def format_list(T, select, query=None):
data = []
if isinstance(query.select, list):
for row in T:
r = Data()
for s in select:
r[s.put.name][s.put.child] = unwraplist(row[s.pull])
data.append(r if r else None)
elif isinstance(query.select.value, LeavesOp):
for row in T:
r = Data()
for s in select:
r[s.put.name][s.put.child] = unwraplist(row[s.pull])
data.append(r if r else None)
else:
for row in T:
r = Data()
for s in select:
r[s.put.child] = unwraplist(row[s.pull])
data.append(r if r else None)
return Data(
meta={"format": "list"},
data=data
)
def format_table(T, select, query=None):
data = []
num_columns = (Math.MAX(select.put.index) + 1)
for row in T:
r = [None] * num_columns
for s in select:
value = unwraplist(row[s.pull])
if value == None:
continue
index, child = s.put.index, s.put.child
if child == ".":
r[index] = value
else:
if r[index] is None:
r[index] = Data()
r[index][child] = value
data.append(r)
header = [None] * num_columns
for s in select:
if header[s.put.index]:
continue
header[s.put.index] = s.name
return Data(
meta={"format": "table"},
header=header,
data=data
)
def error(
cls,
template, # human readable template
default_params={}, # parameters for template
cause=None, # pausible cause
stack_depth=0,
**more_params
):
"""
raise an exception with a trace for the cause too
:param template: *string* human readable string with placeholders for parameters
:param default_params: *dict* parameters to fill in template
:param cause: *Exception* for chaining
:param stack_depth: *int* how many calls you want popped off the stack to report the *true* caller
:param log_context: *dict* extra key:value pairs for your convenience
:param more_params: *any more parameters (which will overwrite default_params)
:return:
"""
if default_params and isinstance(listwrap(default_params)[0], BaseException):
cause = default_params
default_params = {}
params = dict(unwrap(default_params), **more_params)
add_to_trace = False
cause = wrap(unwraplist([Except.wrap(c, stack_depth=1) for c in listwrap(cause)]))
trace = exceptions.extract_stack(stack_depth + 1)
if add_to_trace:
cause[0].trace.extend(trace[1:])
e = Except(exceptions.ERROR, template, params, cause, trace)
raise e
def warning(
cls,
template,
default_params={},
cause=None,
stack_depth=0,
log_context=None,
**more_params
):
"""
:param template: *string* human readable string with placeholders for parameters
:param default_params: *dict* parameters to fill in template
:param cause: *Exception* for chaining
:param stack_depth: *int* how many calls you want popped off the stack to report the *true* caller
:param log_context: *dict* extra key:value pairs for your convenience
:param more_params: *any more parameters (which will overwrite default_params)
:return:
"""
if isinstance(default_params, BaseException):
cause = default_params
default_params = {}
if "values" in more_params.keys():
Log.error("Can not handle a logging parameter by name `values`")
params = dict(unwrap(default_params), **more_params)
cause = unwraplist([Except.wrap(c) for c in listwrap(cause)])
trace = exceptions.extract_stack(stack_depth + 1)
e = Except(exceptions.WARNING, template, params, cause, trace)
Log.note(
"{{error|unicode}}",
error=e,
log_context=set_default({"context": exceptions.WARNING}, log_context),
stack_depth=stack_depth + 1
)
def list2cube(rows, column_names=None):
if column_names:
keys = column_names
else:
columns = set()
for r in rows:
columns |= set(r.keys())
keys = list(columns)
data = {k: [] for k in keys}
output = wrap({
"meta": {"format": "cube"},
"edges": [
{
"name": "rownum",
"domain": {"type": "rownum", "min": 0, "max": len(rows), "interval": 1}
}
],
"data": data
})
for r in rows:
for k in keys:
data[k].append(unwraplist(r[k]))
return output
def get_decoders_by_depth(query):
"""
RETURN A LIST OF DECODER ARRAYS, ONE ARRAY FOR EACH NESTED DEPTH
"""
schema = query.frum
output = FlatList()
for e in wrap(coalesce(query.edges, query.groupby, [])):
if e.value != None and not isinstance(e.value, NullOp):
e = e.copy()
vars_ = e.value.vars()
for v in vars_:
if not schema[v]:
Log.error("{{var}} does not exist in schema", var=v)
e.value = e.value.map({schema[v].name: schema[v].es_column for v in vars_})
elif e.range:
e = e.copy()
min_ = e.range.min
max_ = e.range.max
