Unable to install via cask with emacs 26.1

[bhrgunatha]

I'm using Arch Linux 64 bit with their emacs package. emacs: 26.1 cask : 0.8.4

I'm trying to install pdf-tools via cask which is unsuccessful.

Next I tried to build and install pdf-tools from source which also fails indicating that cask can't install undercover from melpa.

Using Emacs 26.1 
cask install
Dependency undercover failed to install: Variable binding depth exceeds max-specpdl-size
Output:
Setting `package-selected-packages' temporarily since "emacs -q" would overwrite customizations
Contacting host: melpa.org:443
Checking /home/bhrgunatha/scratch/pdf-tools/.cask/26.1/elpa/shut-up-20150423.522...
Compiling /home/bhrgunatha/scratch/pdf-tools/.cask/26.1/elpa/shut-up-20150423.522/shut-up-autoloads.el...
Compiling /home/bhrgunatha/scratch/pdf-tools/.cask/26.1/elpa/shut-up-20150423.522/shut-up-pkg.el...
Compiling /home/bhrgunatha/scratch/pdf-tools/.cask/26.1/elpa/shut-up-20150423.522/shut-up.el...
~/.cask/cask-cli.el:Error: Variable binding depth exceeds max-specpdl-size
Done (Total of 0 files compiled, 2 skipped)
Error loading autoloads: (error Variable binding depth exceeds max-specpdl-size)
Error in package--load-files-for-activation: (error Variable binding depth exceeds max-specpdl-size)

make: *** [Makefile:44: .cask/26.1] Error 255

The Cask file for pdf-tools is:

(source gnu)
(source melpa)

(package-file "lisp/pdf-tools.el")

(files "lisp/*.el"
       "README"
       "server/epdfinfo")

(development
 (depends-on "let-alist")
 (depends-on "tablist")
 (depends-on "ert-runner")
 (depends-on "undercover"))

let-alist, tablist and ert-runner are installed OK but undercover fails.

[sviridov]

Hi @bhrgunatha, sorry for the late response, but I'm afraid I cannot help with this one. The error message means nothing to me. Could you please create an issue in cask repository and link it here?

[macie]

I successfully installed pdf-tools on Arch Linux 64 bit with emacs 26.1 and cask 0.8.4. It looks like resolving metioned cask issue helped.

I think that this issue can be closed.

[sviridov]

@macie, great, thank you!