CN118916031B - Multi-terminal code mapping translation method for Internet medical platform - Google Patents
Multi-terminal code mapping translation method for Internet medical platform Download PDFInfo
- Publication number
- CN118916031B CN118916031B CN202411402188.2A CN202411402188A CN118916031B CN 118916031 B CN118916031 B CN 118916031B CN 202411402188 A CN202411402188 A CN 202411402188A CN 118916031 B CN118916031 B CN 118916031B
- Authority
- CN
- China
- Prior art keywords
- code
- file
- node
- sequence
- key code
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
- 238000000034 method Methods 0.000 title claims abstract description 79
- 238000013507 mapping Methods 0.000 title claims abstract description 53
- 238000013519 translation Methods 0.000 title claims abstract description 47
- 238000004458 analytical method Methods 0.000 claims abstract description 20
- 239000012634 fragment Substances 0.000 claims description 20
- 230000006870 function Effects 0.000 claims description 20
- 238000004364 calculation method Methods 0.000 claims description 14
- 230000009471 action Effects 0.000 claims description 13
- 238000007781 pre-processing Methods 0.000 claims description 10
- 230000008569 process Effects 0.000 claims description 10
- 238000012546 transfer Methods 0.000 claims description 8
- 238000013467 fragmentation Methods 0.000 claims description 7
- 238000006062 fragmentation reaction Methods 0.000 claims description 7
- 238000000605 extraction Methods 0.000 claims description 6
- 238000004519 manufacturing process Methods 0.000 claims description 5
- 238000010276 construction Methods 0.000 claims description 4
- 238000009826 distribution Methods 0.000 claims description 3
- 238000012217 deletion Methods 0.000 claims 2
- 230000037430 deletion Effects 0.000 claims 2
- 238000003780 insertion Methods 0.000 claims 2
- 230000037431 insertion Effects 0.000 claims 2
- 206010048669 Terminal state Diseases 0.000 claims 1
- 230000001105 regulatory effect Effects 0.000 claims 1
- 238000006243 chemical reaction Methods 0.000 description 5
- 238000012163 sequencing technique Methods 0.000 description 4
- 238000012545 processing Methods 0.000 description 3
- 230000007704 transition Effects 0.000 description 3
- 206010063385 Intellectualisation Diseases 0.000 description 2
- 244000182264 Lucuma nervosa Species 0.000 description 2
- 238000004590 computer program Methods 0.000 description 2
- 238000011161 development Methods 0.000 description 2
- 238000010586 diagram Methods 0.000 description 2
- 238000012423 maintenance Methods 0.000 description 2
- 230000007246 mechanism Effects 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 238000005457 optimization Methods 0.000 description 2
- 230000004044 response Effects 0.000 description 2
- 230000006978 adaptation Effects 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000005540 biological transmission Effects 0.000 description 1
- 238000004422 calculation algorithm Methods 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 239000012535 impurity Substances 0.000 description 1
- 238000003672 processing method Methods 0.000 description 1
- 238000013179 statistical model Methods 0.000 description 1
- 230000001360 synchronised effect Effects 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F8/00—Arrangements for software engineering
- G06F8/40—Transformation of program code
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/10—File systems; File servers
- G06F16/16—File or folder operations, e.g. details of user interfaces specifically adapted to file systems
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F40/00—Handling natural language data
- G06F40/20—Natural language analysis
- G06F40/253—Grammatical analysis; Style critique
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F8/00—Arrangements for software engineering
- G06F8/70—Software maintenance or management
- G06F8/73—Program documentation
-
- G—PHYSICS
- G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
- G16H—HEALTHCARE INFORMATICS, i.e. INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR THE HANDLING OR PROCESSING OF MEDICAL OR HEALTHCARE DATA
- G16H40/00—ICT specially adapted for the management or administration of healthcare resources or facilities; ICT specially adapted for the management or operation of medical equipment or devices
- G16H40/40—ICT specially adapted for the management or administration of healthcare resources or facilities; ICT specially adapted for the management or operation of medical equipment or devices for the management of medical equipment or devices, e.g. scheduling maintenance or upgrades
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- General Engineering & Computer Science (AREA)
- Health & Medical Sciences (AREA)
- General Physics & Mathematics (AREA)
- Physics & Mathematics (AREA)
- Software Systems (AREA)
- General Business, Economics & Management (AREA)
- General Health & Medical Sciences (AREA)
- Biomedical Technology (AREA)
- Business, Economics & Management (AREA)
- Data Mining & Analysis (AREA)
- Databases & Information Systems (AREA)
- Human Computer Interaction (AREA)
- Computational Linguistics (AREA)
- Audiology, Speech & Language Pathology (AREA)
- Epidemiology (AREA)
- Medical Informatics (AREA)
- Primary Health Care (AREA)
- Public Health (AREA)
- Library & Information Science (AREA)
- Artificial Intelligence (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Abstract
本发明属于终端代码技术领域，本发明公开了一种用于互联网医疗平台的多终端代码映射转译方法；包括：获取医疗平台待转译的源代码文件，源代码文件包括n段代码片段，对源代码文件进行词法分析，并生成得到抽象语法树；遍历抽象语法树，提取关键代码片段，计算每个关键代码片段的信息熵值和上下文相关性权重；将每个关键代码片段按照对应的信息熵值和上下文相关性权重进行排序，得到有序代码片段序列；将有序代码片段序列进行映射转译，生成目标代码文件；将目标代码文件分发至不同的终端医疗设备，不同的终端医疗设备对运行目标代码文件进行反向映射并运行，极大地降低了人工干预和编码的工作量，提高了转译效率和准确性。
The present invention belongs to the technical field of terminal codes, and discloses a multi-terminal code mapping and translation method for an Internet medical platform; the method comprises: obtaining a source code file to be translated of a medical platform, the source code file comprising n code segments, performing lexical analysis on the source code file, and generating an abstract syntax tree; traversing the abstract syntax tree, extracting key code segments, and calculating the information entropy value and context relevance weight of each key code segment; sorting each key code segment according to the corresponding information entropy value and context relevance weight to obtain an ordered code segment sequence; mapping and translating the ordered code segment sequence to generate a target code file; distributing the target code file to different terminal medical devices, and the different terminal medical devices reversely mapping and running the running target code file, thereby greatly reducing the workload of manual intervention and coding, and improving the translation efficiency and accuracy.
