WO2015100652A1

WO2015100652A1 - Method and apparatus for performing conversion among serialized messages

Info

Publication number: WO2015100652A1
Application number: PCT/CN2013/091193
Authority: WO
Inventors: 吕明伟
Original assignee: 北京新媒传信科技有限公司
Priority date: 2013-12-31
Filing date: 2013-12-31
Publication date: 2015-07-09
Also published as: CN105793842B; CN105793842A

Abstract

Disclosed are a method and an apparatus for performing conversion among serialized messages. The method provided in an embodiment of the present invention comprises: generating, by using a set serialized description entry, a serialized data structure according to a first structure of a first serialized message and a correspondence between the first structure and a second structure of a second serialized message, the serialized description entry comprising a data structure description entry and a field description entry set in the data structure description entry; for a to-be-converted serialized message conforming to a first format of the first structure, parsing the to-be-converted serialized message according to the serialized data structure, and writing parsed data into the serialized data structure, so as to obtain the serialized data structure the data is written into; extracting the data in the serialized data structure the data is written into; and converting the extracted data into a serialized message with a second format of the second structure according to a correspondence recorded in the serialized data structure.

Description

Method and apparatus for converting between serialized messages

The present invention relates to the field of data processing technologies, and in particular, to a method and apparatus for converting between serialized messages.

Background technique

With the development of the Internet and the application of cloud computing, there are more and more large-scale distributed systems. Message serialization is an important part of distributed systems. It is responsible for message delivery between service clusters like blood. Serialization protocols commonly found in message serialization include self-explanatory serialization protocols such as XML (Extensible Markup Language) protocol, binary-based serialization protocol ProtocolBuffer (Protocol Cache, PB for short), and JSON (JavaScript). Object Notation, JavaScript object tagging, etc.

Currently, a server generally supports only one format of a serialization protocol. For example, a WEB server only supports an XML format or a JS0N format protocol, and a service server generally only supports a PB format protocol.

Existing solutions have at least the following drawbacks:

Since the existing server can only support messages in a single format, the message transfer mode between the servers is limited, for example, the message can only be transmitted in the XML format or can only be transmitted in the PB format, and the message body of the XML format message is large. The operation efficiency is poor, and the message in the PB format is poorly readable. In the existing solution, the message delivery mode of the server is inflexible and the message transmission performance is low. Summary of the invention

In view of the above problems, the embodiments of the present invention provide a conversion method between serialized messages and a corresponding conversion device between serialized messages, so as to solve the inflexible manner of message transmission between servers in the existing solution. Low performance issues, etc.

In order to achieve the above objective, the technical solution adopted by the embodiment of the present invention is as follows:

In one aspect, an embodiment of the present invention provides a method for converting between serialized messages, including:

And generating, according to the first structure of the first serialized message and the correspondence between the first structure and the second structure of the second serialized message, the serialized data structure by using the set serialization description item; wherein, the first serialized message Having a first format, the second serialized message has a second format; the serialized description item includes a data structure description item and a field description item set in the data structure description item, a structure of each data structure description item, and a data structure description item The structure of each field description item is consistent with the first structure; for the serialized message to be converted in the first format conforming to the first structure, the serialized message is parsed according to the serialized data structure, and the parsed data is written. In the serialized data structure, the serialized data structure after the data is written is obtained; the data in the serialized data structure after the data is written is extracted; and the extracted data is converted into the data according to the correspondence recorded in the serialized data structure. A serialized message of a second format having a second structure.

On the other hand, an embodiment of the present invention provides a device for converting between serialized messages, including:

a data structure generating unit, configured to generate a serialized data structure by using the set serialized description item according to the first structure of the first serialized message and the correspondence between the first structure and the second structure of the second serialized message; The first serialized message has a first format, and the second serialized message has a second format. The serialized description item includes a data structure description item and a field description item set in the data structure description item, and each data structure description item The structure of each field description item in the structure and data structure description item is consistent with the first structure;

The parsing unit is configured to parse the serialized message to be converted in the first format conforming to the first structure, parse the serialized message according to the serialized data structure, and write the parsed data into the serialized data structure to obtain Serialized data structure after data is written;

a data extracting unit, configured to extract data in the serialized data structure after the data is written;

And a converting unit, configured to convert the extracted data into a serialized message of a second format having the second structure according to the correspondence recorded in the serialized data structure.

As described above, the embodiment of the present invention generates a serialized data structure by using the set serialization description item, and writes the serialized data message to be converted in the first format into the serialized data structure, and then serializes the data after the data is written. The technical means for extracting data into a serialized message of the second format in the structure can enable the server to conveniently and quickly serialize messages of different formats through a serialized data structure compatible with serialized messages of multiple formats. Conversion, according to actual needs, select a serialization protocol of a specific format to transmit serialized messages, for example, use XML format to transmit messages outside the server, to inherit the extensive application of the XML format in the system, and transmit the message in the PB format inside the server. In order to inherit the efficiency of the PB format, thereby improving the flexibility of the server to deliver messages and optimizing the message transmission performance. DRAWINGS

1 shows a flow chart of a method for converting between serialized messages according to an embodiment of the present invention;

2 is a schematic diagram showing a method for parsing an XML serialized message according to a serialized data structure according to still another embodiment of the present invention;

3 illustrates a flow diagram of outputting a PB serialized message from a serialized data structure written by data, in accordance with yet another embodiment of the present invention;

4 is a schematic diagram showing a method of parsing a PB message according to a serialized data structure according to still another embodiment of the present invention; FIG. 5 is a diagram showing a serialized data structure written from data according to still another embodiment of the present invention; Flowchart for outputting an XML serialized message;

FIG. 6 is a block diagram showing the structure of a device for converting between serialized messages according to still another embodiment of the present invention.

detailed description

The embodiments of the present invention will be further described in detail below with reference to the accompanying drawings.

An embodiment of the present invention provides a method for converting between serialized messages. Referring to FIG. 1, the method includes the following steps: S100: following a first structure of a first serialized message, and the first structure and the second sequence Corresponding to the second structure of the message, the serialized data structure is generated using the set serialization description item, wherein the first serialized message has a first format and the second serialized message has a second format.