vars_ = min_.vars() | max_.vars()
for v in vars_:
if not schema[v]:
Log.error("{{var}} does not exist in schema", var=v)
map_ = {schema[v].name: schema[v].es_column for v in vars_}
e.range = {
"min": min_.map(map_),
"max": max_.map(map_)
}
elif e.domain.dimension:
vars_ = e.domain.dimension.fields
e.domain.dimension = e.domain.dimension.copy()
e.domain.dimension.fields = [schema[v].es_column for v in vars_]
elif all(e.domain.partitions.where):
vars_ = set()
for p in e.domain.partitions:
vars_ |= p.where.vars()
try:
depths = set(len(schema[v].nested_path)-1 for v in vars_)
if -1 in depths:
Log.error(
"Do not know of column {{column}}",
column=unwraplist([v for v in vars_ if schema[v]==None])
)
if len(depths) > 1:
Log.error("expression {{expr}} spans tables, can not handle", expr=e.value)
max_depth = Math.MAX(depths)
while len(output) <= max_depth:
output.append([])
except Exception, e:
# USUALLY THE SCHEMA IS EMPTY, SO WE ASSUME THIS IS A SIMPLE QUERY
max_depth = 0
output.append([])
limit = 0
output[max_depth].append(AggsDecoder(e, query, limit))
def update(self, command):
"""
EXPECTING command == {"set":term, "where":where}
THE set CLAUSE IS A DICT MAPPING NAMES TO VALUES
THE where CLAUSE IS AN ES FILTER
"""
command = wrap(command)
schema = self._es.get_schema()
# GET IDS OF DOCUMENTS
results = self._es.search(
{
"fields": listwrap(schema._routing.path),
"query": {
"filtered": {"query": {"match_all": {}}, "filter": jx_expression(command.where).to_esfilter()}
},
"size": 200000,
}
)
# SCRIPT IS SAME FOR ALL (CAN ONLY HANDLE ASSIGNMENT TO CONSTANT)
scripts = FlatList()
for k, v in command.set.items():
if not is_keyword(k):
Log.error("Only support simple paths for now")
if isinstance(v, Mapping) and v.doc:
scripts.append({"doc": v.doc})
else:
scripts.append({"script": "ctx._source." + k + " = " + jx_expression(v).to_ruby()})
if results.hits.hits:
updates = []
for h in results.hits.hits:
for s in scripts:
updates.append(
{
"update": {
"_id": h._id,
"_routing": unwraplist(h.fields[literal_field(schema._routing.path)]),
}
}
)
updates.append(s)
content = ("\n".join(convert.value2json(c) for c in updates) + "\n").encode("utf-8")
response = self._es.cluster.post(
self._es.path + "/_bulk",
data=content,
headers={"Content-Type": "application/json"},
timeout=self.settings.timeout,
params={"consistency": self.settings.consistency},
)
if response.errors:
Log.error(
"could not update: {{error}}",
error=[e.error for i in response["items"] for e in i.values() if e.status not in (200, 201)],
)
def _get_schema_from_list(frum, columns, prefix, nested_path, name_to_column):
"""
SCAN THE LIST FOR COLUMN TYPES
"""
for d in frum:
row_type = _type_to_name[d.__class__]
if row_type != "object":
full_name = join_field(prefix)
column = name_to_column.get(full_name)
if not column:
column = Column(
name=full_name,
table=".",
es_column=full_name,
es_index=".",
type="undefined",
nested_path=nested_path
)
columns[full_name] = column
column.type = _merge_type[column.type][row_type]
else:
for name, value in d.items():
full_name = join_field(prefix + [name])
column = name_to_column.get(full_name)
if not column:
column = Column(
name=full_name,
table=".",
es_column=full_name,
es_index=".",
type="undefined",
nested_path=nested_path
)
columns[full_name] = column
if isinstance(value, list):
if len(value)==0:
this_type = "undefined"
elif len(value)==1:
this_type = _type_to_name[value[0].__class__]
else:
this_type = _type_to_name[value[0].__class__]
if this_type == "object":
this_type = "nested"
else:
this_type = _type_to_name[value.__class__]
new_type = _merge_type[column.type][this_type]
column.type = new_type
if this_type == "object":
_get_schema_from_list([value], columns, prefix + [name], nested_path, name_to_column)
elif this_type == "nested":
np = listwrap(nested_path)
newpath = unwraplist([join_field(split_field(np[0])+[name])]+np)
_get_schema_from_list(value, columns, prefix + [name], newpath, name_to_column)
def _convert_edge(self, edge):
dim = self.dimensions[edge.value]
if not dim:
return edge
if len(listwrap(dim.fields)) == 1:
#TODO: CHECK IF EDGE DOMAIN AND DIMENSION DOMAIN CONFLICT
new_edge = set_default({"value": unwraplist(dim.fields)}, edge)
return new_edge
new_edge.domain = dim.getDomain()
edge = copy(edge)
