def run(self):
super().run()
self.build_config()
# Parse the matcher query
if self.args.qmacro:
strquery = ""
qmacro_query = self.query
else:
if self.query is None:
self.parser.error(
"you need to specify a filter expression or use --file")
strquery = self.query
qmacro_query = None
if strquery:
query = arkimet.Matcher(
arkimet.dataset.http.expand_remote_query(
self.config, strquery))
else:
query = arkimet.Matcher()
with self.outfile() as outfd:
arki_query = arkimet.cmdline.ArkiQuery()
arki_query.set_inputs(self.config)
arki_query.set_processor(
query=query,
outfile=outfd,
yaml=self.args.yaml,
json=self.args.json,
annotate=self.args.annotate,
inline=self.args.inline,
data=self.args.data,
summary=self.args.summary,
summary_short=self.args.summary_short,
summary_restrict=self.args.summary_restrict,
archive=self.args.archive,
postproc=self.args.postproc,
postproc_data=self.args.postproc_data,
sort=self.args.sort,
)
if self.args.stdin:
arki_query.query_file(sys.stdin, self.args.stdin)
elif self.args.merged:
arki_query.query_merged()
elif self.args.qmacro:
arki_query.query_qmacro(self.args.qmacro, qmacro_query)
else:
arki_query.query_sections()
def run(self):
super().run()
if self.args.query:
if not self.args.input:
self.parser.error("--query needs a query on the command line")
matcher = arkimet.Matcher(self.args.input)
print(matcher.expanded)
raise Exit()
if self.args.aliases:
if self.args.input:
sections = arkimet.dataset.http.get_alias_database(
self.args.input)
else:
sections = arkimet.get_alias_database()
if self.args.output:
with open(self.args.output, "wt") as fd:
sections.write(fd)
else:
sections.write(sys.stdout)
raise Exit()
if self.args.config:
self.parser.error(
"please use arki-mergeconf instead of arki-dump --config")
if self.args.info:
import json
cfg = arkimet.config()
print(json.dumps(cfg, indent=1))
raise Exit()
if self.args.bbox:
dump = arkimet.cmdline.ArkiDump()
with self.input("rb") as fd:
bbox = dump.bbox(fd)
with self.output("wt") as fd:
print(bbox, file=fd)
raise Exit()
if self.args.from_yaml_data:
dump = arkimet.cmdline.ArkiDump()
with self.input("rb") as fdin:
with self.output("wb") as fdout:
raise Exit(dump.reverse_data(fdin, fdout))
if self.args.from_yaml_summary:
dump = arkimet.cmdline.ArkiDump()
with self.input("rb") as fdin:
with self.output("wb") as fdout:
raise Exit(dump.reverse_summary(fdin, fdout))
dump = arkimet.cmdline.ArkiDump()
with self.input("rb") as fdin:
with self.output("wb") as fdout:
raise Exit(dump.dump_yaml(fdin, fdout, self.args.annotate))
def test_query_summary_style_yaml(self):
"""
Test style=yaml summary queries
"""
res = requests.post(self.server_url +
"/dataset/test200/summary?style=yaml",
data={
"query": arki.Matcher().expanded,
})
self.assertEqual(res.text[:11], "SummaryItem")
def test_query_global_summary_style_binary(self):
"""
Test style=binary summary queries
"""
res = requests.post(self.server_url + "/summary?style=binary",
data={
"query": arki.Matcher().expanded,
})
res.raise_for_status()
self.assertEqual(res.content[:2], b"SU")
def test_query_summaryshort_style_yaml(self):
"""
Test style=yaml summary queries
"""
res = requests.post(self.server_url +
"/dataset/test200/summaryshort?style=yaml",
data={
"query": arki.Matcher().expanded,
})
res.raise_for_status()
self.assertEqual(res.text[:12], "SummaryStats")
def test_query_summaryshort_style_json(self):
"""
Test style=json summary short queries
"""
res = requests.post(self.server_url +
"/dataset/test200/summaryshort?style=json",
data={
"query": arki.Matcher().expanded,
})
res.raise_for_status()
self.assertIn("items", res.json())
def __init__(self, macro_cfg, datasets_cfg, macro_args, query):
self.macro_cfg = macro_cfg,
self.datasets_cfg = datasets_cfg
# The argument, if provided, is a date used to expand @ in date expressions
macro_args = macro_args.strip()
if macro_args:
self.date_ref = datetime.datetime.strptime(macro_args,
"%Y-%m-%d").date()
else:
self.date_ref = None
# Rows to query
self.rows = []
# Parse query
for idx, line in enumerate(query.splitlines(), start=1):
mo = self.re_line.match(line)
if not mo:
raise RuntimeError("query:{}: line not parsed: {}".format(
idx, repr(line)))
dsname = mo.group("ds")
reftime = datetime.datetime.combine(self._to_date(mo.group("d")),
self._to_time(mo.group("t")))
matcher = arki.Matcher(
"reftime: ={reftime:%Y-%m-%dT%H:%M:%S}Z; "
"timerange: {trange}; level: {level}; product: {product}".
