WO2015188315A1 - Data query method, device and system - Google Patents

Abstract

Provided are a data query method, device and system. The method comprises: after syntax analysis is conducted on a query request, obtaining a relationship tree corresponding to the query request; determining a type corresponding to the relationship tree; then, determining an execution function template corresponding to the type of the relationship tree; and finally, obtaining a final query result by executing the determined execution function template. In this way, it is possible to correspond a complete query request to an execution function template, and the execution function template contains execution functions corresponding to various nodes on the relationship tree of the corresponding type, thereby avoiding the signalling overhead due to the fact that execution functions are repeatedly invoked in the query process, and improving the speed of data query.

Description

Data query method, device and system Technical field

The present invention relates to the field of computer technologies, and in particular, to a data query method, apparatus, and system.

Background technique

Currently, in data warehousing and online online analytical processing (English: Online Analytical Processing, referred to as: OLPA), the speed of data query is particularly important. At the OLAP collar Domain, the database query process needs to traverse a large amount of data to complete the query, in such a large data Under the amount, if you need a database query to have a faster response, this will be very good for database processing performance. Big challenge.

The current database query method is generally based on structured query language (English: Structured Query Language, referred to as: SQL) for query, the execution process of the query is: SQL query request After entering the database system, the database system first converts the SQL query request into a relationship through syntax analysis. Tree, the raw data is obtained in the database through the relation tree, for example, table A, table B, Table C and Table D, through the relationship tree to determine the need to query data in Table A and Table B, then the table will be A and Table B are retrieved from the database as raw data.

Then by calling the execution function corresponding to the relational node on the relation tree, and finally executing the relationship in order The execution function corresponding to the node queries the required query result in the original data.

The existing SQL-based database query execution methods mainly include volcano model engine and column number According to the execution engine. Among them, the volcano model execution process is shown in Figure 1, and its query mode is to perform the relationship. After the execution function corresponding to a relational node on the tree, the next function is called once, thus executing The execution function on a node until the execution function on the last relational node is executed Finally, the query results are finally obtained in the original data. Thus, when there are more relationship nodes in the relationship tree, Calling the execution function on each relational node in turn, the call overhead will be very large, and the query speed is higher. slow.

The column data query engine execution process is shown in Figure 2. The database system receives the SQL query. After the request, the SQL query request is converted into a relation tree through syntax analysis, and the structure of the relationship tree is as shown in FIG. 2 The process of executing the query is: first, the execution function on the first relation node in the relationship tree is retrieved. To perform a basetable operation, so as to filter out a column data and b column data in the database, for column a The data is subjected to a select operation to obtain the qualified a' column data, and the number of eligible a' columns is determined. Temporarily saved in the cache; then enter the next relational node, retrieve and execute the execution on the node Row function to fetchjoin operation of the qualified a' column data and b column data to obtain a matching bar The b' column data of the piece temporarily stores the eligible b' column data in the cache; then proceeds to the next Relational nodes, retrieve and execute the execution function on the node to match the qualified a' column data with The qualified b' column data is subjected to a group aggregation operation to obtain a query result. This columnar data query The main disadvantage of the engine execution mode is that you need to retrieve the execution function on each relational node separately, and And the eligible column data obtained by executing each execution function will be cached in memory. Not only does it cause slower queries, but it also consumes a lot of memory.

So whether you use the volcano model execution engine or the column data execution engine, there is data check. Ask for a slower question.

Summary of the invention

The embodiment of the invention provides a data query method, device and system, which are used to solve the current problem The problem of slower query data in the database.

The specific technical solutions are as follows:

A first aspect of the embodiments of the present invention provides a data query method, including:

Obtain a query request for querying data in the database;

Converting the query request into a corresponding query relationship tree, and in the pre-stored relationship tree set Determining a relationship tree matching the query relationship tree, where the pre-stored relationship tree set includes different pairs Query the different relationship trees obtained by the request conversion;

Determining a relationship tree pair with the matching according to a correspondence between the relationship tree and the relationship tree type Should be related to the tree type;

Retrieving and determining the relationship tree according to the correspondence between the relationship tree type and the execution function template An execution function template corresponding to the type, wherein the execution function template includes a corresponding type of An execution function for querying data on each relationship node in the system tree;

Executing an execution function corresponding to the execution function template of the retrieved database for the database, and obtaining a query result.

In combination with the first aspect, in the first possible implementation, in the pre-stored relationship tree set Determining a relationship tree that matches the query relationship tree, including:

Determining a node location of each relationship node in the query relationship tree, and a type of each relationship node;

Traversing the node locations of the pre-stored relationship tree sets and the determined relationship nodes and the relationship sections A relationship tree in which the types of points match, as a relationship tree matching the query relationship tree.

