From 0e4bfd08e548e90faf5d91bddccf418dc9a62a42 Mon Sep 17 00:00:00 2001
From: Koichi ITO <koic.ito@gmail.com>
Date: Sun, 3 Mar 2024 21:25:10 +0900
Subject: [PATCH] [ruby/prism] Fix an AST and token incompatibility for
 `Prism::Translation::Parser`

Fixes https://github.com/ruby/prism/pull/2506.

This PR fixes an AST and token incompatibility between Parser gem and `Prism::Translation::Parser`
for symbols quoted with line breaks.

https://github.com/ruby/prism/commit/06ab4df8cd
---
 lib/prism/translation/parser/compiler.rb | 16 +++++++++++++++-
 lib/prism/translation/parser/lexer.rb    | 10 ++++++++++
 test/prism/fixtures/dsym_str.txt         |  2 ++
 test/prism/snapshots/dsym_str.txt        | 11 +++++++++++
 4 files changed, 38 insertions(+), 1 deletion(-)
 create mode 100644 test/prism/fixtures/dsym_str.txt
 create mode 100644 test/prism/snapshots/dsym_str.txt

diff --git a/lib/prism/translation/parser/compiler.rb b/lib/prism/translation/parser/compiler.rb
index 8170e1953b..d64e382c78 100644
--- a/lib/prism/translation/parser/compiler.rb
+++ b/lib/prism/translation/parser/compiler.rb
@@ -1528,9 +1528,23 @@ module Prism
               builder.symbol([node.unescaped, srange(node.location)])
             end
           else
+            parts = if node.value.lines.one?
+              [builder.string_internal([node.unescaped, srange(node.value_loc)])]
+            else
+              start_offset = node.value_loc.start_offset
+
+              node.value.lines.map do |line|
+                end_offset = start_offset + line.length
+                offsets = srange_offsets(start_offset, end_offset)
+                start_offset = end_offset
+
+                builder.string_internal([line, offsets])
+              end
+            end
+
             builder.symbol_compose(
               token(node.opening_loc),
-              [builder.string_internal([node.unescaped, srange(node.value_loc)])],
+              parts,
               token(node.closing_loc)
             )
           end
diff --git a/lib/prism/translation/parser/lexer.rb b/lib/prism/translation/parser/lexer.rb
index 9a0a48a00f..62ce17c7ea 100644
--- a/lib/prism/translation/parser/lexer.rb
+++ b/lib/prism/translation/parser/lexer.rb
@@ -291,6 +291,16 @@ module Prism
                 quote = value[2] == "-" || value[2] == "~" ? value[3] : value[2]
                 value = "<<#{quote == "'" || quote == "\"" ? quote : "\""}"
               end
+            when :tSTRING_CONTENT
+              unless (lines = token.value.lines).one?
+                start_offset = offset_cache[token.location.start_offset]
+                lines.map do |line|
+                  end_offset = start_offset + line.length
+                  tokens << [:tSTRING_CONTENT, [line, Range.new(source_buffer, start_offset, offset_cache[end_offset])]]
+                  start_offset = end_offset
+                end
+                next
+              end
             when :tSTRING_DVAR
               value = nil
             when :tSTRING_END
diff --git a/test/prism/fixtures/dsym_str.txt b/test/prism/fixtures/dsym_str.txt
new file mode 100644
index 0000000000..ee68dde88d
--- /dev/null
+++ b/test/prism/fixtures/dsym_str.txt
@@ -0,0 +1,2 @@
+:"foo
+  bar"
diff --git a/test/prism/snapshots/dsym_str.txt b/test/prism/snapshots/dsym_str.txt
new file mode 100644
index 0000000000..33a5e2da21
--- /dev/null
+++ b/test/prism/snapshots/dsym_str.txt
@@ -0,0 +1,11 @@
+@ ProgramNode (location: (1,0)-(2,6))
+├── locals: []
+└── statements:
+    @ StatementsNode (location: (1,0)-(2,6))
+    └── body: (length: 1)
+        └── @ SymbolNode (location: (1,0)-(2,6))
+            ├── flags: forced_us_ascii_encoding
+            ├── opening_loc: (1,0)-(1,2) = ":\""
+            ├── value_loc: (1,2)-(2,5) = "foo\n  bar"
+            ├── closing_loc: (2,5)-(2,6) = "\""
+            └── unescaped: "foo\n  bar"