JPH02255985A - Reserved word retrieving method - Google Patents

Abstract

PURPOSE:To retrieve a reserved word at high speed by expressing reserved word information by three structure, recognizing whether a character string is the reserved word or not and executing the retrieval. CONSTITUTION:Reserved word information 1 are expressed by the tree structure and '0' is defined to the part of a preceding state number 8 in end information 6 and continuation information 7 corresponding to a position number 3 in state information 2. When the retrieval is started, a character position number-th character code is taken out from the character string to be retrieved. The state information 2, for which the position number 3 is equal to the character position, are taken out and association 4, for which a correspondent character code 5 is equal to a code 5, is taken out. Next, the information 6 and 7, for which the preceding state number 8 is equal to a state number, are taken out. When there are the information 7, processing is executed from the state information 2 with next state information 10 as the state number and repeated until the information 7 is eliminated. When there are the information 6, the character string is recognized as the reserved word and retrieved from the tree structure by an identifier 9 at high speed.

Description

[Detailed Description of the Invention] [Field of Commercial Application] The present invention relates to a reserved word search method in one-sentence IF processing.
It relates to a method of searching for a specific character string from a series of characters.

[Conventional technology]

FIG. 4 is a flowchart showing a conventional search algorithm.

When you start a search, first find the variable n? Think about it and set the initial value l to it (Step Sword) Next, refer to all the reserved word libraries in which the character string is registered as a reserved word. Compare whether it matches the first reserved word in the library (
step example).

If there is a match here, it is recognized as a reserved word (step!25) and the search ends1.On the other hand, if there is no match, the order n in which the reserved word is registered is registered in the reserved word library. Compared to the total number of reserved words NK
A check (step @) is made to see if the value is N. At this time, if n is N, then substitute n + 1 for n (step unit), return to step 5!3, and process step RI ~ 2? repeat. If n(N does not hold in step 1), the character string is recognized as not being a reserved word (step -), and the search ends.

This will be explained using a specific example.

For example, consider a case where the words shown in FIG. 6 are registered in the reserved word library and the character strings da A Nn-' and "decrement" are to be searched.

First, it is determined whether the character string "AND" is a reserved word or not. First, it is checked whether it matches the first registered word '9 decode' in the reserved word library. Since there is no match here, the second registered "change)' is checked to see if it matches. It doesn't match here either, so next 8
It is checked whether or not it matches the "AND" registered in the second place. Since they match here, “AN”
D" is recognized as a reserved word.Furthermore, in order to determine whether the string "decrement)) is a reserved word or not, it is necessary to use the l-th character string registered in the reserved word library. Whether or not they match in order is checked in the same way as above.

Since it is checked that there is no match with the 10th "k〈", it is recognized that this character string is not a reserved word.

[Problem to be solved by the invention] In conventional algorithms, when determining whether a character string to be searched is a reserved word, it is necessary to compare the reserved words registered in the reserved word library word by word. Must be. It can only be determined that a string does not correspond to a reserved word after referring to all reserved words registered in the reserved word library.

Therefore, there is a problem in that it takes time to search for reserved words.

The present invention has been made to solve the above-mentioned problems, and an object of the present invention is to provide a high-speed reserved word search method.

[Means to solve the problem]

The reserved word search method according to the present invention includes an expression step of expressing reserved word information in this structure, and a search step of searching for a character string to be searched using the reserved word information expressed by this expression step. We are prepared.

[Effect]

The expression step in the present invention expresses the reserved word information using this structure. Therefore, it has the effect of speeding up the search for a required character string.

〔Example〕

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

FIG. 1 is a diagram showing the structure of reserved word information in one embodiment of the present invention. This expresses the reserved word information il+ as character position status information (hereinafter abbreviated as status information) (21).

Here, the reserved word information fi+ is contained in multiple pieces of status information (2). This status information: 2) is the character position number (hereinafter abbreviated as position number) (31 and The character code (hereinafter abbreviated as the corresponding character code) 16) is within the range of +41 and corresponds to the character position of association 141 (abbreviated as the corresponding character code).
, pattern end information (hereinafter abbreviated as end information) +81
, pattern continuation information (hereinafter abbreviated as continuation information) (71
It has settled down since.

In addition, in the end @ information 16), the state number C specified at the previous character position and the following, consisting of the previous state number) (8) and the pattern identifier (hereinafter abbreviated as the identifier) (9), The continuation information (71 is the previous state number (81 and the state number used in the next character position (hereinafter abbreviated as primary state number)) is included and stands from UO+.

