Save memory by using a singly-linked list for extraction aliases

[wimlib] / include / wimlib / apply.h

diff --git a/include/wimlib/apply.h b/include/wimlib/apply.h
index 2d706b4e47d6e5b9a46cf2158a2b00babcf9cd6c..ad21418ef5f92f6516914dff6c2ebeef5271c894 100644 (file)
--- a/include/wimlib/apply.h
+++ b/include/wimlib/apply.h
@@ -134,10 +134,13 @@ report_apply_error(struct apply_ctx *ctx, int error_code, const tchar *path)
        return report_error(ctx->progfunc, ctx->progctx, error_code, path);
 }
 
-/* Returns any of the aliases of an inode that are being extracted.  */
-#define inode_first_extraction_dentry(inode)           \
-       list_first_entry(&(inode)->i_extraction_aliases,        \
-                        struct wim_dentry, d_extraction_alias_node)
+#define inode_first_extraction_dentry(inode)                           \
+       ((inode)->i_first_extraction_alias)
+
+#define inode_for_each_extraction_alias(dentry, inode)                 \
+       for (dentry = inode_first_extraction_dentry(inode);             \
+            dentry != NULL;                                            \
+            dentry = dentry->d_next_extraction_alias)
 
 extern int
 extract_blob_list(struct apply_ctx *ctx, const struct read_blob_callbacks *cbs);
@@ -199,10 +202,9 @@ struct apply_operations {
         * TODO: really, the extraction backends should be responsible for
         * generating 'd_extraction_name'.
         *
-        * Each dentry will refer to a valid inode in 'd_inode'.
-        * 'd_inode->i_extraction_aliases' will contain a list of just the
-        * dentries of that inode being extracted.  This will be a (possibly
-        * nonproper) subset of the 'd_inode->i_dentry' list.
+        * Each dentry will refer to a valid inode in 'd_inode'.  Each inode
+        * will contain a list of dentries of that inode being extracted; this
+        * list may be shorter than the inode's full dentry list.
         *
         * The blobs required to be extracted will already be prepared in
         * 'apply_ctx'.  The extraction backend should call extract_blob_list()