core/file to 5.40-5

core/file/PKGBUILD

@@ -8,7 +8,7 @@
 
 pkgname=file
 pkgver=5.40
-pkgrel=3
+pkgrel=5
 pkgdesc='File type identification utility'
 arch=('x86_64')
 license=('custom')
@@ -17,10 +17,14 @@ url='https://www.darwinsys.com/file/'
 depends=('glibc' 'zlib' 'xz' 'bzip2')
 provides=('libmagic.so')
 source=("https://astron.com/pub/$pkgname/$pkgname-$pkgver.tar.gz"{,.asc}
+        "pr256.patch" # fix binary detection that breaks subversion tests
+        "pr261.patch" # follow-up patch which restores file 5.39 behavior
         "https://github.com/file/file/commit/9b0459afab309a82aa4e46f73a4e50dd641f3d39.patch")
 validpgpkeys=('BE04995BA8F90ED0C0C176C471112AB16CB33B3A') # Christos Zoulas
 sha256sums=('167321f43c148a553f68a0ea7f579821ef3b11c27b8cbe158e4df897e4a5dd57'
             'SKIP'
+            '5fd8dc41a10ee8097ddbffad37e07ea417b6ea5ff9d416d4aadfe15bc33a46df'
+            'c149311a54ee2539344d6ace2cb733e45bc1c134cd91d50d462dd9edbff916f5'
             'f5d9c964233b75d6b5ac1b95fe4cc23143fbc6070d3b136e91b542021959e7dd')
 
 prepare() {

core/file/pr256.patch

@@ -0,0 +1,23 @@
+From 749e1ecfc3d333e5ec8b1a2e639da7ccb2498c34 Mon Sep 17 00:00:00 2001
+From: Christos Zoulas <christos@zoulas.com>
+Date: Mon, 19 Apr 2021 18:38:04 +0000
+Subject: [PATCH] PR/256: mutableVoid: If the file is less than 3 bytes, use
+ the file length to determine type
+
+---
+ src/encoding.c | 2 +-
+ 1 file changed, 1 insertion(+), 1 deletion(-)
+
+diff --git a/src/encoding.c b/src/encoding.c
+index 0fb1f47b..31d4d125 100644
+--- a/src/encoding.c
++++ b/src/encoding.c
+@@ -285,7 +285,7 @@ looks_ ## NAME(const unsigned char *buf, size_t nbytes, file_unichar_t *ubuf, \
+ 		if (dist[i]) \
+ 			u++; \
+ 	} \
+-	if (u < 3) \
++	if (u < MIN(nbytes, 3)) \
+ 		return 0; \
+ \
+ 	return 1; \

core/file/pr261.patch

@@ -0,0 +1,44 @@
+From c07e242e766242a44ff720c149b1bdd4924ec247 Mon Sep 17 00:00:00 2001
+From: Christos Zoulas <christos@zoulas.com>
+Date: Tue, 27 Apr 2021 19:37:14 +0000
+Subject: [PATCH] Revert the fix for PR/180. It lead to PR/261. Using character
+ count heuristics ends up with confusing behavior, the following should not be
+ producing different results:     echo -n xx | ./file -     echo -n xy |
+ ./file -
+
+---
+ src/encoding.c | 13 +-------------
+ 1 file changed, 1 insertion(+), 12 deletions(-)
+
+diff --git a/src/encoding.c b/src/encoding.c
+index 31d4d1251..3647a481d 100644
+--- a/src/encoding.c
++++ b/src/encoding.c
+@@ -265,9 +265,7 @@ private int \
+ looks_ ## NAME(const unsigned char *buf, size_t nbytes, file_unichar_t *ubuf, \
+     size_t *ulen) \
+ { \
+-	size_t i, u; \
+-	unsigned char dist[256]; \
+-	memset(dist, 0, sizeof(dist)); \
++	size_t i; \
+ \
+ 	*ulen = 0; \
+ \
+@@ -278,16 +276,7 @@ looks_ ## NAME(const unsigned char *buf, size_t nbytes, file_unichar_t *ubuf, \
+ 			return 0; \
+ \
+ 		ubuf[(*ulen)++] = buf[i]; \
+-		dist[buf[i]]++; \
+ 	} \
+-	u = 0; \
+-	for (i = 0; i < __arraycount(dist); i++) { \
+-		if (dist[i]) \
+-			u++; \
+-	} \
+-	if (u < MIN(nbytes, 3)) \
+-		return 0; \
+-\
+ 	return 1; \
+ }
+