JPH01291329A - High speed compile processing system - Google Patents

Abstract

PURPOSE:To omit the determination processing for a kind of a sentence and to execute the compile processing at high speed by designating a classification of the sentence onto a source text and recognizing it by a compiler. CONSTITUTION:A source text is generated, while designating a sentence identifier for showing a kind of a sentence at every sentence to a fixed column which is determined at every compiler. A sentence identifier read-in part 1 in a sentence structure analyzing part 11 of the compiler reads in its sentence identifier, decides whether this sentence identifier coincides with that which is registered or not by a sentence identifier deciding part 2, and when they do not coincide, the second token generating part B6 and an existing system sentence processing part 7 are started and a classification determination processing of the sentence which is being processed at present is executed. Also, when the sentence identifier has coincided with that which is registered, a token generating part A3 and a designated sentence processing part 4 are started, and in an intermediate text output part 9, an intermediate text is generated, and outputted to an intermediate text file 14. In such a way, a determination processing for a kind of a sentence form becomes unnecessary, and the processing can be executed at a high speed.

Description

DETAILED DESCRIPTION OF THE INVENTION [Field of Industrial Application] The present invention relates to a language processor executed in an electronic computer system, and particularly to language processors C and below for programming languages that do not have reserved words, such as PL/I. , compiler) processing speed.

[Conventional technology]

Conventionally, in the compilation processing method of this kind of programming language without reserved words, 92 types of sentences are predicted based on the type of the first token of each sentence in the source text, and the input source text is used to predict the type of the predicted sentence. I was checking to see if it matched the sentence pattern.

[Problem to be solved by the invention]

However, in the above-mentioned conventional compilation processing method of a programming language that does not have reserved words, if the sentence pattern on the input source text does not match the sentence pattern predicted for each sentence, a different sentence pattern is predicted, A search is performed again to determine whether the sentence pattern of the newly predicted sentence matches the sentence pattern from the beginning of the sentence on the input source text.

In this way, if the predicted sentence pattern does not match, another sentence is predicted one after another and the search is re-searched, which has the disadvantage of slowing down the compilation speed.

In view of the above-mentioned drawbacks, the technical object of the present invention is to provide a high-speed compilation processing method that can speed up the compilation processing (by eliminating the need for determining the type of sentence type). .

[Failure to solve the problem]

According to the present invention, there is provided a high-speed compilation processing method for a compiler without reserved words based on the user's statement type specification,
When inputting each sentence in the source text, the syntax analysis unit that makes up the language processor for programming languages that do not have reserved words, such as I, uses predetermined columns from fixed columns determined for each compiler. (User specification, etc.) A sentence identifier reading section that inputs a sentence identifier indicating the type of sentence; and a sentence identifier determining section that determines whether the sentence identifier input in the sentence identifier reading section is a registered sentence identifier. 9 When the sentence identifier is a registered sentence identifier.

A high-speed compilation process characterized by having a token creation unit that reads source text and creates tokens that are processing units of the compiler, and a specified statement processing unit that performs syntactic analysis of a statement specified by a two-sentence identifier. You get speed.

〔Example〕

Next, two embodiments of the present invention will be described with reference to the drawings.

FIG. 1 is a diagram showing the configuration of the present invention. Figure 2 is the first
3 is a flowchart of a process for determining a sentence type in the existing method sentence processing shown in FIG. FIG. 3 is a detailed diagram of the directive processing shown in FIG. 1.

Referring to FIG. 1, the high-speed compilation processing method of the compiler without reserved words based on the user's statement type specification according to an embodiment of the present invention is as follows. Create the source text while specifying the sentence identifier that indicates the type of sentence. The statement identifier reading unit 1 in each compiler reads the statement identifier specified by the user from this fixed column, and the statement identifier determining unit 2 determines whether this statement identifier matches a registered one. .

If they do not match, it is assumed that no sentence identifier has been specified for this corresponding sentence, and the second token creation unit B6 is activated, the existing method sentence processing unit 7 is activated, and the process is explained in detail in FIG. Follow the flow chart.

Performs processing to determine the type of sentence currently being processed. That is, it is determined in step 21 whether or not a token obtained by reading the source text and tokenizing it is retained. If there are, process the tokens sequentially in the input order.

Determine whether the sentence pattern matches the currently predicted sentence.

When the sentence patterns match and the types of the two sentences are determined, the currently processed sentence pattern and determination process 28 are performed, and the intermediate text output section 9 of FIG. 1 is executed to create an intermediate text. If the sentence pattern does not match the currently predicted sentence, a second sentence pattern determination process 29 is performed to newly predict a new sentence pattern, and the process restarts from the beginning of FIG. If there is no tokenized token, the token is read 25 to tokenize the source text, and the same processing as described in 22, 23, and 24 is performed to determine the sentence pattern.

On the other hand, if the user-specified sentence identifier matches the registered one in the two-sentence identifier determination unit 2, the token creation unit A3
Then, the instruction sentence processing 4 is activated, and it is determined at 31 in FIG. 3 whether the syntactic form of the source text matches the sentence type of the sentence specified by the user. - If they do not match, it is assumed that there is an error in the syntax format, and the error output section 32 outputs an error.

If one sentence processing is completed without any error being detected, an intermediate text is created by the intermediate text output unit 9 and output to the intermediate text file 14.

The above-described process is sequentially repeated, and when there is no more input source text, control is transferred to the semantic analysis section 12.

〔Effect of the invention〕

According to two aspects of the present invention, as described in detail above.

Let the user specify the type of sentence on the source text.

When the compiler recognizes this, it becomes possible to omit the process of determining the type of statement to be performed by the compiler, which has the effect of speeding up the compilation process. The column that specifies the statement type can also be specified as a compile-time option.

Note that if you do not want to use this function, you can leave the column for specifying the statement identifier blank, etc., and the compiler can perform the process of determining the statement type in the same way as existing compilers.

[Brief explanation of the drawing]

FIG. 1 is an overall configuration diagram of a high-speed con-file processing method of a compiler without reserved words based on the user's statement type specification according to an embodiment of the present invention, and FIG. A detailed flowchart, FIG. 3, is a detailed diagram of the instruction text processing section of FIG. 1... Sentence identifier reading section, 2... Sentence identifier determining section. 3... Token creation section A, 4... Directive sentence processing section, 5
. . . - Sentence processing end determination unit A, 6 . . . Token creation unit B. 7... Existing method sentence processing unit, 8...-Sentence processing end determination unit B, 9... Intermediate text output unit, 10... Source text end determination unit, 11... Syntax analysis unit, 12・・・
・Semantic analysis section, 13... Source text file, 1
4...Intermediate text file.

Claims (1)

[Claims] 1. When inputting each sentence on the source text, there is a sentence identifier reading section that inputs a sentence identifier indicating a predetermined type of sentence from a fixed column determined for each compiler, and a sentence identifier reading section. A statement identifier determination unit that determines whether a given statement identifier is a registered statement identifier; and a token that reads the source text and creates a token, which is a processing unit of the compiler, when the statement identifier is a registered statement identifier. A high-speed compilation processing method for a compiler, characterized by including a creation section and a specified statement processing section that performs syntactic analysis of a statement specified by a statement identifier.