GNU bug report logs - #78730
31.0.50; Some eglot test failures

Previous Next

To reply to this bug, email your comments to 78730 AT debbugs.gnu.org.

Toggle the display of automated, internal messages from the tracker.

Message #5 received at submit <at> debbugs.gnu.org (full text, mbox):

From: Michael Albinus <michael.albinus <at> gmx.de>
To: bug-gnu-emacs <at> gnu.org
Subject: 31.0.50; Some eglot test failures
Date: Mon, 09 Jun 2025 09:56:59 +0200

[Message part 1 (text/plain, inline)]

Hi,

The regression tests for eglot on emba.gnu.org have been adapted
recently. They run now in a Debian Trixie environment. New language
servers being installed are rust-analyzer, typescript-language-server, and
vscode-json-languageserver. See <https://emba.gnu.org/emacs/emacs/-/jobs/104841>.

As a consequence, previously skipped tests have been applied from
eglot-tests.el. Some of them fail, see
<https://emba.gnu.org/emacs/emacs/-/jobs/104844>. The failed tests are

--8<---------------cut here---------------start------------->8---
   FAILED  eglot-test-javascript-basic  "Timed out waiting for:\n((string= method \"textDocument/publishDiagnostics\"))\n"
   FAILED  eglot-test-project-wide-diagnostics-typescript  "Timed out waiting for:\n((string= method \"textDocument/publishDiagnostics\"))\n"
   FAILED  eglot-test-rust-completion-exit-function  ((should (equal (if (bound-and-true-p yas-minor-mode) "fn test() -> i32 { let v: usize = 1; v.count_ones().1234567890;" "fn test() -> i32 { let v: usize = 1; v.count_ones.1234567890;") (buffer-string))) :form (equal "fn test() -> i32 { let v: usize = 1; v.count_ones.1234567890;" #("fn test() -> i32 { let v: usize = 1; v.count_on1234.1234567890;" 0 63 (fontified nil))) :value nil :explanation (arrays-of-different-length 61 63 "fn test() -> i32 { let v: usize = 1; v.count_ones.1234567890;" #("fn test() -> i32 { let v: usize = 1; v.count_on1234.1234567890;" 0 63 (fontified nil)) first-mismatch-at 47))
--8<---------------cut here---------------end--------------->8---

See also eglot-tests.log:

[eglot-tests(1).log (text/plain, attachment)]

[Message part 3 (text/plain, inline)]

Best regards, Michael.

Message #8 received at 78730 <at> debbugs.gnu.org (full text, mbox):

From: João Távora <joaotavora <at> gmail.com>
To: Michael Albinus <michael.albinus <at> gmx.de>
Cc: 78730 <at> debbugs.gnu.org
Subject: Re: bug#78730: 31.0.50; Some eglot test failures
Date: Mon, 09 Jun 2025 10:29:13 +0100

Michael Albinus <michael.albinus <at> gmx.de> writes:

> Hi,
>
> The regression tests for eglot on emba.gnu.org have been adapted
> recently. They run now in a Debian Trixie environment. New language
> servers being installed are rust-analyzer, typescript-language-server, and
> vscode-json-languageserver. See <https://emba.gnu.org/emacs/emacs/-/jobs/104841>.

We'll that's good news!

> As a consequence, previously skipped tests have been applied from
> eglot-tests.el. Some of them fail, see
> <https://emba.gnu.org/emacs/emacs/-/jobs/104844>. The failed tests are
>
>    FAILED  eglot-test-javascript-basic  "Timed out waiting for:\n((string= method \"textDocument/publishDiagnostics\"))\n"
>    FAILED  eglot-test-project-wide-diagnostics-typescript  "Timed out waiting for:\n((string= method \"textDocument/publishDiagnostics\"))\n"
>    FAILED  eglot-test-rust-completion-exit-function  ((should (equal (if (bound-and-true-p yas-minor-mode) "fn test() -> i32 { let v: usize = 1; v.count_ones().1234567890;" "fn test() -> i32 { let v: usize = 1; v.count_ones.1234567890;") (buffer-string))) :form (equal "fn test() -> i32 { let v: usize = 1; v.count_ones.1234567890;" #("fn test() -> i32 { let v: usize = 1; v.count_on1234.1234567890;" 0 63 (fontified nil))) :value nil :explanation (arrays-of-different-length 61 63 "fn test() -> i32 { let v: usize = 1; v.count_ones.1234567890;" #("fn test() -> i32 { let v: usize = 1; v.count_on1234.1234567890;" 0 63 (fontified nil)) first-mismatch-at 47))
>

I understand.  Well the tests pass locally here, a 2018 Thinkpad 480,
fresh master.  I'm on Arch linux so I'm running close to the bleeding
edge of these language servers, package-system installed.

