From 376e8a11a3282474c84f11ef904040d8b4971574 Mon Sep 17 00:00:00 2001
From: Daniel Oaks <daniel@danieloaks.net>
Date: Thu, 22 Oct 2015 23:19:50 +1000
Subject: [PATCH] Avoid out-of-memory error in two extra places

---
 dumpgenerator.py | 11 ++++++++---
 1 file changed, 8 insertions(+), 3 deletions(-)

diff --git a/dumpgenerator.py b/dumpgenerator.py
index daa65b9..41dcb53 100755
--- a/dumpgenerator.py
+++ b/dumpgenerator.py
@@ -600,8 +600,13 @@ def getXMLPage(config={}, title='', verbose=True, session=None):
         while not truncated and params['offset']:  # next chunk
             # get the last timestamp from the acum XML
             params['offset'] = re.findall(r_timestamp, xml)[-1]
-            xml2 = getXMLPageCore(
-                params=params, config=config, session=session)
+            try:
+                xml2 = getXMLPageCore(
+                    params=params, config=config, session=session)
+            except MemoryError:
+                print "The page's history exceeds our memory, halving limit."
+                params['limit'] = params['limit'] / 2
+                continue
 
             # are there more edits in this next XML chunk or no <page></page>?
             if re.findall(r_timestamp, xml2):
@@ -626,8 +631,8 @@ def getXMLPage(config={}, title='', verbose=True, session=None):
                     """
                     # offset is OK in this wiki, merge with the previous chunk
                     # of this page history and continue
-                    xml2 = xml2.split("</page>")[0]
                     try:
+                        xml2 = xml2.split("</page>")[0]
                         yield '  <revision>' + ('<revision>'.join(xml2.split('<revision>')[1:]))
                     except MemoryError:
                         print "The page's history exceeds our memory, halving limit."