DE3248189A1 - Method for identifying interleaved source program blocks - Google Patents

Abstract

Method for identifying (S) interleaved source program blocks which, composed of steps or rows, are written into registers or stores of an electronic data processing system, particularly a compiler, an identification (6+, 5-) marking such block boundaries and, particularly, corresponding to a change in the depth of interleaving, being inserted in each case in first and, respectively, last steps or rows of each block. The digital identification (6) corresponding to a graphical character, particularly an alphabetic and/or numeric character, is also uniformly inserted within the block (B6) in each case at the same defined position (S) of each step or respectively of each row of the block (B6). <IMAGE>

Description

Procedure for the identification of interchangeable

telten source program blocks.

The invention represents a further development of the in the preamble of Claim 1 defined by DE-OS 29 39 194 known method. The blocks of the known method are each subroutines, more precisely, in general, about deeply nested subroutines of subroutines. Both the first lines and the last lines of each such block or subroutines are each marked by an indicator that indicates the change the nesting depth of the relevant subroutine. This mark is attached there to block processing errors, i.e. subroutines to be able to search for data, with the relevant characteristics there mainly for Limitation of the number of lines of the program to be searched.

In the invention, however, the block in question does not need each add your own filter program. The relevant block can also be on, off for whatever reason, be a special section of the source program, that is E.g. just a loop or a somehow uniform section form within the source program. The invention is also not limited to browsing limited by such individual blocks after errors in the processing of data. In the case of the invention, the mark in question can, for example, also serve to the affiliation to the relevant block immediately based on each individual line to be able to determine, or e.g. to very quickly a whole block against one others, Replace the already prepared block due to mordernization of the source program to be able to. If the indicator has meter reading displays at the beginning and / or End of the relevant block with the help of a counter that counts up and down are acidified, is avoided in the invention that by a one-time Error in a license plate or due to a one-off error in license plate scanning entrained errors occur in the number erstandsanieigen, which errors about the affiliation of a line to a special block, e.g. incorrect identifications and processing of data.

The method according to the invention thus allows such a machine recognizable marking that after completion of the procedure reliable without running Counting at any time, even on small sections of a block, e.g. on one Line that can be determined to belong to a certain block.

The object of the invention is specified in claim 1 Measures resolved.

The additional measures specified in the two subclaims allow additional advantages, namely the measure according to claim 2 the identifier in addition to the nesting depth of the relevant block also the Change direction or size of this nesting depth in relation to the previous one or on the following block or on the while the CODES des Block to indicate possible changes in the nesting depth by jumping, and 3, by means of a special counter, the nesting depth of the relevant Automatically insert blocks as identifiers.

The invention is further based on the example shown in the figure explained.

The example shown in the figure represents a section of a Source- # rogram with very strong nesting of its blocks. This excerpt of the source program written in registers or memories is here in sections reproduced as its characters by a later expression or by a Subsequent screen display, e.g. for visual verification of membership of a special block, can be obtained.

In the example shown in the figure, the point S is where according to the invention In each case the identifier is inserted, always at the beginning of each line Z des relevant block. Not only in the first and last lines Z of each such Each block, see the transition from block BS to block B6, is one of these Block change or identifier marking the block affiliation inserted, cf. the marks 5 and 6+ marking such a change at this transition from B5 to B6. Also within the relevant block, cf.

the block B6, the relevant identifier, here 6, is always uniform at the same defined point S of each of these lines Z of block 36 inserted.

In a corresponding manner, the blocks B5 shown are also replaced by the Marks 5 at the relevant point S of each line Z marked. The concerned Lines Z therefore not only contain some commands and addresses as CODE, but in addition, the relevant identifier of the block on a typical, for all Lines Z uniformly selected position S, in particular at the beginning of the line.

Each individual line Z is therefore immediately associated with one special block, e.g. to block 36, very much easily even if the block in question, see B6 itself many hundreds of lines Z contains. The invention thereby facilitates in particular the automatic complete Exchanging a block for another block that has been provided is very strong.

Not just with one blindly from the middle of the source program line Z read out by machine on a trial basis, but also for screen displays of only short excerpts from the source program, is therefore always itself in the case of comparatively extremely long blocks, the affiliation is immediately reliable Detectable every single line for a specific block.