The position number (1 of 31) in the status information 12) of this structure above
O is defined in the previous state number +81 portion of the end information (6) and continuation information () corresponding to .

Next, the operation will be explained. FIG. 2 is a flowchart showing the search algorithm of the present invention. When the search starts, two variables are set as initial values of state number and character position, each of which is 0.1i. (Step a
1)). Next, the character code at the character position is extracted from the character string to be searched (step u'lJ). Step ll31) extracts the state information (2) where the position number (3) is equal to the character position; extracts the association 141 where the 'R15 character code code is the same as the character code (step U >. In step αG, the previous state number (8 ) is equal to the state number end fii161, continuation information (71
Take out.

At step aα, it is checked whether or not there is continuation information (7), and if there is continuation information (7), the next state information tlol
'i state number (step II?)), returns to step 021, and repeats steps u21 to (no step H).

If there is no continuation information 17), end information 6) in step 4.
If there is end information 1G), it is recognized that the word is a reserved word, and the search ends when the identifier (step 1211 of repeating 9) is found.

If there is no end information (61), the character string that does not correspond to the reserved word is returned (step ■), and the search ends. For example, suppose that the reserved word information shown in FIG. Example 11 AN D"""Decode""Change A, N D"Start"NO"○R"Write "Pabit〃"m〃"<<" is registered.

Find the case of searching for "AND" and "decrement J) as character strings. First, to determine whether "AND" can be used as a reserved word, first select position number 131 (STA
1 from the corresponding "annotations" of
Character code 165' of character lA1 corresponds to character code 1
5) Extract the motsu (65N engineering L ((Ojri))? Since there is continuation information 17+ (0,2), 9 [(STATE)
. 2) Corresponding association (41? see. 2nd character I N l's character code 78'' vs. 15 character code - l51 (mo-') (78 NIL (12, 21) l
Continuation information 17+ whose previous state number (8) is the state number 2 indicated by the next state number (10) of the previous stage.
Since we have (2, 2), we look at the association 14) that corresponds to (STATE, 3). The character code 1881 of the 8th character lDI has the wick character code i51 (68C (
2, AND)))? Get it out. Next state number tl of previous stage
Also change the state number 2 indicated by ol to the previous state number (8)'
:) No continuation information (7), ending emotion ill (2,A
Since there is an identifier (91ANDf!: return)
It is recognized that AND'' is a reserved word.Furthermore, to determine whether ['decrement 1 is a reserved word?
From 4), 1'; C half moon 1 de 1 character code '95
4!9' i compatible character code 15) ni motsu (9543N
Take out L((0,41)l.

Since there is a continuous emotion cap phoenix I (0,41) (8 TATL, 2
) pair H5-r association (41k * r.

The character code of the 2-character month 171 “9519'i diagonal character code 16) has an anciation (41 is missing, so °
Decrement 1 is recognized as a character string corresponding to a reserved word.

By the above method, recognition that a string is a reserved word,
Recognition of non-reserved words becomes faster.

Furthermore, by allocating character position state information (21 to 21 processors), pipeline processing becomes possible.

〔Effect of the invention〕

As described above, by expressing reserved word information in wooden beaks and using A, it is possible to quickly recognize that a 5-character string is a reserved word, and that it is not a reserved word. This has the effect of increasing speed.

[Brief explanation of drawings]

Fig. 1 is a block diagram showing the arrangement of Ichikyuen according to the present invention, Fig. 2 is a flowchart showing the algorithm of reserved word search, Fig. 3 is a diagram showing an example of reserved word information, and Fig. 4 is a conventional diagram. The flowchart showing the algorithm, Figure 5 is a sample of the contents of the reserved word library? FIG. In the figure, ([I is the reserved word information %qυ~(211 is each step. In each figure, the symbol --8' indicates the same or corresponding part. Agent Masuo Oiwa Figure 3 2 Figure 4 Figure 5 Procedural amendment (voluntary) 1. Plain patent application for indication of the case @ 1-・12941 No. 3. Person making the amendment Representative IL, N. Kimoriya 4. Agent 5. Amendment Column for detailed description of the invention in the statement of subject matter, and the drawings.Contents of drawing amendments (1) The description is corrected as follows.(2) Attached documents that correct the third figure of the cylinder in the drawings as shown in the attached sheet. Inventory (1) Corrected drawings (Fig. 3, one or more copies)

Claims (1)

[Claims] A reserved word search method comprising: an expression step of expressing reserved word information in a tree structure; and a search step of searching for a character string to be searched using the reserved word information expressed by this expression step. .