Ran 55 tests, 49 results as expected, 0 unexpected, 6 skipped
(2025-06-09 10:19:40+0100, 34.989411 sec)

6 skipped results:
  SKIPPED  eglot-test-eclipse-connect
  SKIPPED  eglot-test-json-basic
  SKIPPED  eglot-test-path-to-uri-windows
  SKIPPED  eglot-test-snippet-completions
  SKIPPED  eglot-test-snippet-completions-with-company
  SKIPPED  eglot-test-zig-insert-replace-completion

But it's not a huge suprise it happens on virtual machines or containers
with less-than-sane schedulers and system clocks.  Is that the case with
EMBA or is this running on bare metal?  Is there an HEYIMRUNNINGONEMBA
variable?  If there is, I think we could skip those tests for when
truish.

João

Message #11 received at 78730 <at> debbugs.gnu.org (full text, mbox):

From: Michael Albinus <michael.albinus <at> gmx.de>
To: João Távora <joaotavora <at> gmail.com>
Cc: 78730 <at> debbugs.gnu.org
Subject: Re: bug#78730: 31.0.50; Some eglot test failures
Date: Mon, 09 Jun 2025 11:45:44 +0200

João Távora <joaotavora <at> gmail.com> writes:

Hi João,

> I understand.  Well the tests pass locally here, a 2018 Thinkpad 480,
> fresh master.  I'm on Arch linux so I'm running close to the bleeding
> edge of these language servers, package-system installed.

typescript-language-server and vscode-json-languageserver are installed
via npm, so they should be recent. Unfortunately, they aren't offered as
Debian packages, still in trixie, so I decided for this.

See test/infra/Dockerfile.emba:

--8<---------------cut here---------------start------------->8---
RUN npm install -g typescript-language-server typescript
RUN npm install -g vscode-json-languageserver
--8<---------------cut here---------------end--------------->8---

> But it's not a huge suprise it happens on virtual machines or containers
> with less-than-sane schedulers and system clocks.  Is that the case with
> EMBA or is this running on bare metal?

Containers. See admin/notes/emba, it explains how you could reproduce
the docker environment locally.

> Is there an HEYIMRUNNINGONEMBA variable?  If there is, I think we
> could skip those tests for when truish.

(skip-when (getenv "EMACS_EMBA_CI")) shall do. However, not all eglot
tests fail due to a timeout.

> João

Best regards, Michael.

Message #14 received at 78730 <at> debbugs.gnu.org (full text, mbox):

From: Michael Albinus <michael.albinus <at> gmx.de>
To: João Távora <joaotavora <at> gmail.com>
Cc: 78730 <at> debbugs.gnu.org
Subject: Re: bug#78730: 31.0.50; Some eglot test failures
Date: Mon, 09 Jun 2025 12:30:35 +0200

[Message part 1 (text/plain, inline)]

Michael Albinus <michael.albinus <at> gmx.de> writes:

Hi João,

> (skip-when (getenv "EMACS_EMBA_CI")) shall do. However, not all eglot
> tests fail due to a timeout.

What about this?

[Message part 2 (text/x-patch, inline)]

diff --git a/test/lisp/progmodes/eglot-tests.el b/test/lisp/progmodes/eglot-tests.el
index 2b6e09a4d16..2b98da4134b 100644
--- a/test/lisp/progmodes/eglot-tests.el
+++ b/test/lisp/progmodes/eglot-tests.el
@@ -185,7 +185,9 @@ eglot--call-with-timeout
                 (funcall fn)))
       (cancel-timer timer)
       (when (eq retval timed-out)
-        (error "%s" (concat "Timed out " message))))))
+        (if (getenv "EMACS_EMBA_CI")
+            (ert-skip (concat "Timed out " message))
+          (error "%s" (concat "Timed out " message)))))))
 
 (defun eglot--find-file-noselect (file &optional noerror)
   (unless (or noerror

[Message part 3 (text/plain, inline)]

>> João

Best regards, Michael.

Previous Next

GNU bug tracking system
Copyright (C) 1999 Darren O. Benham, 1997,2003 nCipher Corporation Ltd, 1994-97 Ian Jackson.