unarchive: handle content and size differences (#83454)

Consider content differs and size differs while unarchiving
the same tar.gz file

Fixes: #29610

Signed-off-by: Abhijeet Kasurde <akasurde@redhat.com>

changelogs/fragments/unarchive_differs.yml

@@ -0,0 +1,4 @@
+---
+bugfixes:
+  - unarchive - trigger change when size and content differ when other
+                properties are unchanged (https://github.com/ansible/ansible/pull/83454).

lib/ansible/modules/unarchive.py

@@ -275,6 +275,8 @@ ZIP_FILE_MODE_RE = re.compile(r'([r-][w-][SsTtx-]){3}')
 INVALID_OWNER_RE = re.compile(r': Invalid owner')
 INVALID_GROUP_RE = re.compile(r': Invalid group')
 SYMLINK_DIFF_RE = re.compile(r': Symlink differs$')
+CONTENT_DIFF_RE = re.compile(r': Contents differ$')
+SIZE_DIFF_RE = re.compile(r': Size differs$')
 
 
 def crc32(path, buffer_size):
@@ -891,16 +893,15 @@ class TgzArchive(object):
                 out += line + '\n'
             if not self.file_args['mode'] and MODE_DIFF_RE.search(line):
                 out += line + '\n'
-            if MOD_TIME_DIFF_RE.search(line):
-                out += line + '\n'
-            if MISSING_FILE_RE.search(line):
-                out += line + '\n'
-            if INVALID_OWNER_RE.search(line):
-                out += line + '\n'
-            if INVALID_GROUP_RE.search(line):
-                out += line + '\n'
-            if SYMLINK_DIFF_RE.search(line):
-                out += line + '\n'
+            differ_regexes = [
+                MOD_TIME_DIFF_RE, MISSING_FILE_RE, INVALID_OWNER_RE,
+                INVALID_GROUP_RE, SYMLINK_DIFF_RE, CONTENT_DIFF_RE,
+                SIZE_DIFF_RE
+            ]
+            for regex in differ_regexes:
+                if regex.search(line):
+                    out += line + '\n'
+
         if out:
             unarchived = False
         return dict(unarchived=unarchived, rc=rc, out=out, err=err, cmd=cmd)

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

@@ -5,6 +5,8 @@
 - import_tasks: test_tar_gz_creates.yml
 - import_tasks: test_tar_gz_owner_group.yml
 - import_tasks: test_tar_gz_keep_newer.yml
+- import_tasks: test_tar_gz_size_differs.yml
+- import_tasks: test_tar_gz_content_differs.yml
 - import_tasks: test_tar_zst.yml
 - import_tasks: test_zip.yml
 - import_tasks: test_exclude.yml

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

@@ -0,0 +1,56 @@
+- set_fact:
+    dest: '{{remote_tmp_dir}}/test-unarchive-tar-gz-content-differs'
+
+- name: create our tar.gz unarchive destination
+  file:
+    path: "{{ dest }}"
+    state: directory
+
+- name: unarchive a tar.gz file
+  unarchive:
+    src: 'content_differs.tar.gz'
+    dest: '{{ dest }}'
+    remote_src: no
+  register: unarchive_content_differs_01
+
+- name: verify that the task was marked as changed
+  assert:
+    that:
+      - unarchive_content_differs_01.changed
+
+- name: checksum the file after first unarchive
+  stat:
+    path: '{{ dest }}/hello.txt'
+    checksum_algorithm: sha256
+    get_checksum: yes
+  register: first_unarchive_content
+
+- name: unarchive a tar.gz file with same name, size but different content
+  unarchive:
+    src: 'content_differs_2.tar.gz'
+    dest: '{{ dest }}'
+    remote_src: no
+  register: unarchive_content_differs_01
+
+- name: verify that the task was marked as changed
+  assert:
+    that:
+      - unarchive_content_differs_01.changed
+
+- name: checksum the file after second unarchive
+  stat:
+    path: '{{ dest }}/hello.txt'
+    checksum_algorithm: sha256
+    get_checksum: yes
+  register: second_unarchive_content
+
+- name: verify that unarchive extracted file with new content
+  assert:
+    that:
+      - first_unarchive_content.stat.checksum != second_unarchive_content.stat.checksum
+      - first_unarchive_content.stat.size == second_unarchive_content.stat.size
+
+- name: remove our tar.gz unarchive destination
+  file:
+    path: '{{ dest }}'
+    state: absent

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

@@ -0,0 +1,52 @@
+- set_fact:
+    dest: '{{remote_tmp_dir}}/test-unarchive-tar-gz-size-differs'
+
+- name: create our tar.gz unarchive destination
+  file:
+    path: "{{ dest }}"
+    state: directory
+
+- name: unarchive a tar.gz file
+  unarchive:
+    src: 'size_differs.tar.gz'
+    dest: '{{ dest }}'
+    remote_src: no
+  register: unarchive_size_differs_01
+
+- name: verify that the task was marked as changed
+  assert:
+    that:
+      - unarchive_size_differs_01.changed
+
+- name: Check size after first unarchive
+  stat:
+    path: '{{ dest }}/hello.txt'
+  register: first_unarchive
+
+- name: unarchive a tar.gz file with same name but different size
+  unarchive:
+    src: 'size_differs_2.tar.gz'
+    dest: '{{ dest }}'
+    remote_src: no
+  register: unarchive_size_differs_02
+
+- name: verify that the task was marked as changed
+  assert:
+    that:
+      - unarchive_size_differs_02.changed
+
+- name: Check size after unarchive
+  stat:
+    path: '{{ dest }}/hello.txt'
+  register: second_unarchive
+
+- name: verify that unarchive extracted new sized file
+  assert:
+    that:
+      - first_unarchive.stat.size != second_unarchive.stat.size
+      - first_unarchive.stat.size < second_unarchive.stat.size
+
+- name: remove our tar.gz unarchive destination
+  file:
+    path: '{{ dest }}'
+    state: absent