Merge pull request 'sql慢查询注释' (#3) from Eukanj827/openGauss-server:master into master

src/gausskernel/dbmind/tools/app/diagnosis/query/slow_sql/featurelib/__init__.py

@@ -14,11 +14,23 @@ import os
 
 from . import feature_mapping
 from . import features
+# To import file feature_mapping and features from parent folder
 
+#function name: load_feature_lib
+#description: Print the variable FEATURE_LIB in the file-- features
+#return value: The value of FEATURE_LIB
+#date: 2022/8/2
+#contact: 1865997821
 
 def load_feature_lib():
     return features.FEATURE_LIB
 
+#function name: get_feature_mapper
+#description: Get the item and value of a dictionary type in the file-- feature_mapping and output it as a generator.
+#return value: The item and value in _dict_  variable
+#note：Dictionary key-value pairs must start with C then the  item and value will be return.
+#date: 2022/8/2
+#contact: 1865997821
 
 def get_feature_mapper():
     return {

src/gausskernel/dbmind/tools/app/diagnosis/query/slow_sql/featurelib/feature_model.py

@@ -11,22 +11,27 @@
 # MERCHANTABILITY OR FIT FOR A PARTICULAR PURPOSE.
 # See the Mulan PSL v2 for more details.
 import csv
+#import csv packet
 from collections import defaultdict
 from typing import List
+# To import defaultdict in the parent floder collections and List  in the parent floder typing
 
 import numpy as np
+# import numpy packet as the name np
 
 from ..analyzer import _euclid_distance as euclid_distance
 from dbmind.common.utils import ExceptionCatch
+#To import private function-- _euclid_distance as euclid_distance
 
+#function name: calculate_weight
+#description: This function will output feature_weight (= residual_vector / the sum of residual_vector)
+#The data used for the calculation is from the features_labels_dict, and the key value pairs of the features_labels_dict are filtered
+#arguments: np.ndarray and  np.ndarray
+#return value: weight_matrix
+#date: 2022/8/2
+#contact: 1865997821
 
 def calculate_weight(features: np.ndarray, labels: np.ndarray) -> List:
-    """
-    Calculate weight matrix based on feature set
-    :param features: feature set
-    :param labels: label set
-    :return: weight_matrix
-    """
     normalize_features, normalize_labels = [], []
     features_labels_dict = defaultdict(list)
     for i in range(len(labels)):
@@ -56,6 +61,16 @@ def calculate_weight(features: np.ndarray, labels: np.ndarray) -> List:
     return weight_matrix
 
 
+# function name: build_model
+# description: Create two variables-- features and labels.There are refer to two numpy array(all elements are zero)
+# The features array's size is feature_number and dimension is feature_dimension
+# This function will read the two arrays and write it as a matrix in a csv file(the save path is './features_new.npz')
+# And then it will call the function calculate_weight to calculate the matrix
+# arguments: feature_path, feature_number, feature_dimension
+# return value: None
+# note：A ExceptionCatch function modifier is used
+# date: 2022/8/2
+#contact: 1865997821
 @ExceptionCatch(strategy='exit', name='FEATURE')
 def build_model(feature_path: str, feature_number: int, feature_dimension: int,
                 save_path: str = './features_new.npz') -> None:

src/gausskernel/dbmind/tools/app/diagnosis/query/slow_sql/significance_detection/__init__.py

@@ -11,6 +11,13 @@
 # MERCHANTABILITY OR FIT FOR A PARTICULAR PURPOSE.
 # See the Mulan PSL v2 for more details.
 
+#function name: detect
+#description: if the method is "bool" type, then call the functions sum_detect、avg_detect、ks_detect to diagnose errors
+#These functions are in the parent slow_sql/significance_detection
+#arguments: data1(array), data2(array), method
+#return value: bool type
+#date: 2022/8/2
+#contact: 1865997821
 
 def detect(data1, data2, method='bool', threshold=0.01, p_value=0.5):
     if method == 'bool':

src/gausskernel/dbmind/tools/app/diagnosis/query/slow_sql/significance_detection/average_base.py

@@ -12,17 +12,16 @@
 # See the Mulan PSL v2 for more details.
 
 alpha = 1e-10
+#Define a minimum number of errors
 
+#function name: detect
+#description: Calculate whether the data has abrupt changes based on the average value
+#arguments: data1, data2, threshold,method
+#return value: bool
+#date: 2022/8/
+#contact: 1865997821
 
 def detect(data1, data2, threshold=0.5, method='bool'):
-    """
-    Calculate whether the data has abrupt changes based on the average value
-    :param data1: input data array
-    :param data2: input data array
-    :param threshold: Mutation rate
-    :param method: The way to calculate the mutation
-    :return: bool
-    """
     if not isinstance(data1, list) or not isinstance(data2, list):
         raise TypeError("The format of the input data is wrong.")
     avg1 = sum(data1) / len(data1) if data1 else 0