Description
Technical Field
The invention relates to the technical field of terminal codes, in particular to a multi-terminal code mapping translation method for an Internet medical platform.
Background
The patent with the application publication number of CN104360850A discloses a service code processing method and a device, which comprise the steps of receiving a request for acquiring a code description sent by an external application program, requesting associated information containing the code description, judging whether the code type is loaded in a pre-generated code translation mapping relation table, acquiring the code description corresponding to the code value from the code translation mapping relation table if the code type is loaded in the pre-generated code translation mapping relation table, and sending the code description corresponding to the code value acquired from the code translation mapping relation table to the external application program, so that the overhead and redundancy of an application program memory can be reduced, and efficient and convenient translation of the service code and synchronous sharing of the code translation mapping relation are realized.
However, in the internet medical field, because of the huge difference between the hardware architecture and the operating system of the terminal medical equipment, the traditional code translation method often needs to perform a great amount of manual adjustment for each platform system, has low efficiency and is easy to introduce errors, the control codes of the traditional code translation method have uneven running efficiency on different equipment, which seriously affects the accuracy and the safety, on the other hand, the traditional code translation method often has difficulty in accurately evaluating the complexity context correlation of the codes, so that the problem of low efficiency or semantic errors may exist in the translation result, and in addition, the traditional code translation process lacks automation and intelligence, needs a great amount of manual intervention and coding, has low efficiency and is easy to cause human errors.
In view of the above, the present invention proposes a multi-terminal code mapping translation method for an internet medical platform to solve the above-mentioned problems.
Disclosure of Invention
In order to overcome the defects in the prior art, the invention provides a multi-terminal code mapping translation method for an Internet medical platform, which comprises the following steps of S1, acquiring a source code file to be translated of the medical platform, wherein the source code file comprises n sections of code fragments, performing lexical analysis on the source code file, and generating an abstract syntax tree;
S2, traversing an abstract syntax tree, extracting key code segments, and calculating an information entropy value and a context correlation weight of each key code segment;
s3, ordering each key code segment according to the corresponding information entropy value and the context correlation weight to obtain an ordered code segment sequence;
s4, mapping and translating the ordered code segment sequences to generate an object code file;
and S5, distributing the object code file to different terminal medical equipment, and reversely mapping and operating the operation object code file by the different terminal medical equipment.
Further, the generating the abstract syntax tree comprises:
The method comprises the steps of converting a source code file into a character stream, extracting lexical units from the character stream, generating a lexical unit sequence, carrying out grammar analysis on the lexical unit sequence, constructing a grammar tree, preprocessing the grammar tree, and reorganizing the hierarchical structure of nodes in the grammar tree to obtain an abstract grammar tree.
Further, the method for generating the lexical unit sequence comprises the following steps:
initializing an empty sequence shell, reading characters in a character stream one by one, and recording the current read position in real time by using a pointer;
Defining lexical unit types, constructing a corresponding prefix tree for each lexical unit type, sequentially trying to match the prefix tree corresponding to each lexical unit type from the current read position, and extracting the corresponding lexical unit if the prefix tree is successfully matched with the current character;
Determining the type and the value of the lexical unit according to the corresponding prefix tree, recording the position information of the lexical unit in the source code file, constructing a lexical unit example based on the type, the value and the position information of the lexical unit, adding the lexical unit example into a sequence shell, moving a pointer to the end position of the lexical unit extracted at the time, repeating until the character stream is read, and obtaining the final sequence shell as the lexical unit sequence.
Further, the method for converting the source code file into the character stream comprises the following steps:
an initial memory buffer area is initially defined and has a size of Opening a source code file by using file I/O operation, obtaining a file stream object, reading the file stream object byte by byte to obtain m file blocks, sequentially storing the m file blocks in a memory buffer, and dynamically adjusting the size of the memory bufferTo the point of;
Wherein, the method comprises the steps of,For the total size of m file blocks,For the size of the file block that has been read,For the total memory size to be the same,As for the memory that is currently already in use,For a preset upper limit of memory buffer size,AndIn order to adjust the factor(s),AndIs an exponential factor; in order to adjust the function of the function, For the purpose of I/O throughput,For the purpose of CPU utilization,The memory fragmentation degree;
Adjusting function Wherein, the method comprises the steps of,For the throughput adjustment factor to be used,In order to adjust the coefficient of utilization,The memory fragmentation adjustment coefficient is used;
Preprocessing the file blocks in the memory buffer area to obtain preprocessed file blocks, and converting the preprocessed file blocks into character streams, namely reading characters in the preprocessed file blocks one by one and forming a character sequence, namely the character streams.