In conjunction with the first possible implementation in the first aspect, in a second possible implementation, Traversing the pre-stored relationship tree set with the determined node locations of the relationship nodes and the relationship nodes Before the type matches the relationship tree, it also includes:

Determining the total number of relationship nodes in the query relationship tree;

The total number of relationship nodes in the pre-stored relationship tree set is equal to the total number of determined relationship nodes. Relationship tree

The filtered relationship tree is used as a pre-stored set of relationship trees to be traversed.

In combination with the first aspect, in a third possible implementation manner, determining in a pre-stored relationship tree set A relationship tree matching the query relationship tree, including:

Performing a hash calculation on the query relationship tree to obtain a hash value corresponding to the query relationship tree;

In the correspondence between the pre-stored relationship tree and the hash value corresponding to the relationship tree, the hash value is determined The obtained relationship tree of the query relationship tree has the same hash value as the matching relationship with the query relationship tree. Tie the tree.

Combining the first aspect or the first possible implementation of the first aspect or the first aspect The second possible implementation or the third possible implementation of the first aspect, in the fourth In an implementation manner, the execution function corresponding to the execution function template of the database is executed for the database Number, including:

When performing an execution function corresponding to the execution function template of the retrieved database for the database, The database determines the raw data, which is the number that is not executed by the execution function template according to;

Obtaining data features for characterizing the amount of data of the original data;

According to the correspondence between the data features and the execution function template, the execution function module is retrieved. Determining an execution function template corresponding to the data feature in the board, and

Executing an execution function corresponding to the execution function template corresponding to the data feature for the original data number.

A second aspect of the embodiments of the present invention provides a data query apparatus, including:

An obtaining module, configured to obtain a query request for querying data in a database;

a conversion module, configured to convert the query request into a corresponding query relationship tree, and pre-stored A relationship tree that matches the query relationship tree is determined in the relationship tree set, and the pre-stored relationship tree set The combination includes different relational trees obtained by converting different query requests;

a determining module, configured to determine and describe according to a correspondence between a relationship tree and a relationship tree type The relationship tree type corresponding to the matching relationship tree;

Retrieving a module for retrieving a correspondence between a relationship tree type and an execution function template An execution function template corresponding to the determined relationship tree type, wherein the execution function template includes An execution function for querying data on each relationship node in a relation tree of a corresponding type;

An execution module, configured to perform execution of the retrieved execution function template for the database Function, get the query result.

With reference to the second aspect, in a first possible implementation, the converting module includes:

a determining unit, configured to determine a node location of each relationship node in the query relationship tree, and each The type of the relationship node;

a query unit, configured to traverse the node bits of the pre-stored relationship tree set and the determined relationship nodes And a relationship tree matching the types of the relationship nodes, as a match with the query relationship tree Tie the tree.

In conjunction with the first possible implementation in the second aspect, in a second possible implementation, The conversion module further includes:

a screening unit, configured to traverse the pre-stored relationship tree set in the query unit with the determined relationship section The node location of the point and the relationship tree of the type of each relationship node match before the query relationship tree Determine the total number of relationship nodes, and filter out the total number of relationship nodes in the pre-stored relationship tree set and determine The relationship tree with the same total number of relational nodes, and the filtered relationship tree as the pre-stored traversal Tie tree collection.

With reference to the second aspect, in a third possible implementation manner, the determining module is specifically configured to Performing a hash calculation on the query relationship tree to obtain a hash value corresponding to the query relationship tree, and pre-storing In the correspondence between the relationship tree and the hash value corresponding to the relationship tree, the hash value and the obtained query are determined. A relationship tree with the same hash value of the relationship tree as a relationship tree matching the query relationship tree.

Combining the second aspect or the first possible implementation of the second aspect or the second aspect The second possible implementation or the third possible implementation of the second aspect, in the fourth In an implementation manner, the execution module is specifically configured to perform a retrieval on the database. When the execution function corresponding to the function template is executed, the original data is determined in the database, the original The data is data that is not executed by the execution function template, and the number used to represent the original data is obtained. According to the data characteristics of the size, according to the correspondence between the data features and the execution function template, Determining an execution function template corresponding to the data feature in the extracted execution function template, and for The original data executes an execution function corresponding to the execution function template corresponding to the data feature.

A third aspect of the embodiments of the present invention provides a data query system, including:

a memory for storing an execution function template and a database;

a processor for obtaining a query request for querying data in a database stored in the memory, Converting the query request into a corresponding query relationship tree, and determining in the pre-stored relationship tree set a relationship tree matching the query relationship tree, according to a correspondence between the relationship tree and the relationship tree type, Determining a relationship tree type corresponding to the matched relationship tree, according to the relationship tree type and the execution function module The correspondence between the boards, the execution function module corresponding to the determined relationship tree type is retrieved in the memory a board, executing an execution function corresponding to the execution function template for the database stored in the memory Number, get the query result.