The first format and the second format respectively correspond to a data format of a serialization protocol, such as an XML (Extensible Markup Language) format, a PB (Protocol Cache, abbreviated as PB) format, or a JSON (JavaScript Object Notation, JavaScript object tag) format, etc., where the first format is different from the second format.

A serialized data structure is a bridge between serialized messages, which contains several serialized description items. A serialized description item represents a type or field that can be serialized to describe the rules at the time of serialization.

The serialization description item includes a data structure description item and a field description item. A data structure description item is a set of field description items. A data structure description item may include multiple field description items, and another data structure may be nested in the data structure description item, that is, set in the data structure description item. Subdata structure description items support complex data types.

The sub-data structure description item is basically the same as the data structure description item. The main difference is that the sub-data structure description item also needs to include a relationship information, and the sub-data structure description item located in the same data structure description item is established by the relationship information. The field describes the relationship of the items.

The structure of each data structure description item and the structure of each field description item and sub-data description item (if present) in the data structure description item need to be consistent with the first structure. The following is explained in conjunction with the examples.

Illustratively, when the first format is an XML format, an example of a first serialized message (XML serialized message) having a first structure is as follows:

<config> 〃 indicates the root node

<name></name> 〃 indicates node element

<address></address> 〃 indicates node element

<childs childName= " "> // indicates nested child root nodes

<childName></childName> 〃 indicates child attribute

< childAddress ></childAddress> indicates the child element

</childs>

</config>

In the first structure of the above XML serialization message, the root node name is "config", and the root node includes node elements named "name" and "address", and the first structure further includes a nested substructure, substructure The child root node has the name "childs" and the child root node includes the node attribute named "childName" and the node element named "childAddress".

An example of a second serialized message (PB serialized message) having a second structure when the second format is the PB format is as follows:

Message config {

Optional string name = 1; 〃 indicates a data item with sequence number 1

Optional string address = 2; 〃 indicates a data item with sequence number 2

Message childs {

Optional string name = 1; 〃 indicates a data item with sequence number 1

Optional string address = 2; 〃 indicates a data item with sequence number 2 }

Optional childs childsEntity = 3; 〃 indicates a substructure of sequence number 3

}

The second structure of the PB serialization message includes a data item whose sequence number is 1, a data type is a string (string), and a data item whose data type is a character string. The second structure further includes a nested sub-structure of sequence number 3. The sub-structure includes a data item whose sequence number is 1, a data type is a character string, and a data item whose serial number is 2. and the data type is a character string.

Then, according to the first structure of the XML serialization message and the correspondence between the first structure and the second structure of the PB serialization message, the serialized data structure is generated by using the set serialization description item, which can be expressed as follows:

Data structure { // indicates data structure description item

Structure name = "config" 〃 indicates the structure name

Field { sequence number =1, name = "name" , type=String } // indicates field description item

Field { sequence number = 2, name = ' address" , type=String } 〃 indicates field description item

Data structure { 〃 indicates sub-data structure description item

No. =3, 〃 indicates relationship information

Structure name = "childs" 〃 indicates substructure name

Field {serial number =1, name = = " childName" , node type = "property", type=String}〃 indicates the field description item

Field {serial number = 2, name = " childAddress" , type=String } 〃 indicates field description item

}

The serialized data structure includes a data structure description item, and the data structure description item includes a structure name, a field description item, and a sub data structure description item. The structure name is the root node name "config" in the first structure above.

Field description items include data type (type), sequence number, node type, and name, where the node type is the default node type (such as a node element) and can be omitted. The sequence number in the field description item corresponds to the sequence number in the PB format, and the sequence number in the field description item is the sequence number of the data in the field description item in the PB serialization message having the second structure, and the field description item includes the “sequence number. =1 ", indicating that the data in the field description item corresponds to the data item with sequence number 1 in the PB format; the node type in the node type field description item includes the node element and the node attribute, and the default node type is the node element, Omitting or not; the name in the field description item corresponds to the name of the node element or node attribute in the XML format.

The sub-data structure description item includes a sequence number, a sub-structure name, and a sub-field description item. In the above example, the relationship between the sub-data structure description item and the field description item in the same serialized data structure is established by using the sequence number (serial number=3) in the sub-data structure description item, and the first-order data structure is sorted in the first place. It is a field description item with sequence number 1. If the second field is the field description item with sequence number 2, the relationship between the sub-data structure description item with sequence number 3 and the first two field description items is located in the first two. After the field description item is sorted. The substructure name in the subdata structure description item corresponds to the name of the subroot node in the XML format. The subfield description item includes the data type, sequence number, node type, and name. The information in the subfield description item has the same effect as the information in the field description item.

It can be seen from the above that the first structure and the PB sequence of the XML serialized message are established by the structure name in the data structure description item, the substructure name in the subdata structure description item, and the sequence number and the name in the field description item and the sub field description item. Corresponding relationship between the second structure of the message; for the scenario where the nested structure does not exist, the first structure and the PB of the XML serialization message are established by the structure name and the sequence number and the name in the field description item in the data structure description item The correspondence between the second structures of the serialized messages; that is, the correspondence is set in the serialized data structure by the information in the data structure description item and the information in the field description item. In a variant, the correspondence may also be set in the serialized data structure only by the information in the field description item.

The information in the data structure description item, the sub-data structure description item, the field description item, and the sub-field description item are changed according to the format and structure of the serialization message, and are not strictly limited herein.

After the serialization data structure is created by the above operation, the process proceeds to step S101.

S101: parsing the to-be-converted serialized message according to the serialized data structure, and parsing the parsed data into the serialized data structure to obtain data Serialized data structure after writing.

For example, when writing data, you can add a value (value) to the field description item or subfield description item, and write the parsed data into the value.

S102: Extract data in the serialized data structure after the data is written.

S103: Convert the extracted data into a serialized message of a second format having a second structure according to the correspondence recorded in the serialized data structure.

In another embodiment of the present invention, a scenario in which the first format is an XML format and the second format is a PB format is taken as an example, and the foregoing step S 101 is performed. The operation in the description is described in which the serialized message in the XML format has the first structure, and the serialized message in the PB format has the second structure; other contents of the conversion method between the serialized messages are referred to other embodiments of the present invention.

