|
PushPull works optimally with HTTP/1.1 servers. It will also work
effectively with HTTP/1.0 servers that return a Last-Modified: date.
Servers that do not return a Last-Modified date and dynamically
created pages will always appear as though they are new to Puller.
I have created two partial solutions to this problem:
-
If Puller is in offline mode and finds a page with no Last-Modified
date then it will download the entire page and compare it with the
previous version.
- For those using online mode the comparison is not available so you
can specify how often Puller should check that URL (always, daily,
weekly, monthly).
Dynamicaly created HTML pages cause a second problem for PushPull.
Frequently they make minor changes to a page simly to specify the
current date or the number of visitors to the page without changing
the bulk of the page. With this version I have introduced some
experimental code to compare these pages and decide whether a page has
actually been updated.
PushPull will only download images which are on the same server as the
web page currently. This may lead to a few missing images.
PushPull can now cope with the ../ syntax in URLs. But it has only
been tested in a few cases.
There is no current method for removing old images from the cache or
removing dead directories. This will be addressed in future versions.
All HTML links are converted to their full http:// form instead of
local links. This means that you cannot move between two pages from
the same server in your cache.
Note: This is only the case if you are not using the "Download local
links" option, in which case you can move to those downloaded links
but not to other ones.
| 