From 7f646541c3f9ca47e66ed3b71f981287dbcf812c Mon Sep 17 00:00:00 2001
From: Pablo Hoffman <pablo@pablohoffman.com>
Date: Wed, 27 Oct 2010 21:18:58 -0200
Subject: [PATCH] added trackref stats to memory debugger report. closes #272

---
 scrapy/contrib/memdebug.py | 15 +++++++++++----
 scrapy/utils/trackref.py   | 18 ++++++++++--------
 2 files changed, 21 insertions(+), 12 deletions(-)

diff --git a/scrapy/contrib/memdebug.py b/scrapy/contrib/memdebug.py
index f2917e506..2d79d0b80 100644
--- a/scrapy/contrib/memdebug.py
+++ b/scrapy/contrib/memdebug.py
@@ -4,6 +4,7 @@ MemoryDebugger extension
 See documentation in docs/topics/extensions.rst
 """
 
+import os
 import gc
 import socket
 
@@ -12,6 +13,7 @@ from scrapy.xlib.pydispatch import dispatcher
 from scrapy import signals
 from scrapy.exceptions import NotConfigured
 from scrapy.mail import MailSender
+from scrapy.utils.trackref import format_live_refs
 from scrapy.conf import settings
 from scrapy import log
 
@@ -22,7 +24,7 @@ class MemoryDebugger(object):
             import libxml2
             self.libxml2 = libxml2
         except ImportError:
-            raise NotConfigured
+            self.libxml2 = None
         if not settings.getbool('MEMDEBUG_ENABLED'):
             raise NotConfigured
 
@@ -33,7 +35,8 @@ class MemoryDebugger(object):
         dispatcher.connect(self.engine_stopped, signals.engine_stopped)
 
     def engine_started(self):
-        self.libxml2.debugMemory(1)
+        if self.libxml2:
+            self.libxml2.debugMemory(1)
 
     def engine_stopped(self):
         figures = self.collect_figures()
@@ -41,12 +44,13 @@ class MemoryDebugger(object):
         self.log_or_send_report(report)
 
     def collect_figures(self):
-        self.libxml2.cleanupParser()
         gc.collect()
 
         figures = []
         figures.append(("Objects in gc.garbage", len(gc.garbage), ""))
-        figures.append(("libxml2 memory leak", self.libxml2.debugMemory(1), "bytes"))
+        if self.libxml2:
+            self.libxml2.cleanupParser()
+            figures.append(("libxml2 memory leak", self.libxml2.debugMemory(1), "bytes"))
         return figures
 
     def create_report(self, figures):
@@ -54,6 +58,9 @@ class MemoryDebugger(object):
         s += "SCRAPY MEMORY DEBUGGER RESULTS\n\n"
         for f in figures:
             s += "%-30s : %d %s\n" % f
+        if settings.getbool('TRACK_REFS'):
+            s += os.linesep
+            s += format_live_refs()
         return s
 
     def log_or_send_report(self, report):
diff --git a/scrapy/utils/trackref.py b/scrapy/utils/trackref.py
index 6006c1bbe..e28489ff3 100644
--- a/scrapy/utils/trackref.py
+++ b/scrapy/utils/trackref.py
@@ -10,7 +10,7 @@ and no performance penalty at all when disabled (as object_ref becomes just an
 alias to object in that case).
 """
 
-import weakref
+import weakref, os
 from collections import defaultdict
 from time import time
 from operator import itemgetter
@@ -34,12 +34,10 @@ class object_ref(object):
 if not settings.getbool('TRACK_REFS'):
     object_ref = object
 
-def print_live_refs(ignore=NoneType):
+def format_live_refs(ignore=NoneType):
     if object_ref is object:
-        print "The trackref module is disabled. Use TRACK_REFS setting to enable it."
-        return
-    print "Live References"
-    print
+        return "The trackref module is disabled. Use TRACK_REFS setting to enable it."
+    s = "Live References" + os.linesep + os.linesep
     now = time()
     for cls, wdict in live_refs.iteritems():
         if not wdict:
@@ -47,8 +45,12 @@ def print_live_refs(ignore=NoneType):
         if issubclass(cls, ignore):
             continue
         oldest = min(wdict.itervalues())
-        print "%-30s %6d   oldest: %ds ago" % (cls.__name__, len(wdict), \
-            now-oldest)
+        s += "%-30s %6d   oldest: %ds ago" % (cls.__name__, len(wdict), \
+            now-oldest) + os.linesep
+    return s
+
+def print_live_refs(*a, **kw):
+    print format_live_refs(*a, **kw)
 
 def get_oldest(class_name):
     for cls, wdict in live_refs.iteritems():