extractor: filter out non-message paths

We may allow files like "README" to appear in ssoma
repositories to reduce confusion.

diff --git a/lib/Ssoma/Extractor.pm b/lib/Ssoma/Extractor.pm
index 11ad130c93ea5280fbcb74f74dc99b7a3ad0111d..9d7bcceb13915791d669f177006294d4d3d707b3 100644 (file)
--- a/lib/Ssoma/Extractor.pm
+++ b/lib/Ssoma/Extractor.pm
@@ -13,14 +13,17 @@ sub new {
        bless { git => $git, ref => "refs/heads/master" }, $class;
 }
 
-# runs a command which returns a list of files, no file name sanitization
-# here needed since all of the path names stored in git trees are controlled
-# by us (and based on SHA-1 hexdigest)
+# runs a command which returns a list of files belonging to emails
+# This won't prevent invalid/corrupt messages from attempts at
+# being imported, though.  This allows administrators to add things
+# like a top-level README file to avoid confusing folks who may
+# accidentally check out the ssoma repository as a working copy.
 sub _flist {
        my ($cmd) = @_;
        my @rv = `$cmd`;
        $? == 0 or die "$cmd failed: $?\n";
        chomp @rv;
+       @rv = grep(m!\A[a-f0-9]{2}/[a-f0-9]{38}(?:/[a-f0-9]{40})?\z!, @rv);
        \@rv
 }