In combination with the third aspect, in a first possible implementation manner, the processor is specifically configured to be used Determining the node position of each relation node in the query relationship tree, and the type of each relationship node, traversing The location of the node in the pre-stored relationship tree set and the determined relationship nodes and the type of each relationship node A matching relationship tree as a relationship tree that matches the query relationship tree.

In conjunction with the first possible implementation in the third aspect, in a second possible implementation, The processor is further configured to traverse the nodes of the determined relationship nodes in the pre-stored relationship tree set Before determining the location and the relationship tree of the types of the relationship nodes, determining in the query relationship tree The total number of relationship nodes is selected, and the total number of relationship nodes and the determined relationship are filtered out in the pre-stored relationship tree set. A relationship tree with the same number of nodes, and the filtered relationship tree as the pre-stored relationship tree set to be traversed Hehe.

In combination with the third aspect, in a third possible implementation manner, the processor is specifically used to The query relationship tree performs hash calculation, and obtains a hash value corresponding to the query relationship tree, which is pre-stored. In the correspondence between the relationship tree and the hash value corresponding to the relationship tree, the hash value is determined and the obtained check is obtained. The relationship tree of the relationship tree having the same hash value is used as a relationship tree matching the query relationship tree.

In combination with the third aspect or the first possible implementation of the first aspect or the third aspect The second possible implementation or the third possible implementation of the third aspect, in the fourth In an implementation manner, the processor is specifically configured to perform a retrieval on the database. When the row function template corresponds to the execution function, the original data is determined in the database, the original number Acquiring data for characterizing the original data according to data that is not executed by the execution function template The size of the data feature, according to the correspondence between the data characteristics and the execution function template, in the retrieval Determining an execution function template corresponding to the data feature in an execution function template, and for the original The start data executes an execution function corresponding to the execution function template corresponding to the data feature.

The embodiment of the invention provides a data query method, in which the database system acquires After querying the query request for the data, determining the query relationship tree corresponding to the query request, and the query The type corresponding to the relationship tree, and finally the corresponding execution function module is retrieved according to the type of the query relationship tree. a board, in the embodiment of the present invention, because the corresponding function tree is included in the execution function template The execution function used to query the data on each relational node, so as long as the execution function template is determined, The final query result can be obtained by executing the execution function corresponding to the execution function template, thereby It avoids the problem that the query speed is slow due to repeated call of the execution function in the existing data query process. This improves the speed of data queries and also saves network resources.

DRAWINGS

1 is a schematic diagram of a SQL data query mode in the prior art;

2 is a schematic diagram of another SQL data query mode in the prior art;

3 is a flowchart of a data query method according to an embodiment of the present invention;

4a is a schematic structural diagram of a relationship tree according to an embodiment of the present invention;

4b is a second schematic structural diagram of a relationship tree according to an embodiment of the present invention;

4c is a third schematic structural diagram of a relationship tree according to an embodiment of the present invention;

4d is a fourth schematic structural diagram of a relationship tree according to an embodiment of the present invention;

4e is a fifth structural diagram of a relationship tree according to an embodiment of the present invention;

FIG. 5 is a schematic structural diagram of a data query apparatus according to an embodiment of the present invention;

6 is a schematic structural diagram of a conversion module according to an embodiment of the present invention;

FIG. 7 is a schematic structural diagram of a data query system according to an embodiment of the present invention.

detailed description

In the embodiment of the present invention, a data query method is provided, in which a query request is After parsing, the relationship tree corresponding to the query request is obtained, and the class corresponding to the relationship tree is determined. Type, and then determine the execution function template corresponding to the type of the relationship tree, and finally determine the execution by execution. The row function template gets the final query result. This can correspond to a complete query request into a Execution function template, in which the node pair on the relationship tree of the corresponding type is included in the execution function template Should execute the function, thus avoiding the signaling that is caused by repeated calls to the execution function during the query process. Sales, increasing the speed of data queries.

The technical solutions of the present invention will be described in detail below with reference to the accompanying drawings and specific embodiments.

FIG. 3 is a flowchart of a data query method according to an embodiment of the present invention, where the method includes:

S301. Acquire a query request for querying data in a database.

S302. Convert the query request into a corresponding query relationship tree, and in the pre-stored relationship tree set. Determining a relationship tree that matches the query relationship tree;

Different relational trees obtained by different query request conversions are included in the relationship tree set.

S303. Determine a relationship tree according to the correspondence between the relationship tree and the relationship tree type. Corresponding relationship tree type;

In the execution function template, the corresponding type of relationship tree is included in the relationship tree for querying The execution function of the data.

S304. Retrieving and determining according to the correspondence between the relationship tree type and the execution function template. The execution function template corresponding to the relationship tree type;

S305. Perform an execution function corresponding to the execution function template of the retrieved database for the database, and obtain the check function. Ask for results.

