JP2018142126A - Naming device, naming method and naming program - Google Patents

Abstract

PROBLEM TO BE SOLVED: To support consistent naming in spite of variation in naming tendency among individual developers.SOLUTION: A naming device 2 includes: an element division unit 23 which extracts a plurality of elements prescribed by an element division rule 33 from a character string of input data 11 and generates a converted character string with respect to each of the extracted elements in accordance with the element division rule 33; and an element sequence determination unit 24 which acquires corresponding tag information from the element division rule 33 for each of elements extracted by the element division unit 23, rearranges converted character strings generated by the element division unit 23 in accordance with sequence information between pieces of tag information prescribed by a priority rule 34, and then connects the converted character strings to obtain a character string of output data 12.SELECTED DRAWING: Figure 1

Description

  The present invention relates to a name design apparatus, a name design method, and a technology of a name design program.

  An increasing number of businesses provide a processing system such as a map search on the web as a web service and execute the web service. A developer who performs programming requests an external Web service to execute a service of the company and obtains a processing result from the Web service. As a result, it is possible to take in a Web service provided by another business operator and construct and operate its own service.

  When requesting processing to a Web service, it is necessary to describe the processing content to be requested in accordance with an API (Application Programming Interface) published by the Web service. The API is a set of a function name (method name) of a program to be called, a variable name (parameter name) used in the function, a URL (Uniform Resource Locator) for specifying the position of the Web service in the network, and the like. . Therefore, in order to make it easy to call a Web service, it is necessary to keep in mind an easy-to-understand API name.

Many programming language name definitions use alphabetic characters such as English. However, it is difficult for Japanese developers who are unfamiliar with English to name alphabetic variable names.
Therefore, Non-Patent Document 1 provides a naming tool that, when a variable name written in Japanese is input, returns the result of translating the variable name into English. For example, if “customer name” is input to the naming tool, the result “customerName” is output. With this naming tool, the burden of designing the API name can be reduced.

codic project team, "Is the naming negative? VER.3", [online], [Search February 20, 2017], Internet <URL: https://codic.jp/>

  The programmer who designs the name of the API also has individual variations (habits) in the result of the name design due to differences in implementation experience and good language. For example, it is assumed that developer B takes over development of an existing Web service from developer A. This Web service uses a customer's telephone number as a parameter.

Developer A designed a parameter called “TelNumberOfCustomer” and used it as the customer's phone number. However, Developer B who succeeded decided to add a parameter called “UserTelNumber” and use it as the customer's phone number.
Here, “TelNumberOfCustomer” indicating the telephone number of the same customer and “UserTelNumber” are inconsistent (notation fluctuation). In other words, developer A has the habit of placing the element “phone number” in the order ahead of the element “customer”. On the other hand, the developer B has a habit of setting the element “telephone number” after the element “customer”.
Conventionally, it has been necessary to manually check for such inconsistencies in parameter names, and thus checking consistency when adding new parameters has been a heavy burden.

  Therefore, the main object of the present invention is to support consistent name design even when the naming tendency of individual developers varies.

In order to solve the above problems, the name design apparatus of the present invention has the following characteristics.
The present invention relates to an element extraction rule that prescribes a converted character string and tag information for each element that is at least part of a character string of input data, and an order rule that prescribes order information between tag information A storage unit storing
An element extraction unit that extracts a plurality of elements defined in the element extraction rule from a character string of the input data, and creates the converted character string in accordance with the element extraction rule for each extracted element;
For each element extracted by the element extraction unit, corresponding tag information is acquired from the element extraction rule, and the converted character string by the element extraction unit is obtained according to the order information between the tag information defined in the order rule. And an element order determining unit that, after rearranging, concatenates the converted character strings into name design output data.

  As a result, even if there is a variation in the naming tendency of individual developers with respect to the appearance order of the character strings of the input data, consistent name design can be supported by applying the order rules.

In the present invention, the element extraction rule is associated with the part of speech of the element as tag information for each element,
The element order determination unit rearranges the converted character strings in accordance with order information between parts of speech defined in the order rule.

  Thereby, the number of rules prepared in advance can be reduced by describing rules using part-of-speech that has few types such as nouns and adjectives and is highly versatile.

In the present invention, the name design device further includes a rule application unit,
The rule application unit is configured to be able to read and write at least one of the element extraction rules and the order rules stored in the storage unit from the outside of the name design apparatus.

  Thereby, since the same rule can be easily diverted by a plurality of name design apparatuses, a consistent name design is expected even for name designs between different business operators.

