''$var''                        # becomes "'test\$var"
+'''can''t'''                          # becomes "'can't"
+
+# But these stay as indented strings
+''
+  hello
+  world
+''
+''hello "quoted" text''
 ```
 
 #### Interpolations

test/correct/indented-string.nix

@@ -9,6 +9,6 @@
     $'\t'
   ''
 
-  ''ending dollar $''
-  ''$''
+  ''"ending dollar $''
+  ''"$''
 ]

test/diff/idioms_nixos_2/in.nix

@@ -734,6 +734,7 @@ in {
 
     { assertions = [
       { assertion = cfg.database.createLocally -> cfg.config.dbtype == "mysql";
+        # single line idented strings must be reformatted to simple strings 
         message = ''services.nextcloud.config.dbtype must be set to mysql if services.nextcloud.database.createLocally is set to true.'';
       }
     ]; }

test/diff/idioms_nixos_2/out-pure.nix

@@ -794,7 +794,8 @@ in
       assertions = [
         {
           assertion = cfg.database.createLocally -> cfg.config.dbtype == "mysql";
-          message = ''services.nextcloud.config.dbtype must be set to mysql if services.nextcloud.database.createLocally is set to true.'';
+          # single line idented strings must be reformatted to simple strings
+          message = "services.nextcloud.config.dbtype must be set to mysql if services.nextcloud.database.createLocally is set to true.";
         }
       ];
     }

test/diff/idioms_nixos_2/out.nix

@@ -796,7 +796,8 @@ in
       assertions = [
         {
           assertion = cfg.database.createLocally -> cfg.config.dbtype == "mysql";
-          message = ''services.nextcloud.config.dbtype must be set to mysql if services.nextcloud.database.createLocally is set to true.'';
+          # single line idented strings must be reformatted to simple strings
+          message = "services.nextcloud.config.dbtype must be set to mysql if services.nextcloud.database.createLocally is set to true.";
         }
       ];
     }

test/diff/string/in.nix

@@ -92,4 +92,10 @@
   "--${
     "test"
   }"
+  ''''${pkgs.ghostscript}/bin/ps2pdf''
+  '''test''$test''
+  '''test'''quotes''
+  '''plain''
+  '' between spaces ''
+  '' !#$%&'()*+,-./0123456789:;<=>?@ABCDEFGHIJKLMNOPQRSTUVWXYZ[]^_`abcdefghijklmnopqrstuvwxyz{|}~''
 ]

test/diff/string/out-pure.nix

@@ -13,11 +13,11 @@
    b
   "
   ###
-  ''''
+  ""
   ###
-  ''a''
+  "a"
   ###
-  ''${""}''
+  "${""}"
   ###
   ''
     ${""}
@@ -51,7 +51,7 @@
        e
   ''
   ###
-  ''''
+  ""
   ###
   ''
     declare -a makefiles=(./*.mak)
@@ -66,15 +66,15 @@
     [${mkSectionName sectName}]
   ''
   ###
-  ''-couch_ini ${cfg.package}/etc/default.ini ${configFile} ${pkgs.writeText "couchdb-extra.ini" cfg.extraConfig} ${cfg.configFile}''
+  "-couch_ini ${cfg.package}/etc/default.ini ${configFile} ${pkgs.writeText "couchdb-extra.ini" cfg.extraConfig} ${cfg.configFile}"
   ###
   ''exec i3-input -F "mark %s" -l 1 -P 'Mark: ' ''
