get_url: Fix regex for GNU digest line (#86134)

* GNU digest line may contain multiple spaces between
  checksum and filename. Fix regex to handle this situation.

Fixes: #86132

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

changelogs/fragments/get_url_regex.yml

@@ -0,0 +1,3 @@
+---
+bugfixes:
+    - get_url - fix regex for GNU Digest line which is used in comparing checksums (https://github.com/ansible/ansible/issues/86132).

lib/ansible/modules/get_url.py

@@ -477,17 +477,13 @@ def is_url(checksum):
     return urlsplit(checksum).scheme in supported_schemes
 
 
-def parse_digest_lines(filename, lines):
+def parse_digest_lines(filename: str, lines: list[str]) -> list[tuple[str, str]]:
     """Returns a list of tuple containing the filename and digest depending upon
       the lines provided
-
-    Args:
-        filename (str): Name of the filename, used only when the digest is one-liner
-        lines (list): A list of lines containing filenames and checksums
     """
     checksum_map = []
     BSD_DIGEST_LINE = re.compile(r'^(\w+) ?\((?P<path>.+)\) ?= (?P<digest>[\w.]+)$')
-    GNU_DIGEST_LINE = re.compile(r'^(?P<digest>[\w.]+) ([ *])(?P<path>.+)$')
+    GNU_DIGEST_LINE = re.compile(r'^(?P<digest>[\w.]+)\s+(\*|\.\/|\.)?(?P<path>.+)$')
 
     if len(lines) == 1 and len(lines[0].split()) == 1:
         # Only a single line with a single string

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

@@ -1,20 +1,6 @@
 # Test code for the get_url module
 # (c) 2014, Richard Isaacson <richard.c.isaacson@gmail.com>
-
+# GNU General Public License v3.0+ (see COPYING or https://www.gnu.org/licenses/gpl-3.0.txt)
-# This file is part of Ansible
-#
-# Ansible is free software: you can redistribute it and/or modify
-# it under the terms of the GNU General Public License as published by
-# the Free Software Foundation, either version 3 of the License, or
-# (at your option) any later version.
-#
-# Ansible is distributed in the hope that it will be useful,
-# but WITHOUT ANY WARRANTY; without even the implied warranty of
-# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
-# GNU General Public License for more details.
-#
-# You should have received a copy of the GNU General Public License
-# along with Ansible.  If not, see <https://www.gnu.org/licenses/>.
 
 - name: Determine if python looks like it will support modern ssl features like SNI
   command: "{{ ansible_python.executable }} -c 'from ssl import SSLContext'"
@@ -332,13 +318,13 @@
 
 - name: create src file
   copy:
-    dest: '{{ files_dir }}/27617.txt'
+    dest: '{{ files_dir }}/{{ item }}.txt'
-    content: "ptux"
-
-- name: create duplicate src file
-  copy:
-    dest: '{{ files_dir }}/71420.txt'
     content: "ptux"
+  loop:
+    - 27617
+    - 71420
+    - 86132
+    - 86132_single_space
 
 - name: create sha1 checksum file of src
   copy:
@@ -346,6 +332,8 @@
     content: |
       a97e6837f60cec6da4491bab387296bbcd72bdba  27617.txt
       a97e6837f60cec6da4491bab387296bbcd72bdba  71420.txt
+      a97e6837f60cec6da4491bab387296bbcd72bdba       86132.txt
+      a97e6837f60cec6da4491bab387296bbcd72bdba 86132_single_space.txt
       3911340502960ca33aece01129234460bfeb2791  not_target1.txt
       1b4b6adf30992cedb0f6edefd6478ff0a593b2e4  not_target2.txt
 
@@ -355,6 +343,8 @@
     content: |
       b1b6ce5073c8fac263a8fc5edfffdbd5dec1980c784e09c5bc69f8fb6056f006.  27617.txt
       b1b6ce5073c8fac263a8fc5edfffdbd5dec1980c784e09c5bc69f8fb6056f006.  71420.txt
+      b1b6ce5073c8fac263a8fc5edfffdbd5dec1980c784e09c5bc69f8fb6056f006.     86132.txt
+      b1b6ce5073c8fac263a8fc5edfffdbd5dec1980c784e09c5bc69f8fb6056f006. 86132_single_space.txt
       30949cc401e30ac494d695ab8764a9f76aae17c5d73c67f65e9b558f47eff892  not_target1.txt
       d0dbfc1945bc83bf6606b770e442035f2c4e15c886ee0c22fb3901ba19900b5b  not_target2.txt
 