The present invention defines an element extraction rule that defines a converted character string and tag information for each element, and tag information that is a target of hierarchization, for elements that are at least part of a character string of input data A storage unit in which the hierarchization rules to be stored are stored;
Extracting a plurality of elements defined in the element extraction rule from the character string of the input data, and creating the converted character string in accordance with the element extraction rule for each extracted element;
For each element extracted by the element extraction unit, corresponding tag information is acquired from the element extraction rule, and when the acquired tag information is a target to be hierarchized in the hierarchization rule, the conversion by the element extraction unit And a hierarchizing unit that outputs output data of a name design obtained by hierarchizing a concatenated character string obtained by concatenating subsequent character strings.

  Thereby, even if individual developers vary in the design concept of the hierarchical structure of the character string of the input data, it is possible to support consistent name design by applying the hierarchization rule.

  According to the present invention, consistent name design can be supported even when the naming tendency of individual developers varies.

It is a block diagram of the name design apparatus concerning this embodiment. This is a configuration in which an external customization function is added to the name design apparatus of FIG. 1 according to the present embodiment. It is table data which shows the detail of each rule concerning this embodiment. It is a 1st example which shows the execution result of the name design apparatus concerning this embodiment. It is a 2nd example which shows the execution result of the name design apparatus concerning this embodiment. It is a 3rd example which shows the execution result of the name design apparatus concerning this embodiment.

  Hereinafter, an embodiment of the present invention will be described in detail with reference to the drawings.

FIG. 1 is a configuration diagram of the name design device 2. The name design device 2 is configured as a computer having a CPU (Central Processing Unit), a memory, storage means (storage unit) such as a hard disk, and a network interface.
In this computer, the CPU executes a program (also referred to as an application or its abbreviated application) read on the memory, thereby operating a control unit (control means) configured by each processing unit.

The name design device 2 includes an input unit 22, an element division unit (element extraction unit) 23, an element order determination unit 24, and a hierarchization unit 25 as processing units.
The storage unit of the name design device 2 stores an element division rule (element extraction rule) 33, a priority rule (order rule) 34, and a hierarchization rule 35. Further, the storage unit stores input data 11, output data 12, and intermediate data in the process of processing from input data 11 to output data 12.

Hereinafter, an outline of each processing unit of the name design device 2 will be described. Details of each processing unit will be clarified in the description using specific examples in FIG.
For example, when converting “customer name” to “customerName”, the input unit 22 accepts “customer name” as the input data 11.
The element division unit 23 refers to the element division rule 33 and divides the input data 11 as an element that is a minimum unit.
The element order determining unit 24 refers to the priority rule 34 and determines the order of each element divided by the element dividing unit 23. Further, the element order determination unit 24 creates one name by concatenating each element of the input data 11 according to the determined order. For example, “customerName” is an example of a name creation result.
The hierarchization unit 25 bundles one or more names into one hierarchy (group) according to the hierarchization rule 35 for the names created by the element order determination unit 24. This hierarchical structure is embodied in the API design as, for example, one class (object) including three member variables. The hierarchization unit 25 sets the hierarchized data as output data 12.

FIG. 2 shows a configuration in which an external customization function is added to the name design apparatus 2 of FIG. In the name design apparatus 2 of FIG. 2, a rule application unit 21 is added.
The rule application unit 21 customizes the name design process of each processing unit according to the user's request by reflecting the rule designation data 13 input by the user in each rule. The rule designation data 13 is at least one of the element division rule 33, the priority rule 34, and the hierarchization rule 35.

Further, the rule designation data 13 may replace the common rules prepared in advance with the rule designation data 13 as they are (all designation). On the other hand, part of the common rules prepared in advance may be overwritten by the rule designation data 13, and the remaining part that is not overwritten may use the common rule (partial designation). By rewriting a part of the common rule as a local rule, an API design in accordance with the culture of each business operator / organization can be realized separately from the API design common policy.
Furthermore, the rule application unit 21 not only inputs (imports) the rule designation data 13 into the name design apparatus 2 but also outputs (exports) each rule in the name design apparatus 2 as the rule designation data 13. A plurality of users may share the same rule designation data 13.