   ###
   ''exec i3-input -F '[con_mark="%s"] focus' -l 1 -P 'Go to: ' ''
   ###
   ''"${pkgs.name or "<unknown-name>"}";''
   ###
-  ''${pkgs.replace-secret}/bin/replace-secret '${placeholder}' '${secretFile}' '${targetFile}' ''
+  "${pkgs.replace-secret}/bin/replace-secret '${placeholder}' '${secretFile}' '${targetFile}' "
   ###
   ''
     mkdir -p "$out/lib/modules/${kernel.modDirVersion}/kernel/net/wireless/"
@@ -92,4 +92,10 @@
   ''
 
   "--${"test"}"
+  "\${pkgs.ghostscript}/bin/ps2pdf"
+  "'test\$test"
+  "'test''quotes"
+  "'plain"
+  "between spaces "
+  "!#$%&'()*+,-./0123456789:;<=>?@ABCDEFGHIJKLMNOPQRSTUVWXYZ[]^_`abcdefghijklmnopqrstuvwxyz{|}~"
 ]

test/diff/string/out.nix

@@ -13,11 +13,11 @@
    b
   "
   ###
-  ''''
+  ""
   ###
-  ''a''
+  "a"
   ###
-  ''${""}''
+  "${""}"
   ###
   ''
     ${""}
@@ -51,7 +51,7 @@
        e
   ''
   ###
-  ''''
+  ""
   ###
   ''
     declare -a makefiles=(./*.mak)
@@ -66,15 +66,15 @@
     [${mkSectionName sectName}]
   ''
   ###
-  ''-couch_ini ${cfg.package}/etc/default.ini ${configFile} ${pkgs.writeText "couchdb-extra.ini" cfg.extraConfig} ${cfg.configFile}''
+  "-couch_ini ${cfg.package}/etc/default.ini ${configFile} ${pkgs.writeText "couchdb-extra.ini" cfg.extraConfig} ${cfg.configFile}"
   ###
   ''exec i3-input -F "mark %s" -l 1 -P 'Mark: ' ''
   ###
   ''exec i3-input -F '[con_mark="%s"] focus' -l 1 -P 'Go to: ' ''
   ###
   ''"${pkgs.name or "<unknown-name>"}";''
   ###
-  ''${pkgs.replace-secret}/bin/replace-secret '${placeholder}' '${secretFile}' '${targetFile}' ''
+  "${pkgs.replace-secret}/bin/replace-secret '${placeholder}' '${secretFile}' '${targetFile}' "
   ###
   ''
     mkdir -p "$out/lib/modules/${kernel.modDirVersion}/kernel/net/wireless/"
@@ -92,4 +92,10 @@
   ''
 
   "--${"test"}"
+  "\${pkgs.ghostscript}/bin/ps2pdf"
+  "'test\$test"
+  "'test''quotes"
+  "'plain"
+  "between spaces "
+  "!#$%&'()*+,-./0123456789:;<=>?@ABCDEFGHIJKLMNOPQRSTUVWXYZ[]^_`abcdefghijklmnopqrstuvwxyz{|}~"
 ]

test/diff/string_interpol/out-pure.nix

@@ -2,9 +2,9 @@
   "${
     /* a */ "${/* b */ "${c}"}" # d
   }"
-  ''${
-    /* a */ ''${/* b */ ''${c}''}'' # d
-  }''
+  "${
+    /* a */ "${/* b */ "${c}"}" # d
+  }"
   {
     ExecStart = "${pkgs.openarena}/bin/oa_ded +set fs_basepath ${pkgs.openarena}/openarena-0.8.8 +set fs_homepath /var/lib/openarena ${
       concatMapStringsSep (x: x) " " cfg.extraFlags

test/diff/string_interpol/out.nix

@@ -2,9 +2,9 @@
   "${
     /* a */ "${/* b */ "${c}"}" # d
   }"
-  ''${
-    /* a */ ''${/* b */ ''${c}''}'' # d
-  }''
+  "${
+    /* a */ "${/* b */ "${c}"}" # d
+  }"
   {
     ExecStart = "${pkgs.openarena}/bin/oa_ded +set fs_basepath ${pkgs.openarena}/openarena-0.8.8 +set fs_homepath /var/lib/openarena ${
       concatMapStringsSep (x: x) " " cfg.extraFlags