JPH04153880A - Partial string extraction processing system for character string - Google Patents

Abstract

PURPOSE:To divide an input character string into specific partial strings and extract a partial string which meets specific requirements by storing attributes in an array type storage means in specific order by an attribute decision part and comparing the storage contents of the array type storage area with an extraction condition decision string by an extraction part. CONSTITUTION:A division part 2 divides the input character string which is read by a reader 7 into character primitive strings according to predetermined conditions and the attribute decision part 3 decides the attributes of the respective divided character primitive strings by using predetermined conditions and stores the results in the array type storage area 6 in the specific order. An extraction condition generation part 4 generate the extraction condition decision string from extraction conditions read by the reader 7 and the extraction part 5 extracts the character primitive string corresponding to a string of matching attributes obtained by comparing the storage contents of the array type storage area 6, i.e. attributes with the extraction condition string. Consequently, only the character string matching the specific extraction requirements can automatically be extracted.

Description

DETAILED DESCRIPTION OF THE INVENTION [Summary] This invention relates to a character string substring extraction processing method for extracting a substring that satisfies a predetermined condition from a character primitive string.

A dividing unit that divides an input character string into character primitive strings for the purpose of dividing an input character string into predetermined substrings and extracting substrings that meet predetermined conditions; and an attribute that determines the attributes of the character primitive strings. a discriminator, an extraction condition generation section that generates an extraction condition discriminant string from the extraction conditions, an extraction section that extracts the character primitive string that matches the extraction condition, and an array type storage area that stores the attributes in a predetermined order. The attribute discriminator stores the attributes in the array type storage area in the predetermined order, and the extraction unit compares the contents stored in the array type storage area with the extraction condition discrimination string. Accordingly, the above extraction is performed.

[Industrial application field]

The present invention relates to a substring extraction processing method for character strings.

More specifically, the present invention relates to a character string substring extraction processing method for extracting a substring that satisfies a predetermined condition from a single character primitive string.

From among strings created according to certain rules.

If it is possible to automatically extract a subsequence that meets a predetermined condition, it is effective for checking whether the contents of the subsequence are correct or not.

[Conventional technology]

For example, if a product is manufactured by assembling multiple parts, one assembly drawing is usually created.

The assembly drawings include information on which parts and how many are to be used (parts information) in addition to the specifications in m-stand. Part information includes: 1. Quantity (how many pieces to use) 1. Product name (general name of the part) 2. Drawing number (number that specifies the details of the part)
Consists of etc.

On the other hand, a 3-ordinary-2 parts list (SK) is also created for the purpose of purchasing, managing, etc. parts used for m-setting. The parts list consists of five parts information.

Here, let us consider the relationship between the 1 assembly drawing and the parts list.9 Originally, the 1 parts list is created based on the 1Mi vertical drawing after completion. However, since this cannot shorten the time required to complete the manufacturing process, a parts list is created and an order is placed after the assembly drawings have been created to a certain extent. Then, after completing one assembly drawing, the parts list is revised.

In order to make this correction, it is necessary to check the 1 part information by comparing the 1Mi elevation and the part 7 nost.

[Problem to be solved by the invention]

In the prior art described above, checking by comparing information on one part is performed manually by a designer or the like.

This caused the following problems. That is, the component information in one assembly drawing is written near (at a predetermined position) the component, so it is scattered throughout the two drawings, and its position is difficult to understand. Also, since there is a mixture of figures and text, it is difficult to understand from this point of view as well. Furthermore, in addition to single part information, there are various pieces of information in the assembly diagram, and since both of these pieces of information take the form of "character strings," it is difficult to determine which character string is part information. Furthermore, there are several types of formats for character strings of the 5-component information, and the formant may be changed.

In such a case, checking by one person cannot accurately respond to changes, etc., and there is a possibility that errors may occur.

SUMMARY OF THE INVENTION An object of the present invention is to provide a character string substring extraction processing method that makes it possible to divide a character string manually into predetermined substrings and extract those that meet predetermined conditions.

[Means to solve the problem]

FIG. 1 is a basic configuration diagram of the present invention, and shows a data processing apparatus according to the present invention.

In FIG. 1, 1 is a processing device, 2 is a division unit 3 is an attribute discrimination unit, 4 is an extraction condition generation unit, 5 is an extraction unit, 6 is an array type storage area, 7 is a reading device, and 8 is an output device. .

The processing device l consists of a central processing unit (CPU) and a memory.