First of all, different relational trees and relationships are stored in the database system of the embodiment of the present invention. The type corresponding to the tree, for example, the relationship tree type corresponding to the relationship tree shown in Figure 4a is of the Select type. The relationship tree type corresponding to the relationship tree shown in Figure 4b is a Group type, and the relationship tree class shown in Figure 4c The type is a TOPN type, and the relationship tree type corresponding to the relationship tree shown in FIG. 4d is a TOPN_Group type. The relationship tree type corresponding to the relationship tree shown in Figure 4e is of the TOPN_NO_Group type, and the above five The relationship tree type is determined by the relationship node in the relationship tree, and of course not in the embodiment of the present invention. The definition can only define the relationship tree type through the relationship node. In the actual application, other types can be used. The implementation method to define the relationship tree type corresponding to the relationship tree is not exhaustive here.

When the database system receives the query request input by the user, the database system will request the query. Semantic analysis is performed to obtain a query relationship tree corresponding to the query request.

For example, the query request received by the database system is "from the table T, the number of columns is greater than 5 Data", after the query request has been analyzed by the language structure, you can get the SQL statement, namely: Select A from T, where A>5. According to the SQL statement, each relationship section in the relationship tree can be obtained. Point, and each relationship node performs filtering according to the corresponding execution function, firstly Basetable The relational node and the execution filtering operation corresponding to the Basetable relational node are to find the table T from the database. Since the table T has not undergone any processing, the table T is the original data, and the need to explain here is: The initial data is extracted from the database, and the table T is not sorted by data, data filtered, data Update and other processing.

Then the Select relationship node, the corresponding filtering operation corresponding to the Select relationship node is in the table T Filter out the data with column number greater than 5; finally, the function of Project relationship node and Project relationship node Is to get the final query result, according to these relationship nodes can get the corresponding query of the query request The relationship tree is consulted, and the query relationship tree is the relationship tree shown in Figure 4a. In the relationship tree shown in Figure 4a The relationship nodes correspond to their own types.

After obtaining the query relationship tree, the database system will analyze the tree structure of the query relationship tree. The structural feature characterizes the type of each relational node in the query relationship tree and the relationship node is inspected. Ask for the location of the node in the relationship tree.

For example, the relationship tree corresponding to the query request is the query relationship tree shown in FIG. 4a, and the query relationship is The relation node corresponding to the tree head in the tree is the "Basetable" relation node, and the tree header represents "Basetable" The location of the relational node in the query relationship tree, "Basetable" represents the type of the relational node; The relationship node in the tree after the "Basetable" relationship node is the "Select" relationship node, "Select" Represents the type of relational node; the relational node corresponding to the end of the tree is the "Project" relational node, "Project" Characterizes the type of the relational node, by querying the relationship node in the query relationship tree in the query relationship tree The location and the type of each relationship node can be used to obtain the tree structure feature corresponding to the query relationship tree.

After the system gets the tree structure feature of the query relationship tree, the system will get the query relationship tree. The tree structure features are matched with the tree structure features of the pre-stored relationship tree in the system. The tree structure feature matching of the phylogenetic tree is the type of the relational node and the node location in the query relationship tree. Match the type of each relationship node in the pre-stored relationship tree set with the node location.

For example, when the query relationship tree is the relationship tree shown in FIG. 4a, the relationship node of the tree head of the query relationship tree is queried. For Basetable, the relationship node after the tree header is Select, the tree tail node is Project, and the database system is The system will determine whether to save the tree with the query relationship tree shown in Figure 4a in the pre-stored relationship tree set. A relational tree with the same structural characteristics, if there is a tree structure with a query relationship tree in the database system If the relationship tree has the same characteristics, the database system determines that the relationship tree type corresponding to the relationship tree is The Select type, that is, the relation tree type of the query relationship tree is of the Select type.

Optionally, in order to reduce the number of times the relationship tree matches the screening, in the embodiment of the invention, the data The library system can pass the total number of relational nodes in the relationship tree before performing tree structure feature matching. A part of the relationship tree is initially selected in the pre-stored relationship tree set.

Specifically, first obtain the total number of relational nodes in the query relationship tree, and the database system will The total number of nodes in the pre-stored relationship tree set is equal to the total number of relationship nodes in the query relationship tree. The phylogenetic tree, for example, the total number of relational nodes in the query relationship tree is three, and the database system will be pre-stored. In the relationship tree, the relationship tree with 3 total number of relation nodes is traversed. At this time, the filtered relationship tree is preliminary. The filtered relationship tree cannot be used as a relationship tree that matches the query relationship tree.

After filtering out the relationship tree equal to the total number of nodes in the query relationship tree in the pre-stored relationship tree, Then, in the initially selected relationship tree, the node locations of the relationship nodes of the query relationship tree are filtered out. And a relationship tree in which the types of the relationship nodes match. This way you can avoid having to query the relationship tree The node location of the relational node and the type of each relational node and each relationship in the pre-stored relational tree set The node position of the tree and the type of the relationship node are matched, thereby reducing the number of matches and improving The speed and efficiency of data queries.

