multiarchanalyze: detect some form wrong M-A:foreign

When an arch:any package ships a .so file in a public library search
path (e.g. a symlink as many lib*-dev packages do) it most likely
shouldn't be M-A:foreign. A common exception is plugins loaded into
programs, so exclude that case.

Many thanks to Johannes Schauer and Guillem Jover for helping discover
this pattern of Multi-Arch: foreign abuse.

2 files changed, 33 insertions, 0 deletions

diff --git a/multiarchanalyze.py b/multiarchanalyze.py
index 201ffc3..f00c66a 100755
--- a/multiarchanalyze.py
+++ b/multiarchanalyze.py
@@ -2,6 +2,7 @@
 
 import argparse
 import os.path
+import re
 import sqlite3
 import sys
 
@@ -77,6 +78,9 @@ def show_files(filenames):
         return "%d files starting with %s" % (len(filenames), prefix)
     return "%d files" % len(filenames)
 
+def sqlite_regexp(pattern, value):
+    return re.match(pattern, value) is not None
+
 def main():
     parser = argparse.ArgumentParser()
     parser.add_argument("-d", "--database", action="store",
@@ -85,6 +89,7 @@ def main():
     args = parser.parse_args()
     hints = []
     db = sqlite3.connect(args.database)
+    db.create_function("REGEXP", 2, sqlite_regexp)
     cur = db.cursor()
     cur.execute("SELECT name1, architecture1, architecture2, filename FROM masame_conflict;")
     same_conflicts = {}
@@ -146,6 +151,14 @@ def main():
             link="https://wiki.debian.org/MultiArch/Hints#dep-any",
             severity="normal"))
 
+    cur.execute("SELECT name FROM maforeign_library ORDER BY name;")
+    for name, in fetchiter(cur):
+        hints.append(dict(
+            binary=name,
+            description="%s is wrongly marked Multi-Arch: foreign" % name,
+            link="https://wiki.debian.org/MultiArch/Hints#ma-foreign-library",
+            severity="high"))
+
     for hint in hints:
         if "source" not in hint:
             cur.execute("SELECT distinct(source) FROM package WHERE name = ?;",
diff --git a/multiarchanalyze.sql b/multiarchanalyze.sql
index 95ebd7d..b98092f 100644
--- a/multiarchanalyze.sql
+++ b/multiarchanalyze.sql
@@ -196,4 +196,24 @@ CREATE VIEW colonany_candidate AS
 	HAVING count(dependee) = 1
 		AND min(CASE multiarch WHEN 'allowed' THEN 1 ELSE 0 END) = 1;
 
+/* Packages that are wrongly marked Multi-Arch:foreign, because they
+ *  * are architecture dependent
+ *  * ship a shared library in a public path
+ *  * and the library is not a plugin for a contained program
+ */
+DROP VIEW IF EXISTS maforeign_library;
+CREATE VIEW maforeign_library AS
+	SELECT DISTINCT p.name
+	FROM package AS p
+	WHERE architecture != 'all'
+		AND multiarch = 'foreign'
+		AND EXISTS (
+			SELECT 1 FROM content
+			WHERE pid = p.id
+				AND filename REGEXP '^\./(usr/)?lib/([a-z0-9_]*-linux-gnu[a-z0-9_*]*/)?lib.*\.so$')
+		AND NOT EXISTS (
+			SELECT 1 FROM content
+			WHERE pid = p.id
+				AND filename REGEXP '^\./(usr/)?s?bin/.*');
+
 COMMIT;