The dividing unit 2 divides the input character string into character primitive strings (ie, substrings).

The attribute determination unit 3 determines the attribute of each character primitive string divided into two.

The extraction condition generation unit 4 generates an extraction condition discriminant string from the extraction conditions.

The extracting unit 5 extracts a string of character primitives that meets the extraction conditions from among the seven divided character primitive strings.

The array type storage area 6 is provided on the memory of the processing device 1 and stores attributes of a two-character primitive string in a predetermined order. That is, for this purpose, an array type having a predetermined size is used.

The reading device 7 reads input character strings and extraction conditions on one drawing.

The output device 8 outputs the extracted character primitive string.

[Effect]

FIG. 2 is an explanatory diagram of the operation of the present invention, and shows a processing flow of the present invention.

■ The dividing unit 2 divides the input character string read by the reading device 7 into a one-character primitive string (
subsequence).

■ The attribute determination unit 3 determines the attribute of each of the character primitive strings divided by the first division unit 2 using predetermined conditions, and stores the results in the array type storage area 6 in a predetermined order (array). do.

This provides an attribute string that describes the attributes.

(2) The extraction condition generation unit 4 generates an extraction condition discrimination string from the extraction conditions read by the reading device 7.

■ The extraction unit 5 compares the contents stored in the array type storage area 6, that is, each of the attribute strings, with the extraction condition discrimination string, and selects a character primitive string corresponding to the matching attribute string as one that meets the extraction conditions. Extract as. Then, the output device FB outputs the character primitive string by, for example, printing.

According to the above, character strings on one drawing can be input without distinction, and only a character string (partial string) 1, for example, 1 part information that meets a predetermined extraction condition can be automatically extracted from among the character strings.

Furthermore, by inputting extraction conditions and generating a discriminant string, it is possible to easily respond to changes in the format of component information, for example.

〔Example〕

FIG. 3 is a block diagram of an embodiment, showing details of the data processing apparatus shown in FIG.

Input character string reading device 10 and extraction condition reading device 14
constitutes the reading device 7.

The character primitive decomposition unit 11 corresponds to the two-part dividing unit 2, and stores the character primitives divided into one into the character primitive storage area 1.
Store in 2. The character primitive storage area 12 is provided at a predetermined location in the memory of the processing device 1.

The discriminant value determining unit 13 corresponds to the two-attribute discriminating unit 3, takes out a character primitive from the character primitive storage area 12, determines its attributes, that is, two discriminant values, and stores it in an array type IJI.
Store in area 6.

The extraction condition discriminant string generation section 15 corresponds to the extraction condition generation section 4, and stores the generated extraction condition discriminant string in the extraction condition discriminant string storage area 16. This area 16 is.

It is provided at a predetermined location in the memory of the processing device 1.

The array type storage area extraction condition discriminant string comparison unit 17 is.

The extraction section 5 is configured together with the character primitive extraction section 18, and compares each of the contents of the array type storage area 6 with each of the extraction condition discrimination strings. Based on the result of this comparison, the two-character primitive extraction unit 18 extracts the one-character primitive storage area 1.
Extract matching character primitives from 2.

The character primitive output device 19 corresponds to the output device 8 and outputs the extracted character primitives.
．． 13, 15, 17 and 18.

This is realized by the CPU of the processing bag N1 and the processing program existing on its memory.

FIG. 4 is an explanatory diagram of character primitives.

In FIG. 4(a), each of ■ to [phase] is a character primitive, and "!" is a character primitive string.

Character primitive string! is a predetermined symbol (delimiter) 1 that is equal to the input string, for example, "mouth (shall represent a blank)", r, (period).

Separated by ", (comma)" etc. Each of the separated character strings from ■ to [phase] are character primitives. Therefore, the one-character primitive ■ to 0.

It may contain multiple characters, and it would be meaningless to divide it further. for example.

If you break down ``Koneji'' in ■ into ``koJ, r$ji'', it loses its meaning.

Note that this character primitive string 1 mainly consists of parts information (2 quantities, 1 product name, 1 drawing number).

Further, such division processing is executed by the 1 division unit 2 according to the division processing flow shown in FIG.

That is, the input character string read by the reading device 7 is temporarily stored in a predetermined storage area (not shown).

The dividing unit 2 determines whether or not each character extracted from the input character string is a single delimiter, and uses this to decompose the character into character primitives.

In addition, in FIG. 8, Ne is the number of input characters.

Np is the number of character primitives.

