def ets_internal(table, key):
schema = utils.load_schema(table)
if schema is None: return None
lines = []
_append_ets_internals(lines, table, key, schema)
return lines
def erl_db_accessor_update_cache_fields_footer(table):
schema = utils.load_schema(table)
if schema is None: return None
lines = []
lines.append(' _ -> none')
lines.append(' end.')
return lines
def erl_mod_c2s_to_pb(module, table, pb_type):
schema = utils.load_schema(table)
if schema is None: return None
pb_schema = schema['pb_types'][pb_type]
lines = []
_append_to_pb(lines, module, table, schema, pb_type, pb_schema)
return lines
def ets_external(table, key):
schema = utils.load_schema(table)
if schema is None: return None
lines = []
_append_common_exports(lines, table, schema)
_append_common_defines(lines, table, schema)
_append_ets_defines(lines, table, schema)
return lines
def sql_create_table(table):
schema = utils.load_schema(table)
if schema is None: return None
lines = []
_append_create_table(lines, table, schema)
_append_primary_key(lines, table, schema)
if schema.has_key('autoIncrement'):
_append_auto_increment(lines, table, schema)
return lines
def hrl_record(table, is_partial):
is_partial = is_partial.strip() == 'true'
schema = utils.load_schema(table)
if schema is None: return None
lines = []
lines.append('%%%% %s基本信息(修改数据库时，需要修改此record)' %
utils.get(schema, 'description', table))
_hrl_record_with_schema(lines, table, schema, is_partial)
return lines
def erl_db_accessor_extension(table, x, encode_decode='false'):
schema = utils.load_schema(table)
if schema is None: return None
encode_decode = encode_decode.strip() == 'true'
lines = []
_append_accessor_export_common(lines, table, schema)
_append_common_db_crud_export(lines, table, schema)
_append_extension_db_crud_export(lines, table, x, schema)
_append_accessor_common(lines, table, encode_decode, schema)
_append_extension_db_crud(lines, table, x, schema)
return lines
def erl_db_accessor_update_cache_fields_update(table):
schema = utils.load_schema(table)
if schema is None: return None
lines = []
lines.append('New_%s = %s#%s{' % (table.title(), table.title(), table))
def get_line(field_name, is_last):
has_comma = (not is_last)
return ' %s = %s%s' % (field_name, field_name.title(), has_comma and ',' or '')
utils.append_single(lines, table, schema['cache_fields'], get_line)
lines.append('},')
lines.append('ok = update(New_%s),' % table.title())
lines.append('New_%s;' % table.title())
return lines
def erl_db_accessor_update_cache_fields_header(table):
schema = utils.load_schema(table)
if schema is None: return None
params = schema['primaryKey']
if type(params) == list:
params = ', '.join(['%s' % key.title() for key in params])
else:
params = params.title()
lines = []
lines.append('%% @doc Calculate the cache fields #%s' % table)
lines.append('update_cache_fields(%s) ->' % params)
lines.append(' case load(%s) of' % params)
lines.append(' %s = #%s{' % (table.title(), table))
def get_line_key(field_name, is_last):
return ' %s = %s,' % (field_name, field_name.title())
utils.append_single(lines, table, schema['cache_keys'], get_line_key)
def get_line_field(field_name, is_last):
has_comma = (not is_last)
return ' %s = _Old_%s%s' % (field_name, field_name.title(), has_comma and ',' or '')
utils.append_single(lines, table, schema['cache_fields'], get_line_field)
lines.append(' } ->')
return lines
from pyspark.sql import SparkSession
from utils import spark_schema_from_json, load_schema
spark = SparkSession.builder.master("spark://vm1:7077").appName(
"Language_forks").getOrCreate()
db_schema = load_schema()
df_projects = spark.read.csv(
# path="/home/hthuwal/Desktop/Ghtorrent_heads/projects.csv",
# schema=spark_schema_from_json(db_schema["projects.csv"]),
path="hdfs:/projects.csv",
schema=spark_schema_from_json(db_schema["projects_new.csv"]),
multiLine=True,
nullValue="\\N",
)
# df = df_projects.where(df_projects.forked_from.isNotNull()).groupBy(df_projects.language).agg(sf.count().alias('num_forks')).sort(df_projects.num_forks.desc())
df_projects.createOrReplaceTempView("temp")
query = """
SELECT language, count(*) as num_forks
FROM temp
WHERE forked_from is NOT NULL AND language is NOT NULL
GROUP BY language
ORDER BY num_forks DESC
"""
df = spark.sql(query)
df.coalesce(1).write.json("Language_forks", mode="overwrite")
# spark-submit --master spark://vm1:7077 --deploy-mode client --executor-memory 3G --driver-memory 4G --conf spark.driver.maxResultSize=3g language_forks.py
def erl_mod_c2s_record_to_fields(table, record):
schema = utils.load_schema(table)
if schema is None: return None
lines = []
_append_record_to_fields_match(lines, table, schema, record)
return lines
def erl_mod_c2s_helpers(table):
schema = utils.load_schema(table)
if schema is None: return None
lines = []
_append_records_to_lists(lines, table, schema)
return lines