After the serialized data structure is obtained through the above step S100, all the XML serialized messages conforming to the first structure can be parsed as the serialized data structure to be converted, and the parsing process is shown in FIG. 2, which shows The process of parsing an XML serialized message according to a serialized data structure includes the following steps:

S200: The server (or other device) loads the XML serialization message to be converted.

S201: The server corrects the root node according to the XML serialization message to be converted.

The server uses the stream analysis method to read the XML elements in the XML serialization message to be converted according to the serialized data structure, and processes the XML elements in the read XML serialization message to be processed one by one. When the name in the read XML element is the same as the structure name in the data structure description item, it is determined that the XML node corresponding to the XML element is the root node.

By correcting the root node, the nodes of the XML and the corresponding field description items in the serialized data structure can be made consistent.

S202: The server parses the node attribute in the current node, and writes the data of the XML element as the node attribute into the corresponding field description item in the serialized data structure, and proceeds to step S203.

The current node is an XML node that is currently being processed or indirectly dependent on the root node, and the XML element in the current node is dependent or indirectly dependent on the root node. The data of the XML element in the current node is parsed one by one, and the parsed data is written into the serialized data structure. When the name in the XML element is the same as the name in the field description item and the type of the XML element is the same as the node type in the field description item, the parsed data of the XML element is written in the field description item.

The type of the XML element includes a node attribute and a node element. In this embodiment, when the parsing is performed, the parsing of the node attribute in the currently processed current node is performed, and then the parsing of the node element is performed. However, this embodiment does not apply to the node. The order in which attributes and node elements are parsed is strictly limited.

For example, describe the following field descriptions in the serialized data structure:

Field {serial number =1, name = "childName" , node type = "property", type=String} After writing the parsed node attribute data into the field description item, the details are as follows:

Field {serial number = 1, name = "childName", node type = "property", type=String, value= "XXX" } where XXX written in the above value represents the parsed data of the XML element as the node attribute .

S203: The server parses the node element in the current node, and proceeds to step S204.

Reads the name of the XML element currently in the XML node and the data of the XML element.

S204: The server determines whether there is a nested structure, and if yes, returns to step S201, and if no, proceeds to step S205 to determine whether there is a nested structure according to the name of the XML element, and the name and sub-data structure description item of the read XML element When the substructure names in the same are the same, confirm that there is a nested structure.

For the scenario in which the nested structure exists, after returning to step S201, the currently processed XML element is used as a child root node (replace the root node with the child root node), and each XML element in the child root node is read, and the child root is read. The name of the XML element in the node will be the same as the name in the subfield description item (substituting the field description item with the subfield description item) and the type of the XML element is the same as the node type in the subfield description item. The data of the XML element is written in the subfield description item.

S205: The server writes the data of the XML element as the node element into the corresponding field descriptor, and the name of the XML element read in step S206 is the same as the name in the field descriptor, and the node type in the field descriptor When it is a node element, it is confirmed that the XML element corresponds to one node element, and the data of the XML element is read and written into the corresponding field descriptor.

S206: determining whether the data of the XML element in the read XML serialization message to be converted has been written to the corresponding field description item or sub-field description item, and if yes, proceeding to step S207, if no, returning to step S201, Repeating the above steps S201 to S206, writing the unwritten XML element into the field description item or the sub-field description item until it is determined that all the data in the XML serialization message is written in the serialization structure, and proceeding to step S207

S207: The serialized data structure after the data is written is obtained by the field description item or the sub-field description item after the data of the XML element is written, and the operation ends.

In another embodiment of the present invention, a scene in which the first format is an XML format and the second format is a PB format is taken as an example, and data is extracted from a serialized structure after data writing and converted into a PB sequence having a second structure. The operation of the message is explained. Further details of the method of converting between serialized messages are found in other embodiments of the invention.

After the data stored in the XML serialization message is written into the serialized data structure, the data in the data structure needs to be output in PB format, and finally the serialized message from the XML format to the serialized message in the PB format is converted. Follow the PB protocol specification when performing the conversion. Referring to FIG. 3, a flow chart of outputting a PB serialization message from a serialized data structure after data writing according to the embodiment is shown, which specifically includes the following steps:

S300: The server (or other device) starts the output of the PB serialization message, and proceeds to step S301.

The server enters an output mode of the PB format, in which the server outputs a PB serialization message. Server load count The serialized data structure is written to perform format conversion.

S301: The server extracts the data in the field description item and the data in the sub-data structure description item (if any) from the serialized data structure after the data is written, and proceeds to step S302.

S302: The server performs different operations according to different types of extracted data.

When the extracted data is a numeric type of data, the process proceeds to step S303;

When the extracted data is string type data, the process proceeds to step S304;

When the extracted data is a sub data structure description item, the process proceeds to step S305.

S303: When the extracted data is a numeric type of data, the first manner is used for processing: first, in a variable (Varint) manner, the serial number and the data type of the data recorded in the field description item are first written into the PB serialized message. In the data item corresponding to the same serial number, data of the numerical type is written into the data item.

S304: When the extracted data is a string type data, the following method is used to process: first, in the Varint manner, the sequence number of the data recorded in the field description item is the same as the data type written in the PB serialization message. In the data item corresponding to the serial number, the string length of the data recorded in the field description item is written into the data item, and then the data of the string type is written into the data item.

S305: When the extracted data is a sub-data structure description item, the sequence number and the structure type (the structure type indication sub-data structure) in the sub-data structure description item are first written into the PB serialization message in the Varint manner. In the corresponding data item, the data length in the sub-data structure description item is written into the data item, and then returns to step S301, and the data in the sub-data structure description item is processed in the same manner as in steps S301 to S304. The difference is mainly to replace the field description item in steps S301 to S304 with the sub-field description item, that is, when the extracted content is numeric type data, according to the first manner, the sub-field description item is firstly introduced in the Varint manner. The serial number of the data recorded in the data type is written in the data item corresponding to the same serial number in the PB serialization message, and the data of the numerical type is written into the data item; when the extracted data is the data of the string type According to the second manner described above, the sequence number and data type of the data recorded in the subfield description item are first described in the Varint manner. PB message into sequences of the same sequence number corresponding to the data item, then the length of the substring description entry field data recorded in the data item is written, and then writes the character string data type in the data item.

