I keep needing to open pages at the Wayback Machine in Links (graphical mode) because Dillo won't load images there. I know this is because a while ago they started using redirects on the image links in their archived pages, and the old Dillo FAQ says: "If Dillo requests an image and receives a response containing a redirection (i.e., pointing to a different URL), the redirection is not followed. These have a strong tendency to be advertisements." https://dillo-browser.github.io/old/FAQ.html#q28 However it's a real pain since most of the good stuff on the Web seems to have gone offline. ~/.dillo/domainrc was added to solve similar problems with over-active blocking of images at different domains (next part of that FAQ answer). Could some sort of allow rule for image redirects be added in that too? Or is there an existing work-around for the Wayback Machine images that I haven't figured out? Example: The image link at the top of this archived webpage: http://web.archive.org/web/20001009182503/http://www.abc.net.au/defaultlo.ht... http://web.archive.org/web/20001009182503im_/http://www.abc.net.au/img/abcbl... Redirects to: http://web.archive.org/web/20000822094140im_/http://www.abc.net.au/img/abcbl... which isn't loaded unless opened on its own via the right-click menu in Dillo.
Hi, On Tue, Mar 25, 2025 at 09:20:25AM +1100, Kevin Koster wrote:
I keep needing to open pages at the Wayback Machine in Links (graphical mode) because Dillo won't load images there. I know this is because a while ago they started using redirects on the image links in their archived pages, and the old Dillo FAQ says:
"If Dillo requests an image and receives a response containing a redirection (i.e., pointing to a different URL), the redirection is not followed. These have a strong tendency to be advertisements." https://dillo-browser.github.io/old/FAQ.html#q28
However it's a real pain since most of the good stuff on the Web seems to have gone offline. ~/.dillo/domainrc was added to solve similar problems with over-active blocking of images at different domains (next part of that FAQ answer). Could some sort of allow rule for image redirects be added in that too? Or is there an existing work-around for the Wayback Machine images that I haven't figured out?
It is not implemented yet: https://github.com/dillo-browser/dillo/issues/304#issuecomment-2486646504 See Cache_redirect() in src/cache.c if you want to take a look. Best, Rodrigo.
On Tue, 25 Mar 2025 00:13:57 +0100 Rodrigo Arias <rodarima@gmail.com> wrote:
"If Dillo requests an image and receives a response containing a redirection (i.e., pointing to a different URL), the redirection is not followed. These have a strong tendency to be advertisements." https://dillo-browser.github.io/old/FAQ.html#q28
However it's a real pain since most of the good stuff on the Web seems to have gone offline. ~/.dillo/domainrc was added to solve similar problems with over-active blocking of images at different domains (next part of that FAQ answer). Could some sort of allow rule for image redirects be added in that too? Or is there an existing work-around for the Wayback Machine images that I haven't figured out?
It is not implemented yet:
https://github.com/dillo-browser/dillo/issues/304#issuecomment-2486646504
Oh right, I saw that but thought it must have been something else. In the past I believe it's been considered as intended behaviour rather than a bug (eg. in earlier mailing list discussions which I can't find now). Hence my suggestion of a domainrc type whitelist rather than an outright 'fix'.
See Cache_redirect() in src/cache.c if you want to take a look.
Thanks, I've poked around there before but not yet invested the time to study how it all fits together. I hadn't discovered the developer docs the first time though (I assumed there was only the info that's in doc/). Well, I'm still not sure if there are any docs covering the network-related code...
Hi,
Thanks, I've poked around there before but not yet invested the time to study how it all fits together. I hadn't discovered the developer docs the first time though (I assumed there was only the info that's in doc/). Well, I'm still not sure if there are any docs covering the network-related code...
There are not many, but you can find some in devdoc/. I should add them somehow to the website as well. Essentially the cache keeps a list of entries for connections in progress. When data is received, the callback for that entry is called to forward the data. But when a redirect is received, currently Dillo navigates to the new URL, as if you typed it in the location bar. What we need to support first is to allow the cache to fetch the new URL instead of changing the root URL of the current document. I believe we can do this by adding a new entry to the cache with the new URL and copying the callback and other related data, or simply reusing the same entry while we update the URL. I took a look at the time but I wasn't able to come up with a quick solution so I didn't look further. I will eventually take a look at it, but feel free to poke with the cache code in the meanwhile. Best, Rodrigo.
participants (2)
-
Kevin Koster -
Rodrigo Arias