| __init__.py |
Mozlog aims to standardize log handling and formatting within Mozilla.
It implements a JSON-based structured logging protocol with convenience
facilities for recording test results.
The old unstructured module is deprecated. It simply wraps Python's
logging_ module and adds a few convenience methods for logging test
results and events.
|
1054 |
- |
| capture.py |
Wrapper for file like objects to redirect output to logger
instead |
3150 |
- |
| commandline.py |
a subset of formatters for non test harnesses related applications |
12274 |
- |
| formatters |
|
|
- |
| handlers |
|
|
- |
| logtypes.py |
A DataType that contains other DataTypes.
ContainerTypes must specify which other DataType they will contain. ContainerTypes
may contain other ContainerTypes.
Some examples:
List(Int, 'numbers')
Tuple((Unicode, Int, Any), 'things')
Dict(Unicode, 'config')
Dict({TestId: Status}, 'results')
Dict(List(Unicode), 'stuff')
|
8475 |
- |
| proxy.py |
A ProxyLogger behaves like a
:class:`mozlog.structuredlog.StructuredLogger`.
Each method and attribute access will be forwarded to the underlying
StructuredLogger.
RuntimeError will be raised when the default logger is not yet initialized.
|
2426 |
- |
| pytest_mozlog |
|
|
- |
| reader.py |
Return a generator that will return the entries in a structured log file.
Note that the caller must not close the file whilst the generator is still
in use.
:param log_f: file-like object containing the raw log entries, one per line
:param raise_on_error: boolean indicating whether ValueError should be raised
for lines that cannot be decoded. |
2838 |
- |
| scripts |
|
|
- |
| stdadapter.py |
Adapter for stdlib logging so that it produces structured
messages rather than standard logging messages
:param logger: logging.Logger to wrap |
1488 |
- |
| structuredlog.py |
Structured Logging for recording test results.
Allowed actions, and subfields:
suite_start
tests - List of test names
name - Name for the suite
run_info - Dictionary of run properties
add_subsuite
name - Name for the subsuite (must be unique)
run_info - Updates to the suite run_info (optional)
suite_end
test_start
test - ID for the test
path - Relative path to test (optional)
subsuite - Name of the subsuite to which test belongs (optional)
test_end
test - ID for the test
status [PASS | FAIL | OK | ERROR | TIMEOUT | CRASH |
ASSERT PRECONDITION_FAILED | SKIP] - test status
expected [As for status] - Status that the test was expected to get,
or absent if the test got the expected status
extra - Dictionary of harness-specific extra information e.g. debug info
known_intermittent - List of known intermittent statuses that should
not fail a test. eg. ['FAIL', 'TIMEOUT']
subsuite - Name of the subsuite to which test belongs (optional)
test_status
test - ID for the test
subtest - Name of the subtest
status [PASS | FAIL | TIMEOUT |
PRECONDITION_FAILED | NOTRUN | SKIP] - test status
expected [As for status] - Status that the subtest was expected to get,
or absent if the subtest got the expected status
known_intermittent - List of known intermittent statuses that should
not fail a test. eg. ['FAIL', 'TIMEOUT']
subsuite - Name of the subsuite to which test belongs (optional)
process_output
process - PID of the process
command - Command line of the process
data - Output data from the process
test - ID of the test that the process was running (optional)
subsuite - Name of the subsuite that the process was running (optional)
assertion_count
count - Number of assertions produced
min_expected - Minimum expected number of assertions
max_expected - Maximum expected number of assertions
subsuite - Name of the subsuite for the tests that ran (optional)
lsan_leak
frames - List of stack frames from the leak report
scope - An identifier for the set of tests run during the browser session
(e.g. a directory name)
allowed_match - A stack frame in the list that matched a rule meaning the
leak is expected
subsuite - Name of the subsuite for the tests that ran (optional)
lsan_summary
bytes - Number of bytes leaked
allocations - Number of allocations
allowed - Boolean indicating whether all detected leaks matched allow rules
subsuite - Name of the subsuite for the tests that ran (optional)
mozleak_object
process - Process that leaked
bytes - Number of bytes that leaked
name - Name of the object that leaked
scope - An identifier for the set of tests run during the browser session
(e.g. a directory name)
allowed - Boolean indicating whether the leak was permitted
subsuite - Name of the subsuite for the tests that ran (optional)
log
level [CRITICAL | ERROR | WARNING |
INFO | DEBUG] - level of the logging message
message - Message to log
Subfields for all messages:
action - the action type of the current message
time - the timestamp in ms since the epoch of the log message
thread - name for the thread emitting the message
pid - id of the python process in which the logger is running
source - name for the source emitting the message
component - name of the subcomponent emitting the message
|
27968 |
- |
| unstructured |
|
|
- |