@@ -364,6 +354,8 @@
     content: |
       b1b6ce5073c8fac263a8fc5edfffdbd5dec1980c784e09c5bc69f8fb6056f006.  ./27617.txt
       b1b6ce5073c8fac263a8fc5edfffdbd5dec1980c784e09c5bc69f8fb6056f006.  ./71420.txt
+      b1b6ce5073c8fac263a8fc5edfffdbd5dec1980c784e09c5bc69f8fb6056f006.     ./86132.txt
+      b1b6ce5073c8fac263a8fc5edfffdbd5dec1980c784e09c5bc69f8fb6056f006. ./86132_single_space.txt
       30949cc401e30ac494d695ab8764a9f76aae17c5d73c67f65e9b558f47eff892  ./not_target1.txt
       d0dbfc1945bc83bf6606b770e442035f2c4e15c886ee0c22fb3901ba19900b5b  ./not_target2.txt
 
@@ -373,6 +365,8 @@
     content: |
       b1b6ce5073c8fac263a8fc5edfffdbd5dec1980c784e09c5bc69f8fb6056f006. *27617.txt
       b1b6ce5073c8fac263a8fc5edfffdbd5dec1980c784e09c5bc69f8fb6056f006. *71420.txt
+      b1b6ce5073c8fac263a8fc5edfffdbd5dec1980c784e09c5bc69f8fb6056f006.    *86132.txt
+      b1b6ce5073c8fac263a8fc5edfffdbd5dec1980c784e09c5bc69f8fb6056f006. *86132_single_space.txt
       30949cc401e30ac494d695ab8764a9f76aae17c5d73c67f65e9b558f47eff892 *not_target1.txt
       d0dbfc1945bc83bf6606b770e442035f2c4e15c886ee0c22fb3901ba19900b5b *not_target2.txt
 
@@ -551,6 +545,61 @@
     path: "{{ remote_tmp_dir }}/27617.txt"
   register: stat_result_sha256_checksum_only
 
+- name: download 86132.txt with sha1 checksum url
+  get_url:
+    url: 'http://localhost:{{ http_port }}/86132.txt'
+    dest: '{{ remote_tmp_dir }}'
+    checksum: 'sha1:http://localhost:{{ http_port }}/sha1sum.txt'
+  register: result_sha1_86132
+
+- stat:
+    path: "{{ remote_tmp_dir }}/86132.txt"
+  register: stat_result_sha1_86132
+
+- name: download 86132.txt with sha256 checksum url
+  get_url:
+    url: 'http://localhost:{{ http_port }}/86132.txt'
+    dest: '{{ remote_tmp_dir }}/86132sha256.txt'
+    checksum: 'sha256:http://localhost:{{ http_port }}/sha256sum.txt'
+  register: result_sha256_86132
+
+- stat:
+    path: "{{ remote_tmp_dir }}/86132.txt"
+  register: stat_result_sha256_86132
+
+- name: download 86132.txt with sha256 checksum url with dot leading paths
+  get_url:
+    url: 'http://localhost:{{ http_port }}/86132.txt'
+    dest: '{{ remote_tmp_dir }}/86132sha256_with_dot.txt'
+    checksum: 'sha256:http://localhost:{{ http_port }}/sha256sum_with_dot.txt'
+  register: result_sha256_with_dot_86132
+
+- stat:
+    path: "{{ remote_tmp_dir }}/86132sha256_with_dot.txt"
+  register: stat_result_sha256_with_dot_86132
+
+- name: download 86132.txt with sha256 checksum url with asterisk leading paths
+  get_url:
+    url: 'http://localhost:{{ http_port }}/86132.txt'
+    dest: '{{ remote_tmp_dir }}/86132sha256_with_asterisk.txt'
+    checksum: 'sha256:http://localhost:{{ http_port }}/sha256sum_with_asterisk.txt'
+  register: result_sha256_with_asterisk_86132
+
+- stat:
+    path: "{{ remote_tmp_dir }}/86132sha256_with_asterisk.txt"
+  register: stat_result_sha256_with_asterisk_86132
+
+- name: download 86132_single_space.txt with sha256 checksum url
+  get_url:
+    url: 'http://localhost:{{ http_port }}/86132_single_space.txt'
+    dest: '{{ remote_tmp_dir }}/86132_single_space.txt'
+    checksum: 'sha256:http://localhost:{{ http_port }}/sha256sum.txt'
+  register: result_sha256_86132_single_space
+
+- stat:
+    path: "{{ remote_tmp_dir }}/86132_single_space.txt"
+  register: stat_result_sha256_86132_single_space
+
 - name: Assert that the file was downloaded
   assert:
     that:
@@ -579,6 +628,16 @@
       - "stat_result_sha256_with_asterisk_71420.stat.exists == true"
       - "stat_result_sha256_with_file_scheme_71420.stat.exists == true"
       - "stat_result_sha256_checksum_only.stat.exists == true"
+      - result_sha1_86132 is changed
+      - result_sha256_86132 is changed
+      - result_sha256_with_dot_86132 is changed
+      - result_sha256_with_asterisk_86132 is changed
+      - "stat_result_sha1_86132.stat.exists == true"
+      - "stat_result_sha256_86132.stat.exists == true"
+      - "stat_result_sha256_with_dot_86132.stat.exists == true"
+      - "stat_result_sha256_with_asterisk_86132.stat.exists == true"
+      - result_sha256_86132_single_space is changed
+      - "stat_result_sha256_86132_single_space.stat.exists == true"
 