S306: determining whether the extracted data has been written into the corresponding data item in the PB serialization message, and if yes, proceeding to step S307, if not, returning to step S302, repeating the above steps S302 to S306, according to the unwritten data The data type, in the first mode or the second mode, writes the unwritten data into the corresponding data item until all the data in the serialized data structure is processed.

S307: Obtain a PB serialized message having a second structure from each data item after the data is written.

At this point, the conversion from the XML serialization message to the PB serialization message is completed, and the operation ends.

In another embodiment of the present invention, a scenario in which the first format is a PB format and the second format is an XML format is used as an example, and the data in the serialized message of the PB format of the first structure is parsed, and the parsed data is written into the sequence. The operation of the data structure is explained. Further details of the method of converting between serialized messages are found in other embodiments of the invention.

After obtaining the serialized data structure, all PB serialized messages conforming to the first structure can be parsed by using the serialized data structure. The parsing process is shown in FIG. 4, which shows that the PB serialized message is performed according to the serialized data structure. The process of parsing includes the following steps:

S400: The server loads the binary data stream of the PB to be converted serialized message.

S401: The server reads the variable (Varint) value in the binary data stream of the PB to be converted serialized message one by one.

S402: The server performs different operations according to different types of extracted Varint values.

The server first determines the data type of the extracted data according to the correspondence recorded in the serialized data structure. For example, the server can know the data type of the data written in the field description item according to the data type in the field description item. In the specific operation, the server obtains the data type and sequence number corresponding to the variable value from the read variable value. According to the Varint coding standard, the data of the last 3 bits of the Varint value is placed in the data type, and the remaining bits are placed in the sequence number.

When the read Varint value is a numeric type, the process proceeds to step S403;

When the read Varint value is a string type, proceed to step S404;

When the read Varint value is a sub data structure description item, the process proceeds to step S405.

Here, when the sequence number of the read Varint value is the same as the sequence number of the sub data structure description item, it is determined that the read Varint value is a sub data structure description item.

S403: When the read Varint value is a numeric type, the following method is used: continue to read the next Varint value (the next Varint value is a numeric type data), and write the next Varint value. In the serialized data structure, the sequence number is the same as the sequence number of the Varint value. Since the sequence number and the data type of the data item in the PB format are recorded in the serialized data structure, and the order of each data item in the serialized data structure is also the same as the order of each data item in the PB serialized message to be converted, Therefore, according to the serialization structure, each data item in the PB serialization message can be quickly read out, and the read data is written into the serialized data structure according to the data type and the serial number.

S404: When the read Varint value is a string type, the fourth method is used for processing: continue to read the next Varint value, obtain the length of the string, and then read the byte length of the string, The read byte is encoded in UTF8 (8-bit Unicode Transformation Format) to obtain a string, which is written in the field description item whose sequence number is the same as the serial number of the string in the serialized data structure.

S405: when the sequence number of the read Varint value is the same as the sequence number of the sub-data structure description item, continue to read the next Varint value, obtain the data length, and then read the byte of the data length size to obtain the bytecode set. Go back to step S401. The data in the sub-data structure description item is processed in the same manner as in the steps S401 to S404, and the difference mainly lies in replacing the field description item in the steps S401 to S404 with the sub-field description item, that is, when the byte code is from the byte code When the Varint value read in the set is a numeric type, the third method is used for processing, and the third method is used for processing: continue reading the next Varint value, and writing the next Varint value to the serialized data structure. The subfield is in the same subfield description as the sequence number of the Varint value. When the Varint value read from the bytecode set is a string type, the fourth method is used for processing: continue to read the next Varint value, obtain the length of the string, and then read the length of the string. Byte, UTF8 encoding the read byte, obtaining a string, and writing the string into the subfield description item whose sequence number is the same as the serial number of the string in the serialized data structure.

S406: determining whether all the Varint values in the PB to be converted serialized message have been written in the field description item or the subfield description item, if yes, proceeding to step S407, if not, returning to step S402, repeating the above steps S402 to S406, According to the data type of the unwritten Varint value, the unwritten variable value is written into the field description item or the subfield description item in the third mode or the fourth mode until all the data in the PB serialization message is written. In the serialized structure, proceed to step S407.

S407: The serialized data structure after the data is written by the field description item or the sub-field description item after the Varint value is written. End the operation.

In another embodiment of the present invention, a scenario in which the first format is a PB format and the second format is an XML format is taken as an example, and the extracted data is converted into an XML serialization of the second structure according to the correspondence in the serialized data structure. The operation of the message is explained. Further details of the method of converting between serialized messages are found in other embodiments of the invention.

After the data stored in the PB serialization message is written into the serialized data structure, the data in the data structure needs to be output in XML format, and finally the conversion from the PB format to the XML format is completed. Referring to FIG. 5, the present embodiment is shown. The flowchart for outputting the XML serialization message from the serialized data structure after data writing provided by the embodiment includes the following steps:

S500: The server (or other device) starts the output of the XML serialization message, and proceeds to step S501.

The server enters an output mode in XML format, in which the server outputs an XML serialized message. The server loads the serialized data structure after the data is written to perform format conversion.

S501: The server creates an XML root node, and proceeds to step S502.

The server first determines whether the extracted data is the structure name and the node type for judging the extracted data according to the correspondence recorded in the serialized data structure. When the name in the extracted data is the same as the structure name of the serialized data structure record, the root node in the XML serialization message is created with the data and structure name.

S502: The server extracts data and sub-data structure description items (if any) in the field description item from the serialized data structure after the data is written, and proceeds to step S503.

S503: The server creates an XML node attribute, and writes data into the XML node attribute, and proceeds to step S504 to learn whether the data corresponds to the node element by using the field description item or the node type recorded in the sub-field description item to which the extracted data belongs. Node properties.

When the extracted data is a node attribute, the following sixth manner is used for processing: Creating an XML node with the name in the field description item of the node attribute under the current node (the currently processed node directly or indirectly subordinate to the root node) Attribute, and write the data in the field description item to the XML node attribute.

