diff --git a/swh/model/identifiers.py b/swh/model/identifiers.py
index cf3b3265f1e0392b76430b880ffaee2c8cd2171b..c53513ae6bfbb9798a0845b7d1b38c2399b8c62a 100644
--- a/swh/model/identifiers.py
+++ b/swh/model/identifiers.py
@@ -171,19 +171,21 @@ def format_date(date):
microseconds (postgres type "datetime with timezone").
Therefore, we print timestamps with no microseconds as integers, and
- timestamps with microseconds as floating point values.
+ timestamps with microseconds as floating point values. We elide the
+ trailing zeroes from microsecond values, to "future-proof" our
+ representation if we ever need more precision in timestamps.
"""
- if isinstance(date, datetime.datetime):
- if date.microsecond == 0:
- date = int(date.timestamp())
- else:
- date = date.timestamp()
- return str(date).encode()
+ if not isinstance(date, dict):
+ raise ValueError('format_date only supports dicts, %r received' % date)
+
+ seconds = date.get('seconds', 0)
+ microseconds = date.get('microseconds', 0)
+ if not microseconds:
+ return str(seconds).encode()
else:
- if date == int(date):
- date = int(date)
- return str(date).encode()
+ float_value = ('%d.%06d' % (seconds, microseconds))
+ return float_value.rstrip('0').encode()
@lru_cache()
@@ -221,8 +223,9 @@ def normalize_timestamp(time_representation):
Returns: a normalized dictionary with three keys
- - timestamp: a number of seconds since the UNIX epoch (1970-01-01 at 00:00
- UTC)
+ - timestamp: a dict with two optional keys:
+ - seconds: the integral number of seconds since the UNIX epoch
+ - microseconds: the integral number of microseconds
- offset: the timezone offset as a number of minutes relative to UTC
- negative_utc: a boolean representing whether the offset is -0000 when
offset = 0.
@@ -235,12 +238,23 @@ def normalize_timestamp(time_representation):
negative_utc = False
if isinstance(time_representation, dict):
- timestamp = time_representation['timestamp']
+ ts = time_representation['timestamp']
+ if isinstance(ts, dict):
+ seconds = ts.get('seconds', 0)
+ microseconds = ts.get('microseconds', 0)
+ elif isinstance(ts, int):
+ seconds = ts
+ microseconds = 0
+ else:
+ raise ValueError(
+ 'normalize_timestamp received non-integer timestamp member:'
+ ' %r' % ts)
offset = time_representation['offset']
if 'negative_utc' in time_representation:
negative_utc = time_representation['negative_utc']
elif isinstance(time_representation, datetime.datetime):
- timestamp = time_representation.timestamp()
+ seconds = int(time_representation.timestamp())
+ microseconds = time_representation.microsecond
utcoffset = time_representation.utcoffset()
if utcoffset is None:
raise ValueError(
@@ -250,12 +264,20 @@ def normalize_timestamp(time_representation):
# utcoffset is an integer number of minutes
seconds_offset = utcoffset.total_seconds()
offset = int(seconds_offset) // 60
- else:
- timestamp = time_representation
+ elif isinstance(time_representation, int):
+ seconds = time_representation
+ microseconds = 0
offset = 0
+ else:
+ raise ValueError(
+ 'normalize_timestamp received non-integer timestamp:'
+ ' %r' % time_representation)
return {
- 'timestamp': timestamp,
+ 'timestamp': {
+ 'seconds': seconds,
+ 'microseconds': microseconds,
+ },
'offset': offset,
'negative_utc': negative_utc,
}
diff --git a/swh/model/tests/test_identifiers.py b/swh/model/tests/test_identifiers.py