Further, the method for preprocessing the file blocks in the memory buffer area includes:
Defining a five-tuple Wherein, the method comprises the steps of, wherein,As a set of states,In order to input the alphabet list of the letters,As a function of the state transition(s),Is in an initial state, and;Is a group of termination states, and;
State transfer functionWherein, the method comprises the steps of, wherein,In the event of a current state,In order to input the character(s),Is in a new state;
defining a state set based on m file blocks And inputting an alphabet;
Reading characters from file blocks of a memory buffer one by oneAccording to the current stateAnd the character readNew state is calculated by state transfer functionIf the state is newBelonging toThen execute corresponding preprocessing operation to update the current state toRepeating until all characters in the file block of the memory buffer are read, and obtaining the preprocessed file block.
Further, the method for analyzing the lexical unit sequence comprises the following steps:
Defining grammar rules, namely analyzing grammar of a programming language, initializing a grammar tree based on the defined grammar rules, creating a root node, extracting lexical units from a lexical unit sequence, matching the lexical units with the grammar rules, creating a new node serving as a left sub-node of the root node, matching operators by using the first lexical unit as a sub-node of the newly created left sub-node, creating a new node serving as a right sub-node of the root node, continuing to match according to a programming generation formula, creating a new node serving as a right sub-node te of the right sub-node of the root node, matching a second lexical unit serving as a right sub-node of the newly created te, matching operators, creating a new node serving as a left sub-node tw of the right sub-node te, continuing to match a third lexical unit according to the programming generation formula, creating a new node serving as a right sub-node of tw, repeating until the extraction of the lexical units in the lexical unit sequence is finished, and completing the construction of the grammar tree.
Further, the method for reorganizing the hierarchical structure of the nodes in the syntax tree includes:
Defining a tree reconstruction rule as a group of pattern matching rules, wherein each pattern matching rule consists of a matching pattern and a reconstruction action;
The matching modes have m2 types, and define the structure of the subtree to be reconstructed, wherein the reconstruction actions comprise node lifting, node merging and node splitting;
Matching rules for each pattern Assigning a priority weightFor any two pattern matching rulesAndThey are respectively matched with grammar treeSubtrees of (3)AndIf there is a nesting relationship between the matched patterns, calculating the nesting value between them;
If it isThen reconstructIf (1)Then reconstructIf (1)Then compareAndThe weights of the two are reconstructed firstly with higher weight;
;
Wherein, 、AndIs the adjustment coefficient of the light source,Is thatAndIs used to determine the most recent common ancestor node of (c),Is thatIs provided with a root node of (c),Is thatAndIs provided for the difference in size of the (c) in the (c),Is thatAndIs a degree of overlap of (2); Is a grammar tree In (a)Is provided with a depth of (a),Is a grammar treeIn (a)Is a depth of (2);
The reconstruction process comprises the following steps:
Constructing a priority queue Q, initially placing all pattern matching rules into the priority queue Q according to descending priority, and when traversing to a node n', extracting the pattern matching rule corresponding to the highest priority from the Q Matching the matching modes of the subtrees taking n' as the root, if the matching is successful, executing the reconstruction action to obtain the reconstructed subtrees, and adjusting other mode matching rules in Q according to the reconstruction action;
If the matching fails, continuing to match the pattern matching rule corresponding to the next priority level until all pattern matching rules are traversed, and replacing the original grammar tree with all reconstructed subtrees to finish the generation of the abstract grammar tree.
Further, the calculation mode of the information entropy value and the context correlation weight of each key code segment comprises the following steps:
representing each key code segment as a corresponding sequence of tokens, wherein each token is a key, identifier or literal quantity;
Counting each unique token In key code segmentFrequency of occurrence in (a)Frequency-basedCalculating to obtain key code fragmentsInformation entropy value of (2);
,
Wherein, Is thatIs used to determine the complexity of the code segment of (c),Is thatIs a data stream complexity of (1); And Adjusting coefficients for the information;
wherein, the method comprises the steps of, Is thatIs used to determine the degree of complexity of the cycle,Is thatIs used for the branching complexity of the (c) signal,Is thatIs a length of (2);、 And Adjusting parameters for the weights;
Loop complexity refers to the sum of the layers of nested loops in key code segments
The impurity degree refers to the sum of the number of conditional branch sentences in the code segment;
wherein, the method comprises the steps of, Is thatIs used to determine the scope complexity of the (c) system,Is thatIs used to determine the degree of complexity of the dependency of (1),Is thatThe scope complexity refers to the sum of nesting layers of the scope of the variable in the key code segment;
wherein, the method comprises the steps of, Is the firstThe weight coefficient of the individual variable is determined,Is the firstThe depth of dependence of the individual variables,To rely on adjustment parameters; Is the first Complexity of individual control dependencies;
fragmenting key code And its context is converted into a vector representation and the cosine similarity between them is calculated as semantic similarity;
Computing key code snippetsStructural similarity mean with other key code segmentsComputing statistical similarity using an n-gram model;
Based on semantic similaritySimilarity of structuresAnd statistical similarityCalculating to obtain key code fragmentsContext correlation weight of (c),
Wherein, the method comprises the steps of,Is a first adjustment coefficient; Is the second adjustment coefficient.
Further, the calculating method of the structural similarity mean value includes:
Traversing the abstract syntax tree, and extracting the subtrees corresponding to each key code segment according to the position information of the extracted key code segment.