In this embodiment, when performing the conversion, the node attribute in the XML serialization message is created first, and then the node element is created. However, this embodiment does not strictly define the node attribute and the order in which the node element is created.

S504: The server creates an XML node element.

When the extracted data is a node element, the scene with no nested structure is processed in the following fifth way: Under the root node, an XML node is created with the name in the field description item of the node element, and the field description is performed. The data in the item or subfield description item is written to the XML node.

For a scenario with a nested structure, in this step, after creating an XML node element under the root node with the name in the field description item of the node element, proceed to step S505.

S505: The server determines whether the node is a nested structure, and if yes, returns to step S502, and if no, proceeds to step S506. When the extracted structure includes the substructure name, it is known that the data corresponds to the sub data structure description item. When the extracted data is a sub-data structure description item, that is, when the node includes a nested structure, the sub-root node is created by using the name in the sub-data structure description item, and the data in the sub-data structure description item is extracted, and step S502 is adopted. The same method in S504 processes the data in the sub-data structure description item, the main difference is that the field description items in steps S501 to S504 are replaced with sub-field description items, that is, when the extracted data is a node element, An XML node is created under the child root node with the name in the subfield description item of the node element, and the data in the subfield description item is written into the XML node. When the extracted data is a node attribute, an XML node attribute is created under the current node with the name in the sub-field description item of the node attribute, and the data in the sub-field description item is written into the XML node attribute.

S506: The server writes data in the field description item (or sub-field description item) into the XML node.

S507: determining whether the extracted data has been written into the corresponding XML node element or the XML node attribute, and if yes, proceeding to step S508, if not, returning to step S502, repeating the above steps S502 to S507, according to the unwritten The node type of the data, the unwritten data is written into the corresponding XML node element or the XML node attribute by using the fifth mode or the sixth mode described above, until all the data in the serialized data structure after the data is written is processed. Go to step S508.

S 508: The XML serialization message with the second structure is obtained from the XML node element or the XML node attribute after the data is written, and the operation ends. Yet another embodiment of the present invention provides a conversion apparatus 600 between serialized messages. Referring to Figure 6, the apparatus 600 includes a data structure generating unit 601, a parsing write unit 602, a data extracting unit 603, and a converting unit 604. The data structure generating unit 601 includes a correspondence relationship setting unit 601 1 and a sub-structure setting unit 6012.

The data structure generating unit 601 is configured to generate a serialized data structure by using the set serialization description item according to the first structure of the first serialization message and the correspondence between the first structure and the second structure of the second serialization message; The first serialized message has a first format, and the second serialized message has a second format. The serialized description item includes a data structure description item and a field description item set in the data structure description item, and each data structure description item The structure of each field description item in the structure and data structure description item is consistent with the first structure. Specifically, the data structure generating unit 601 includes a correspondence relationship setting unit 601 1 for using the sequence according to the first structure of the first serialization message and the correspondence between the first structure and the second structure of the second serialization message. The information in the data structure description item of the description item and the information in the field description item are set in the generated serialized data structure. When the first format is the XML format, the second format is the PB format, or the first format is the PB format, and the second format is the XML format, the correspondence relationship setting unit 601 1 is configured to set the structure name in the data structure description item. And a field description item, the structure name is a name of a root node in the XML serialization message having the first structure, the field description item includes a data type, a sequence number, a node type, and a name, and the serial number is a data in the field description item. The sequence number in the PB serialization message of the two structure; the correspondence relationship is set in the generated serialized data structure by the structure name in the data structure description item, the sequence number and the name in the field description item.

The data structure generating unit 601 further includes a sub-structure setting unit 6012, configured to set a sub-data structure description item in the data structure description item, where the sub-data structure description item is provided with a sub-field description item, and the corresponding relationship includes the same data structure. The relationship between the subdata structure description item in the description item and the field description item. The correspondence relationship setting unit 601 1 is configured to set a structure name, a field description item, and a sub data structure description item in the data structure description item, where the structure name is a name of the root node in the XML serialization message having the first structure, and the sub data structure description The item includes a sequence number, a substructure name, and a subfield description item, and the field description item or the subfield description item includes a data type, a sequence number, a node type, and a name, and the serial number is a field description item or the data in the subfield description item has a second The sequence number in the structured PB serialization message; the corresponding relationship is set in the generated structure by the structure name in the data structure description item, the substructure name in the subdata structure description item, the field description item, and the sequence number and name in the subfield description item. Serialized in a data structure.

When the first format is the XML format and the second format is the PB format, the parsing and writing unit 602 is configured to read the XML element in the XML serialized message to be converted according to the serialized data structure, when the read XML element When the name in the name is the same as the structure name in the data structure description item, the XML node corresponding to the XML element is determined to be the root node; the XML elements that are dependent or indirectly subordinate to the root node are read, and the name and the field description item in the XML element are When the names in the same are the same and the type of the XML element is the same as the node type in the field description item, the data of the parsed XML element is written in the field description item; when the name in the XML element and the sub-data structure description item When the substructure names are the same, the XML element is used as a sub-root node, and each XML element subordinate to or indirectly subordinate to the sub-root node is read, and the name and sub-field description item in the sub-field description item are subordinate or indirectly subordinate to the sub-root node. When the name is the same and the type of the XML element is the same as the node type in the subfield description item, the data of the parsed XML element is written in the subfield. In the description item, determining whether the data of the XML element in the read XML serialization message to be converted has been written to the corresponding field description item or sub-field description item, if yes, after writing the data of the XML element The field description item or the sub-field description item obtains the serialized data structure after the data is written. If not, the unwritten XML element is written in the field description item or the sub-field description item, and the field description item or sub-field description item The name and type of the node are the same as the name and type of the unwritten XML element.

