JPH11224200A - Program converting method, device therefor, and storage medium having stored program converting program - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a program converting method and device, by which an existing program is automatically converted at the time of defining a new type and calculation is enabled even for the new type in a formula manipulation system which is not perfectly object-oriented, and to provide a storage medium in which a program converting program is stored. SOLUTION: The syntax analysis of the inputted existing program is performed (S2), the syntax analysis of the existing program subjected to the syntax analysis is performed (S3) and the existing program subjected to the syntax analysis is converted based on the definition of the new type through the use of a conversion rule for executing conversion from the type defined in an existing numerical formula processing system into the newly defined type (S4) and outputted (S5). Thus, when the new type is defined in the numerical formula processing system, only a basic and arithmetic part and a type converting program are previously prepared so that the remaining programs are used as they are only by converting them through the use of a program converting means.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a program conversion method and apparatus, and a storage medium storing a program conversion program. More particularly, the present invention relates to a program conversion method and a program conversion method for converting a program in order to extend the functions of a mathematical expression processing system. The present invention relates to an apparatus and a storage medium storing a program conversion program.

[0002]

2. Description of the Related Art In a mathematical processing system, when a number of types other than those defined in advance is handled, if the system is completely object-oriented, an existing program can be used as it is by defining a class for the type. Available.

[0003]

However, some mathematical processing systems are not sufficiently object-oriented. In such a case, various types of program conversion and the like are performed together with new type definitions and operation definitions. Must be performed manually. An example is shown below.

[0004] In a mathematical processing system, consider a case where a new type is defined and a number having the new type is calculated. Here, as an example, in the mathematical processing system, basic operations of rational numbers and addition / subtraction / multiplication / division between them, magnitude comparison, and absolute value are defined. It is assumed that the operation is defined and rounding, -∞ rounding, and + ∞ rounding are provided as floating-point rounding modes.

A case will be described in which a new type of "section number" is defined in this system, and further a calculation of a polynomial of the section number coefficient is to be performed. The number of sections is represented by a floating-point pair [a, b] where a ≦ b, and a number c where a ≦ c ≦ b
Means a set of The definition of the basic calculation of the number of sections is
Suppose that it is as shown in FIG. However, in the calculation of the _{float, o, o d, o u} (o is +, -, *, either /) as the rounding mode, respectively, rounded,
Represents −∞ rounding and + ∞ rounding. In addition, subtraction, multiplication and division of the section coefficient polynomial, coefficients are grouped for each degree and calculated by calculating the number of sections.

[0006] For example, FIG. 19, or rational coefficients, floating point computing the L ₁ norm of univariate polynomial f (x) of the coefficients and there is a program called L ₁ (polynomial f (x) = c _n x ⁿ + c _n-1 x ^n-1 + ... + c ₀
Is of the L _{1 norm, | c n | + | c} n-1 | + ... + | c
₁ |). In the figure, “L ₁ : = proc (f, x)” on the first line means that the dummy arguments of L ₁ are f and x, and “local” on the second line
"c, cs, res;" are declarations of local variables used in this function. The third line “coeffs (f,
“x)” is a function that returns a list including coefficients of the polynomial f (x). In the fifth line, "for c in c
s do. . . "od" indicates that c repeats across all components in list cs.

Regardless of which method is used, when a new type is defined, it is necessary to write a program defining at least the basic operation. FIG. 20 shows an example of a program for calculating the addition of two section numbers, and "ad" is a program for defining the addition between the two section numbers shown in FIG.
d2interval ".

FIG. 21 shows an example of a program for calculating the L ₁ norm of the interval coefficient polynomial. If you write a program "L1interval" shown in the figure, the calculation of the L ₁ norm interval coefficient polynomial it becomes possible. However, this method requires rewriting all programs each time a new type is defined. If a completely object-oriented mathematical processing system is used, only the basic operations are defined, and FIG.
The calculation of the L ₁ norm of the section number coefficient polynomial is also possible with the program shown in FIG.

