JPS5854450A - Multiprocessing system for text incorporation - Google Patents

Abstract

PURPOSE:To minimize a stack source and ease multiplexed levels by processing incorporated instructions in each incorporation level and releasing a stack means in every processing of one text incorporating information. CONSTITUTION:Incorporating instructions INC1, INC2, INC3 in an input text a01 are successively processed to prepare an intermediate text obtained by incorporating external texts a11, a12 into the input text a01. At that time, incorporating instructions INC4, INC5, INC6 having a level ''1'' are left unprocessed. In the next stage, incorporating instructions INC4, INC5, INC6 in said intermediate text are processed to prepare a new intermediate text in which texts a21, a22, a23 having a level ''2'' are incorporated. Incorporating instructions INC7-INC10 in the new intermediate text are left unprocessed. By repeating said operation also in levels ''3'' and ''4'', the whole incorporation of texts is completed.

Description

DETAILED DESCRIPTION OF THE INVENTION In language processing or text processing, the present invention provides text embedding processing for text embedding structures or nested structures that include instructions for embedding external text into text without limiting the multiplexing level. The present invention relates to a text multiple embedding processing method that can perform multiple embedding processes.

Conventionally, when performing text multiplex embedding processing, a method was used in which processing was performed sequentially from the beginning of the text while eliminating nesting at all levels, so the stack was only as deep as the multiplexing level. Capacity was needed. Additionally, there is a problem in that the level of text h multiplex embedding is conversely restricted due to the size of the stack capacity.

FIG. 1 shows an example of the conventional processing method described above. In the figure, asj represents text, and i is the number of multiplexing levels (i-1-0, 1, 2,
...) within the same level.3iIJ
li indicates ordinal number (j-1, ls,...).

Further, INC, to INC,, indicate text embedded instructions, and ■ to [phase] indicate the aSS order of the embedded instructions.

The text a, , is the main input stream and is the level @1'' built-in instruction INC,,INC,.

In order to process the 0 built-in instruction INC, which meets INC, k is an external text) 'll needs to be called.

text) 'll further includes the built-in instruction INC,, INC, of level @2'', and then the built-in instruction INC4■NC8, up to level @41 to J611 the built-in instruction I NO, ■NC2,III
Processed M book INC,, outside 11 Af't strike '11
# 'ml '1m' MI I '410 installation must be done.

After processing instruction,INC,, another instruction,I,of text,a,
N.C.

INC, is processed sequentially, but at this time, the instructions are processed in order from ■ to 00. Although a stack of a certain length is used for depthwise processing of these levels, it is impractical to create the input text by considering the number of multiplexing levels allowed in advance.
Conventionally, when the multiplexing level cannot be restricted, the problem has been solved by performing complicated processing.

In order to solve the nine problems mentioned above, the present invention has the following features:
The built-in instructions are processed for each built-in level, and the number of multiplexing levels allowed is virtually unlimited.

As a configuration for this purpose, the present invention provides a data processing system that includes a stack means, an external text storage means, and an external text storage means for processing an input text that internally includes external text embedded instructions in a multilayer structure. The first processing means for creating an intermediate text capable of text embedding, and the first processing means for creating an intermediate text capable of text embedding; extracting the text embedding information, reading the necessary external text from the external text storage means, and detecting a text embedding command similar to the processing performed on the input text by the first processing means for the external text that has appeared in the armpit;
a second processing means for loading a stack of text embedding information, creating an intermediate text into which the text can be embedded, and then performing embedding processing; and by repeatedly operating the second processing means, a multiple structure of text embedding is created. The stacking means are sequentially resolved and are characterized in that the stack means is freed as soon as one text embedding information has been processed to enable reuse for the processing of the next embedding level.

FIG. 2 is a detailed explanatory diagram of the present invention, and is shown in comparison with FIG. In the method of the present invention, in the first step, the built-in command lN of the input text a@1 is
Cl, INC,, INC, tl [form part NIL, embedding level @1'' external text a., ao input cat text)
'Create intermediate text embedded in @1. However, at this time, the level 11'' built-in instructions lNC4, INC,
, INC6 is left unprocessed without any CF1 solution ff4f. In the next step, the embedded instructions INC,,INC, in the previous intermediate text.

INC, and text of 12") 'U l
Create a new intermediate text that incorporates a□ and am, but as before, insert the embedded instruction INC in it.

1NC1 to INC1 remain unprocessed. When such operations are repeated at level 13'' and level 14'', all text incorporation is completed from K.