The converting unit 604 is configured to determine, according to the correspondence relationship recorded in the serialized data structure, the data type of the extracted data; when the extracted data is a numerical type data, the first manner is performed by using the following method: First describe the item or sub The sequence number of the data recorded in the field description item is written in the data item corresponding to the same sequence number in the PB serialization message, and the data of the value type is written into the data item; when the extracted data is a string type data The second method is used to process: first, in the variable manner, first input the sequence number and data type of the data recorded in the field description item or the sub-field description item into the data item corresponding to the same serial number in the PB serialization message, and then The string length of the data recorded in the field description item or subfield description item is written in the data item, and then the data of the string type is written into the data item; when the extracted data is a sub data structure description item, the variable is The method first writes the sequence number and structure type in the sub-data structure description item into the data item corresponding to the same serial number in the PB serialization message, and then writes the data length in the sub-data structure description item into the data item, and then extracts the sub-data structure. The sub-field description item of the item describes the data in the item. When the extracted data is a numeric type of data, the first method is used for processing. When the extracted data is a string type data, the second method is used for processing; whether the extracted data has been written into the corresponding data item in the PB serialization message, and if so, the data after the data is written The item obtains the serialized message of the second format having the second structure. If not, according to the data type of the unwritten data, the unwritten data is written into the corresponding data item according to the first mode or the second mode. .

When the first format is the PB format and the second format is the XML format, the parsing and writing unit 602 is configured to read the variable values in the binary data stream of the PB to be converted serialized message one by one; from the read variable values Obtain the data type and serial number corresponding to the variable value; When the read variable value is a numeric type, use the following third method to process: continue reading the next variable value, and write the next variable value to the serialized data structure. In the field description item or sub-field description item whose serial number is the same as the serial number of the variable value, when the read variable value is a string type, the following fourth method is used for processing: continue to read the next variable value, and obtain The length of the string, and then the byte of the length of the string is read, the string is obtained, and the string is written into the field description item or the subfield description item of the sequence number in the serialized data structure; When the sequence number of the extracted variable value is the same as the sequence number of the sub data structure description item, the value of the next variable is continuously read, the data length is obtained, and then the number is read. The length of the byte gets the bytecode set, and reads the variable value in the bytecode set. When the read variable value is a numeric type, the third way is used to process the read variable value. When the string type is used, the fourth method is used for processing; determining whether all variable values in the PB to be converted serialized message have been written in the field description item or the sub field description item, and if so, the field after the variable value is written The description item or subfield description item obtains the serialized data structure after the data is written. If not, according to the data type of the unwritten variable value, the unwritten variable value is written into the field by the third mode or the fourth mode. Description item or subfield description item.

The converting unit 604 is configured to determine, according to the correspondence relationship recorded in the serialized data structure, whether the extracted data is a structure name and determine a node type of the extracted data; when the extracted data includes a structure name, the data and The structure name creates the root node in the XML serialization message; when the extracted data is a node element, it is processed in the following fifth way: under the root node, the field description item or subfield description item name of the node element Create an XML node element and write the data in the field description item or sub-field description item into the XML node element; when the extracted data is a node attribute, the following sixth manner is used to process: Create an XML node attribute in the field description item or the name in the subfield description item, and write the data in the field description item or subfield description item into the XML node attribute; when the extracted data is a sub data structure description item , create a child root node with the name in the child data structure description item, extract the number in the child data structure description item When the extracted data is a node element, the fifth method is used for processing under the child root node, and when the extracted data is a node attribute, the sixth method is used for processing; determining whether the extracted data has been written To the corresponding XML node element or XML node attribute, if yes, the XML node element or the XML node attribute written by the data obtains the serialized message of the second format having the second structure, and if not, according to the unwritten data The node type is written into the corresponding XML node element or XML node attribute by using the fifth mode or the sixth mode described above.

The data extracting unit 603 is for extracting data in the serialized data structure after the data is written.

For a specific working mode of each unit in the device embodiment of the present invention, reference may be made to the method embodiment of the present invention.

As described above, the embodiment of the present invention generates a serialized data structure by using the set serialization description item, and writes the serialized message to be converted in the first format into the serialized data structure, and then serializes after the data is written. The technical means for extracting data into a serialized message of the second format in the data structure can enable the server to facilitate serialized messages of different formats by using a serialized data structure compatible with serialized messages of multiple formats. Fast conversion, according to actual needs, select a serialization protocol of a specific format to transmit messages, for example, use XML format to transmit messages outside the server, to inherit the extensive application of the XML format in the system, and transmit messages in the PB format inside the server. Inheriting the efficiency of the PB format, thereby improving the flexibility of the server to deliver messages and optimizing the message transmission performance.

In order to facilitate the clear description of the technical solutions of the embodiments of the present invention, in the embodiments of the present invention, the words "first", "second," and the like are used to distinguish the same items or similar items whose functions and functions are substantially the same, in the field. The skilled person can understand that the words "first", "second", etc. do not limit the quantity and execution order.

The above is only the preferred embodiment of the present invention and is not intended to limit the scope of the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and scope of the invention are intended to be included within the scope of the invention.

Claims

claims

1. A conversion method between serialized messages, including:

According to the first structure of the first serialized message and the corresponding relationship between the first structure and the second structure of the second serialized message, a serialized data structure is generated using the set serialization description item; wherein, the first sequence The serialization message has a first format, and the second serialization message has a second format; the serialization description item includes a data structure description item and a field description item set in the data structure description item, and the The structure of each field description item among the structure and data structure description items is consistent with the first structure;

For a serialized message to be converted that conforms to the first format of the first structure, the serialized message to be converted is parsed according to the serialized data structure, and the parsed data is written into the serialized data structure. , get the serialized data structure after data is written;

Extract the data in the serialized data structure after the data is written;

According to the corresponding relationship recorded in the serialized data structure, the extracted data is converted into a serialized message in a second format with a second structure.

2. The method according to claim 1, wherein, according to the first structure of the first serialized message and the corresponding relationship between the first structure and the second structure of the second serialized message, using the set Generating the serialized data structure from the serialized description item includes: according to the first structure of the first serialized message and the corresponding relationship between the first structure and the second structure of the second serialized message, using the serialized description item The information in the data structure description item and the information in the field description item set the corresponding relationship in the generated serialized data structure.

3. The method according to claim 2, wherein the first format is XML format, the second format is PB format, or the first format is PB format, and the second format is XML format. ,

The use of the information in the data structure description item and the information in the field description item of the serialization description item to set the corresponding relationship in the generated serialization data structure includes:

A structure name and a field description item are set in the data structure description item. The structure name is the name of the root node in the XML serialized message with the first structure. The field description item includes data type, sequence number, node type and name, The sequence number is the sequence number of the data in the field description item in the PB serialization message with the second structure;

The corresponding relationship is set in the generated serialized data structure through the structure name in the data structure description item and the sequence number and name in the field description item.

