JPH02255976A - Information processor - Google Patents

Abstract

PURPOSE:To use a storage area with high efficiency by producing as necessary a byte attribute table to the bytes except the least significant one of a character of plural bytes and then an attribute bit table to the least significant type respectively. CONSTITUTION:An identifier showing the presence/absence of a special character attribute is registered into a byte attribute table 4. A byte attribute check part 3 retrieves the table 4 and inspects the presence of the special character attribute. In addition, the attribute bit tables 6, 7 and 8 store the numerical value showing the type of the special character attribute. Then NO is decided when a NULL pointer is confirmed by reference to the elements set on the table 4 and showing the most significant bit and its subsequent ones except the least significant bit. When the NULL pointer does not exist till the bit next to the least significant one, the attribute is known by reference to the tables 6, 7 and 8 shown by the least significant bit. As a result, the character attribute can be referred to an updated at a high speed.

Description

DETAILED DESCRIPTION OF THE INVENTION [Industrial Application Field] The present invention is applied to a computer system that defines and processes character attributes. In particular, it relates to language processing means.

It is especially suitable for Japanese language processing.

〔overview〕

The present invention, in a language processing means that defines attributes of multi-byte character codes, does not require processing involving hash calculations and eliminates unnecessary storage areas, thereby allowing character attributes to be referenced and updated at high speed. It has been made possible.

[Conventional technology]

Processing Japanese data has become indispensable in recent information processing systems, but various problems have arisen in creating processing systems for characters that cannot be expressed in one byte.

For example, spaces in 1-byte characters serve as word delimiters, but this is not the case with spaces in 2-byte characters. To solve this problem, it becomes necessary to give multi-byte characters special character attributes.

Conventionally, to manage this type of character attributes, a bit table has been prepared for one-byte characters to indicate the presence or absence of various attributes for each character. For example, if you try to manage 32 types of attributes using characters with 8 bits per byte, you will need to prepare a table with 256 elements and 4 bytes per element. Then, the contents of this attribute table are referred to and updated using the character code as the element number. Furthermore, for multi-byte characters, another attribute table of a certain size is prepared in advance, and the elements of the attribute table are referenced and updated using the hash-calculated value of the character code. At this time, the attribute table for multi-byte characters does not have enough elements to register all multi-byte characters,
When character attributes were referenced, if the character was not registered, it was determined that the character did not have special character attributes. In addition, there are some that prepare attribute tables that have elements for all characters, including multi-byte characters, especially double-byte characters.

[Problem that the invention seeks to solve]

In such a conventional example, a separate attribute table is provided for multi-byte characters, and the element of the attribute table for each character code is obtained by calculating the hash value of the character code, so different characters obtain the same hash value. There may be cases where processing is required. Also, if you prepare an attribute table for all double-byte characters, you will have bit strings even for characters that do not have special character attributes. For example, if 1 byte is 8 bits, the number of elements will be 65,536, which is a large number. storage space is required.

In this way, in the conventional example, attribute reference and update processing for multi-byte characters is complicated and slow, or
The disadvantage is that it requires a large amount of storage space.

SUMMARY OF THE INVENTION An object of the present invention is to eliminate such drawbacks, and to provide an information processing device that does not require special processing and has attribute management means that can save storage space.

[Means for solving problems]

The present invention provides an information processing device including a definition means for defining a special character attribute for a character code expressed in multiple bytes, wherein the definition means registers an identifier indicating the presence or absence of the special character attribute. a first table; a byte attribute checking means for searching the first table using a combination of numerical values of bytes excluding the least significant byte of the character code to test the existence of a special character attribute; and a numerical value indicating the type of the special character attribute. When the test result by the above-mentioned byte attribute checking means indicates the presence of a special character attribute, the second table is searched using the numerical value of the lowest byte of the above character code to check the special character attribute. The present invention is characterized by comprising an attribute bit check means for verifying a numerical value indicating the type.

[Effect]

First, refer to the element on the byte attribute table indicated by each bit sequentially from the most significant bit excluding the least significant bit, and if it is a NULL pointer, select N.

If there is no NULL pointer up to the bit immediately above the least significant bit, the attribute is determined by referring to the attribute bit table indicated by this least significant bit.

〔Example〕

An embodiment of the present invention will be described below with reference to the drawings.

FIG. 1 is a block diagram showing the configuration of this embodiment. As shown in FIG. 1, this embodiment includes byte attribute checking means 1, attribute bit table means 2,
Here, the byte attribute checking means 1 includes a byte attribute table 4, a byte attribute table creation unit 5, and a byte attribute check unit 3 that manages the byte attribute table 4 and the byte attribute table creation unit 5, and The attribute bit table means 2 includes an attribute bit table 6.71,
and 8, an attribute bit creation unit 9 that creates the attribute bit tables 6.7, and 8, an attribute bit check unit 10 that references and updates the elements of the attribute bit tables 6.7, 8, and attributes. Bit update section 1
1.