As described above, the configuration in which the rule designation data 13 can be input / output via the rule application unit 21 is particularly useful when designing different APIs for a plurality of operators.
For example, in a user system designed by company C, it is assumed that both a web service SA provided by company A and a web service SB provided by company B are called. If the developer of company A and the developer of company B share the same rule specification data 13 and design parameter names, the parameter names used by the web service SA and the web service SB Is consistent. Therefore, the developer of company C can construct a system that calls both the web service SA and the web service SB with a small learning cost.
On the other hand, when the developer of company A and the developer of company B design the parameter names according to different rules, the developer of company C constructs the system after grasping each inconsistent parameter group. It must be.

FIG. 3 is table data showing details of each rule.
The element division rule 33 associates the element of the input data 11 before conversion, the element after conversion, and one or more tags attached to the element. As the element division rule 33, for example, a morphological analysis dictionary in which an element before conversion is a morpheme and a part of speech for each morpheme is a tag can be used. Further, as the element division rule 33, a translation dictionary that associates a first language word indicating an element before conversion with a second language word indicating an element after conversion may be used.
Here, in addition to the above-mentioned part of speech, the tag of the element division rule 33 may include information indicating the meaning content of the element (customer or the like) and information indicating “prohibition” that does not perform conversion.

For example, the element dividing unit 23 refers to the record “customer” before conversion and the record “name” before conversion of the element division rule 33 with respect to the input data 11 “customer name”. Is divided into a first element called “name” and a second element called “name”.
Here, the pre-conversion-to-conversion correspondence data of the element division rule 33 is not limited to 1: 1, but may be associated with M: 1. For example, the element division rule 33 in FIG. 3 is 3: 1 data in which “customer” before conversion, “customer” before conversion, and “user” before conversion are associated with the same “customer” after conversion. As a result, even if “customer” or “customer” is input by a different business operator, it is converted to the same “customer”, so that it is possible to prevent a difference in API name due to the notation fluctuation.

The priority rule 34 is data that lists the characteristics (tags) of the elements in each order (first order, second order, third order,...) After the element order determination unit 24 rearranges the elements.
For example, the element order determination unit 24 regards the first element “customer” and the second element “name” as they are because the corresponding record does not exist in the priority rule 34.
Further, although not shown, the priority rule 34 may also include a connection rule for connecting the converted character strings of the elements after the elements are rearranged. For example, when the element “name” is connected after the element “customer”, the connection rule “snake case” is specified as “customer_name”, and when the connection rule “camel case” is specified, “ CustomerName ".

The hierarchization rule 35 is data that associates the tag of each element with the hierarchy name to which the tag belongs.
For example, the hierarchization unit 25 does not correspond to any tag “contract, maintenance, operation” for the first element “customer” and the second element “name”, so “customer name” is hierarchized. Not subject to
In the hierarchization rule 35, the tag and the hierarchy are not limited to 1: 1. Even two different tags of maintenance and operation may be associated with the same layer (operator (maintenance staff)).

FIG. 4 is a first example showing the execution result of the name design device 2.
The input unit 22 receives the character string “customer telephone number” as the input data 11 (processing of reference numeral 101).
The element dividing unit 23 obtains two elements “customer” and “phone number” by searching from the pre-conversion column of the element dividing rule 33 using at least a part of the character string “customer telephone number” as a search key. Therefore, the element dividing unit 23 divides the “customer telephone number” before the division into two elements “customer” after the conversion “string” and “phone number” after the conversion “string”. (Process of reference numeral 102).

  In addition, in the pre-conversion column of the element division rule 33, there are not only an element “phone number” but also an element “phone” and an element “number”. When there are a plurality of candidates in this way, the element dividing unit 23 may preferentially adopt an entry close to the upper line (first line) of the element dividing rule 33. Alternatively, the element division unit 23 may preferentially adopt an element having a longer character string length that matches the character string of the input data 11 and the character string described in the pre-conversion column of the element division rule 33.

The element order determination unit 24 determines the order among the plurality of divided elements according to the priority rule 34. For example, the element division rule 33 defines a tag for each element, and the tag “noun, customer” for the element “customer” and the tag “noun, general” for the element “phone number” can be acquired.
The element order determination unit 24 searches the order information corresponding to the current tag information divided by the process of reference numeral 102 from the order information for each tag specified in the element division rule 33 in the priority rule 34. Here, since both of the two elements are “nouns” and do not correspond to the second line “first order = department (noun), second order = name (noun)” of the priority rule 34, this time. There is no order information corresponding to the tag information. Therefore, the element order determination unit 24 does not change the order of the two elements “customer” and “telephone number” and connects them as “CustomerContactTel” as it is (processing of reference numeral 103).