FIG. 4(b) shows the attributes of each of the two-character primitives ``■'' to ``phase'' using the discriminant value ``1'' or ``0''. In this case, a one-character primitive string! Since we are trying to extract component information from , the attribute indicates which of the quantity, product name, and drawing number of the component information.

For example, "Screw" in ■ is a "product name", so the discriminant values for 2 product names are set to "1", and the discriminant values for 5 quantities and drawing numbers are set to "OJ". ] The discriminant value is determined in the same way. For example, 2
If the character primitive consists of katakana, it is determined that it is a "product name," if it consists of numbers, it is determined that it is a "quantity," and if it consists of an alphanumeric character, it is determined that it is a "picture number."

Note that the determination method is not limited to this.

In addition, in general, if the number of attributes (discriminant values) of a one-character primitive is Na, then the attribute is (al+a!+...a N11-1+ a m
It is represented by the set al. When a certain character primitive has the nth attribute (the nth discriminant value is "1"), 5 ak =1 (k=n) =O(kin).

FIG. 5 is an explanatory diagram of the extraction condition discrimination sequence.

Assume that you are currently trying to extract information on two parts.

It is also assumed that the two-part information is composed of only three items, one quantity, one item name, and one drawing number, written in this order.

In this case, 1 parts information 2 essentially consists of all of the above three items filled in, so the extraction condition is ``
Quantity: 1 item name: 1 drawing number.

is input. This results in 5 extraction condition discrimination columns.

rll, l". In other words, three character strings exist and have been entered: quantity 1 product name 5 drawing number).

As a result, those whose discriminant value is "l" are extracted,
The character string 2, that is, the 2-part information, that meets the extraction conditions is set.

Furthermore, when the quantity of three parts information is "1", this may be omitted and entered. Therefore, when extracting such information as part information, the extraction condition is specified as ``[omitted] 10 product names and 10 drawing numbers.'' This results in
The extraction condition discriminant string is "rol,l".

Furthermore, since two-part information may require only one drawing number and no product name, this may be omitted and entered. Therefore, in this case, the extraction condition is "Quantity + (omitted) 10 drawing number j," and the extraction condition discrimination column is ro, 0.IJ or rl,
0.0".

According to the present invention, various extraction conditions can be set as desired in this way. That is, by providing the extraction condition reading device 14, depending on the extraction target.

Flexible extraction conditions can be set. By setting various extraction conditions in this manner, the extraction rate of the extraction target can be made extremely high.

Note that the extraction condition may consist of four or more items. For example, suppose that the extraction condition consists of eight items and the third, fifth, and sixth items are omitted. In this case, the extraction condition determination column is.

＊＋＋kOO011 $$00001* *0O100**.

11000***.

There are four.

Further, such a generation process is executed by the extraction condition generation unit 4 according to the generation process flow shown in FIG.

In FIG. 9, element dij is "0" or "1".

If omitted, "*" is stored.

Determine the extraction condition discriminant column. Here, j is "1" to rN
a", which corresponds to the X direction in FIG. 5, and i is "1".
” to rNdJ (number of extraction condition discriminant column),
This corresponds to the X direction in FIG.

If all the elements of the i-th column in the X direction are represented by Di, then 2 This Di indicates one extraction condition discriminant string.

FIG. 6 is an explanatory diagram of the stored state.

Now, if we focus on the two-character primitive ■, there is a possibility that this is part of the component information, for example.

(Quantity: 10 product names, 10 drawing numbers) - (omitted + ■ + ■) (Quantity: 10 product names, 10 drawing numbers) - (■ + ■ + ■) (Quantity: 0.
There are three combinations: 10■+■). Therefore, for each of these extraction condition discriminant columns r1.1. It is necessary to check whether it matches with IJ or rO, 1°1.

Therefore, first, an array type storage area 6 is prepared.

The number of horizontal elements in this array is the number Na of parts information items, and the number of vertical elements is (number of character primitives Np+(number of items - 1)). Then, the discrimination values of the attributes of each of the one-character primitives ``■'' to ``stock'' are stored in this array in a predetermined order. Looking at the character primitive ■(n), its drawing number, product name, and quantity are each in the order of combination "2".
(n)” “’3 (n+1)” and “4 (n+2)”
is stored in the area of the corresponding item. Similarly for the others, it is possible to create all attribute strings of combinations that may constitute one part information for each of the one-character primitives ① to [phase] using 0 or more.

Such discrimination and storage processing is carried out by the attribute discriminator 3 in the first
This is executed according to the discrimination storage processing flow shown in FIG.

