From 52df48ca169e07caa5c726d51ed4ed83aed748a5 Mon Sep 17 00:00:00 2001
From: Michael Vrable <mvrable@cs.ucsd.edu>
Date: Mon, 12 May 2008 12:57:49 -0700
Subject: [PATCH] Report compressed size of data written in a backup as well as
 uncompressed.

When exiting, a summary of the size of all segments (grouped by type: data,
metadata, ...) is printed.  Extend this so that both the uncompressed and
compressed sizes of the segments are printed, and to do so now also keep
track of the compressed size of data.
---
 store.cc | 17 +++++++++++++----
 store.h  |  1 +
 2 files changed, 14 insertions(+), 4 deletions(-)

diff --git a/store.cc b/store.cc
index e6e5641..00d0878 100644
--- a/store.cc
+++ b/store.cc
@@ -29,6 +29,7 @@
 using std::max;
 using std::list;
 using std::map;
+using std::pair;
 using std::set;
 using std::string;
 
@@ -212,7 +213,8 @@ size_t Tarfile::size_estimate()
 
 static const size_t SEGMENT_SIZE = 4 * 1024 * 1024;
 
-static map<string, int64_t> group_sizes;
+/* Backup size summary: segment type -> (uncompressed size, compressed size) */
+static map<string, pair<int64_t, int64_t> > group_sizes;
 
 ObjectReference TarSegmentStore::write_object(const char *data, size_t len,
                                               const std::string &group)
@@ -225,6 +227,7 @@ ObjectReference TarSegmentStore::write_object(const char *data, size_t len,
         segment = new segment_info;
 
         segment->name = generate_uuid();
+        segment->group = group;
         segment->basename = segment->name + ".tar";
         segment->basename += filter_extension;
         segment->count = 0;
@@ -245,7 +248,7 @@ ObjectReference TarSegmentStore::write_object(const char *data, size_t len,
     segment->count++;
     segment->size += len;
 
-    group_sizes[group] += len;
+    group_sizes[group].first += len;
 
     ObjectReference ref(segment->name, id_buf);
 
@@ -266,9 +269,10 @@ void TarSegmentStore::sync()
 void TarSegmentStore::dump_stats()
 {
     printf("Data written:\n");
-    for (map<string, int64_t>::iterator i = group_sizes.begin();
+    for (map<string, pair<int64_t, int64_t> >::iterator i = group_sizes.begin();
          i != group_sizes.end(); ++i) {
-        printf("    %s: %lld\n", i->first.c_str(), i->second);
+        printf("    %s: %lld (%lld compressed)\n", i->first.c_str(),
+               i->second.first, i->second.second);
     }
 }
 
@@ -285,6 +289,11 @@ void TarSegmentStore::close_segment(const string &group)
             db->SetSegmentChecksum(segment->name, segment->basename, checksum,
                                    segment->size);
         }
+
+        struct stat stat_buf;
+        if (stat(segment->rf->get_local_path().c_str(), &stat_buf) == 0) {
+            group_sizes[segment->group].second += stat_buf.st_size;
+        }
     }
 
     segment->rf->send();
diff --git a/store.h b/store.h
index c358770..1f1edf8 100644
--- a/store.h
+++ b/store.h
@@ -116,6 +116,7 @@ public:
 private:
     struct segment_info {
         Tarfile *file;
+        std::string group;
         std::string name;           // UUID
         int count;                  // Objects written to this segment
         int size;                   // Combined size of objects written
-- 
2.20.1