Optionally, in the embodiment of the present invention, in addition to the matching of the node location and the node type, In addition to the type of the query relationship tree, you can also determine the query by calculating the hash value of the query relationship tree. The type of relationship tree.

Specifically, each type of relationship tree is pre-stored in the database system, such as Select. Type relationship tree, Group type relationship tree, TOPN type relationship tree, etc. A type of relationship tree performs hash calculation to obtain a hash value corresponding to each type of relationship tree, After saving each type of relationship tree in the database system and corresponding to each type of relationship tree Hash value.

After obtaining the query relationship tree corresponding to the query request, performing a hash calculation on the query relationship tree, Thereby obtaining a hash value corresponding to the query relationship tree, and then according to the hash value of the query relationship tree and the advance The stored hash value is matched, if there is a hash value in the database system and the query relationship tree When the same hash value is used, the type of the relation tree corresponding to the hash value is used as the type of the query relationship tree. For example, the correspondence between the relationship tree type and the hash value shown in Table 1 is stored in the database system:

Hash value Relationship tree type a Select type b Group type c TOPN type

Table 1

It can be concluded from Table 1 that if the hash value of the query relationship tree is a, then the query relationship tree The type is the Select type, and the relation tree corresponding to the Select type is the relationship tree shown in Figure 4a. If the hash value of the query relationship tree is b, the type of the query relationship tree is the Group type, and the group The relationship tree corresponding to the type is the relationship tree shown in Figure 4b. If the hash value of the query relationship tree is c, Then, the type of the query relationship tree is the TOPN type, and the relationship tree corresponding to the TOPN type is as shown in FIG. 4c. The relationship tree shown.

By querying the hash value of the relation tree, you can quickly locate the type of the query relationship tree, even if you check The tree structure of the query tree is more complicated, and the query relationship tree can be quickly determined by the hash value. type.

Further, if the hash value of the query relationship tree and the hash value pre-stored in the database system are When there is a hash conflict between them, it is also necessary to determine the query based on the tree structure characteristics of the query relationship tree. The type corresponding to the tree.

For example, the query relationship tree is hashed after the hash function is calculated and the hash in Table 1. If the value a has a hash conflict, then the relationship tree corresponding to the hash value a is determined, and then the tree of the relationship tree is queried. The structural feature of the shape is matched with the tree structure feature of the relation tree corresponding to the hash value a, if the query is off The tree structure characteristic of the relationship tree corresponding to the hash value a is matched, and the relation tree corresponding to the hash value a The type of the query relationship tree, if it does not match, it is determined that the query relationship tree has no match. Relationship tree.

Matching the tree structure feature of the relation tree or by querying the hash corresponding to the relationship tree After the values are matched to get the type of the query relationship tree, the system will retrieve the pre-stored relationship tree. The correspondence between the type and the execution function template, specifically, pre-stored in the database system a correspondence table between each type of relationship tree and an execution function template, the correspondence table may But not limited to Table 2:

Relationship tree type Execution function template Select type Template 1 Group type Template 2 TOPN type Template 3

Table 2

Based on the correspondence in Table 2, as long as the type of the query relationship tree is determined, Determine the corresponding execution function template. For example, if the query relationship tree type is Group type, then The execution function template corresponding to the query relationship tree is "template 2", where the function template can be executed but not limited. In the following code:

After determining the execution function template, the database system will execute the determined execution function template Corresponding execution function, so as to query the required query result in the database. In the execution function mode The board contains the execution function corresponding to each node in the relation tree of the corresponding type, that is, executed. The determined execution function template is equivalent to executing each node on the corresponding type of relationship tree. Executing the function, so in the embodiment of the present invention, therefore, the data query process no longer needs to be repeated. Retrieve the execution function, which improves the speed and efficiency of data query, and also saves system resources source.

Further, in the embodiment of the present invention, a type of relationship tree may correspond to multiple execution function modules. Board, so if you can't accurately determine an execution function template by the type of the relationship tree, then The execution function template can be further determined in conjunction with the data characteristics of the original data, where the original number is According to the data that is not executed by the function template found in the database, such as query request To "query data from column T with columns greater than 5", in response to the query request, the first is the number According to the table T found in the library, when the table T is found, the database system does not have a table T Do any processing, so the table T is the original data.

After obtaining the original data, it is necessary to further determine the data characteristics of the original data, here The data characteristics include at least the amount of data of the original data, and the column data to be searched in the original data. Type, one of the attributes of the column data to be searched in the original data, or a plurality of numbers According to the combination of features, the attributes of the column data represent whether the column data in the original data is sorted or Whether the column data in the original data is a loop data or the like.

In the embodiment of the present invention, the amount of data of the original data is used, that is, the query relationship When the tree corresponds to multiple execution function templates, the data size of the original data is determined, and then pre-stored according to the data. Correspondence between the amount of data and the execution function template, determining the amount of data corresponding to the original data Execution function template.