The definition tree editing operation comprises node renaming, node deleting and node inserting, and a cost value is respectively assigned to the node renaming, the node deleting and the node inserting;
for the following Corresponding subtreeAnd any sub-treeThe calculation willConversion toMinimum editing cost, i.e. tree editing distance;
The calculation mode of the minimum editing cost comprises the following steps:
initializing a two-dimensional table, the rows and columns respectively corresponding to AndFrom bottom to top, calculates the value of each two-dimensional table cell, representing the node to beIs converted into (a) subtrees ofMinimum edit cost of the subtree of (a);
Then Wherein, the method comprises the steps of,For the number of all sub-trees,AndRespectively representAndIs the number of nodes;
the acquisition mode of the ordered code segment sequence comprises the following steps:
The information entropy value and the context correlation weight of each key code segment are weighted and summed to obtain a comprehensive grading value of each key code segment;
the acquisition mode of the target code file comprises the following steps:
Traversing the ordered code segment sequence, mapping each key code segment to an equivalent code segment of the target language, and fusing the mapped target code segments according to the context information of the key code segments in the source code file to obtain a target code file;
The reverse mapping is performed by converting the object code file into machine code or intermediate code directly executed by the terminal medical equipment.
The multi-terminal code mapping and translating system for the Internet medical platform is used for realizing the multi-terminal code mapping and translating method for the Internet medical platform and comprises a source code acquisition and analysis module, a processing module and a processing module, wherein the source code acquisition and analysis module is used for acquiring a source code file to be translated of the medical platform, the source code file comprises n sections of code fragments, performing lexical analysis on the source code file and generating an abstract syntax tree;
The segment extraction analysis module is used for traversing the abstract syntax tree, extracting key code segments and calculating the information entropy value and the context correlation weight of each key code segment;
the segment sequencing module is used for sequencing each key code segment according to the corresponding information entropy value and the context correlation weight to obtain an ordered code segment sequence;
The code mapping translation module is used for mapping and translating the ordered code fragment sequences to generate an object code file;
And the distribution mapping module is used for distributing the object code file to different terminal medical equipment, and the different terminal medical equipment reversely maps and operates the operation object code file.
The invention relates to a multi-terminal code mapping translation method for an Internet medical platform, which has the technical effects and advantages that:
The invention improves the quality and efficiency of code translation through deep lexical analysis and abstract grammar tree generation, different terminal medical equipment can execute the translated and optimized target code with high efficiency, improves the performance and reliability of a medical informatization system, introduces an information entropy value and a context correlation weight calculation mechanism, generates a better and efficient target code, is beneficial to improving the running speed of the system, and ensures the semantic correctness of the code in various heterogeneous environments, thereby reducing potential errors and potential safety hazards, realizing flexible code conversion and optimization, remarkably reducing the development and maintenance cost, realizing the automation and intellectualization of the code translation process, greatly reducing the workload of manual intervention and coding, improving the translation efficiency and accuracy, and effectively improving the response speed and user experience of a medical platform.
Drawings
FIG. 1 is a diagram of a multi-terminal code mapping translation method for an Internet medical platform according to the present invention;
Fig. 2 is a schematic diagram of a multi-terminal code mapping translation system for an internet medical platform according to the present invention.
Detailed Description
The following description of the embodiments of the present invention will be made clearly and completely with reference to the accompanying drawings, in which it is apparent that the embodiments described are only some embodiments of the present invention, but not all embodiments. All other embodiments, which can be made by those skilled in the art based on the embodiments of the invention without making any inventive effort, are intended to be within the scope of the invention.
Example 1
Referring to fig. 1, a multi-terminal code mapping translation method for an internet medical platform according to the present embodiment includes:
s1, acquiring a source code file to be translated of a medical platform, wherein the source code file comprises n sections of code fragments, performing lexical analysis on the source code file, and generating an abstract syntax tree;
S2, traversing an abstract syntax tree, extracting key code segments, and calculating an information entropy value and a context correlation weight of each key code segment;
s3, ordering each key code segment according to the corresponding information entropy value and the context correlation weight to obtain an ordered code segment sequence;
s4, mapping and translating the ordered code segment sequences to generate an object code file;
s5, distributing the object code file to different terminal medical equipment, and reversely mapping and operating the operation object code file by the different terminal medical equipment, so that cross-platform code translation is realized, and the operation efficiency of the code in a heterogeneous environment is improved.
Further, the generating the abstract syntax tree includes:
The source code file is converted into a character stream (sequence of characters organized in sequence), lexical units are extracted from the character stream, and a sequence of lexical units is generated.
Specifically, an empty sequence shell is initialized, characters in the character stream are read one by one, and a pointer is used for recording the current read position in real time.
The method comprises the steps of defining lexical unit types (such as keywords, identifiers, numerical values and the like), constructing a corresponding prefix tree (the character string set is organized and stored according to the prefix paths of the character strings) for each lexical unit type, and storing all possible character strings (such as keyword lists) contained in each lexical unit type in one prefix tree.
And sequentially trying to match the prefix tree corresponding to each lexical unit type from the current read position, and extracting the corresponding lexical unit if the prefix tree is successfully matched with the current character.
Determining the type and the value of the lexical unit according to the corresponding prefix tree, recording the position information (such as line number, column number and the like) of the lexical unit in the source code file, constructing a lexical unit example based on the type, the value and the position information of the lexical unit, adding the lexical unit example into a sequence shell, moving a pointer to the end position of the lexical unit extracted at the time, repeating until the character stream is read, and obtaining the final sequence shell as the lexical unit sequence.