However, this assumption does not hold, ie,
In some cases, the user cannot change the definition of the basic operation. At this time, as shown in FIG. 22, if there is an addition “+” in the program, the portion is added to, for example, “add”
It is necessary to rewrite the function as a two-argument function “2” and prepare a program “add2.” In addition, functions abs (or functions called there) that return an absolute value usually include large or small functions. Because there is a comparison, it needs to be rewritten.

The present invention has been made in view of the above points, and in a mathematical processing system that is not completely object-oriented, when a new type is defined, an existing program is automatically converted to a new type. It is another object of the present invention to provide a program conversion method and device which can be calculated by the user and a storage medium storing the program conversion program.

[0011]

FIG. 1 is a diagram for explaining the principle of the present invention. The present invention (claim 1) provides a program conversion method for converting a basic operation part of an existing program associated with a definition of a new type in a mathematical processing system. The program is parsed (step 2), the grammatical analysis is performed on the parsed existing program (step 3), and a program that can be calculated with the type defined in the existing mathematical processing system is converted into a newly defined type. Using a program conversion rule for performing a conversion that enables calculation, an existing program that has been subjected to grammatical analysis is converted so that a newly defined type can be calculated (step 4) and output (step 5).

In the present invention (claim 2), the basic operation to be converted by the conversion rule includes any of four arithmetic operations, magnitude comparison, and zero judgment. The present invention (claim 3) is a program conversion device for converting a basic operation part of an existing program associated with a definition of a new type in a mathematical processing system, wherein the program input means 1 inputs an existing program.
Syntactic analysis means 2 for analyzing the syntax of the input existing program, grammatical analysis means 3 for analyzing the grammar of the existing program parsed by the syntax analysis means, and a type defined in the existing mathematical processing system. A conversion rule 4 for converting a program which can be calculated by a new type to be able to be calculated, a conversion means 5 for converting an existing program grammatically analyzed using the conversion rule 4, and a conversion means 5. Output means 6 for outputting the converted program
And

In the present invention (claim 4), the basic operation to be converted by the conversion rule 4 includes any of four arithmetic operations, magnitude comparison, and zero judgment. The present invention (claim 5) is a storage medium for storing a program conversion program for converting a basic operation part of an existing program associated with a definition of a new type in a mathematical processing system, wherein the program input causes the existing program to be input. Process, a parsing process for parsing the input existing program, a grammar parsing process for parsing the existing program parsed by the parsing process, and a type defined in the existing mathematical processing system. A conversion process that converts an existing program that has been grammatically analyzed using a program conversion rule that converts a computable program into a newly defined type that can be calculated, and an output that outputs the program converted by the conversion process Process.

According to the present invention (claim 6), the basic operation to be converted in the conversion process includes any of four arithmetic operations, magnitude comparison, and zero judgment. As described above, in the present invention, when a new type is defined, a basic operation part and a program for type conversion are prepared in advance as conversion rules, and by referring to this, an existing program is converted to a new program. It can be converted.

[0015]

FIG. 3 shows the configuration of a program conversion device according to the present invention. The program conversion device shown in FIG.
It comprises a program input unit 1, a syntax analysis unit 2, a grammar analysis unit 3, a conversion rule database 4, a program conversion unit 5, and an output unit 6.

The program input unit 1 inputs an existing program. The syntax analysis unit 2 analyzes the syntax of the input program by an existing method, and transfers it to the grammar analysis unit 3. The grammar analyzer 3 performs grammar analysis on the result obtained by the syntax analysis with reference to a grammar dictionary (not shown) or the like, and transfers the result to the program converter 5.

Note that the syntactic analysis unit 2 and the grammar analysis unit 3 use the existing technology, and the present invention does not specify the method. The conversion rule database 4 stores the conversion rules of the basic operation part. Basic operations to be converted include:
For example, there are addition of two arguments, comparison of magnitude, determination of zero, and the like.
The program conversion unit 5 stores the conversion rule database 4
The program is converted based on the conversion rules registered in, and the result of the grammatical analysis, and transferred to the output unit 6.

The output unit 6 outputs the program converted by the program conversion unit 5.

