4 ms·
Another solution we implemented was to have a WordPress site up, but only serving a spidered copied HTML version of it in the public_html directory. Every 2 mi
by limeyy 12y ago
Another solution we implemented was to have a WordPress site up, but only serving a spidered copied HTML version of it in the public_html directory.
Every 2 minutes the site gets fully spidered, dumped into flat HTML files. And if the dump is not successful, because of a database connection or any problem at all, it doesn't get saved and doesn't overwrite the public_html dir.
- sea6ear 12y agoThis is interesting. Do the internal links get updated to point to the static spidered version or do they still point to the dynamic website? I've been toying with the idea of creating a product / service that would let you choose which internal links went to the static copy and which would be forwarded to the original site so that dynamic stuff like commenting could still work (although perhaps display would be delayed), but everything else could be static / cached. Honestly I think a proof of concept could be done using something like wget to mirror the site and a little ruby or awk to do the specific link rewriting.