From 3507b6b918a443bc5603b1a1f31b5157339198df Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Adri=C3=A1n=20Enr=C3=ADquez?= <adrian.enriquez@serokell.io>
Date: Tue, 24 Jan 2023 19:22:27 +0100
Subject: [PATCH] [#239][#249] Parse quoted git ls-files

Problem: When a file name contains unusual characters, its output line
in git ls-files is quoted and Xrefcheck parses the file name wrong.

Solution: If a git ls-files output line starts with a double quote, use
the String Read instance for parsing it.
---
 src/Xrefcheck/Scan.hs | 7 ++++++-
 1 file changed, 6 insertions(+), 1 deletion(-)

diff --git a/src/Xrefcheck/Scan.hs b/src/Xrefcheck/Scan.hs
index 2cdee8c1..444cd998 100644
--- a/src/Xrefcheck/Scan.hs
+++ b/src/Xrefcheck/Scan.hs
@@ -181,7 +181,7 @@ readDirectoryWith
   -> CanonicalPath
   -> IO [(CanonicalPath, a)]
 readDirectoryWith mode config scanner root = do
-  relativeFiles <- L.lines <$> getFiles
+  relativeFiles <- fmap filePath . L.lines <$> getFiles
   canonicalFiles <- mapM (root </) relativeFiles
   traverse scanFile $ filter (not . isIgnored) canonicalFiles
 
@@ -197,6 +197,11 @@ readDirectoryWith mode config scanner root = do
     getUntrackedFiles = readCreateProcess
       (shell "git ls-files --others --exclude-standard"){cwd = Just $ unCanonicalPath root} ""
 
+    filePath :: String -> FilePath
+    filePath = \case
+      quoted@('\"' : _) -> fromMaybe quoted $ readMaybe quoted
+      unquoted -> unquoted
+
     scanFile :: CanonicalPath -> IO (CanonicalPath, a)
     scanFile c = (c,) <$> scanner c