format(reftime=reftime,
trange=self._to_matcher(mo.group("s")),
level=self._to_matcher(mo.group("l")),
product=self._to_matcher(mo.group("v"))))
self.rows.append(Row(idx, line, reftime, dsname, matcher))
# Sort rows by reftime
self.rows.sort(key=lambda x: x.reftime)
def test_empty(self):
matcher = arki.Matcher()
self.assertEqual(str(matcher), "")
def test_docs(self):
"""
Run the doc/matcher/*.txt as doctests
"""
re_given = re.compile(
r"^Given (\w+ ``[^`]+``(?:\s*,\s*\w+ ``[^`]+``)*)")
re_given_item = re.compile(r"(\w+) ``([^`]+)``")
re_matches = re.compile(r"^\* ``(.+)`` matches\b")
re_not_matches = re.compile(r"^\* ``(.+)`` does not match\b")
re_doctest_info = re.compile(r".. doctest (\w+): (.+)")
docdir = os.path.join(os.environ["TOP_SRCDIR"], "doc")
doctests = [os.path.join(docdir, "matcher.rst")]
doctests += glob.glob(os.path.join(docdir, "matcher", "*.rst"))
self.assertTrue(doctests)
for pathname in doctests:
md = None
expected_matches = 0
expected_not_matches = 0
count_matches = 0
count_not_matches = 0
with open(pathname, "rt", encoding="utf-8") as fd:
for lineno, line in enumerate(fd):
mo = re_given.match(line)
if mo:
md = arki.Metadata()
for m in re_given_item.finditer(mo.group(1)):
try:
md[m.group(1)] = m.group(2)
except Exception as e:
self.fail(
"{}:{}: cannot parse metadata in line {}: {}"
.format(pathname, lineno, line.rstrip(),
e))
continue
mo = re_matches.match(line)
if mo:
if md is None:
self.fail(
"{}:{}: `matches` line found before Given line"
.format(pathname, lineno))
matcher = arki.Matcher(mo.group(1))
self.assertTrue(matcher.match(md))
count_matches += 1
continue
mo = re_not_matches.match(line)
if mo:
if md is None:
self.fail(
"{}:{}: `does not match` line found before Given line"
.format(pathname, lineno))
matcher = arki.Matcher(mo.group(1))
self.assertFalse(matcher.match(md))
count_not_matches += 1
continue
mo = re_doctest_info.match(line)
if mo:
tag = mo.group(1)
if tag == "matched":
expected_matches = int(mo.group(2))
elif tag == "not_matched":
expected_not_matches = int(mo.group(2))
continue
if (expected_matches != count_matches
or expected_not_matches != count_not_matches):
self.fail(
"{}: test count expectations failed: matches {}/{}, not matches {}/{}"
.format(pathname, count_matches, expected_matches,
count_not_matches, expected_not_matches))
def test_parseerror(self):
with self.assertRaises(ValueError):
arki.Matcher("invalid")
def test_parse(self):
matcher = arki.Matcher("reftime:=today")
self.assertEqual(str(matcher), "reftime:=today")
def run(self):
super().run()
# Process --config options
if self.args.config is not None:
for pathname in self.args.config:
cfg = arki.dataset.read_configs(pathname)
for name, section in cfg.items():
self.add_config_section(section, name)
# Read the config files from the remaining commandline arguments
for path in self.args.sources:
if path.startswith("http://") or path.startswith("https://"):
sections = arki.dataset.http.load_cfg_sections(path)
for name, section in sections.items():
self.add_config_section(section, name)
else:
section = arki.dataset.read_config(path)
self.add_config_section(section)
if not self.config:
raise Fail(