The hierarchizing unit 25 determines to which hierarchy the “CustomerContactTel” connected in the process of reference numeral 103 belongs, according to the hierarchization rule 35. However, the tag “noun, customer” of the element “customer” and the tag “noun, general” of the element “phone number” do not correspond to any tag “contract, maintenance, operation” of the hierarchization rule 35. . Therefore, the hierarchizing unit 25 sets “CustomerContactTel” out of the hierarchy and outputs it as the output data 12 as it is (processing of reference numeral 104).
The first example in which the order change by the priority rule 34 is not performed and the layering by the layering rule 35 is not performed has been described above with reference to FIG.

FIG. 5 is a second example showing the execution result of the name design device 2. Compared to the first example of FIG. 4, the second example of FIG. 5 is an example in which the order is changed by the priority rule 34 and the hierarchization is also performed by the hierarchization rule 35.
The input unit 22 accepts the character string “contact emergency” as input data 11 (the process of reference numeral 201).
The element dividing unit 23 searches the character string “contact emergency” from the pre-conversion column of the element division rule 33, thereby converting the “contact” converted character string “ContactTel” and the “emergency” converted character string. Two elements “emergency” are obtained (processing of reference numeral 202).

The element order determination unit 24 determines the order between the elements based on the tag “noun, general” of the element “contact” acquired from the element division rule 33 and the tag “adjective, contract” of the element “emergency”. Determine according to the priority rule 34. Here, since there is one “noun” element and one “adjective” element, the first line of the priority rule 34 “first order = adjective, second order = noun” is included in this tag information. Is applicable.
Therefore, the order is changed so that the element “emergency” of the tag “adjective” is in the first order (first) and the element “contact” of the tag “noun” is in the second order. In accordance with the changed order, the element order determination unit 24 generates “EmergencyContactTel” by concatenating the character strings after the conversion of the two elements (processing of reference numeral 203).

The hierarchizing unit 25 searches the tag set “noun, general, adjective, contract” of the components of “EmergencyContactTel” connected by the process of reference numeral 203 from the tag string of the hierarchization rule 35. This time corresponds to the tag “contract” on the first line of the hierarchization rule 35. Therefore, the hierarchization unit 25 hires (groups) the character string “EmergencyContactTel” into the hierarchy name “contractor” in the first line of the hierarchization rule 35 (processing of reference numeral 204). Here, the JSON (JavaScript Object Notation) format is used as a notation for layering. In this way, the character string “contact emergency” was added to the hierarchy “contractor” as “EmergencyContactTel”.
Further, “EmergencyContactTel” hierarchized by the process of reference numeral 204 is associated with a sample value “03xxxxxxxx”. The sample values associated with the element “contact” as the element division rule 33 are read and used by the hierarchizing unit 25 (not shown).

In addition, in the hierarchy “contractor”, an entry of “weekdayContactTel” indicating “ordinary contact” was previously present as another character string. The name design device 2 sets each of the hierarchized data as output data 12 at the stage where the processing of the input data 11 is completed (processing of reference numeral 205).
As described above, in FIG. 5, the rules are applied in the order of the element division rule 33 → the priority rule 34 → the hierarchization rule 35. On the other hand, only the element division rule 33 → priority rule 34 may be applied, or only the element division rule 33 → hierarchization rule 35 may be applied.

FIG. 6 is a third example showing the execution result of the name design device 2. This third example explains the effect of consistent name design when the hierarchization rule 35 is used based on the processing result by the hierarchization unit 25.
First, it is assumed that developers A and B have received requests to incorporate four character strings (parameters) denoted by reference numeral 301 into the API design as input data 11.
Developer A created hierarchized output data 12 focusing on the persons "customer" and "trouble inquiry" (reference numeral 302). On the other hand, developer B created hierarchized output data 12 focusing on the data format of “telephone number” and “mail address” (reference numeral 303). As described above, even when the same data is handled, the API design results vary due to individual differences among developers.

Therefore, the developers A and B are allowed to use the name design device 2 in which the hierarchization rule 35 indicated by reference numeral 304 is prepared in advance. This hierarchization rule 35, when the input data 11 includes the element of the first column of the hierarchization rule 35, groups the input data 11 into the hierarchy indicated by the hierarchy name of the second column, This indicates that the character string indicating the element in the first column is deleted (when the deletion flag in the third column is ◯).
For example, since the element “customer” is included in the “customer telephone number” denoted by reference numeral 301, the hierarchizing unit 25 changes the “phone number” obtained by deleting “customer” from the “customer telephone number” to the hierarchical level. Place under "Customer".
As a result, even if the developer B has a habit of hierarchizing focusing on the data format, by using the name design device 2, what is consistent with the hierarchization result of the developer A indicated by reference numeral 302 can be obtained. Can be obtained.