edge.value = None
edge.domain = dim.getDomain()
return edge
def list2table(rows, column_names=None):
if column_names:
keys = list(set(column_names))
else:
columns = set()
for r in rows:
columns |= set(r.keys())
keys = list(columns)
output = [[unwraplist(r.get(k)) for k in keys] for r in rows]
return wrap({
"meta": {"format": "table"},
"header": keys,
"data": output
})
def wrap(cls, e, stack_depth=0):
if e == None:
return Null
elif isinstance(e, (list, Except)):
return e
elif isinstance(e, Mapping):
e.cause = unwraplist([Except.wrap(c) for c in listwrap(e.cause)])
return Except(**e)
else:
if hasattr(e, "message") and e.message:
cause = Except(ERROR, unicode(e.message), trace=_extract_traceback(0))
else:
cause = Except(ERROR, unicode(e), trace=_extract_traceback(0))
trace = extract_stack(stack_depth + 2) # +2 = to remove the caller, and it's call to this' Except.wrap()
cause.trace.extend(trace)
return cause
def fatal(
cls,
template, # human readable template
default_params={}, # parameters for template
cause=None, # pausible cause
stack_depth=0,
log_context=None,
**more_params
):
"""
SEND TO STDERR
:param template: *string* human readable string with placeholders for parameters
:param default_params: *dict* parameters to fill in template
:param cause: *Exception* for chaining
:param stack_depth: *int* how many calls you want popped off the stack to report the *true* caller
:param log_context: *dict* extra key:value pairs for your convenience
:param more_params: *any more parameters (which will overwrite default_params)
:return:
"""
if default_params and isinstance(listwrap(default_params)[0], BaseException):
cause = default_params
default_params = {}
params = dict(unwrap(default_params), **more_params)
cause = unwraplist([Except.wrap(c) for c in listwrap(cause)])
trace = exceptions.extract_stack(stack_depth + 1)
e = Except(exceptions.ERROR, template, params, cause, trace)
str_e = unicode(e)
error_mode = cls.error_mode
with suppress_exception:
if not error_mode:
cls.error_mode = True
Log.note(
"{{error|unicode}}",
error=e,
log_context=set_default({"context": exceptions.FATAL}, log_context),
stack_depth=stack_depth + 1
)
cls.error_mode = error_mode
sys.stderr.write(str_e.encode('utf8'))
def map_edge(e, map_):
partitions = unwraplist([
set_default(
{"where": p.where.map(map_)},
p
)
for p in e.domain.partitions
])
domain = copy(e.domain)
domain.where = e.domain.where.map(map_)
domain.partitions = partitions
edge = copy(e)
edge.value = e.value.map(map_)
edge.domain = domain
if e.range:
edge.range.min = e.range.min.map(map_)
edge.range.max = e.range.max.map(map_)
return edge
def _normalize_edge(edge, schema=None):
if not _Column:
_late_import()
if edge == None:
Log.error("Edge has no value, or expression is empty")
elif isinstance(edge, basestring):
if schema:
try:
e = schema[edge]
except Exception, e:
e = None
e = unwraplist(e)
if e and not isinstance(e, (_Column, set, list)):
if isinstance(e, _Column):
return Data(
name=edge,
value=jx_expression(edge),
allowNulls=True,
domain=_normalize_domain(domain=e, schema=schema)
)
elif isinstance(e.fields, list) and len(e.fields) == 1:
return Data(
name=e.name,
value=jx_expression(e.fields[0]),
allowNulls=True,
domain=e.getDomain()
)
else:
return Data(
name=e.name,
allowNulls=True,
domain=e.getDomain()
)
return Data(
name=edge,
value=jx_expression(edge),
allowNulls=True,
domain=_normalize_domain(schema=schema)
)
def to_dict(self):
return wrap({
"meta": {"format": "list"},
"data": [{k: unwraplist(v) for k, v in row.items()} for row in self.data]
})
def es_aggsop(es, frum, query):
select = wrap([s.copy() for s in listwrap(query.select)])
es_column_map = {c.name: unwraplist(c.es_column) for c in frum.schema.all_columns}
es_query = Data()
new_select = Data() #MAP FROM canonical_name (USED FOR NAMES IN QUERY) TO SELECT MAPPING
formula = []
for s in select:
if s.aggregate == "count" and isinstance(s.value, Variable) and s.value.var == ".":
s.pull = "doc_count"
elif isinstance(s.value, Variable):
if s.value.var == ".":
if frum.typed:
# STATISITCAL AGGS IMPLY $value, WHILE OTHERS CAN BE ANYTHING
if s.aggregate in NON_STATISTICAL_AGGS:
#TODO: HANDLE BOTH $value AND $objects TO COUNT
Log.error("do not know how to handle")
else:
s.value.var = "$value"
new_select["$value"] += [s]
