Package Details: uefi-shell-git 30713.edk2.stable202211.40.g1fd8d08970-1

Git Clone URL: https://aur.archlinux.org/uefi-shell-git.git (read-only, click to copy)
Package Base: uefi-shell-git
Description: UEFI Shell v2 - from Tianocore EDK2 - GIT Version
Upstream URL: https://github.com/tianocore/edk2
Licenses: BSD
Conflicts: uefi-shell, uefi-shell-svn
Provides: uefi-shell
Submitter: ka2107
Maintainer: Flubbadub
Last Packager: Flubbadub
Votes: 51
Popularity: 0.000000
First Submitted: 2015-07-22 00:17 (UTC)
Last Updated: 2022-12-13 21:11 (UTC)

Dependencies (3)

Required by (0)

Sources (6)

Latest Comments

1 2 3 4 5 Next › Last »

gesh commented on 2024-03-05 18:07 (UTC)

find: cannot delete ‘./CryptoPkg/Library/MbedTlsLib/mbedtls/ChangeLog.d’: Directory not empty

Fixed by adding -type f to the find -delete invocation

gesh commented on 2024-03-05 15:28 (UTC) (edited on 2024-03-05 18:10 (UTC) by gesh)

Missed this error, so I don't know since when it started

==> Extracting sources...
  -> Creating working copy of edk2 git repo...
fatal: bad object refs/remotes/origin/dependabot/github_actions/actions/upload-artifact-4
error: /home/gesh/.local/var/cache/pacman/sources/edk2 did not send all necessary objects

Any idea what might be wrong?

UPDATE: For some reason, after nuking the pacman cache several times, this suddenly worked. No idea what changed, perhaps an upstream difference.

Flubbadub commented on 2022-12-13 21:14 (UTC)

Thanks for reporting that @Gadgethm. I have pushed a fix for that now which seems to be in line with what other projects & Arch packages have adopted. It's working okay to update it for me but please let me know if you get any further issues.

Gadgethm commented on 2022-11-20 17:59 (UTC)

This package no longer builds for me (and hasn't for a while). I get the following error when cloning certain submodules:

fatal: transport 'file' not allowed

I believe this may be related to this bug: https://vielmetti.typepad.com/logbook/2022/10/git-security-fixes-lead-to-fatal-transport-file-not-allowed-error-in-ci-systems-cve-2022-39253.html

Flubbadub commented on 2021-08-15 07:34 (UTC)

Thanks for the patch gesh - I agree it makes sense to always have the consistent format.

gesh commented on 2021-08-14 18:35 (UTC) (edited on 2021-08-14 18:38 (UTC) by gesh)

Please add --long to the git --describe command in pkgver() -- I use the installed commitish in my "what's new" script. Patch included below

commit 0806a2b2a7345683f032352176b846e0e1af57fb (HEAD -> local)
Author: gesh <gesh@gesh.uni.cx>
Date:   Sat Aug 14 00:00:00 2021 +0000

    Always include commitish in pkgver

diff --git a/PKGBUILD b/PKGBUILD
index 0cf6334..071858e 100644
--- a/PKGBUILD
+++ b/PKGBUILD
@@ -65,7 +65,7 @@ sha1sums=(

 pkgver() {
        cd "${srcdir}/${_TIANO_DIR_}/"
-       echo "$(git rev-list --count HEAD).$(git describe --always --tags)" | sed -e 's|-|\.|g'
+       echo "$(git rev-list --count HEAD).$(git describe --long --always --tags)" | sed -e 's|-|\.|g'
 }

 _setup_env_vars() {

GeneArch commented on 2021-05-04 16:35 (UTC)

Looks like its pulling older tag using git describe. Maybe annotated vs annotated issues. Changing to 'git describe --tags' works for me and pulls the later tag : edk2-stable202102-188-g1e6b0394d6

gesh commented on 2021-03-08 15:21 (UTC) (edited on 2021-03-08 15:22 (UTC) by gesh)

Noticed that submodules didn't get cached (I have SRCDEST=$XDG_CACHE_HOME/pacman/sources), so I wrote this patch. Basically, declarations at toplevel of PKGBUILD aren't passed through to functions:

From 80a3d3ea401897e9e3e1b6a27f78f6bccf00af47 Mon Sep 17 00:00:00 2001
From: gesh <gesh@gesh.uni.cx>
Date: Mon, 8 Mar 2021 15:00:00
Subject: [PATCH] Move _submod_path to _prepare_tianocore_sources()

---
 PKGBUILD | 20 ++++++++++----------
 1 file changed, 10 insertions(+), 10 deletions(-)

diff --git a/PKGBUILD b/PKGBUILD
index 7fcb285..e6d839b 100644
--- a/PKGBUILD
+++ b/PKGBUILD
@@ -45,14 +45,6 @@ provides=('uefi-shell')

 install="${_pkgname}.install"

-declare -A _submod_path
-_submod_path["CryptoPkg/Library/OpensslLib/openssl"]="openssl"
-_submod_path["SoftFloat"]="softfloat"
-_submod_path["UnitTestFrameworkPkg/Library/CmockaLib/cmocka"]="cmocka"
-_submod_path["MdeModulePkg/Universal/RegularExpressionDxe/oniguruma"]="oniguruma"
-_submod_path["MdeModulePkg/Library/BrotliCustomDecompressLib/brotli"]="brotli"
-_submod_path["BaseTools/Source/C/BrotliCompress/brotli"]="brotli"
-
 source=(
    "${_TIANO_DIR_}::git+https://github.com/tianocore/edk2.git#branch=master"
    "brotli::git+https://github.com/google/brotli"
@@ -85,6 +77,14 @@ _setup_env_vars() {
 _prepare_tianocore_sources() {
    cd "${_UDK_DIR}/"

+    declare -A _submod_path
+    _submod_path["CryptoPkg/Library/OpensslLib/openssl"]="openssl"
+    _submod_path["SoftFloat"]="softfloat"
+    _submod_path["UnitTestFrameworkPkg/Library/CmockaLib/cmocka"]="cmocka"
+    _submod_path["MdeModulePkg/Universal/RegularExpressionDxe/oniguruma"]="oniguruma"
+    _submod_path["MdeModulePkg/Library/BrotliCustomDecompressLib/brotli"]="brotli"
+    _submod_path["BaseTools/Source/C/BrotliCompress/brotli"]="brotli"
+
    msg "Updating submodules"
    git submodule init
    for _module in "${!_submod_path[@]}"; do
-- 
2.30.1

Flubbadub commented on 2020-11-10 23:51 (UTC)

It was actually down to spacing in the patch (not sure if it's on my end or the AUR has done it). I manually fixed that up and got it applied now. Thanks for your help.

gesh commented on 2020-11-10 21:48 (UTC)

I think the sticking point is the pkgver edit, try removing it. Can't test right now.