In the invention, a frequent major difficulty of the complex is thus Nesting of the individual blocks of the program solved, namely that with the Creation, modification, maintenance and / or maintenance of already existing source programs, i.e. in the end with the entire software support, complex block nesting can only be broken down into their individual blocks with great difficulty.

It is true that some of the program languages have to write a certain identifier of blocks, especially of the ren nesting depths, e.g. in the form of heavy countable space bar presses, e.g. Algol, Pascal or Chill. Still is it has so far also been the case with this program language, if the invention is not used, still too cumbersome, e.g. find out the nesting depth of blocks to make, or immediately determine the block size, especially because the number the space bar presses are difficult to count on the screen, especially when one is displayed Block cutout is much shorter than the total length of the block. Moreover, have to in this prior art too many storage locations at that line position S of can be reserved in advance for printing spaces. In the invention, however, in a few storage locations per line, the marking indicator simple Way, where you have to go into each line or

in each step of the block the respective, easily retrievable, comparatively very short number plates, e.g.

as a numeric or alphabetic value.

If the relevant identifier specifies the nesting depth you can use the entire source program to automatically write the identifier Search line by line for CODES or its typical commands or addresses, each signifying the beginning or the end of such a block. If such a Step or such a line is found, the counter reading of a counter is namely a nesting depth counter, forwards or backwards accordingly controlled, the counter reading of this counter then as an indicator where S is inserted automatically.

In addition, you can add an additional symbol to this indicator, that shows the direction or size of the change, in particular the fact whether the nesting depth is now larger or smaller.

However, if the nesting depth remains with the relevant line or Step Z remains constant, then such an additional symbol can be completely omitted become, cf.

the lines within block B6 in the figure. So if a block see B6, a search is carried out for CODES which indicate the beginning or the end of a Blockes mean, and if in this search in one line or in one step No such CODE is found, the old counter reading of the relevant nesting depth is used Counter as an identifier at the relevant point S of this line or of this step inserted.

Additional symbols can also be attached which include a Indicate that a jump command belongs to the relevant line Z, which enables a jump to another block with a different nesting depth - whereby even the nesting depth of this other block is also included the position S of the line or of the step Z can be inserted, see in the figure the identifier 65 3 in the CODE OPQ in block B6. Such additional characters become the automatic and optical readability of the source program even stronger relieved.

All blocks, which according to the invention or according to the described developments of the invention were marked '*> ind furthermore without restriction in other languages translatable, especially into machine language, without having to exchange two Blocks a corresponding additional maintenance of the source program or the machine language program is needed. After the new source program has been translated, the translation protocol Include the identifier, which in particular can indicate the nesting depth, so that a post-processing for documentation purposes at least with regard to this Marking can be omitted. If for some reason an additional one If post-processing is desired, there is another advantage: by inserting the identifier at the firmly defined points 5 can do this post-processing without large expenditure of computation time can be carried out because belonging to the individual blocks can be determined immediately.

The invention is also not restricted to a specific programming language bound for the source program. You can *) with the code in each case as well can be treated as a mere commentary namely even with everyone free-formatted higher program language can be used, e.g. also with the Language chill for source programs for telephone switching technology.

3 claims 1 figure - blank page -

Claims (3)

Claims. @ Procedure for marking (s) nested Source program blocks, which are composed of steps or lines in registers or storage of an EDP, in particular a compiler, are written in, wherein In the first or last steps or lines of each block there is one such block boundary Marking identifier, in particular corresponding to the change in the nesting depth (6+, 5-) is inserted, that is, that the one graphic, in particular alphabetical and / or numerical, characters digital identification (6) uniformly also within the block (36) in each case on the same defined point (S) of each scrap or each line of the Block (36) is inserted. 2. The method according to claim 1, d a d u r c h g e k e n n z e i c h n e t that, if applicable, the indicator indicates the direction of a change The additional character (+, -,>, <) indicating the nesting is added, that specifies whether in the relevant step or in the relevant line a branch (>, <) (: and / or and / or a jump to a lower one (-) or greater (+) nesting depth is performed. 3. The method according to claim 1 or 2, d a d u r c h g e k e n n z e i c h n e t that no or not all marks (5) The source program contained in the EDP automatically searches for such special terms or commands are searched which indicate the beginning (HIK) and / or the end (RST) of the Block (B6) indicate that these special terms resp. Commands control the count (6) of a nesting depth counter, and that the count (6) each as a mark (S) on the relevant Position of each step or each line is inserted.