else:
if s.aggregate in NON_STATISTICAL_AGGS:
#TODO: WE SHOULD BE ABLE TO COUNT, BUT WE MUST *OR* ALL LEAF VALUES TO DO IT
Log.error("do not know how to handle")
else:
Log.error('Not expecting ES to have a value at "." which {{agg}} can be applied', agg=s.aggregate)
elif s.aggregate == "count":
s.value = s.value.map(es_column_map)
new_select["count_"+literal_field(s.value.var)] += [s]
else:
s.value = s.value.map(es_column_map)
new_select[literal_field(s.value.var)] += [s]
else:
formula.append(s)
for canonical_name, many in new_select.items():
representative = many[0]
if representative.value.var == ".":
Log.error("do not know how to handle")
else:
field_name = representative.value.var
# canonical_name=literal_field(many[0].name)
for s in many:
if s.aggregate == "count":
es_query.aggs[literal_field(canonical_name)].value_count.field = field_name
s.pull = literal_field(canonical_name) + ".value"
elif s.aggregate == "median":
# ES USES DIFFERENT METHOD FOR PERCENTILES
key = literal_field(canonical_name + " percentile")
es_query.aggs[key].percentiles.field = field_name
es_query.aggs[key].percentiles.percents += [50]
s.pull = key + ".values.50\.0"
elif s.aggregate == "percentile":
# ES USES DIFFERENT METHOD FOR PERCENTILES
key = literal_field(canonical_name + " percentile")
if isinstance(s.percentile, basestring) or s.percetile < 0 or 1 < s.percentile:
Log.error("Expecting percentile to be a float from 0.0 to 1.0")
percent = Math.round(s.percentile * 100, decimal=6)
es_query.aggs[key].percentiles.field = field_name
es_query.aggs[key].percentiles.percents += [percent]
s.pull = key + ".values." + literal_field(unicode(percent))
elif s.aggregate == "cardinality":
# ES USES DIFFERENT METHOD FOR CARDINALITY
key = literal_field(canonical_name + " cardinality")
es_query.aggs[key].cardinality.field = field_name
s.pull = key + ".value"
elif s.aggregate == "stats":
# REGULAR STATS
stats_name = literal_field(canonical_name)
es_query.aggs[stats_name].extended_stats.field = field_name
# GET MEDIAN TOO!
median_name = literal_field(canonical_name + " percentile")
es_query.aggs[median_name].percentiles.field = field_name
es_query.aggs[median_name].percentiles.percents += [50]
s.pull = {
"count": stats_name + ".count",
"sum": stats_name + ".sum",
"min": stats_name + ".min",
"max": stats_name + ".max",
"avg": stats_name + ".avg",
"sos": stats_name + ".sum_of_squares",
"std": stats_name + ".std_deviation",
"var": stats_name + ".variance",
"median": median_name + ".values.50\.0"
}
elif s.aggregate == "union":
# USE TERMS AGGREGATE TO SIMULATE union
stats_name = literal_field(canonical_name)
es_query.aggs[stats_name].terms.field = field_name
es_query.aggs[stats_name].terms.size = Math.min(s.limit, MAX_LIMIT)
s.pull = stats_name + ".buckets.key"
else:
# PULL VALUE OUT OF THE stats AGGREGATE
es_query.aggs[literal_field(canonical_name)].extended_stats.field = field_name
s.pull = literal_field(canonical_name) + "." + aggregates1_4[s.aggregate]
for i, s in enumerate(formula):
canonical_name = literal_field(s.name)
abs_value = s.value.map(es_column_map)
if s.aggregate == "count":
es_query.aggs[literal_field(canonical_name)].value_count.script = abs_value.to_ruby()
s.pull = literal_field(canonical_name) + ".value"
elif s.aggregate == "median":
# ES USES DIFFERENT METHOD FOR PERCENTILES THAN FOR STATS AND COUNT
key = literal_field(canonical_name + " percentile")
es_query.aggs[key].percentiles.script = abs_value.to_ruby()
es_query.aggs[key].percentiles.percents += [50]
s.pull = key + ".values.50\.0"
elif s.aggregate == "percentile":
# ES USES DIFFERENT METHOD FOR PERCENTILES THAN FOR STATS AND COUNT
key = literal_field(canonical_name + " percentile")
percent = Math.round(s.percentile * 100, decimal=6)
es_query.aggs[key].percentiles.script = abs_value.to_ruby()
es_query.aggs[key].percentiles.percents += [percent]
s.pull = key + ".values." + literal_field(unicode(percent))
elif s.aggregate == "cardinality":
# ES USES DIFFERENT METHOD FOR CARDINALITY
key = canonical_name + " cardinality"
es_query.aggs[key].cardinality.script = abs_value.to_ruby()
s.pull = key + ".value"
elif s.aggregate == "stats":
# REGULAR STATS
stats_name = literal_field(canonical_name)
es_query.aggs[stats_name].extended_stats.script = abs_value.to_ruby()
