From 4cdff74e9b953118c3afb6bddaefeb6a266a2995 Mon Sep 17 00:00:00 2001
From: Dan Helfman <witten@torsion.org>
Date: Wed, 18 Sep 2019 16:52:27 -0700
Subject: [PATCH] Support for Borg check --verify-data flag via borgmatic
 "data" consistency check (#210).

---
 NEWS                          |  1 +
 borgmatic/borg/check.py       | 20 ++++++++++++--------
 borgmatic/config/schema.yaml  | 13 +++++++------
 tests/unit/borg/test_check.py | 24 ++++++++++++++++++++++++
 4 files changed, 44 insertions(+), 14 deletions(-)

diff --git a/NEWS b/NEWS
index 6a10c2a..6304ced 100644
--- a/NEWS
+++ b/NEWS
@@ -1,4 +1,5 @@
 1.3.16.dev0
+ * #210: Support for Borg check --verify-data flag via borgmatic "data" consistency check.
  * When generating sample configuration with generate-borgmatic-config, add a space after each "#"
    comment indicator.
 
diff --git a/borgmatic/borg/check.py b/borgmatic/borg/check.py
index 393ab57..08924c2 100644
--- a/borgmatic/borg/check.py
+++ b/borgmatic/borg/check.py
@@ -24,14 +24,17 @@ def _parse_checks(consistency_config):
 
     If no "checks" option is present, return the DEFAULT_CHECKS. If the checks value is the string
     "disabled", return an empty tuple, meaning that no checks should be run.
+
+    If the "data" option is present, then make sure the "archives" option is included as well.
     '''
-    checks = consistency_config.get('checks', []) or []
+    checks = [check.lower() for check in (consistency_config.get('checks', []) or [])]
     if checks == ['disabled']:
         return ()
 
-    return (
-        tuple(check for check in checks if check.lower() not in ('disabled', '')) or DEFAULT_CHECKS
-    )
+    if 'data' in checks and 'archives' not in checks:
+        checks.append('archives')
+
+    return tuple(check for check in checks if check not in ('disabled', '')) or DEFAULT_CHECKS
 
 
 def _make_check_flags(checks, check_last=None, prefix=None):
@@ -68,13 +71,14 @@ def _make_check_flags(checks, check_last=None, prefix=None):
                 'Ignoring consistency prefix option, as "archives" is not in consistency checks.'
             )
 
+    common_flags = last_flags + prefix_flags + (('--verify-data',) if 'data' in checks else ())
+
     if set(DEFAULT_CHECKS).issubset(set(checks)):
-        return last_flags + prefix_flags
+        return common_flags
 
     return (
         tuple('--{}-only'.format(check) for check in checks if check in DEFAULT_CHECKS)
-        + last_flags
-        + prefix_flags
+        + common_flags
     )
 
 
@@ -91,7 +95,7 @@ def check_archives(
     check_last = consistency_config.get('check_last', None)
     lock_wait = None
 
-    if set(checks).intersection(set(DEFAULT_CHECKS)):
+    if set(checks).intersection(set(DEFAULT_CHECKS + ('data',))):
         remote_path_flags = ('--remote-path', remote_path) if remote_path else ()
         lock_wait = storage_config.get('lock_wait', None)
         lock_wait_flags = ('--lock-wait', str(lock_wait)) if lock_wait else ()
diff --git a/borgmatic/config/schema.yaml b/borgmatic/config/schema.yaml
index 521dce7..34f8bee 100644
--- a/borgmatic/config/schema.yaml
+++ b/borgmatic/config/schema.yaml
@@ -290,14 +290,15 @@ map:
             checks:
                 seq:
                     - type: str
-                      enum: ['repository', 'archives', 'extract', 'disabled']
+                      enum: ['repository', 'archives', 'data', 'extract', 'disabled']
                       unique: true
                 desc: |
-                    List of one or more consistency checks to run: "repository", "archives", and/or
-                    "extract". Defaults to "repository" and "archives". Set to "disabled" to disable
-                    all consistency checks. "repository" checks the consistency of the repository,
-                    "archive" checks all of the archives, and "extract" does an extraction dry-run
-                    of the most recent archive.
+                    List of one or more consistency checks to run: "repository", "archives", "data",
+                    and/or "extract". Defaults to "repository" and "archives". Set to "disabled" to
+                    disable all consistency checks. "repository" checks the consistency of the
+                    repository, "archives" checks all of the archives, "data" verifies the integrity
+                    of the data within the archives, and "extract" does an extraction dry-run of the
+                    most recent archive. Note that "data" implies "archives".
                 example:
                     - repository
                     - archives
diff --git a/tests/unit/borg/test_check.py b/tests/unit/borg/test_check.py
index 2dc0e11..3802cee 100644
--- a/tests/unit/borg/test_check.py
+++ b/tests/unit/borg/test_check.py
@@ -46,12 +46,36 @@ def test_parse_checks_with_disabled_returns_no_checks():
     assert checks == ()
 
 
+def test_parse_checks_with_data_check_also_injects_archives():
+    checks = module._parse_checks({'checks': ['data']})
+
+    assert checks == ('data', 'archives')
+
+
+def test_parse_checks_with_data_check_passes_through_archives():
+    checks = module._parse_checks({'checks': ['data', 'archives']})
+
+    assert checks == ('data', 'archives')
+
+
 def test_make_check_flags_with_repository_check_returns_flag():
     flags = module._make_check_flags(('repository',))
 
     assert flags == ('--repository-only',)
 
 
+def test_make_check_flags_with_archives_check_returns_flag():
+    flags = module._make_check_flags(('archives',))
+
+    assert flags == ('--archives-only',)
+
+
+def test_make_check_flags_with_data_check_returns_flag():
+    flags = module._make_check_flags(('data',))
+
+    assert flags == ('--verify-data',)
+
+
 def test_make_check_flags_with_extract_omits_extract_flag():
     flags = module._make_check_flags(('extract',))