4. The method according to claim 2, wherein a sub-data structure description item is set in the data structure description item, a sub-field description item is set in the sub-data structure description item, and the corresponding relationship includes a sub-data structure description item located in the same data structure. The relationship between the sub-data structure description items and field description items in the structure description item;

When the first format is XML format and the second format is PB format, or when the first format is PB format and the second format is XML format,

A structure name, a field description item and a sub-data structure description item are set in the data structure description item. The structure name is the name of the root node in the XML serialization message with the first structure. The sub-data structure description item includes a sequence number. , substructure name and subfield description item, the field description item or the subfield description item includes data type, serial number, node type and name, the serial number is the data in the field description item or subfield description item The sequence number in the PB serialized message with the second structure; through the structure name in the data structure description item, the substructure name in the sub-data structure description item, the field description item and the sub-field description The serial number and name in the item set the corresponding relationship in the generated serialized data structure.

5. The method according to claim 4, wherein: when the first format is XML format, and when the second format is PB format, the pair of serialization to be converted in the first format that conforms to the first structure message, parse the serialized message to be converted according to the serialized data structure, write the parsed data into the serialized data structure, and obtain the serialized data structure after the data is written, including:

The XML element in the XML serialized message to be converted is read according to the serialized data structure. When the name in the read XML element is the same as the structure name in the data structure description item, it is determined that the XML element corresponds to The XML node is the root node; Read each XML element that is subordinate or indirectly subordinate to the root node. When the name in the XML element is the same as the name in the field description item and the type of the XML element is the same as the node type in the field description item, it will be parsed The data of the XML element is written in the field description item; when the name in the XML element is the same as the substructure name in the subdata structure description item, the XML element is used as a child root node, Read each XML element that is subordinate or indirectly subordinate to the sub-root node, when the name of the XML element that is subordinate or indirectly subordinate to the sub-root node is the same as the name in the sub-field description item and the type of the XML element is the same as the sub-field description item. When the node types in are the same, write the parsed XML element data into the subfield description item; determine whether the read XML element data in the XML serialization message to be converted are all has been written to the corresponding field description item or subfield description item. If so, the serialized data structure after the data is written is obtained from the field description item or subfield description item after the data of the XML element is written. If No, write the unwritten XML element into the field description item or subfield description item, and the name and node type of the field description item or subfield description item are the same as the name and type of the unwritten XML element. .

6. The method according to claim 5, wherein, according to the corresponding relationship recorded in the serialized data structure, converting the extracted data into a serialized message in a second format with a second structure includes :

Determine the data type of the extracted data according to the corresponding relationship recorded in the serialized data structure;

When the extracted data is numerical type data, the following first method is used for processing: first write the sequence number and data type of the data recorded in the field description item or subfield description item into PB in variable mode into the data item corresponding to the same sequence number in the serialized message, and then write the numerical type data into the data item;

When the extracted data is string type data, the following second method is used for processing: first write the sequence number and data type of the data recorded in the field description item or subfield description item in a variable way In the data item corresponding to the same sequence number in the PB serialized message, the string length of the data recorded in the field description item or subfield description item is written into the data item, and then the string type is The data is written into the data item;

When the extracted data is a sub-data structure description item, first write the sequence number and structure type in the sub-data structure description item into the data item corresponding to the same sequence number in the PB serialization message in a variable manner, and then write the The data length in the sub-data structure description item is written into the data item, and then the data in the sub-field description item of the sub-data structure description item is extracted. When the extracted content is numerical type data, the first The first method is used for processing. When the extracted content is string type data, the second method is used for processing;

Determine whether the extracted data has been written into the corresponding data items in the PB serialization message. If so, obtain the serialized message in the second format with the second structure from each data item after the data is written. If not, according to For the data type of the unwritten data, the first method or the second method is used to write the unwritten data into the corresponding data item.

7. The method according to claim 4, wherein: when the first format is PB format, and when the second format is XML format, the serialized message to be converted is processed according to the serialized data structure. Parse and write the parsed data into the serialized data structure. The serialized data structure obtained after the data is written includes:

Read the variable values in the binary data stream of the PB serialized message to be converted one by one;

Obtain the data type and serial number corresponding to the variable value from the read variable value;

When the read variable value is a numeric type, the following third method is used for processing: continue to read the next variable value, and write the next variable value into the serialized data structure between the serial number and the variable value. Among field description items or subfield description items with the same serial number,

When the read variable value is a string type, the following fourth method is used for processing: continue to read the next variable value to obtain the string length, and then read bytes equal to the string length to obtain the string , write the string into the field description item or subfield description item whose serial number is the same as the serial number of the string in the serialized data structure;

When the sequence number of the read variable value is the same as the sequence number of the sub-data structure description item, continue to read the next variable value to obtain the data length, and then read bytes of the data length size to obtain the bytecode set, and, Read the variable values in the bytecode set. When the read variable value is of numeric type, the third method is used for processing. When the read variable value is of string type, the above method is used. The fourth method is processed;

Determine whether all variable values in the PB serialized message to be converted have been written into field description items or subfield description items. If so, obtain data from the field description items or subfield description items after writing variable values. The serialized data structure after writing, if not, uses the third or fourth method to write the unwritten variable value into the field description item or subfield according to the data type of the unwritten variable value. in the description item.

8. The method according to claim 7, wherein, according to the corresponding relationship recorded in the serialized data structure, converting the extracted data into a serialized message in a second format with a second structure includes : According to the corresponding relationship recorded in the serialized data structure, determine whether the extracted data is a structure name and determine the node type of the extracted data;

When the extracted data includes a structure name, create the root node in the XML serialized message using the data and the structure name; when the extracted data is a node element, use the following fifth method for processing: At the root Create an XML node element under the node with the name in the field description item or subfield description item of the node element, and write the data in the field description item or subfield description item into the XML node element;

When the extracted data is a node attribute, the following sixth method is used for processing: Create an XML node attribute under the current node with the name in the field description item or subfield description item of the node attribute, and change the field description item Or the data in the subfield description item is written into the XML node attribute;

When the extracted data is a sub-data structure description item, a sub-root node is created with the name in the sub-data structure description item, and the data in the sub-data structure description item is extracted. When the extracted data is a node element, in The fifth method is used for processing under the child root node. When the extracted data is a node attribute, the sixth method is used for processing;

Determine whether the extracted data has been written into the corresponding XML node element or XML node attribute. If so, obtain a serialized message in the second format with the second structure from the XML node element or XML node attribute after the data is written. , if not, according to the node type of the unwritten data, use the above fifth method or the sixth method to write the unwritten data into the corresponding XML node element or XML node attribute.