 - name: Test for incomplete data read (issue 85164)
   get_url:

test/units/modules/test_get_url.py

@@ -8,22 +8,59 @@ import pytest
 
 from ansible.modules.get_url import parse_digest_lines
 
+FILENAME = "sample.txt"
+
 
 @pytest.mark.parametrize(
     ("lines", "expected"),
     [
+        pytest.param(
+            [
+                "2a32d433bf82355a3f78318a5affa21866c9a98b151785494b386e6b08f40b25 sample.txt",
+            ],
+            [("2a32d433bf82355a3f78318a5affa21866c9a98b151785494b386e6b08f40b25", FILENAME)],
+            id="single-line-digest-single-space",
+        ),
+        pytest.param(
+            [
+                "2a32d433bf82355a3f78318a5affa21866c9a98b151785494b386e6b08f40b25         sample.txt",
+            ],
+            [("2a32d433bf82355a3f78318a5affa21866c9a98b151785494b386e6b08f40b25", FILENAME)],
+            id="single-line-digest-multiple-spaces",
+        ),
+        pytest.param(
+            [
+                "2a32d433bf82355a3f78318a5affa21866c9a98b151785494b386e6b08f40b25         .sample.txt",
+            ],
+            [("2a32d433bf82355a3f78318a5affa21866c9a98b151785494b386e6b08f40b25", FILENAME)],
+            id="single-line-digest-multiple-spaces-with-dot",
+        ),
+        pytest.param(
+            [
+                "2a32d433bf82355a3f78318a5affa21866c9a98b151785494b386e6b08f40b25         *sample.txt",
+            ],
+            [("2a32d433bf82355a3f78318a5affa21866c9a98b151785494b386e6b08f40b25", FILENAME)],
+            id="single-line-digest-multiple-spaces-with-asterisk",
+        ),
+        pytest.param(
+            [
+                "2a32d433bf82355a3f78318a5affa21866c9a98b151785494b386e6b08f40b25         ./sample.txt",
+            ],
+            [("2a32d433bf82355a3f78318a5affa21866c9a98b151785494b386e6b08f40b25", FILENAME)],
+            id="single-line-digest-multiple-spaces-with-dot-and-slash",
+        ),
         pytest.param(
             [
                 "a97e6837f60cec6da4491bab387296bbcd72bdba",
             ],
-            [("a97e6837f60cec6da4491bab387296bbcd72bdba", "sample.txt")],
+            [("a97e6837f60cec6da4491bab387296bbcd72bdba", FILENAME)],
             id="single-line-digest",
         ),
         pytest.param(
             [
                 "a97e6837f60cec6da4491bab387296bbcd72bdba  sample.txt",
             ],
-            [("a97e6837f60cec6da4491bab387296bbcd72bdba", "sample.txt")],
+            [("a97e6837f60cec6da4491bab387296bbcd72bdba", FILENAME)],
             id="GNU-style-digest",
         ),
         pytest.param(
@@ -33,7 +70,7 @@ from ansible.modules.get_url import parse_digest_lines
             [
                 (
                     "b1b6ce5073c8fac263a8fc5edfffdbd5dec1980c784e09c5bc69f8fb6056f006.",
-                    "sample.txt",
+                    FILENAME,
                 )
             ],
             id="BSD-style-digest",
@@ -41,5 +78,4 @@ from ansible.modules.get_url import parse_digest_lines
     ],
 )
 def test_parse_digest_lines(lines, expected):
-    filename = "sample.txt"
+    assert parse_digest_lines(filename=FILENAME, lines=lines) == expected
-    assert parse_digest_lines(filename, lines) == expected