System and method for translating source code

[0002] 1. Technical Field of the Invention

[0003] This invention pertains to source code translation. More particularly, it relates to a safe system and method for reversibly translating source code between unlike character sets.

[0004] 2. Background Art

[0005] More and more often companies are using workstation tools to change or maintain mainframe software source code. However, the mainframes use the IBM EBCDIC character encoding system while workstations use the ASCII system. For the most used characters, there are safe and reversible mappings between the two encoding systems. Unfortunately, this is not true for all possible characters, because there are no equivalents in both systems, because of varying locations of characters within variations of the same system, and because some characters have special control meanings in each system.

[0006] Generally, each programming language uses a set of characters for its operational syntax. By ensuring these characters are mapped reversibly, a translated program can be made to compile on both the mainframe and the workstation. However, mainframe software source code may contain any of the 256 possible characters in program comments and program text literals. Although the usual frequency of occurrence of unsafe characters is very small, if any of those characters are translated to workstation control characters, the program may not compile or be editable or even viewable. If any of those characters are not mapped in a reversible way, they will be different when the program is returned to the mainframe. At best, this may make a comment less understandable; at worst it will change the data in a program literal with unpredictable and possibly disastrous results if not corrected.

[0007] Generally, the upload and download character translation mapping tables are synchronized as much as possible for the most used characters. This often fails because it is not always possible to use the same tool or synchronized tables for both the upload and download. It also doesn't allow all 256 possible values to be safely translated. Usually, separate fix-up work is required to detect and correct translation errors.

[0008] Consequently, there is a need in the art for a translation system and method which eliminates the need for maintaining synchronized translation mapping tables and the need for separate fix-up work.

[0009] Another solution is to use a special escape character before an untranslated value and then convert the untranslated value to a two character value in hexadecimal format. This solution does not work with source code because it changes the length of certain records and, more importantly, it changes the length of program literals, which could make the program not compile.

[0010] Consequently, there is a need in the art for a translation system and method which does not change the length of source code records or program text literals.

[0011] It is an object of the invention to provide an improved system and method for translating source code.

[0012] It is a further object of the invention to provide a system and method for translating source code such that it is possible to move mainframe software source code to a workstation, operate on the code, and move it back to the mainframe, knowing that the character translations are safe and reversible.

[0013] It is a further object of the invention to provide a system and method for translating source code which eliminates errors due to unsafe and un-reversible translations both on the workstation and on the mainframe.

[0014] A system and method is provide for translating source code. A mapping table of syntactically necessary characters is provided for translating source characters from a first format to a second format. A marker character is used to replace other characters in the source code while translating and preserving the original, replaced source character in a comment file. Upon reverse translating, syntactically necessary characters are translated back into the first format using the mapping table, and marker characters are replaced by reverse translated original source characters from the comment file.

[0015] In accordance with a more specific embodiment of the invention, original characters replaced by the marker character are translated into a two-character hexadecimal representation, and then reverse translated from that two-character hexadecimal representation back into original source characters, in order to keep the original characters replaced by the marker character from preventing compilation or editing.

[0016] In accordance with an aspect of the invention, there is provided a computer program product configured to be operable to translate source code using a mapping table, a marker character, and a comment file of source code characters replaced by the marker character.

[0017] Other features and advantages of this invention will become apparent from the following detailed description of the presently preferred embodiment of the invention, taken in conjunction with the accompanying drawings.

[0018] FIG. 1 is a high level system diagram.

[0019] FIG. 2 is a diagrammatic representation of source code translation in accordance with a preferred embodiment of the invention.

[0020] FIG. 3 is a diagrammatic representation of source code reverse translation in accordance with a preferred embodiment of the invention.

[0021] FIG. 4 is a flow diagram of the method of the invention for source code translation.

[0022] FIG. 5 is a flow diagram of the method of the invention for source code reverse translation.

[0023] FIG. 6 is an example of the translation and a line in the comment file.

[0024] Referring to FIG. 1, host main frame 20 is operable to download source code to workstation 22.

[0025] Referring to FIG. 2, system components for code translation are shown. Source code 30 includes a plurality of lines of code, each line including a line indicia field 32 (for example, line number) and a character field including a plurality of characters 36 (in, for example, EBCDIC format).

[0026] Translated file 60 includes a plurality of source code lines, each line carrying either line indicia 42 corresponding to line indicia 32 for lines containing only translated characters 44, or mark indicia (also referred to as distinguishing value) 46 replacing line indicia 32 in those lines containing one or more marker characters 50 typically in addition to translated characters 48 (in, for example, ASCII format).

[0027] Source characters 36 are individually examined by test 40 to determine if character 36 is included in mapping table 74. If so, character 36 is translated using table 74 and the translated character put into translated character position 44 or 48; if not, character 36 is replaced by marker character 75 and put into marker character field 50, and the original source character 36 put into comment file 28 marker replaced source character field 58. Comment file 28 includes a commented (hence, not compilable) line for each source line having a mark indicia 46, and includes line indicia field 52 which is the line indicia field 42 for lines where a mark indicia 46 has replaced field 42, mark indicia field 54 corresponding to mark indicia 46, marker replaced source characters field 58, in two-character hexadecimal representation, and comment file controls 56, which includes parameters indicating the number of lines in commented file 28 corresponding to this mark indicia 54, the number of marker replaced source characters corresponding to this mark indicia 54, and the number of marker replaced source characters 58 on this line of comment file 28.

[0028] Referring to FIG. 3, system components for reverse code translation are illustrated. Translated source code 60 includes a plurality of lines of code corresponding to translated code lines 60. Each line of code in source file 60 includes an indicia field 62 containing either a line indicia or a mark indicia, and a character string selectively including translated characters 44, 48 and marker characters 50.

[0029] Reverse translated source file 80 includes a plurality of lines of source code, each line including line indicia field 76 and source characters 78. Mark indicia test 70 and marker character test block 72 control the manner in which line indicia field 76 and source character field 78 are populated. As each line of source code in translated source file 60 is parsed, if line/mark indicia field 62 contains a line indicia, that line indicia 62 is placed in line indicia field 76 for the line of reverse translated file 80 corresponding to this line of translated source file 60, and each character 66 in character string 68 is translated using table 74 and loaded into the corresponding character location in source characters field 78. If line/mark indicia field 62 contains a mark indicia, which indicates one or more characters 64 in character string 68 are mark characters, then each character 64, 66 is examined by block 72 to determine if it is a marker character. If it is not, the character 64 is translated using mapping table 74 back into the original source character and put to source character field 78 in the corresponding character position. If character 64 is a marker character, then commented file 28 is accessed using mark indicia 62 to locate the corresponding mark indicia field 54, and marker replaced source characters field 58 accessed to determine the original source code character corresponding to this marker character, which is then moved into source character field 78 in the corresponding character position. If indicia field 62 for this source line contains a mark indicia, then commented file 28 is accessed to locate and load into line indicia field 76 the line indicia 52 for this mark indicia 54.

[0030] Referring to FIG. 4 in connection with FIG. 2, in accordance with the preferred embodiment of the method of the invention, in step 100 source code files 30 are downloaded from the mainframe 20 to the workstation 22 in binary format (no character translation is performed by the download process). In the exemplary embodiment described hereafter, EBCDIC is translated to ASCII, and ASCII is reverse translated to EBCDIC. Of course, as will be apparent to those skilled in the art, the translation/reverse translation process described hereafter may be applied to most any two data formats.

[0031] Translating from EBCDIC to ASCII

[0032] In step 110, the EBCDIC and ASCII code page identifiers and the syntactically necessary characters and other commonly used characters are provided as input.

[0033] In step 120, reversible mapping table 74 for the syntactically necessary and common character is constructed.

[0034] In step 130, source code files 30 are examined file by file, line by line, translating the syntactically necessary and common characters 36 according to the constructed mapping table 74.

[0035] In step 140, a special marker character 75 is ii substituted for characters 36 that are not translated; this marker character is not one of the syntactically necessary or common characters.

[0036] In step 150, the value of the serial or sequence number field 32 of each line with marker characters 50 is saved in commented file 28 field 52, and replaced with a distinguishing value (mark indicia) 46 that identifies that particular line.

[0037] At the end of the source code file (or in a separate file), program comments 28 are added that save, in original program order, the original contents 52 of the serial or sequence number fields 32 and the original contents, as two character hexadecimal numbers, of all characters 36 that had the marker character 75 substituted in marker character 50, as well as the specified code page identifiers and marker character.

[0038] The output source files 60 can now be processed on the workstation.

[0039] Simple manual procedures are used if source code lines 60 with marker characters 50 are changed, duplicated, commented out, or reordered.

[0040] Translating from ASCII to EBCDIC

[0041] Referring to FIG. 5 in connection with FIG. 3, in steps 170 and 200, the source files 60 are examined file by file, line by line, translating the syntactically necessary characters 64, 66 according to the constructed mapping table 74.

[0042] In step 210, when the distinguishing values (mark indicia) in the serial or sequence number fields 62 are encountered in translated source file 60, their original values are retrieved from the program comments 28 as well as the original values 58 of the special marker characters 64, 66.

[0043] In step 220, if in file 60, marker characters 50 are missing or added or if distinguishing values 46 are missing, out of order, or added, error messages are produced.

[0044] In step 230, the added comments 28 at the end of the source code files are removed.

[0045] In step 240, the source code files 80 are uploaded to the mainframe 20 from the workstation 22 in binary format.

[0046] In accordance with an aspect of the invention, the need for maintaining synchronized translation mapping tables and the need for separate fix-up work is eliminated by providing that all character translations use the same mapping tables 74 and are done on the same platform 22.

[0047] Referring to FIG. 6 in connection with Tables 1 and 2 a before and after translation example is illustrated. In this example, the marker character is the tilde “{tilde over ( )}”. In FIG. 6, source file 30 corresponds to two source lines from Table 1, and source line 60 and comment line 28 represent two of the lines in Table 2.

[0048] Source line 250 includes line indicia 002600 in field 252, followed by source characters 05 WS-PWD-TABLE-1 PIC X(50) VALUE in field 254. All of the characters in field 254 are syntactically necessary or commonly used characters, and will all appear in mapping table 74. In accordance with the exemplary embodiment of the invention for translating from EBCDIC to ASCII, each EBCDIC character 254 in code line corresponding to line indicia 002600 in Table 1 is translated using mapping table 74 into an ASCII character in code line corresponding to line indicia 002600 in Table 2.

[0049] Source code line 260 includes line indicia 002700 in line indicia field 256, and a character string including characters 263, 265, 267. The accent grave “{grave over ( )}” 263, tilde “{tilde over ( )}” 265, and caret “{circumflex over ( )}” are examples of characters not included in mapping table 74 and these, among others, must be replaced in translated source file 60 by the marker character, the tilde “{tilde over ( )}” in corresponding character positions 273, 275 and 277, respectively. Because translated source code line 60 includes marker characters 273, 275, 277, etc., line indicia field 32 is loaded with a mark indicia CT0010258, and a corresponding comment line provided in comment area or file 28 into which the original line indicia for this line 002700 is placed in field 52, the mark indicia CT0010 placed in field 54, and the original source characters 263, 265, 267 put in marker replaced source characters field 58, as hex values Al (the hex designation for the EBCDIC value for the accent grave) in character position 264, hex value B0 (the tilde) in character position 266, hex value C0 (the caret) in character position 268. Comment code field 56 contains the values 10 01 10 which, in this example, represent the following: 10 marker characters (tilde) correspond to the line identified by mark indicia CT0010, 01 lines in this file 28 contain the marker replaced source characters, and there are on this comment line 10 such characters 264, 266, 268, 270, 274, . . . , respectively.

[0050] Field 52 contains the original line indicia (sequence number) 002700 to be restored to source file 80 (line indicia 76 in FIG. 3) during reverse translation when the marker characters 273, 275 and 277 . . . are replaced by corresponding hex values 264, 266, 268, and the other (non-marker) values translated back from ASCII to EBCDIC using mapping table 74 in steps 200 and 210.

[0051] Referring to Table 3, the invention may be used, for example, to download and upload mainframe 20 Cobol source code to and from a Windows 95 workstation 22, where the source code is processed by a non-IBM Year 2000 tool set.

[0052] It is an advantage of the invention that there is provided an improved system and method for translating source code.

[0053] It is an advantage of the invention that there is provided a system and method for translating source code such that it is possible to move mainframe software source code to a workstation, operate on the code, and move it back to the mainframe, knowing that the character translations are safe and reversible.

[0054] It is an advantage of the invention that there is provided a system and method for translating source code which eliminates errors due to unsafe and un-reversible translations both on the workstation and on the mainframe.

[0055] It will be appreciated that, although specific embodiments of the invention have been described herein for purposes of illustration, various modifications may be made without departing from the spirit and scope of the invention. In particular, it is within the scope of the invention to provide a computer program product or program element, or a program storage or memory device such as a solid or fluid transmission medium, magnetic or optical wire, tape or disc, or the like, for storing signals readable by a machine, for controlling the operation of a computer according to the method of the invention and/or to structure its components in accordance with the system of the invention.

[0056] Further, each step of the method may be executed on any general computer, such as an IBM System 390, AS/400, PC or the like and pursuant to one or more, or a part of one or more, program elements, modules or objects generated from any programming language, such as C++, Java, Pl/1, Fortran or the like. And still further, each said step, or a file or object or the like implementing each said step, may be executed by special purpose hardware or a circuit module designed for that purpose.

[0057] Also, while the use of line and mark indicia are used in the preferred embodiment to improve error detection, usability, and performance, in an alternative embodiment a safe and reversible translation still occurs where the translation saves marked characters in the order encountered in the source file and the reverse translation examines every character in the processed file for a marker character.

[0058] Accordingly, the scope of protection of this invention is limited only by the following claims and their equivalents.