9. A conversion device between serialized messages, including:

A data structure generation unit configured to generate serialized data using the set serialization description item according to the first structure of the first serialized message and the corresponding relationship between the first structure and the second structure of the second serialized message. structure; wherein, the first serialized message has a first format, and the second serialized message has a second format; the serialized description item includes a data structure description item and a field description item set in the data structure description item, so The structure of each data structure description item and the structure of each field description item in the data structure description item are consistent with the first structure;

A parsing and writing unit, configured to parse the serialized message to be converted in the first format that conforms to the first structure, parse the serialized message to be converted according to the serialized data structure, and write the parsed data into In the serialized data structure, the serialized data structure after the data is written is obtained;

A data extraction unit, used to extract the data in the serialized data structure after the data is written;

A conversion unit configured to convert the extracted data into a serialized message in a second format with a second structure according to the corresponding relationship recorded in the serialized data structure.

10. The device according to claim 9, wherein the data structure generating unit includes a correspondence relationship setting unit configured to determine the relationship between the first structure and the second serialized message according to the first structure of the first serialized message. The corresponding relationship between the second structures is set in the generated serialized data structure by using the information in the data structure description item and the information in the field description item of the serialization description item.

1 1. The device according to claim 10, wherein: the first format is XML format, the second format is PB format, or, the first format is PB format, and the second format is XML Format,

The correspondence relationship setting unit is used to set a structure name and a field description item in the data structure description item. The structure name is the name of the root node in the XML serialization message with the first structure. In the field description item Including data type, sequence number, node type and name, the sequence number is the sequence number of the data in the field description item in the PB serialization message with the second structure; through the structure name in the data structure description item, the The serial number and name in the field description item set the corresponding relationship in the generated serialized data structure.

12. The device according to claim 10, wherein the data structure generating unit includes a substructure setting unit for setting a subdata structure description item in the data structure description item, wherein the subdata structure description The item is provided with a subfield description item, and the corresponding relationship includes a relationship between a subdata structure description item and a field description item located in the same data structure description item;

The correspondence relationship setting unit is used to set a structure name, a field description item and a sub-data structure description item in the data structure description item, where the structure name is the name of the root node in the XML serialization message with the first structure, so The sub-data structure description item includes a serial number, a sub-structure name and a sub-field description item, the field description item or the sub-field description item includes a data type, a serial number, a node type and a name, and the serial number is the field description The item or subfield describes the sequence number of the data in the item in the PB serialized message with the second structure; the structure name in the data structure description item, the substructure in the subdata structure description item ized data structure.

13. The device according to claim 12, wherein the first format is XML format, and when the second format is PB format,

The parsing and writing unit is used to read the XML elements in the XML serialized message to be converted according to the serialized data structure. When the name in the read XML element is the same as the structure name in the data structure description item, When the same, determine that the XML node corresponding to the XML element is the root node;

Read each XML element that is subordinate or indirectly subordinate to the root node. When the name in the XML element is the same as the name in the field description item and the type of the XML element is the same as the node type in the field description item, it will be parsed The data of the XML element is written in the field description item; when the name in the XML element is the same as the substructure name in the subdata structure description item, the XML element is used as a child root node, Read each XML element that is subordinate or indirectly subordinate to the sub-root node, when the name of the XML element that is subordinate or indirectly subordinate to the sub-root node is the same as the name in the sub-field description item and the type of the XML element is the same as the sub-field description item. When the node types in are the same, write the parsed XML element data into the subfield description item; determine whether the read XML element data in the XML serialization message to be converted are all has been written to the corresponding field description item or subfield description item. If so, the serialized data structure after the data is written is obtained from the field description item or subfield description item after the data of the XML element is written. If No, write the unwritten XML element into the field description item or subfield description item, and the name and node type of the field description item or subfield description item are the same as the name and type of the unwritten XML element. .

14. The device according to claim 13, wherein the conversion unit is configured to determine the data type of the extracted data according to the corresponding relationship recorded in the serialized data structure; when the extracted data is When numeric type data is used, the following first method is used for processing: First, the sequence number and data type of the data recorded in the field description item or subfield description item are written into the same sequence number in the PB serialized message in variable mode. In the corresponding data item, the numerical type data is then written into the data item;

When the extracted data is a sub-data structure description item, first write the sequence number and structure type in the sub-data structure description item into the data item corresponding to the same sequence number in the PB serialization message in a variable manner, and then write the The data length in the sub-data structure description item is written into the data item, and then the data in the sub-field description item of the sub-data structure description item is extracted. When the extracted data is numerical type data, the first Process in one way, when the extracted data is string type data, use the second way to process; Determine whether the extracted data has been written into the corresponding data items in the PB serialization message, if so, A serialized message in the second format with the second structure is obtained from each data item after the data is written. If not, according to the data type of the unwritten data, the first method or the second method is used to obtain the serialized message in the second format. Unwritten data is written to the corresponding data item.

15. The device according to claim 12, wherein the first format is PB format, and when the second format is XML format, the parsing and writing unit is used to read PB serialized messages to be converted one by one. variable values in the binary data stream;

16. The device according to claim 15, wherein the conversion unit is configured to determine whether the extracted data is a structure name and determine whether the extracted data is a structure name according to the corresponding relationship recorded in the serialized data structure. The node type; when the extracted data includes a structure name, create the root node in the XML serialization message with the data and structure name;

When the extracted data is a node element, the following fifth method is used for processing: Create an XML node element under the root node with the name in the field description item or subfield description item of the node element, and add the field The data in the description item or subfield description item is written into the XML node element;