unarchive: fix non-english locales (#76542)

* unarchive: fix non-english locales

For GNU Gettext, the LANGUAGE environment variable takes precedence over LANG or LC_ALL. On systems where LANGUAGE was set to a non-english locale, the output of the tar command therefore not understood and the module failed silently ("changed": false, but the archive was not extracted).

* add tests

* changelog

changelogs/fragments/76542-fix-unarchive-on-nenglish-locales.yml

@@ -0,0 +1,2 @@
+bugfixes:
+- unarchive - Make extraction work when the LANGUAGE environment variable is set to a non-English locale.

lib/ansible/modules/unarchive.py

@@ -786,7 +786,7 @@ class TgzArchive(object):
             cmd.extend(self.include_files)
 
         locale = get_best_parsable_locale(self.module)
-        rc, out, err = self.module.run_command(cmd, cwd=self.b_dest, environ_update=dict(LANG=locale, LC_ALL=locale, LC_MESSAGES=locale))
+        rc, out, err = self.module.run_command(cmd, cwd=self.b_dest, environ_update=dict(LANG=locale, LC_ALL=locale, LC_MESSAGES=locale, LANGUAGE=locale))
         if rc != 0:
             raise UnarchiveError('Unable to list files in the archive')
 
@@ -831,7 +831,7 @@ class TgzArchive(object):
         if self.include_files:
             cmd.extend(self.include_files)
         locale = get_best_parsable_locale(self.module)
-        rc, out, err = self.module.run_command(cmd, cwd=self.b_dest, environ_update=dict(LANG=locale, LC_ALL=locale, LC_MESSAGES=locale))
+        rc, out, err = self.module.run_command(cmd, cwd=self.b_dest, environ_update=dict(LANG=locale, LC_ALL=locale, LC_MESSAGES=locale, LANGUAGE=locale))
 
         # Check whether the differences are in something that we're
         # setting anyway
@@ -885,7 +885,7 @@ class TgzArchive(object):
         if self.include_files:
             cmd.extend(self.include_files)
         locale = get_best_parsable_locale(self.module)
-        rc, out, err = self.module.run_command(cmd, cwd=self.b_dest, environ_update=dict(LANG=locale, LC_ALL=locale, LC_MESSAGES=locale))
+        rc, out, err = self.module.run_command(cmd, cwd=self.b_dest, environ_update=dict(LANG=locale, LC_ALL=locale, LC_MESSAGES=locale, LANGUAGE=locale))
         return dict(cmd=cmd, rc=rc, out=out, err=err)
 
     def can_handle_archive(self):

test/integration/targets/unarchive/tasks/main.yml

@@ -17,3 +17,4 @@
 - import_tasks: test_symlink.yml
 - import_tasks: test_download.yml
 - import_tasks: test_unprivileged_user.yml
+- import_tasks: test_different_language_var.yml

test/integration/targets/unarchive/tasks/test_different_language_var.yml

@@ -0,0 +1,41 @@
+- name: test non-ascii with different LANGUAGE
+  when: ansible_os_family == 'Debian'
+  block:
+    - name: install de language pack
+      apt: 
+        name: language-pack-de 
+        state: present
+      
+    - name: create our unarchive destination
+      file:
+        path: "{{ remote_tmp_dir }}/test-unarchive-nonascii-くらとみ-tar-gz"
+        state: directory
+
+    - name: test that unarchive works with an archive that contains non-ascii filenames
+      unarchive:
+        # Both the filename of the tarball and the filename inside the tarball have
+        # nonascii chars
+        src: "test-unarchive-nonascii-くらとみ.tar.gz"
+        dest: "{{ remote_tmp_dir }}/test-unarchive-nonascii-くらとみ-tar-gz"
+        mode: "u+rwX,go+rX"
+        remote_src: no
+      register: nonascii_result0
+
+    - name: Check that file is really there
+      stat:
+        path: "{{ remote_tmp_dir }}/test-unarchive-nonascii-くらとみ-tar-gz/storage/àâæçéèïîôœ(copy)!@#$%^&-().jpg"
+      register: nonascii_stat0
+
+    - name: Assert that nonascii tests succeeded
+      assert:
+        that:
+          - "nonascii_result0.changed == true"
+          - "nonascii_stat0.stat.exists == true"
+
+    - name: remove nonascii test
+      file:
+        path: "{{ remote_tmp_dir }}/test-unarchive-nonascii-くらとみ-tar-gz"
+        state: absent
+
+  environment:
+    LANGUAGE: de_DE:en