JPH03251929A - Qualification name settling system for translation system - Google Patents

Abstract

PURPOSE:To reduce a token store area and to shorten the translating time by setting a qualification name at the time point of a word decomposion processing. CONSTITUTION:A translation system 1 reads a source program 5 and calls out a vocabulary analyzing subsystem 2. This subsystem 2 reads one line of the program 5 and decomposes a word to generate plural tokens. If the labels in the tokens are identical with the definition names, the subsystem 2 calls out a definition name register mechanism 3. The mechanism 3 retrieves a definition name table 7 via the definition names. If the same definition name is not contained in the table 7, the retrieving definition name is registered in the table 7 as it is. Meanwhile the definition names are unified with the additional information and then registered in the table 7 is the same definition name is contained in the table 7. When the label is identical with a definition name, the subsystem 2 calls out a definition name solution mechanism 4. The mechanism 4 retrieves the table 7 to delete all qualification information after replacing them with the definition name unified with the additional information given to the same names. Finally the subsystem 2 outputs the token having no unified qualification information to a token store area 6. Thus the token store area can be reduced together with reduction of the translating time.

Description

DETAILED DESCRIPTION OF THE INVENTION [Field of Industrial Application] The present invention relates to a qualified name resolution method in a source program translation system.

[Conventional technology]

Conventionally, the qualified name resolution method in this type of translation system is performed simultaneously at the time of name reference resolution processing.

[Problem to be solved by the invention]

The conventional qualified name resolution method described above requires qualified name information at the time of name reference resolution processing, so the token storage area increases in proportion to the number of qualified names and word length, and translation time decreases. It has the disadvantage of increasing.

[Means to solve the problem]

The qualified name resolution method in the translation system of the present invention reads a source program, performs word decomposition processing, and stores the generated tokens in a token storage area. A definition name registration mechanism that searches a definition name table using a definition name, and if there is an identical name, uniqueizes the definition name using additional information and registers this definition name in the definition table, and a qualified name of the name in the token. and a modification group resolution mechanism that deletes modification information by replacing it with a definition name unified by additional information for the same name.

〔Example〕

Next, the present invention will be explained with reference to the drawings.

FIG. 1 is a block diagram of an embodiment of the present invention.

In FIG. 1, a translation system 1 reads a source program 5 and performs word decomposition processing using an N-string analysis subsystem 2.
Then, the vocabulary analysis system 2 first reads one line of the source program 5, performs word decomposition processing, and generates a plurality of tokens.

If the name in the token is a definition name, the vocabulary analysis subsystem 2 calls the definition name registration mechanism 3. The definition name registration mechanism 3 first searches the definition name table 7 using the definition name to determine whether or not the same name exists, and if the same name does not exist, it is left as is, and if the same name exists, the definition name is made unique using additional information. Later, this definition name is registered in the definition name table 7.

Next, if the name tag is a qualified name, the lexical analysis subsystem 2 calls the modifier resolution mechanism 4. The modification group resolution mechanism 4 searches the definition name table 7 and performs modification name resolution, that is, deletes all modification information by replacing the same name with a definition name made unique by additional information.

Finally, the vocabulary analysis subsystem 2 outputs the unique tokens without modification information to the token storage area 6.

〔Effect of the invention〕

As explained above, the present invention has the effect of reducing the token storage area and shortening the translation time by performing qualified name resolution at the time of word decomposition processing.

[Brief explanation of drawings]

FIG. 1 is a block diagram of an embodiment of the present invention. 1... Translation system, 2... Language analysis subsystem, 3... Definition name registration mechanism, 4... Modifier resolution mechanism,
5... Source program, 6... Token storage area,
7...Definition name table.

Claims (1)

[Claims] In the lexical analysis subsystem of the translation system, which reads the source program, performs word decomposition processing, and stores the generated tokens in the token storage area, the definition name table is searched using the definition name of the name mark in the generated token, and the same name is searched. a definition name registration mechanism that uniqueizes a definition name using additional information and registers this definition name in the definition table;
A qualified name resolution method in a translation system characterized by having a qualified name resolution mechanism that replaces the qualified name of the name mark in the token with a defined name made unique by additional information for the same name and deletes the qualified information. .