In FIG. 10, an attribute string is generated by storing attribute aV in element txy. Here, X is rl
, to "Na", which corresponds to the X direction in FIG. 6, and y takes values from "1" to "rNp +Na-1", and corresponds to the X direction in FIG. Let Ti represent all the elements in the i-th column in the X direction.

This Ti indicates one attribute string.

FIG. 7 is an explanatory diagram of the extraction process, and shows all of FIG. 3 to FIG. 6 together.

As described above, the input character string is decomposed into the character primitives ``■'' to ``stock'', their attributes are determined, and 1 discriminant values are stored in the array type storage area 6 in a predetermined order. On the other hand, from the extraction conditions, the extraction condition discriminant string rl, 1. IJ and ro, 1. Get IJ.

Then, each combination order in the array type storage area 6 is compared with the extraction condition discrimination string. As a result, 1 combination order “6” and 13”
matches. The former is part information whose quantity, two product name, and drawing number are each composed of two-character primitives ■■ and ■. For the latter, the quantity 1 product name and drawing number are respectively
This is part information composed of the omitted one-character primitive @ and ■.

In this way, two parts information from one character primitive string is
One is extracted. That is, 7 "3. Four screws M3 x 30" is 2 drawing number rM3X3. This means that the number of "screws" specified by is "3", and "four screws M7 x 70" is one drawing number "M7 x
7" means that the number of "screws" specified is "1". The parts information extraction rate in this case is 100%. Therefore, subsequent work can be done more efficiently.

The extraction process described above is executed by the extraction unit 5 according to the extraction process flow shown in FIG.

In FIG. 11, the extraction condition discrimination column Dj and the attribute column T+
If they match and the attribute is specified, the character primitive corresponding to the attribute in the attribute string Ti is output as the attribute. Here, if qh=0, the th attribute is specified and Qk
=1 means that the attribute is omitted.

〔Effect of the invention〕

As explained above, according to the present invention, attributes of character primitives obtained by dividing one character string are arranged in a predetermined order and extracted in a substring extraction process of a character string that extracts substrings that meet predetermined conditions. By comparing the condition discriminant string, character primitives that meet the desired extraction conditions can be extracted, which can automate the process of extracting the subsequence when performing work using two subsequences, increasing the efficiency of one task. can be significantly improved.

[Brief explanation of drawings]

FIG. 1 is a diagram showing the principle configuration of the present invention. FIG. 2 is an explanatory diagram of the operation of the present invention. FIG. 3 is a diagram showing the configuration of an embodiment. FIG. 4 is an explanatory diagram of character primitives. FIG. 5 is an explanatory diagram of extraction condition discrimination strings. FIG. 6 is a diagram illustrating the storage state; FIG. 7 is a diagram illustrating the extraction process. FIG. 8 is a division processing flow. FIG. 9 is a generation processing flow. FIG. 10 is a discrimination storage processing flow. FIG. 11 is an extraction processing flow. , 5 is an extraction unit, 6 is an array type storage area, 7 is a reading device, and 8 is an output device. Patent Applicant: BFN Co., Ltd. Agent Patent Attorney: Yo Hase Kojo (2 others) Action explanatory diagram of the present invention Figure 3 Embodiment Figure 3 Generation process flow Figure 9 (Part 1) Housing process flow diagram (Part 2) Generation Processing Flow Section 9 (Part 3) Generation Processing Flow Fig. 9 (Part 4) Separate Storage Processing Flow Fig. 10 Extraction Process 11 Process Flow Diagram (Part 1) Extraction Process 11 1 ! Flow diagram (Part 2)

Claims (2)

[Claims] (1) A dividing unit (2) that divides an input character string into character primitive strings; and an attribute determination unit (2) that determines the attributes of the character primitive strings.
3), an extraction condition generation unit (4) that generates an extraction condition discriminant string from the extraction conditions, an extraction unit (5) that extracts the character primitive string that matches the extraction condition, and stores the attributes in a predetermined order. an array type storage area (6) in which the attribute discriminator (3) stores the attributes in the array type storage area (6) in the predetermined order, and the extraction unit (5) stores the attributes in the array type storage area (6); A substring extraction processing method for a character string, characterized in that the extraction is performed by comparing the content stored in a type storage area (6) with the extraction condition discriminant string. (2) The input character string and extraction conditions are sent to the reading device (7
2. The method for extracting a substring of a character string according to claim 1, wherein the character string is read as follows.