Em 19 de março de 2025 13:32:50 BRT, Liliana Marie Prikler <liliana.prik...@gmail.com> escreveu: >I would strongly advise against using an old unstable Webkit.
I picked this version because it was the latest one before 2.41.92. But yes, as you commented on further, if this version is unstable, then the one that was confirmed to already work in Guix is the better alternative. >We had > 2.40.5 in 782ef67a59f4b564f16101cf23c30a3777b3f734, Yes, I saw it in the logs of webkit.scm. I'll even try building it again now. Here is the definition: >(define-public webkitgtk-for-gtk3-2.40 > (package/inherit webkitgtk-for-gtk3 > (name "webkitgtk-for-gtk3") > (version "2.40.5") > (source (origin > (method url-fetch) > (uri (string-append "https://www.webkitgtk.org/releases/"; > "webkitgtk-" version ".tar.xz")) > (sha256 > (base32 > "0zq32rn34v7hzr53s100r77aglbi6wffp9b13bcj31k6cfi53q3x")))))) I inherited from webkitgtk-for-gtk3 to make it shorter. >We could check whether we can relax the version test and allow newer >versions (we *are* currently at 2.46.6 and the next stable is 2.48); Not sure if that's a viable alternative. The reason Emacs places this check is because it depends on a deprecated feature, and it still hasn't been updated: <https://mail.gnu.org/archive/html/bug-gnu-emacs/2023-09/msg02680.html> Despite that, we could still try to remove the check just to see what happens currently. This bug is ~2 years old, so maybe something changed. I'll try building webkitgtk first, then try to find where this check is defined in Emacs later. Regards, -- Gabriel Santos