The method for converting the source code file into the character stream comprises the following steps:
an initial memory buffer area is initially defined and has a size of (The initial memory buffer size may be set to a smaller value, e.g., 4KB or 8 KB), opening the source code file using a file I/O operation, obtaining a resulting file stream object, reading the file stream object byte by byte, obtaining m file blocks (the multiple small blocks of data into which the source code file is divided), sequentially storing the m file blocks in the memory buffer, and dynamically adjusting the size of the memory bufferTo the point of。
Wherein, the method comprises the steps of,For the total size of m file blocks,For the size of the file block that has been read,For the total memory size (total memory of the system),As for the memory that is currently already in use,For a preset upper limit of memory buffer size,AndFor adjusting the factor, for controlling the proportional relation between the new buffer size and the file remaining size and the current available memory,AndThe index factor is used for adjusting the influence degree of the residual size of the file and the current available memory on the size of the buffer area; in order to adjust the function of the function, For the purpose of I/O throughput,For the purpose of CPU utilization,Is the degree of memory fragmentation.
The memory fragmentation level is 1 minus the size of the largest consecutive idle block in the available memory divided by the total available memory size.
Adjusting functionWherein, the method comprises the steps of,For the throughput adjustment factor to be used,In order to adjust the coefficient of utilization,The memory fragmentation adjustment coefficient is used for controlling the influence degree of each factor on the size of the buffer area.
Preprocessing the file blocks in the memory buffer area to obtain preprocessed file blocks, and converting the preprocessed file blocks into character streams.
The method for preprocessing the file blocks in the memory buffer area comprises the following steps:
Defining a five-tuple Wherein, the method comprises the steps of, wherein,As a set of states,In order to input the alphabet list of the letters,As a function of the state transition(s),Is in an initial state, and;Is a group of termination states, and。
State transfer functionWherein, the method comprises the steps of, wherein,In the event of a current state,In order to input the character(s),Is in a new state.
Defining a state set based on m file blocksAnd inputting an alphabetSpecifically, initializing a state set to be a null set, initializing an input alphabet to be null, traversing m file blocks, and executing the following operations on each file block:
Scanning each character w in the file block, adding the character w to the input alphabet if the character w appears for the first time, adding the corresponding state of the character w to the state set according to the type of the character w (such as letters, numbers, special characters and the like), for example, if the character w is a letter, the character w may correspond to an identifier starting state, the character w may correspond to a numerical value starting state and the like, defining possible subsequent state transition of each new de state according to the lexical rule of the language, and constructing the state set and the input alphabet after traversing all the file blocks.
Reading characters from file blocks of a memory buffer one by oneAccording to the current stateAnd the character readNew state is calculated by state transfer functionIf the state is newBelonging toA corresponding preprocessing operation (delete comment, delete blank character, blank character including space, tab, and line feed) is performed.
Updating the current state toRepeating until all characters in the file block of the memory buffer are read, and obtaining the preprocessed file block.
And converting the preprocessed file blocks into character streams, namely reading characters in the preprocessed file blocks one by one, and forming a character sequence, namely the character streams.
Carrying out grammar analysis on the lexical unit sequence, constructing a grammar tree, and representing a grammar structure of a source code;
Specifically, grammar rules are defined, i.e. the grammar of the programming language (terminators, non-terminators and authoring production formulas) is analyzed.
Initializing a grammar tree based on a defined grammar rule, creating a root node, extracting a lexical unit from a lexical unit sequence, matching with the grammar rule, creating a new node as a left sub-node of the root node, using the matched first lexical unit as a sub-node of the newly created left sub-node, matching operators (terminators, non-terminators and other operators), creating a new node as a right sub-node of the root node, continuing matching according to a writing production formula, creating a new node as a right sub-node te of the right sub-node of the root node, matching a second lexical unit as a right sub-node of the newly created te, matching operators, creating a new node as a left sub-node tw of the right sub-node te, continuing matching a third lexical unit according to the writing production formula, creating a new node as a right sub-node of tw, repeating until the extraction of the lexical unit in the sequence is finished, and completing the construction of the grammar tree.
In the construction process, each time a writing generating formula is matched, a corresponding node is created in the grammar tree, the child node is correctly connected with the father node, a terminal (lexical unit) is used as a leaf node, and a non-terminal is used as an internal node.
Meanwhile, it should be noted that if any of the writing formulas cannot be matched, a grammar error is represented, and for ambiguous grammars, multiple grammar trees may exist, and ambiguity resolution is required according to the priority and the combinability rules of the language.
The method comprises the steps of generating an abstract syntax tree based on the syntax tree, specifically, preprocessing the syntax tree, namely traversing each node in the syntax tree, and removing nonsensical nodes such as notes, blank characters and the like.
And specifically, defining a tree reconstruction rule as a group of pattern matching rules for guiding adjustment of the node hierarchical structure of the abstract syntax tree, wherein each pattern matching rule consists of a matching pattern and a reconstruction action.
There are m2 types of matching patterns defining the structure of the subtree to be reconstructed, for example, matching pattern BinaryExpr (op, real) represents a binary operation expression, both operands of which are Literal quantities.
The reconstruction actions include node lifting, node merging and node splitting, matching rules for each patternAssigning a priority weightThe higher the priority, the earlier the reconstruction is made.