[0019]

Embodiments of the present invention will be described below with reference to the drawings. Here, the type name of the number of sections is “interval”, and a basic operation between the number of sections is defined according to FIG. 18 described above. FIG. 4 is a diagram illustrating an example of a program conversion rule according to an embodiment of the present invention. In the figure, if there is the pattern (a), it is rewritten to the pattern (b), and ‖ indicates that the pattern may be omitted. This is a part of the language grammar of the mathematical processing system written in BNF. In this language, if for <name> is specified in the repetition part,
When m <expr> is omitted, the initial value of <name> is 1, and when by <expr> is omitted, <name> increases by one.

Among the definition programs for the basic operation, FIG. 5 shows an example of adding two arguments, FIG. 6 shows a magnitude comparison, and FIG. 7 shows a zero determination program. These do not need to be prepared by the user. The type (xn) shown in FIGS.
“umber” returns “true” only when x is a rational number or a floating point number.

Further, there are add2. n, large. n, zero. "." in n, represents a concatenation of character strings. For example,
In the case of FIG. 5, when “n” is a character string “interval”, “add2.n” is “add2interva”.
l ".

Further, “evalb (x>” in FIG.
y) ”and“ evalb (x = 0) ”in FIG. 7 are true or false of the result of evaluating x> y and x = 0, respectively, ie,
It is a function that returns “true” or “false”. With the definition of the new type, the user prepares a program for converting the type originally defined in the mathematical processing system to the newly defined type, and a program for defining the basic operation shown in FIG. Here, as an example, FIG. 8 shows a type conversion program, and FIG. 9 shows a two-argument addition add2.
FIG. 10 shows a program for defining magnitude comparison, and FIG. 11 shows a program for zero determination. Note that types other than those originally defined in the system are represented using a list, and the first component has a type name,
It is assumed that the data itself is included in the component.

In FIG. 8, "evalfd
(X) ”and“ evalfu (x) ”convert the value of x into a floating-point number having a predetermined precision. In the case of exchange,“ evalfd (x) ”is obtained by rounding -∞ and“ evalfu (x) ”.
fu (x) performs + ∞ rounding. Here, description will be made assuming that the programs of FIGS. 12, 14, and 16 are given.

First, an example including addition will be described. The existing program shown in FIG. 12 is input from the program input unit 1, and “a +
It can be seen that the portion “b” corresponds to FIG. 4 (3. (a)). According to conversion rule 4, the place of addition “+” is replaced by “a”.
dd2 "is converted to a two-argument function, converted into a program shown in FIG. 13, and output.

Next, an example including a conditional branch will be described. The existing program shown in FIG.
And the syntax analysis unit 2 and the grammar analysis unit 3 find that the part “n> a” in the program corresponds to FIG. 4 (10. (a)) of the conversion rule 4. According to the conversion rule 4, the conversion unit 5 replaces the inequality sign “>” with a two-argument function “larger”.
5 and output.

Further, an example including a repeated portion will be described. The existing program shown in FIG. 16 is input from the program input unit 1, and the syntax analysis unit 2 and the grammar analysis unit 3
"For i from a by" in the program
s to b do. . . It can be seen that the part “od” corresponds to the conversion rule 4 (FIG. 4 (17. (a)). Therefore, according to the conversion rule 4, the program is converted into the program shown in FIG.

Although the above embodiment has been described based on the apparatus configuration shown in FIG. 3, the present invention is not limited to this example, and each component is constructed as a program to realize the program conversion apparatus. The present invention can be easily implemented by storing the program in a portable storage medium such as a disk device, a floppy disk, or a CD-ROM connected to a computer, and installing the program when implementing the present invention.

It should be noted that the present invention is not limited to the above-described embodiment, and various modifications and applications are possible within the scope of the claims.

[0029]

As described above, according to the present invention, when a new type is defined in a mathematical expression processing system, if only a basic operation part and a program for type conversion are prepared in advance, the remaining program Can be used as it is simply by conversion by the program conversion means of the present invention.

The program once converted can be used as it is even when a new type is defined.

[Brief description of the drawings]

FIG. 1 is a diagram for explaining the principle of the present invention.

FIG. 2 is a principle configuration diagram of the present invention.

FIG. 3 is a configuration diagram of a program conversion device of the present invention.

FIG. 4 is a diagram showing an example of a program conversion rule according to an embodiment of the present invention.