In the present embodiment described above, the name design apparatus 2 is made to create the output data 12 from the input data 11 in accordance with various rules (element division rule 33, priority rule 34, hierarchization rule 35). Thereby, even if the habit of the developer is included in the input data 11, it is possible to create the output data 12 with consistency at the stage of applying various rules.
Furthermore, the definition number of various rules (the number of records in the table) can be reduced by using parts of speech such as adjectives and verbs and abstract definitions such as time and place as tags used for various rules of the name design device 2. The prior burden when operating the name design apparatus 2 can be reduced.
On the other hand, in the conventional technology such as Non-Patent Document 1, since the elements are simply converted from Japanese to English, the habit of the developer such as the order of elements remains after conversion.

2 Name design device 11 Input data 12 Output data 13 Rule designation data 21 Rule application part 22 Input part 23 Element division part (element extraction part)
24 element order determination unit 25 hierarchization unit 33 element division rule (element extraction rule)
34 Priority rules (order rules)
35 Tiering rules

Claims (7)

For an element that is at least part of the character string of the input data, an element extraction rule that defines the converted character string and tag information for each element and an order rule that defines the order information between the tag information are stored. A storage unit,
Extracting a plurality of elements defined in the element extraction rule from the character string of the input data, and creating the converted character string in accordance with the element extraction rule for each extracted element;
For each element extracted by the element extraction unit, corresponding tag information is acquired from the element extraction rule, and the converted character string by the element extraction unit is obtained according to the order information between the tag information defined in the order rule. An element order determination unit comprising: an element order determining unit that, after rearranging, concatenates the converted character strings into output data for name design. The element extraction rule is associated with the part of speech of the element as tag information for each element,
The name design apparatus according to claim 1, wherein the element order determination unit rearranges the converted character strings in accordance with order information between parts of speech defined in the order rule. The name design device further includes a rule application unit,
The rule application unit is configured to read and write at least one of the element extraction rule and the order rule stored in the storage unit from the outside of the name design device. The name design apparatus of Claim 1 or Claim 2. For elements that are at least part of the character string of the input data, an element extraction rule that specifies the converted character string and tag information for each element, and a layering rule that specifies tag information that is subject to layering A storage unit storing
Extracting a plurality of elements defined in the element extraction rule from the character string of the input data, and creating the converted character string in accordance with the element extraction rule for each extracted element;
For each element extracted by the element extraction unit, corresponding tag information is acquired from the element extraction rule, and when the acquired tag information is a target to be hierarchized in the hierarchization rule, the conversion by the element extraction unit A name design apparatus comprising: a hierarchization unit that outputs output data of name design obtained by hierarchizing a concatenated character string obtained by concatenating subsequent character strings. The name design device includes a storage unit, an element extraction unit, and an element order determination unit,
For the element that is at least part of the character string of the input data, the storage unit defines an element extraction rule that defines the converted character string and tag information for each element, and order information between the tag information Order rules are remembered,
The element extraction unit extracts a plurality of elements defined in the element extraction rule from the character string of the input data, creates the converted character string according to the element extraction rule for each extracted element,
The element order determination unit acquires corresponding tag information from the element extraction rule for each element extracted by the element extraction unit, and performs the element extraction unit according to the order information between the tag information defined in the order rule. After rearranging the converted character strings, the converted character strings are connected to form output data for name design. The name design device includes a storage unit, an element extraction unit, and a hierarchization unit.
In the storage unit, with respect to an element that is at least a part of a character string of input data, an element extraction rule that defines a converted character string and tag information for each element, and tag information that is a target of layering The stratification rule that prescribes
The element extraction unit extracts a plurality of elements defined in the element extraction rule from the character string of the input data, creates the converted character string according to the element extraction rule for each extracted element,
The hierarchization unit acquires corresponding tag information from the element extraction rule for each element extracted by the element extraction unit, and when the acquired tag information is a target to be hierarchized in the hierarchization rule, A name design method characterized in that a concatenated character string obtained by concatenating the converted character strings by an element extraction unit is used as name design output data in a hierarchical structure.   A name design program for causing the name design apparatus, which is a computer, to execute the name design method according to claim 5 or 6.

Images