"you need to specify at least one config file or dataset")
# Remove unallowed entries
if self.args.restrict:
self.filter_restrict(self.args.restrict)
if self.args.ignore_system_datasets:
to_remove = []
for name, section in self.config.items():
type = section.get("type")
name = section.get("name")
if (type == "error" or type == "duplicates"
or (type == "remote" and
(name == "error" or name == "duplicates"))):
to_remove.append(name)
for name in to_remove:
del self.config[name]
if not self.config:
raise Fail("none of the configuration provided were useable")
# Validate the configuration
has_errors = False
for name, section in self.config.items():
# Validate filters
filter = section.get("filter")
if filter is None:
continue
try:
arki.Matcher(filter)
except ValueError as e:
print("{}: {}".format(name, e), file=sys.stderr)
has_errors = True
if has_errors:
raise Fail("Some input files did not validate.")
# If requested, compute extra information
if self.args.extra:
for name, section in self.config.items():
# Instantiate the dataset
ds = arki.dataset.Reader(section)
# Get the summary
summary = ds.query_summary()
# Compute bounding box, and store the WKT in bounding
bbox = summary.get_convex_hull()
if bbox:
section["bounding"] = bbox
# Output the merged configuration
if self.args.output:
with open(self.args.output, "wt") as fd:
self.config.write(fd)
else:
self.config.write(sys.stdout)
def run(self):
super().run()
self.build_config()
with self.outfile() as outfd:
arki_scan = arkimet.cmdline.ArkiScan()
arki_scan.set_inputs(self.config)
arki_scan.set_processor(
query=arkimet.Matcher(),
outfile=outfd,
yaml=self.args.yaml,
json=self.args.json,
annotate=self.args.annotate,
inline=self.args.inline,
data=self.args.data,
summary=self.args.summary,
summary_short=self.args.summary_short,
summary_restrict=self.args.summary_restrict,
archive=self.args.archive,
postproc=self.args.postproc,
postproc_data=self.args.postproc_data,
sort=self.args.sort,
)
if self.args.dispatch or self.args.testdispatch:
kw = dict(
copyok=self.args.copyok,
copyko=self.args.copyko,
validate=self.args.validate,
flush_threshold=(self.parse_size(self.args.flush_threshold)
if self.args.flush_threshold is not None
else 0),
)
if self.args.dispatch:
dispatch_cfg = arkimet.cfg.Sections()
for source in self.args.dispatch:
self.merge_config(dispatch_cfg,
arkimet.cfg.Sections.parse(source))
kw["dispatch"] = dispatch_cfg
elif self.args.testdispatch:
dispatch_cfg = arkimet.cfg.Sections()
for source in self.args.testdispatch:
self.merge_config(dispatch_cfg,
arkimet.cfg.Sections.parse(source))
kw["testdispatch"] = dispatch_cfg
arki_scan.set_dispatcher(**kw)
if self.args.stdin:
all_successful = arki_scan.dispatch_file(
sys.stdin,
self.args.stdin,
ignore_duplicates=self.args.ignore_duplicates,
status=self.args.status)
else:
all_successful = arki_scan.dispatch_sections(
moveok=self.args.moveok,
moveko=self.args.moveko,
movework=self.args.movework,
ignore_duplicates=self.args.ignore_duplicates,
status=self.args.status)
else:
if self.args.stdin:
all_successful = arki_scan.scan_file(
sys.stdin, self.args.stdin)
else:
all_successful = arki_scan.scan_sections()
if not all_successful:
raise Exit(posix.EX_DATAERR)