At each stage, unprocessed built-in instructions are loaded onto the stack, but when those unprocessed instructions are executed in the next stage, the data on that stack is no longer needed, so the stack is further It can be used repeatedly for the next stage of built-in instructions. Therefore, the depth of multiple embedding levels is not affected by the capacity of the stack.

FIG. 3 is a configuration diagram of this embodiment. In the figure, 1 is the text processing section of the data tin salt system, and phase! processing and phase processing. In addition, 2 is an endless structure stack that stores the information necessary to read external text and embed it in the text according to the text embedding instruction, and is used to process the queue of access requests to external text files. be done. Reference numeral 3 denotes a release pointer indicating the position of the previous entry at the head of the queue released upon completion of processing of the stack 2, that is, the last position of available free entries.

4 is a pointer indicating the earliest position among free entries; T
obl New file access request information is written here. 5 is the external text library 7'-ile.

6 to 8 correspond to FIG. 2, with one illustrative example showing text being processed in memory. 6 is input text 'II, text segment A.

, B, C, and D, external text embedding commands INC, , INC, , INCJ are inserted between fllt and t.
u are there.

In operation, the text processing unit l first performs phase! start processing. First text) 'II, then text)' @l to built-in instruction lNC5゜INC,,I
Take out NC, sequentially. Then, for each command, the requesting external text name '111 'l! t'II and the installation information such as the installation position of these external texts are written one after another to the work/entry position indicated by the free entry pointer 4 of the stack 2. Pointer 4 is incremented by +1 with the information of one built-in instruction. Furthermore, the release pointer 3 holds the eye of the eye one position before the start position (a, , )O. Text 6.1 In the embedding instruction location, if necessary, an indication of embedding requested, a stack entry address, etc. are entered, and an intermediate store capable of embedding processing is created.

In this way, the text '@IK related level @l'
When all of the text embedded information has been loaded onto the stack 2, the text processing unit 2 moves to the phase 3II.
Enter the IC.

In the phase seal, the external text embedding in stack 2 created in the previous 7A
Start from the entry whose release pointer 3 is +I. First, from the request information of the first en)+7, external text aml
and extract the text '0 from file 5.
Here, text a1. First, the external text embedding instruction INC is scanned.

Detects the external text) '11 set ＾ included information, 9
The same process is performed for the 0th next detected INC, which is written to the staff Z position indicated by the % entry pointer 4, and is repeated for the text 'jl @ '1m. A second intermediate text 8 is created in the same manner as the intermediate text γ was created.

The intermediate text 8 of jI2 is the level 2 external text embedding position (I-NC4)t (INC,), (IN
C,), but is structurally the same as Text 7.

In the figure, 'H-'N' indicates some segments of '11, towns,
# is a part of a segment. At this time, release pointer 3 points to entry (ao), indicating that only entry (a, . . . ) 1 has been released.

The level @21 external text embedding process is performed again in phase processing. When processing up to level 14'' is completed, a complete text 8 with no nesting is obtained.

As described above, according to the present invention, it is possible to minimize the stack resources used and relax the text embedding multiplexing level by J+14, thereby increasing the efficiency of text processing.

[Brief explanation of drawings]

FIG. 1 is an explanatory diagram of a conventional text multiple embedding system, FIG. 2 is an explanatory diagram of the text multiple embedding system of the present invention, and FIG. 3 is a configuration diagram of an embodiment. In the figure, l is the text processing unit, 2 is the stack, 3 is the free entry pointer, 4 is the free entry pointer, 5 is the external text library file, 6 is the input text, 72 and 8Fi processed intermediate text show. Patent applicant Fujiwaku, agent patent attorney Mori Co., Ltd. 1) Kan 71, born in 2 countries

Claims (1)

[Scope of Claims] A data processing system comprising: a stack means, an external text storage means, and a text embedding instruction for scanning an input text for processing an input text that includes an external text embedding instruction in a multiple structure; a first processing means that sequentially detects the information necessary for embedding the wrinkled text and sequentially loads the information necessary for incorporating the wrinkled text into the stacking means, and creates an intermediate text that can be embedded into the text; and a first processing means that extracts the text embedding information from the stacking means. The necessary external text is read from the external text storage means, and the first processing means detects a text embedding instruction similar to the processing performed on the input text for the read external text, and stacks the text embedding information. a second processing means for loading and creating an intermediate text that can be incorporated into the text, and then performing the embedding process, and by repeatedly operating the second processing means, the multiple structure of text embedding is sequentially resolved. The stacking means is characterized in that each time one text embedding information is processed, it is released and made available for reuse for processing the next embedding level. Embedded processing method. formula.