Specifically, a base weight is assigned to each pattern matching rule, which may range from 1 to 100. The basic weight reflects the basic importance of the rule in the reconstruction process, and the basic weight is adjusted in real time according to the complexity of the mode.
Rule matching for any two patternsAndThey are respectively matched with grammar treeSubtrees of (3)AndIf there is a nesting relationship between the matched patterns, calculating the nesting value between them;
If it isThen reconstructIf (1)Then reconstructIf (1)Then compareAndThe weights of the two are reconstructed earlier with higher weight.
Wherein, the method comprises the steps of,、AndIs an adjusting coefficient for controlling the influence degree of each factor on the nesting depth, the value ranges are 0,1,Is thatAndIs used to determine the most recent common ancestor node of (c),Is thatIs provided with a root node of (c),Is thatAndExpressed as the difference in node number or subtree height,Is thatAndIs expressed as a proportion of their number of overlapping nodes to the total number of nodes; Is a grammar tree In (a)Is (root node depth is 0),Is a grammar treeIn (a)Is a depth of (c).
The reconstruction process comprises the following steps:
constructing a priority queue Q, initially placing all pattern matching rules into the priority queue Q in descending order of priority (obtained based on nested values), and when traversing to a node n', taking out the pattern matching rule corresponding to the highest priority from the Q And matching the matching modes of the subtrees taking n' as the root, if the matching is successful, executing a reconstruction action to obtain a reconstructed subtree, and adjusting other mode matching rules in Q according to the reconstruction action (if the reconstruction action breaks the matching modes of other mode matching rules, the mode matching rules need to be put into a queue again).
If the matching fails, continuing to match the pattern matching rule corresponding to the next priority level until all pattern matching rules are traversed, and replacing the original grammar tree with all reconstructed subtrees to finish the generation of the abstract grammar tree.
Further, the method for extracting the key code segments comprises the following steps:
the key code snippets include the content of the code snippet (in the form of a string), the code snippet type (function definition, loop statement, conditional statement, etc.), the location information of the code snippet in the source code (file name, line number, column number range), and parent node information of the code snippet (for determining context).
Traversing each node of the abstract syntax tree using a depth-first traversal or breadth-first traversal algorithm, and for each node, identifying key parts in the code, such as function definitions, function calls, variable assignments, control flow statements, etc., according to predefined rules, the key parts typically corresponding to a particular node type or node combination pattern in the abstract syntax tree.
For the identified key code snippets, the corresponding source code text is extracted (the corresponding source code location and scope can be traced back from the grammar tree node, and the corresponding code text is extracted).
The calculation mode of the information entropy value and the context correlation weight of each key code segment comprises the following steps:
each key-code snippet is represented as a corresponding sequence of tokens, where each token is a key, identifier, or literal quantity.
Counting each unique tokenIn key code segmentFrequency of occurrence in (a)Frequency-basedCalculating to obtain key code fragmentsInformation entropy value of (2);
Wherein, the method comprises the steps of, wherein,Is thatIs used to determine the complexity of the code segment of (c),Is thatIs a data stream complexity of (1); And And the information adjustment coefficient is used for controlling the influence degree of the complexity of the code segment and the complexity of the data stream on the information entropy value.
Wherein, the method comprises the steps of,Is thatIs used to determine the degree of complexity of the cycle,Is thatIs used for the branching complexity of the (c) signal,Is thatLength (total number of tokens);、 And For weight adjustment parameters, the influence of variables on the complexity of the code segment is controlled.
The loop complexity refers to the sum of the layers of nested loops in the key code segment, and the branch complexity refers to the sum of the number of conditional branch sentences in the code segment.
Wherein, the method comprises the steps of,Is thatIs used to determine the scope complexity of the (c) system,Is thatIs used to determine the degree of complexity of the dependency of (1),Is thatNumber of medium variables.
It should be explained that, in the abstract syntax subtree, all identifier nodes serving as leaf nodes correspond to variables, specifically, traversing the abstract syntax subtree corresponding to the key code segment to find all the leaf nodes, where the leaf nodes are either constant values or identifiers (variable names), and for the leaf nodes are identifiers, obtaining the variable names of the identifiers, and counting to obtain the total number of the variable names.
Scope complexity refers to the sum of the nesting layers of variable scopes in a key code segment, and dependency complexity refers to the complexity of data dependencies between variables in a key code segment.
Wherein, the method comprises the steps of,Is the firstThe weight coefficient of each variable reflects the importance of the variable in the key code segment,Is the firstThe dependency depth of a variable, i.e. the number of other variables on which the variable depends,Controlling the influence degree of the dependency term on the dependency complexity for the dependency adjustment parameter; Is the first Complexity of individual control dependency, for each control dependency in a key code snippet (e.g., conditional statement, loop statement, etc.), its complexity is calculatedThe complexity of the control dependency is evaluated by weighting according to the number of nested layers and the number of branches.
Fragmenting key codeAnd its context is converted into a vector representation and the cosine similarity between them is calculated as semantic similarity。
Computing key code snippetsStructural similarity mean with other key code segmentsComputing statistical similarity using an n-gram model or other statistical model。
The calculation method of the structural similarity mean value comprises the following steps:
Traversing the abstract syntax tree, and extracting the subtrees corresponding to each key code segment according to the position information of the extracted key code segment.
The definition tree editing operation comprises node renaming, node deleting and node inserting, and a cost value is respectively assigned to the node renaming, the node deleting and the node inserting, wherein the cost value can be fixed or can be obtained by dynamic calculation according to factors such as node types, subtree sizes and the like.