# GET MEDIAN TOO!
median_name = literal_field(canonical_name + " percentile")
es_query.aggs[median_name].percentiles.script = abs_value.to_ruby()
es_query.aggs[median_name].percentiles.percents += [50]
s.pull = {
"count": stats_name + ".count",
"sum": stats_name + ".sum",
"min": stats_name + ".min",
"max": stats_name + ".max",
"avg": stats_name + ".avg",
"sos": stats_name + ".sum_of_squares",
"std": stats_name + ".std_deviation",
"var": stats_name + ".variance",
"median": median_name + ".values.50\.0"
}
elif s.aggregate=="union":
# USE TERMS AGGREGATE TO SIMULATE union
stats_name = literal_field(canonical_name)
es_query.aggs[stats_name].terms.script_field = abs_value.to_ruby()
s.pull = stats_name + ".buckets.key"
else:
# PULL VALUE OUT OF THE stats AGGREGATE
s.pull = canonical_name + "." + aggregates1_4[s.aggregate]
es_query.aggs[canonical_name].extended_stats.script = abs_value.to_ruby()
decoders = get_decoders_by_depth(query)
start = 0
vars_ = query.where.vars()
#<TERRIBLE SECTION> THIS IS WHERE WE WEAVE THE where CLAUSE WITH nested
split_where = split_expression_by_depth(query.where, schema=frum, map_=es_column_map)
if len(split_field(frum.name)) > 1:
if any(split_where[2::]):
Log.error("Where clause is too deep")
for d in decoders[1]:
es_query = d.append_query(es_query, start)
start += d.num_columns
if split_where[1]:
#TODO: INCLUDE FILTERS ON EDGES
filter_ = simplify_esfilter(AndOp("and", split_where[1]).to_esfilter())
es_query = Data(
aggs={"_filter": set_default({"filter": filter_}, es_query)}
)
es_query = wrap({
"aggs": {"_nested": set_default(
{
"nested": {
"path": frum.query_path
}
},
es_query
)}
})
else:
if any(split_where[1::]):
Log.error("Where clause is too deep")
for d in decoders[0]:
es_query = d.append_query(es_query, start)
start += d.num_columns
if split_where[0]:
#TODO: INCLUDE FILTERS ON EDGES
filter = simplify_esfilter(AndOp("and", split_where[0]).to_esfilter())
es_query = Data(
aggs={"_filter": set_default({"filter": filter}, es_query)}
)
# </TERRIBLE SECTION>
if not es_query:
es_query = wrap({"query": {"match_all": {}}})
es_query.size = 0
with Timer("ES query time") as es_duration:
result = es09.util.post(es, es_query, query.limit)
try:
format_time = Timer("formatting")
with format_time:
decoders = [d for ds in decoders for d in ds]
result.aggregations.doc_count = coalesce(result.aggregations.doc_count, result.hits.total) # IT APPEARS THE OLD doc_count IS GONE
formatter, groupby_formatter, aggop_formatter, mime_type = format_dispatch[query.format]
if query.edges:
output = formatter(decoders, result.aggregations, start, query, select)
elif query.groupby:
output = groupby_formatter(decoders, result.aggregations, start, query, select)
else:
output = aggop_formatter(decoders, result.aggregations, start, query, select)
output.meta.timing.formatting = format_time.duration
output.meta.timing.es_search = es_duration.duration
output.meta.content_type = mime_type
output.meta.es_query = es_query
return output
except Exception, e:
if query.format not in format_dispatch:
Log.error("Format {{format|quote}} not supported yet", format=query.format, cause=e)
Log.error("Some problem", e)