index e1adfea34f08ac419af4fc424fb4460e3eaabc6e..16a34bb9c76561bde5e97e42e2298f973b37e75e 100644
--- a/swh/model/tests/test_identifiers.py
+++ b/swh/model/tests/test_identifiers.py
@@ -55,15 +55,23 @@ class UtilityFunctionsIdentifier(unittest.TestCase):
class UtilityFunctionsDateOffset(unittest.TestCase):
def setUp(self):
- self.date = datetime.datetime(
- 2015, 11, 22, 16, 33, 56, tzinfo=datetime.timezone.utc)
- self.date_int = int(self.date.timestamp())
- self.date_repr = b'1448210036'
-
- self.date_microseconds = datetime.datetime(
- 2015, 11, 22, 16, 33, 56, 2342, tzinfo=datetime.timezone.utc)
- self.date_microseconds_float = self.date_microseconds.timestamp()
- self.date_microseconds_repr = b'1448210036.002342'
+ self.dates = {
+ b'1448210036': {
+ 'seconds': 1448210036,
+ 'microseconds': 0,
+ },
+ b'1448210036.002342': {
+ 'seconds': 1448210036,
+ 'microseconds': 2342,
+ },
+ b'1448210036.12': {
+ 'seconds': 1448210036,
+ 'microseconds': 120000,
+ }
+ }
+ self.broken_dates = [
+ 1448210036.12,
+ ]
self.offsets = {
0: b'+0000',
@@ -73,12 +81,14 @@ class UtilityFunctionsDateOffset(unittest.TestCase):
@istest
def format_date(self):
- for date in [self.date, self.date_int]:
- self.assertEqual(identifiers.format_date(date), self.date_repr)
+ for date_repr, date in self.dates.items():
+ self.assertEqual(identifiers.format_date(date), date_repr)
- for date in [self.date_microseconds, self.date_microseconds_float]:
- self.assertEqual(identifiers.format_date(date),
- self.date_microseconds_repr)
+ @istest
+ def format_date_fail(self):
+ for date in self.broken_dates:
+ with self.assertRaises(ValueError):
+ identifiers.format_date(date)
@istest
def format_offset(self):
@@ -285,7 +295,7 @@ dg1KdHOa34shrKDaOVzW
'email': b'robot@softwareheritage.org',
},
'date': {
- 'timestamp': 1437047495.0,
+ 'timestamp': {'seconds': 1437047495},
'offset': 0,
'negative_utc': False,
},
@@ -349,7 +359,7 @@ dg1KdHOa34shrKDaOVzW
'fullname': b'Jiang Xin <worldhello.net@gmail.com>',
},
'date': {
- 'timestamp': '1428538899',
+ 'timestamp': 1428538899,
'offset': 480,
},
'committer': {
@@ -357,7 +367,7 @@ dg1KdHOa34shrKDaOVzW
'email': b'worldhello.net@gmail.com',
},
'committer_date': {
- 'timestamp': '1428538899',
+ 'timestamp': 1428538899,
'offset': 480,
},
'metadata': {
@@ -383,7 +393,7 @@ dg1KdHOa34shrKDaOVzW
'fullname': b'Jiang Xin <worldhello.net@gmail.com>',
},
'date': {
- 'timestamp': '1428538899',
+ 'timestamp': 1428538899,
'offset': 480,
},
'committer': {
@@ -391,7 +401,7 @@ dg1KdHOa34shrKDaOVzW
'email': b'worldhello.net@gmail.com',
},
'committer_date': {
- 'timestamp': '1428538899',
+ 'timestamp': 1428538899,
'offset': 480,
},
'message': None,
@@ -408,7 +418,7 @@ dg1KdHOa34shrKDaOVzW
'fullname': b'Jiang Xin <worldhello.net@gmail.com>',
},
'date': {
- 'timestamp': '1428538899',
+ 'timestamp': 1428538899,
'offset': 480,
},
'committer': {
@@ -416,7 +426,7 @@ dg1KdHOa34shrKDaOVzW
'email': b'worldhello.net@gmail.com',
},
'committer_date': {
- 'timestamp': '1428538899',
+ 'timestamp': 1428538899,
'offset': 480,
},
'message': b'',
@@ -592,7 +602,7 @@ o6X/3T+vm8K3bf3driRr34c=
'target': '54e9abca4c77421e2921f5f156c9fe4a9f7441c7',
'target_type': 'revision',
'date': {
- 'timestamp': 1225281976.0,
+ 'timestamp': {'seconds': 1225281976},
'offset': 0,
'negative_utc': True,
},