FIG. 5 is an example of a predefined conversion rule (a program defining addition of two arguments) according to an embodiment of the present invention.

FIG. 6 is an example of a predefined conversion rule (program for comparing magnitude) according to an embodiment of the present invention.

FIG. 7 is an example of a predefined conversion rule (zero determination program) according to an embodiment of the present invention.

FIG. 8 is an example of a conversion program according to an embodiment of the present invention.

FIG. 9 is an example of a program for defining addition of two section numbers according to an embodiment of the present invention.

FIG. 10 is an example of a program for comparing the size of two sections according to an embodiment of the present invention.

FIG. 11 is an example of a program for performing zero determination of the number of sections according to an embodiment of the present invention.

FIG. 12 is an example of a program including a part of the addition according to the embodiment of the present invention.

FIG. 13 is an example of a program obtained by converting FIG. 12 according to an embodiment of the present invention.

FIG. 14 is an example of a program including a conditional branch as a part of the embodiment of the present invention.

FIG. 15 is an example of a program obtained by converting FIG. 14 according to an embodiment of the present invention;

FIG. 16 is an example of a program including a repetition part in a part of one embodiment of the present invention.

FIG. 17 is an example of a program obtained by converting FIG. 16 according to an embodiment of the present invention;

FIG. 18 is a diagram showing a definition of a basic calculation of the number of sections.

FIG. 19 is an example of a program for calculating an L ₁ norm for a _one- variable polynomial having a rational or floating-point coefficient.

FIG. 20 is an example of a program for calculating addition of two sections.

FIG. 21 is an example of a program for calculating an L ₁ norm of a section number coefficient polynomial.

FIG. 22 is an example of an object-oriented program when it is impossible for a user to redefine a basic operation.

[Explanation of symbols]

 DESCRIPTION OF SYMBOLS 1 Program input means, program input part 2 Syntax analysis means, syntax analysis part 3 Grammar analysis means, grammar analysis part 4 Conversion rule 5 Conversion means, program conversion part 6 Output means, output part

Claims (6)

[Claims] 1. A program conversion method for converting a basic operation part of an existing program according to a definition of a new type in a mathematical processing system, wherein the existing program is input, and a syntax analysis of the input existing program is performed. A program conversion rule for performing a grammatical analysis on the parsed existing program and performing a conversion that allows a program that can be calculated in a type defined in an existing mathematical processing system to be able to calculate in a newly defined type is also provided. A program conversion method comprising: converting a grammatically analyzed existing program so that a newly defined type can be calculated; and outputting the calculated program. 2. The program conversion method according to claim 1, wherein the basic operation to be converted by the conversion rule includes any of four arithmetic operations, magnitude comparison, and zero judgment. 3. A program conversion device for converting a basic operation part of an existing program according to a definition of a new type in a mathematical processing system, comprising: a program input means for inputting the existing program; Syntactic analysis means for analyzing the syntax of the program; grammatical analysis means for performing a grammatical analysis on the existing program parsed by the syntax analysis means; and a program which can be calculated in a type defined in the existing mathematical processing system. A conversion rule for performing a conversion that enables a newly defined type to be calculated, a conversion unit for converting the existing program grammatically analyzed using the conversion rule, and a program converted by the conversion unit. A program conversion device comprising output means. 4. The program conversion device according to claim 3, wherein the conversion rule includes one of four arithmetic operations, a magnitude comparison, and a zero determination as basic operations to be converted. 5. A storage medium storing a program conversion program for converting a basic operation part of an existing program associated with a definition of a new type in a mathematical processing system, a program input process for inputting the existing program, A syntax analysis process for parsing the input existing program; a grammar analysis process for performing grammar analysis on the existing program parsed by the syntax analysis process; and a type defined in the existing mathematical processing system. A conversion process of converting the existing program that has been grammatically analyzed using a program conversion rule for performing a conversion that allows a computable program to be calculated even in a newly defined type; and outputting a program converted by the conversion process. Output process Storage medium storing a program transformation program. 6. A storage medium storing the program conversion program according to claim 4, wherein the basic operation to be converted in the conversion process includes any of four arithmetic operations, magnitude comparison, and zero determination.