For the followingCorresponding subtreeAnd any sub-treeThe calculation willConversion toMinimum editing cost, i.e. tree editing distance。
The calculation mode of the minimum editing cost comprises the following steps:
initializing a two-dimensional table, the rows and columns respectively corresponding to AndFrom bottom to top, calculates the value of each two-dimensional table cell, representing the node to beIs converted into (a) subtrees ofMinimum edit cost of the subtree of (a);
Then Wherein, the method comprises the steps of,For the number of all sub-trees,AndRespectively representAndIs a node number of (a) in the network.
Based on semantic similaritySimilarity of structuresAnd statistical similarityCalculating to obtain key code fragmentsContext correlation weight of (c);
Wherein, the method comprises the steps of,The first adjustment coefficient is used for controlling the influence degree of the semantic similarity on the context correlation weight; And the second adjustment coefficient is used for controlling the influence degree of the structural similarity and the statistical similarity on the context correlation weight.
By adjusting the first adjusting coefficient and the first adjusting coefficient, the influence degree of semantic similarity, structural similarity and statistical similarity on the context correlation weight is adjusted, the correlation between the code segment and the context at the semantic, structural and statistical feature level is considered, the correlation degree between the code segment and the context can be evaluated more comprehensively, and therefore a more accurate result is provided for the calculation of the context correlation weight.
The acquisition mode of the ordered code segment sequence comprises the following steps:
And carrying out weighted summation on the information entropy value and the context correlation weight of each key code segment to obtain a comprehensive grading value of each key code segment, and carrying out descending order arrangement on all the key code segments according to the size of the comprehensive grading value to obtain an ordered code segment sequence.
The acquisition mode of the target code file comprises the following steps:
Traversing the ordered code segment sequence, mapping each key code segment to an equivalent code segment of the target language, and fusing the mapped target code segments according to the context information (such as variable scope, function call relation and the like) of the key code segments in the source code file to obtain the target code file.
Specifically, for each key code segment, it is mapped into equivalent target language code segments (equivalent code segments) according to the grammar and API of the target language (for each medical device), and it is noted that this process needs to consider the differences between languages, such as grammar structure, keywords, data types, built-in functions, use equivalent alternative implementations for key code segments that cannot be mapped directly, or insert necessary auxiliary codes (such as type conversion, function wrapping, etc.).
The specific method for fusing the mapped target code segments comprises the following steps:
According to the scope of the variable, correctly declaring and referring to the variable, inserting function declaration, parameter transfer and return value processing codes according to the function call relation, generating codes such as class definition, method rewriting and the like according to the inheritance relation of the class, and organizing code fragments according to control flow structures such as conditional statements, circulating statements and the like.
Further, the reverse mapping is performed by converting the object code file into a machine code or an intermediate code directly executed by the terminal medical device.
Different terminal medical devices can reversely map and execute the mapped object code files to realize cross-platform code translation and operation, and it should be noted that, because the hardware and software environments of different devices are greatly different, the reverse mapping and executing processes may need to be optimized and adjusted pertinently to ensure efficient operation of codes in various heterogeneous environments.
According to the embodiment, through deep lexical analysis and abstract grammar tree generation, the quality and efficiency of code translation are improved, different terminal medical devices can execute the translated and optimized target codes efficiently, the performance and reliability of a medical informatization system are improved, an introduced information entropy value and a context correlation weight computing mechanism generate more high-quality and efficient target codes, the running speed of the system is improved, the semantic correctness of the codes in various heterogeneous environments can be ensured, potential errors and potential safety hazards are reduced, flexible code conversion and optimization are realized, development and maintenance cost is remarkably reduced, automation and intellectualization of a code translation process are greatly reduced, the workload of manual intervention and coding is improved, the translation efficiency and accuracy are improved, and the response speed and user experience of a medical platform are effectively improved.
Example 2
Referring to fig. 2, the detailed description of the embodiment is not shown in the description of embodiment 1, and a multi-terminal code mapping translation system for an internet medical platform is provided, which includes:
the source code acquisition and analysis module is used for acquiring a source code file to be translated of the medical platform, wherein the source code file comprises n sections of code fragments, performing lexical analysis on the source code file and generating an abstract syntax tree;
The segment extraction analysis module is used for traversing the abstract syntax tree, extracting key code segments and calculating the information entropy value and the context correlation weight of each key code segment;
the segment sequencing module is used for sequencing each key code segment according to the corresponding information entropy value and the context correlation weight to obtain an ordered code segment sequence;
The code mapping translation module is used for mapping and translating the ordered code fragment sequences to generate an object code file;
The distribution mapping module is used for distributing the object code files to different terminal medical equipment, the different terminal medical equipment reversely maps and operates the operation object code files, and the modules are connected in a wired and/or wireless mode to realize data transmission among the modules.
Example 3
The embodiment discloses an electronic device, which comprises a memory, a processor and a computer program stored on the memory and capable of running on the processor, wherein the processor realizes the running mode of the multi-terminal code mapping translation method for the internet medical platform when executing the computer program.
Since the electronic device described in this embodiment is an electronic device for implementing a multi-terminal code mapping translation method for an internet medical platform according to the embodiment of the present application, based on the multi-terminal code mapping translation method for an internet medical platform described in the embodiment of the present application, those skilled in the art can understand the specific implementation of the electronic device and various modifications thereof, so how to implement the method in the embodiment of the present application in this electronic device will not be described in detail herein. It is within the scope of the present application to provide an electronic device for implementing a multi-terminal code mapping translation method for an internet medical platform according to the embodiments of the present application.