That is, in this embodiment, byte attribute table 4, which is the first table in which identifiers indicating the presence or absence of special character attributes are registered, and the numerical values of the bytes excluding the least significant byte of the character code, are used. a byte attribute checking unit 3 which is a byte attribute checking means for searching a table and verifying the existence of special character attributes; an attribute bit table 6.7 which is a second table storing numerical values indicating the type of special character attributes; , and 8, when the test result by the byte attribute checking means indicates the presence of a special character attribute, the second table is searched using the value of the least significant byte of the character code to indicate the type of special character and attribute. It also includes an attribute bit check section 10, which is an attribute bit check means for testing numerical values.

FIG. 3 is a diagram showing the storage state of special character attribute information of multi-byte characters. The n-byte characters 12 shown in Figure 2 are bytes b+, b2, , l), -1, bh in order from the beginning.
have. Here, in a system that handles characters with different byte lengths, it is assumed that the upper byte of a character with a shorter byte length is "0". The special characters and attributes of the n-byte characters 12 are stored as follows.

The first element of the first byte attribute table 13 points to the second byte attribute table 14, and the b2 element of the second byte attribute table 14 points to the next byte attribute table. -, byte attribute table 15,
, , Obtain the attribute bit table 16 from the l elements. n
The special character attribute of byte character 12 is attribute bit table 1.
It is stored in the b9th element of G as a bit string. The presence or absence of an attribute can be determined by turning each bit on or off. Here, the number of elements in byte attribute table 4 and attribute pit pulls 6.71 and 8 is the number of code types that can be expressed in one byte.

Next, taking a 2-byte character where 1 byte is 8 bits as an example,
The actual process flow will be explained. First, the process of checking whether attribute X exists for a character will be explained. However, the upper byte of the character A is A□, and the lower byte is still A. 0, and the bit representing attribute X is the X-th bit. FIG. 4 is a flowchart showing this process.

The byte attribute check section 3 checks No. A of the byte attribute table 4.
□References an element, and if it is a NULL pointer, the result is No. Also, if it is not a NULL pointer, the A-th of the attribute bit table pointed to by the A-th up element,
. .

Refer to the element, if the Xth bit is on, YES, and
If it is off, a NO result is returned.

For example, the character A is 1 piece, the upper byte AuP of the character A is 7A, and the lower byte AIow is 73. The byte attribute check section 3 checks 7 of the byte attribute table 4.
The A-th element is referred to, and if this value is a NULL pointer, the character attribute "saku" does not exist or has a default value. Also, if it is not a N U LI7 pointer, refer to the 73rd attribute bit table and set this value to rooo
If oo:3t Is, it can be seen that the 0th attribute and the 1st attribute exist.

However, this value may be "0".

Here, the attribute bit table contains 2 with the same first byte.
This is a table representing attribute information for 56 characters, of which 25
If the six characters do not have a special attribute, a NULL pointer is assigned to the corresponding element in the first byte attribute table without allocation. Therefore, there are 0 to '256 J attributes in the attribute bit table depending on the attribute of each character.The process of assigning attribute X to character A will be explained. FIG. 5 represents this process and is a flow chart. The byte attribute checking unit 3 refers to the AuP element of the byte attribute table 4, and if it is a NULL pointer, requests the attribute bit table creation unit 91 to create an attribute bit table, and writes the Stores a pointer to the newly created attribute bit table (all element bit strings are off). Then, when the attribute bit table is obtained, the attribute bit updating unit 11 updates the attribute bit table A. , turns on the Xth bit of the bit string of the element.

〔Effect of the invention〕

As explained above, the present invention creates a byte attribute table for bytes other than the lowest byte of a multi-byte character, and an attribute bit table for the lowest byte, respectively, as necessary, so hash calculations are performed. This method allows character attributes to be referenced and updated at high speed, and storage space is used efficiently, without complex processing such as There is an effect that can be done.

The present invention consumes a large amount of storage space when all characters have special character attributes, but in reality the number of characters with special character attributes is extremely small compared to all characters. You can expect it.

[Brief explanation of drawings]

FIG. 1 is a block configuration diagram showing the configuration of an embodiment of the present invention. FIG. 2 is a configuration diagram of a multi-byte character. FIG. 3 is an explanatory diagram showing a special character attribute search procedure for multi-byte characters. The fourth mechanism and FIG. 5 are flowcharts showing attribute management processing operations for multiple characters. l... Byte attribute checking means, 2... Attribute bit table means, 3... Byte attribute checking unit, 4.1
3.14.15... Byte attribute table, 5... Byte attribute table creation section, 6.7,..., 8.16...
Attribute pit table notebook 9...Attribute bit table creation section, 10...Attribute bit checking section, 11...Attribute bit updating section, 12...N-byte character.

Claims (1)

[Scope of Claims] 1. In an information processing device including definition means for defining special character attributes for character codes expressed in multiple bytes, the definition means: an identifier indicating the presence or absence of special character attributes; a first table in which is registered; a byte attribute check means for searching the first table using a combination of numerical values of bytes excluding the lowest byte of the character code to verify the existence of special character attributes; A second table in which a numerical value indicating the type is stored, and when the test result by the byte attribute checking means indicates the presence of a special character attribute, the second table is searched using the numerical value of the lowest byte of the character code. An information processing device comprising attribute bit checking means for verifying a numerical value indicating a type of special character attribute.