Send If-modified-since header on downloading images

sphinx/transforms/post_transforms/images.py

@@ -10,6 +10,7 @@
 """
 
 import os
+from math import ceil
 from hashlib import sha1
 
 from six import text_type
@@ -17,6 +18,7 @@ from docutils import nodes
 
 from sphinx.transforms import SphinxTransform
 from sphinx.util import logging, requests
+from sphinx.util import epoch_to_rfc1123, rfc1123_to_epoch
 from sphinx.util.images import guess_mimetype, get_image_extension, parse_data_uri
 from sphinx.util.osutil import ensuredir
 
@@ -70,8 +72,13 @@ class ImageDownloader(BaseImageConverter):
         ensuredir(os.path.join(self.imagedir, dirname))
         path = os.path.join(self.imagedir, dirname, basename)
         try:
-            r = requests.get(node['uri'])
-            if r.status_code != 200:
+            headers = {}
+            if os.path.exists(path):
+                timestamp = ceil(os.stat(path).st_mtime)
+                headers['If-Modified-Since'] = epoch_to_rfc1123(timestamp)
+
+            r = requests.get(node['uri'], headers=headers)
+            if r.status_code >= 400:
                 logger.warning('Could not fetch remote image: %s [%d]' %
                                (node['uri'], r.status_code))
             else:
@@ -80,6 +87,11 @@ class ImageDownloader(BaseImageConverter):
                 with open(path, 'wb') as f:
                     f.write(r.content)
 
+                last_modified = r.headers.get('last-modified')
+                if last_modified:
+                    timestamp = rfc1123_to_epoch(last_modified)
+                    os.utime(path, (timestamp, timestamp))
+
                 mimetype = guess_mimetype(path, default='*')
                 node['candidates'].pop('?')
                 node['candidates'][mimetype] = path

sphinx/util/__init__.py

@@ -19,12 +19,15 @@ import posixpath
 import traceback
 import unicodedata
 from os import path
+from time import mktime, strptime
 from codecs import BOM_UTF8
+from datetime import datetime
 from collections import deque
 
 from six import text_type, binary_type, itervalues
 from six.moves import range
 from six.moves.urllib.parse import urlsplit, urlunsplit, quote_plus, parse_qsl, urlencode
+from babel.dates import format_datetime
 from docutils.utils import relative_path
 
 from sphinx.errors import PycodeError, SphinxParallelError, ExtensionError
@@ -615,3 +618,14 @@ def status_iterator(iterable, summary, color="darkgreen", length=0, verbosity=0,
         yield item
     if l > 0:
         logger.info('')
+
+
+def epoch_to_rfc1123(epoch):
+    """Convert datetime format epoch to RFC1123."""
+    dt = datetime.fromtimestamp(epoch)
+    fmt = 'EEE, dd LLL yyyy hh:mm:ss'
+    return format_datetime(dt, fmt, locale='en') + ' GMT'
+
+
+def rfc1123_to_epoch(rfc1123):
+    return mktime(strptime(rfc1123, '%a, %d %b %Y %H:%M:%S %Z'))