For example, the relationship tree of the Group type corresponds to 3 different execution function templates, and 3 types are not. There is a correspondence between the same execution function template and the data size of the original data, namely: the original number The execution function template corresponding to the data size of 1 byte, 2 bytes, and 3 bytes is Do_bte_group, do_sht_group, do_hash_group; therefore, the query relationship tree is determined to be a Group Type, the database system will further determine the amount of data of the original data, if it is the original number When the amount of data is 1 byte, the execution function template is do_bte_group, if it is the number of original data. When the data is 2 bytes, the execution function template is do_sht_group. In this way, you can pass If the type of the relationship tree cannot be accurately determined by executing the function template, it can be further based on The amount of data of the original data to determine the execution function template, and finally further determined by execution The execution function corresponding to the execution function template accurately queries the result of the query.

In addition to the execution of the function template by the amount of data of the original data, it is also possible to pass The type or attribute of the column data to be searched in the original data to establish and execute the function template The correspondence is no longer an example.

By establishing the correspondence between the data characteristics of the original data and the execution function template, In the case of multiple execution function templates corresponding to the relationship tree, the data characteristics of the original data are accurately determined The corresponding execution function template is determined to be executed, thereby improving the speed and accuracy of the data query.

Corresponding to a data query method in the foregoing embodiment, an embodiment of the present invention further provides a The device for querying data, as shown in FIG. 5 is a structural diagram of a data query device according to an embodiment of the present invention. Intended, the device includes:

An obtaining module 501, configured to acquire a query request for querying data in a database;

The conversion module 502 is configured to convert the query request into a corresponding query relationship tree, and pre-store Determining a relationship tree matching the query relationship tree in the relationship tree set, the pre-stored relationship tree The set includes different relation trees obtained by converting different query requests;

a determining module 503, configured to determine, according to a correspondence between the relationship tree and the relationship tree type, The relationship tree type corresponding to the matching relationship tree;

The calling module 504 is configured to adjust according to the correspondence between the relationship tree type and the execution function template Taking an execution function template corresponding to the determined relationship tree type, wherein the execution function template includes a package An execution function for querying data on each relational node in the relation tree of the corresponding type;

An execution module 505, configured to execute the retrieved execution function template for the database The line function gets the result of the query.

Further, as shown in FIG. 6, the conversion module 502 includes:

a determining unit 601, configured to determine a node location of each relationship node in the query relationship tree, and The type of each relationship node;

The query unit 602 is configured to traverse the nodes of the pre-stored relationship tree set and the determined relationship nodes a relationship tree matching the location and the type of each relationship node as a match with the query relationship tree Relationship tree.

Further, the conversion module 502 further includes:

The screening module 603 is disposed between the determining module 601 and the query module 602, and is configured to perform the check Determine the total number of relationship nodes in the query relationship tree, and filter out the total relationship nodes in the pre-stored relationship tree set. A relationship tree whose number is equal to the total number of determined relationship nodes, and the filtered relationship tree is used as the traversal Pre-stored relationship tree collection.

Further, the determining module 503 in the device is further configured to perform a hash calculation on the query relationship tree. Calculating, obtaining a hash value corresponding to the query relationship tree, corresponding to the pre-stored relationship tree and the relationship tree In the correspondence between the hash values, a relationship tree having the same hash value as the hash value of the query relationship tree is determined, As a relationship tree that matches the query relationship tree.

Further, the execution module 505 in the device is specifically configured to perform the retrieval on the database. When the execution function corresponding to the execution function template is executed, the original data is determined in the database, The raw data is data that is not executed by the execution function template, and is acquired to represent the original data. The data characteristics of the data size, according to the correspondence between the data characteristics and the execution function template, Determining an execution function template corresponding to the data feature in the retrieved execution function template, and targeting The original data executes an execution function corresponding to the execution function template corresponding to the data feature.

In addition, in the embodiment of the present invention, a data query system is further provided, as shown in FIG. Schematic diagram of the system, the system includes:

a memory 701, configured to store an execution function template and data to be queried;

The processor 702 is configured to obtain a query request for querying data in a database, where the query is The request is converted into a corresponding query relationship tree, and the query is determined in the pre-stored relationship tree set The relationship tree matched by the relationship tree is determined according to the correspondence between the relationship tree and the relationship tree type. The relationship tree type corresponding to the matching relationship tree, according to the relationship between the relationship tree type and the execution function template Corresponding relationship, extracting an execution function template corresponding to the determined relationship tree type, for the database Execute the execution function corresponding to the executed execution function template to obtain the query result.

Further, the processor 702 is specifically configured to determine a section of each relationship node in the query relationship tree. Point location, and the type of each relationship node, traversing the pre-existing relationship tree set with the determined relationships The node location of the node and the relationship tree of the types of the relationship nodes match as the query A tree that matches the tree.