The above formulas are all formulas with dimensionality removed and numerical calculation, the formulas are formulas with the latest real situation obtained by software simulation through collecting a large amount of data, and preset parameters and threshold selection in the formulas are set by those skilled in the art according to the actual situation.
The above description is only a preferred embodiment of the present invention, and the protection scope of the present invention is not limited to the above examples, and all technical solutions belonging to the concept of the present invention belong to the protection scope of the present invention. It should be noted that modifications and adaptations to those skilled in the art without departing from the principles of the present invention are intended to be comprehended within the scope of the present invention.
Claims (8)
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202411402188.2A CN118916031B (en) | 2024-10-09 | 2024-10-09 | Multi-terminal code mapping translation method for Internet medical platform |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202411402188.2A CN118916031B (en) | 2024-10-09 | 2024-10-09 | Multi-terminal code mapping translation method for Internet medical platform |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN118916031A CN118916031A (en) | 2024-11-08 |
| CN118916031B true CN118916031B (en) | 2024-12-31 |
Family
ID=93309013
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202411402188.2A Active CN118916031B (en) | 2024-10-09 | 2024-10-09 | Multi-terminal code mapping translation method for Internet medical platform |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN118916031B (en) |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111708539A (en) * | 2020-06-17 | 2020-09-25 | 腾讯科技（深圳）有限公司 | Application program code conversion method and device, electronic equipment and storage medium |
| US11392373B1 (en) * | 2019-12-10 | 2022-07-19 | Cerner Innovation, Inc. | System and methods for code base transformations |
Family Cites Families (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP3430252B2 (en) * | 2000-01-24 | 2003-07-28 | 独立行政法人産業技術総合研究所 | Source code conversion method, recording medium recording source code conversion program, and source code conversion device |
| CN1516009A (en) * | 2003-01-08 | 2004-07-28 | 深圳市中兴通讯股份有限公司上海第二 | Efficient Optimization Method of Speech Codec Based on Digital Signal Processor |
| US10157055B2 (en) * | 2016-09-29 | 2018-12-18 | Microsoft Technology Licensing, Llc | Code refactoring mechanism for asynchronous code optimization using topological sorting |
| CN113791757B (en) * | 2021-07-14 | 2023-08-22 | 北京邮电大学 | Software requirement and code mapping method and system |
| CN115016793A (en) * | 2022-04-25 | 2022-09-06 | 中国平安人寿保险股份有限公司 | Code generation method and device based on syntax tree, electronic equipment and storage medium |
| CN118113264A (en) * | 2023-12-13 | 2024-05-31 | 天翼云科技有限公司 | A SQL code hint method based on keyword backtracking and token sorting |
-
2024
- 2024-10-09 CN CN202411402188.2A patent/CN118916031B/en active Active
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US11392373B1 (en) * | 2019-12-10 | 2022-07-19 | Cerner Innovation, Inc. | System and methods for code base transformations |
| CN111708539A (en) * | 2020-06-17 | 2020-09-25 | 腾讯科技（深圳）有限公司 | Application program code conversion method and device, electronic equipment and storage medium |
Also Published As
| Publication number | Publication date |
|---|---|
| CN118916031A (en) | 2024-11-08 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN108717470B (en) | A Code Snippet Recommendation Method with High Accuracy | |
| CN112035511A (en) | Target data searching method based on medical knowledge graph and related equipment | |
| CN111191002B (en) | Neural code searching method and device based on hierarchical embedding | |
| CN114924741B (en) | A code completion method based on structural features and sequence features | |
| CN1661593B (en) | Method for translating computer language and translation system | |
| CN111124487B (en) | Code clone detection method and device and electronic equipment | |
| CN115438709A (en) | Code Similarity Detection Method Based on Code Attribute Graph | |
| CN112306494A (en) | Code classification and clustering method based on convolution and cyclic neural network | |
| CN116719520B (en) | Code generation method and device | |
| CN112379917A (en) | Browser compatibility improving method, device, equipment and storage medium | |
| CN113918512A (en) | System and method for constructing knowledge map of power grid operation rules | |
| CN110597847A (en) | SQL statement automatic generation method, device, equipment and readable storage medium | |
| CN111831624A (en) | Data table creating method and device, computer equipment and storage medium | |
| CN116225933A (en) | Program code checking method and checking device | |
| CN118916031B (en) | Multi-terminal code mapping translation method for Internet medical platform | |
| Hu et al. | Deep-autocoder: Learning to complete code precisely with induced code tokens | |
| Xu et al. | Tree2tree structural language modeling for compiler fuzzing | |
| CN118484465A (en) | A method and device for generating SQL statements from natural language statements | |
| CN117828024A (en) | Plug-in retrieval method, device, storage medium and equipment | |
| CN117573084A (en) | Code complement method based on layer-by-layer fusion abstract syntax tree | |
| CN117390130A (en) | A code search method based on multimodal representation | |
| KR20050065015A (en) | System and method for checking program plagiarism | |
| CN114647418A (en) | Software code recommendation method for tree serialization embedding | |
| CN112988952A (en) | Multi-level-length text vector retrieval method and device and electronic equipment | |
| CN111797624A (en) | An automatic extraction method of drug business card based on NPL |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |