Computer file comparison method

A method performed on a programmed computer for determining whether two computer programs process data equivalently, comprising the steps of:

• operating a first computer program on a computer in response to a first input data set to produce a first output data set;
• altering said first input data set by a predetermined quantity to produce a second input data set;
• operating a second computer program on the computer in response to said second input data set to produce a second output data set;
• comparing data items in said first output data set to data items in said second output data set in a predetermined sequence to determine if at least one mismatch exists between data items of the first and second output data sets, said first output data set accessed by said programmed computer simultaneously with said second output data set;
• parsing a first group of data items at each location of mismatch of the at least one mismatch to produce a plurality of tokens, said tokens representing quantities corresponding to said first group of data items;
• and
• performing a group comparison process by making continuous forward progress in a quantity of data which includes a token from said first group of data items and data from a second group of data items located in predetermined relation to the each location of mismatch to find matching data in one of the first and second groups, said comparison performed in response to successive tokens corresponding to said first group.

A method performed on a programmed computer for determining whether a remediated computer program and an original computer program are functionally equivalent, comprising the steps of:

• operating said original computer program in response to a first input data file to produce a first output data file;
• operating said remediated computer program in response to said first input data file to produce a second output data file, said second output data file having a percentage of structural elements identical to structural elements of said first output file;
• comparing data items in said first output data file to data items in said second output data file in said predetermined sequence to determine if mismatches exist between data items in said first and second output data files, said programmed computer simultaneously accessing said first and second output data files;
• parsing data items at each single location of mismatch in said first output data file into a first plurality of tokens and data items at the each single location of mismatch in said second output data file into a second plurality of tokens, said first and second plurality of tokens representing units of time corresponding to said data items;
• and
• determining a difference between a unit of time represented by a token of the first plurality of tokens to a unit of time represented by a token of the second plurality of tokens.

A method performed on a programmed computer for determining whether a remediated computer program and an original computer program are functionally equivalent, comprising the steps of:

• operating said original computer program on a computer having a system date set to a first date, said computer program operating in response to a first input data file to produce a first output data file;
• aging said first input data file by a quantity representing a predetermined amount of time to produce a second input data file;
• changing said system date of said computer by said predetermined amount of time;
• operating said remediated computer program on a computer having a system date set to a second date differing from said first date by said predetermined amount of time, said computer program operating in response to said second input data file to produce a second output data file;
• comparing data items in said first output data file to data items in said second output data file in a predetermined sequence to determine if a mismatch exists between data items at a location in a file;
• parsing data items at a location of mismatch in said first file into a plurality of tokens and data items at a location of mismatch in said second file into a plurality of tokens, said tokens representing dates corresponding to said data items;
• and
• determining a difference between a date represented by a token into which data items in said first file at a location of mismatch have been parsed to a date represented by a token into which data items in said second file at a location of mismatch have been parsed.

A computer program product for use in a computer system, said computer program product comprising a computer-readable data storage medium carrying thereon:

• a file reader for reading data items from a first data file in a predetermined sequence and for reading data items from a second data file in said predetermined sequence, said first data file simultaneously accessed with said second data file, said second data file having a percentage structural elements identical to structural elements of the first data file;
• a file comparer for comparing data items in said first data file to data items in said second data file to determine whether mismatches exist between data items of said first and second data files;
• a parser for parsing a first group of data items in said first data file located in predetermined relation to an each single location of mismatch a plurality of times to generate a first list of a plurality of tokens, each token in said first list representing a quantity corresponding to said first group of data items;
• and
• a group comparer for performing a comparison which includes a token corresponding to said first group of data items and data from a second group of data items located in predetermined relation to the each single location of mismatch to find matching data, said comparison performed in response to successive tokens in said first list.

A computer program product for use in a computer system, said computer program product comprising a computer-readable data storage medium carrying thereon:

• file reader means for reading data items from a first data file in a predetermined sequence and for reading data items from a second data file in said predetermined sequence, wherein said first and second data files are accessed simultaneously, and wherein said second data file has no more than two percent of its structural elements identical to said first data file;
• file comparer means for comparing data items in said first data file to data items in said second data file in said predetermined sequence to determine if at least one mismatch exists between data items of said first and second data files;
• parser means for parsing a first group of data items in said first data file located in predetermined relation to each single location of mismatch of the at least one mismatch a plurality of times to generate a first list of a plurality of tokens, each token in said first list representing a quantity corresponding to said first group of data items;
• and
• group comparer means for performing a comparison including a token from said first group of data items and data from a second group of data items located in predetermined relation to the each single location of mismatch to find matching data in one of the first and second groups, said comparison performed in response to successive tokens in said first list.

A method performed on a programmed computer for determining whether a remediated computer program and an original computer program are functionally equivalent, comprising the steps of:

• operating said original computer program on a computer having a system date set to a first date, said computer program operating in response to a first input data file to produce a first output data file;
• aging said first input data file by a quantity representing a predetermined amount of time to produce a second input data file;
• changing said system date of said computer by said predetermined amount of time;
• operating said remediated computer program on a computer having a system date set to a second date differing from said first date by said predetermined amount of time, said computer program operating in response to said second input data file to produce a second output data file;
• comparing data items in said first output data file to data items in said second output data file in a predetermined sequence to determine if a mismatch exists between data items at a single location;
• parsing data items at the single location of mismatch in said first output data file into a first plurality of tokens and data items at the single location of mismatch in said second output data file into a second plurality of tokens, said first and second plurality of tokens representing units of time corresponding to said data items;
• and
• determining a difference between a date represented by a token into which data items in said first file at a single location of mismatch have been parsed to a date represented by a token into which data items in said second file at a single location of mismatch have been parsed.

A method performed on a programmed computer for comparing a first and second data file to find matching data, comprising the steps of:

• reading data items from the first data file in a predetermined sequence;
• reading data items from the second data file in said predetermined sequence;
• comparing data items in said first data file to data items in said second data file to determine if at least one mismatch exists between data items of the first and second data files;
• parsing a first group of data items in said first data file located in predetermined relation to a location of mismatch a plurality of times to generate a first list of a plurality of tokens, each token in said first list representing an exchange conversion between monetary amounts corresponding to said first group of data items;
• and
• performing a group comparison by making continuous forward progress in the quantity of data compared which includes a token from said first group of data items and data from a second group of data items located in predetermined relation to the location of mismatch to find the matching data, said comparison performed in response to successive tokens in said first list.

A method performed on a programmed computer for comparing a first and second data file to find matching data, comprising the steps of:

• reading data items from the first data file in a predetermined sequence;
• reading data items from the second data file in said predetermined sequence;
• comparing data items in the first data file to data items in the second data file to determine if at least one mismatch exists between data items;
• parsing a first group of data items in said first data file located in predetermined relation to a location of mismatch a plurality of times to generate a first list of a plurality of tokens, each token in said first list representing a quantity corresponding to said first group of data items;
• and
• performing a group comparison comprising the steps of:
• determining a test group of data items in response to each successive token of a plurality of successive tokens in said first list;
• and
• comparing said test group of data items to a second group of data items located in predetermined relation to the location of mismatch to find the matching data;
• wherein said test group of data items corresponds to a quantity differing from said quantity represented by said each successive token by a predetermined amount, whereby said group comparison identifies whether said quantity corresponding to said first group of data items and said quantity corresponding to the second group of data items differ by said predetermined amount.

A method performed on a programmed computer for comparing a first and second data file to find matching data, comprising the steps of:

• reading data items from the first data file in a predetermined sequence;
• reading data items from the second data file in said predetermined sequence;
• comparing data items in the first data file to data items in the second data file to determine if mismatches exists between data items of the first and second data files, wherein a byte of data of the data items that corresponds to a character selected from a group consisting of tab character and space character are skipped;
• parsing a first group of data items in the first data file located in predetermined relation to a location of mismatch a plurality of times to generate a first list of a plurality of tokens, each token in said first list representing a unit of time corresponding to said first group of data items;
• and
• performing a group comparison by making continuous forward progress in the quantity of data compared which includes a token from said first group of data items and data from a second group of data items located in predetermined relation to the location of mismatch to find matching data in one of the first and second groups, said comparison performed in response to successive tokens in said first list.

A method performed on a programmed computer for comparing a first data file and a second data file, comprising the steps of:

• reading data items simultaneously from said first data file and said second data file in a predetermined sequence;
• comparing data items in said first data file to data items in said second data file to determine whether at least one correspondence exists between data items of the first and second data files;
• parsing a first group of data items in said first data file located in predetermined relation to an each single location of the at least one correspondence a plurality of times to generate a first list of a plurality of tokens, each token in said first list representing a quantity;
• and
• performing a group comparison by making continuous forward progress in the quantity of data compared which includes a token generated from said first group of data items and data from a second group of data items located in predetermined relation to the each single location of correspondence to find matching data in one of the first and second groups, said comparison performed in response to successive tokens in said first list.