Further, the processor 702 is further configured to determine a total number of relationship nodes in the query relationship tree. In the pre-stored relationship tree set, the total number of relationship nodes is selected to be equal to the total number of determined relationship nodes. The tree is used to use the filtered relationship tree as a pre-stored set of relationship trees to be traversed.

Further, the processor 702 is further configured to perform hash calculation on the query relationship tree to obtain a The hash value corresponding to the query relationship tree, and the correspondence between the pre-stored relationship tree and the hash value corresponding to the relationship tree In the relationship, a relationship tree having the same hash value as the hash value of the query relationship tree is determined as the context The relationship tree in which the query relationship tree is matched.

Further, the processor 702 is specifically configured to execute an execution function for the database. When the execution function corresponding to the template is used, the original data is determined in the database, and the original data is The data that is not executed by the execution function template acquires a large amount of data used to represent the original data. Small data features, according to the correspondence between the data features and the execution function template, are retrieved Determining an execution function template corresponding to the data feature in an execution function template, and for the original number The execution function corresponding to the execution function template corresponding to the data feature is executed.

The present invention is directed to a method, apparatus (system), and computer program according to an embodiment of the present invention. The flow chart and/or block diagram of the product is described. It should be understood that the flow chart can be implemented by computer program instructions And/or each flow and/or block in the block diagram, and the flow in the flowchart and/or block diagram And/or a combination of boxes. These computer program instructions can be provided to a general purpose computer, a special purpose computer, An embedded processor or processor of another programmable data processing device to generate a machine Instructions executed by a processor of a computer or other programmable data processing device are generated for implementation in the stream a process or a process and/or a block diagram of a function specified in a box or blocks Device.

These computer program instructions can also be stored in a bootable computer or other programmable data processing device. In a computer readable memory that operates in a particular manner such that it is stored in the computer readable memory The instruction in the production includes an article of manufacture of the instruction device, the instruction device being implemented in a flow of the flowchart or Multiple processes and/or block diagrams The functions specified in one or more boxes.

These computer program instructions can also be loaded onto a computer or other programmable data processing device. Enabling a series of operational steps on a computer or other programmable device to produce a computer implemented Processing instructions that are executed on a computer or other programmable device are provided for implementation in the flowchart a step in a process or multiple processes and/or block diagrams of a function specified in a box or blocks Step.

Although a preferred embodiment of the invention has been described, one of ordinary skill in the art will recognize Additional changes and modifications to these embodiments can be made in the basic inventive concept. Therefore, the right to attach All changes and modifications are intended to be included within the scope of the invention.

Obviously, those skilled in the art can make various modifications and variations to the present invention without departing from the present invention. The spirit and scope of the invention. Thus, if such modifications and variations of the present invention are claimed in the present invention The invention is also intended to cover such modifications and variations within the scope of the invention.

Claims (15)

A data query method, comprising: Obtain a query request for querying data in the database; Converting the query request into a corresponding query relationship tree, and in the pre-stored relationship tree set Determining a relationship tree matching the query relationship tree, where the pre-stored relationship tree set includes different pairs Query the different relationship trees obtained by the request conversion; Determining a relationship tree pair with the matching according to a correspondence between the relationship tree and the relationship tree type Should be related to the tree type; Retrieving and determining the relationship tree according to the correspondence between the relationship tree type and the execution function template An execution function template corresponding to the type, wherein the execution function template includes a corresponding type of An execution function for querying data on each relationship node in the system tree; Executing an execution function corresponding to the execution function template of the retrieved database for the database, and obtaining a query result. The method of claim 1 wherein determining in a pre-stored set of relationship trees A relationship tree matching the query relationship tree, including: Determining a node location of each relationship node in the query relationship tree, and a type of each relationship node; Traversing the node locations of the pre-stored relationship tree sets and the determined relationship nodes and the relationship sections A relationship tree in which the types of points match, as a relationship tree matching the query relationship tree. The method of claim 2, wherein traversing the pre-stored set of relationship trees Before the relationship tree that matches the determined node position of each relationship node and the type of each relationship node, Also includes: Determining the total number of relationship nodes in the query relationship tree; The total number of relationship nodes in the pre-stored relationship tree set is equal to the total number of determined relationship nodes. Relationship tree The filtered relationship tree is used as a pre-stored set of relationship trees to be traversed. The method of claim 1 wherein determining in a pre-stored set of relationship trees A relationship tree matching the query relationship tree, including: Performing a hash calculation on the query relationship tree to obtain a hash value corresponding to the query relationship tree; In the correspondence between the pre-stored relationship tree and the hash value corresponding to the relationship tree, the hash value is determined The obtained relationship tree of the query relationship tree has the same hash value as the matching relationship with the query relationship tree. Tie the tree. The method according to any one of claims 1 to 4, wherein the database is Execute the execution function corresponding to the executed execution function template, including: When performing an execution function corresponding to the execution function template of the retrieved database for the database, The database determines the raw data, which is the number that is not executed by the execution function template according to; Obtaining data features for characterizing the amount of data of the original data; According to the correspondence between the data features and the execution function template, the execution function module is retrieved. Determining an execution function template corresponding to the data feature in the board, and Executing an execution function corresponding to the execution function template corresponding to the data feature for the original data number. A data query device, comprising: An obtaining module, configured to obtain a query request for querying data in a database; a conversion module, configured to convert the query request into a corresponding query relationship tree, and pre-stored A relationship tree that matches the query relationship tree is determined in the relationship tree set, and the pre-stored relationship tree set The combination includes different relational trees obtained by converting different query requests; a determining module, configured to determine and describe according to a correspondence between a relationship tree and a relationship tree type The relationship tree type corresponding to the matching relationship tree; Retrieving a module for retrieving a correspondence between a relationship tree type and an execution function template An execution function template corresponding to the determined relationship tree type, wherein the execution function template includes An execution function for querying data on each relationship node in a relation tree of a corresponding type; An execution module, configured to perform execution of the retrieved execution function template for the database Function, get the query result. The device of claim 6, wherein the conversion module comprises: a determining unit, configured to determine a node location of each relationship node in the query relationship tree, and each The type of the relationship node; a query unit, configured to traverse the node bits of the pre-stored relationship tree set and the determined relationship nodes And a relationship tree matching the types of the relationship nodes, as a match with the query relationship tree Tie the tree. The device of claim 7, wherein the conversion module further comprises: a screening unit, configured to traverse the pre-stored relationship tree set in the query unit with the determined relationship section The node location of the point and the relationship tree of the type of each relationship node match before the query relationship tree Determine the total number of relationship nodes, and filter out the total number of relationship nodes in the pre-stored relationship tree set and determine The relationship tree with the same total number of relational nodes, and the filtered relationship tree as the pre-stored traversal Tie tree collection. The apparatus according to claim 6, wherein said determining module is specifically for Performing a hash calculation on the query relationship tree to obtain a hash value corresponding to the query relationship tree, and pre-storing In the correspondence between the relationship tree and the hash value corresponding to the relationship tree, the hash value and the obtained query are determined. A relationship tree with the same hash value of the relationship tree as a relationship tree matching the query relationship tree. The apparatus according to any one of claims 6 to 9, wherein the execution module, Specifically, when performing an execution function corresponding to the execution function template of the retrieved for the database, Determining raw data in the database, the raw data being not executed by the execution function template Data, obtaining data characteristics for characterizing the amount of data of the original data, according to the data Corresponding to the execution function template, determining the number in the retrieved execution function template Performing a function template corresponding to the feature, and performing the data feature corresponding to the original data Execute the execution function corresponding to the function template. A data query system, comprising: a memory for storing an execution function template and a database; a processor for obtaining a query request for querying data in a database stored in the memory, Converting the query request into a corresponding query relationship tree, and determining in the pre-stored relationship tree set a relationship tree matching the query relationship tree, according to a correspondence between the relationship tree and the relationship tree type, Determining a relationship tree type corresponding to the matched relationship tree, according to the relationship tree type and the execution function module The correspondence between the boards, the execution function module corresponding to the determined relationship tree type is retrieved in the memory a board, executing an execution function corresponding to the execution function template for the database stored in the memory Number, get the query result. The system of claim 11 wherein said processor is specifically Determining the node position of each relation node in the query relationship tree, and the type of each relationship node, traversing The location of the node in the pre-stored relationship tree set and the determined relationship nodes and the type of each relationship node A matching relationship tree as a relationship tree that matches the query relationship tree. The system of claim 12 wherein said processor is further operative The pre-existing relationship between the node set and the determined node of each relationship node and the class of each relationship node Before the type matching tree is matched, the total number of relation nodes is determined in the query relationship tree, and is pre-stored. In the relationship tree set, the relationship tree with the total number of relationship nodes and the total number of relationship nodes determined is selected. The filtered relationship tree is used as a pre-stored set of relationship trees to be traversed. The system of claim 11 wherein said processor is specifically for Performing a hash calculation on the query relationship tree to obtain a hash value corresponding to the query relationship tree, and pre-storing In the correspondence between the relationship tree and the hash value corresponding to the relationship tree, the hash value is determined and the obtained Query the relationship tree of the relationship tree with the same hash value as the relationship tree matching the query relationship tree. The system according to any one of claims 11 to 14, wherein said processor is specifically used In the execution function corresponding to the execution function template of the retrieved execution of the database, The database determines raw data, which is data that is not executed by the execution function template, Obtaining data features for characterizing the amount of data of the original data, according to data characteristics and execution Corresponding relationship between function templates, determining the data feature pair in the retrieved execution function template Executing a function template and executing an execution function corresponding to the data feature for the original data The execution function corresponding to the template.

Images