Method for data compression and decompression using decompression instructions

A character searcher for identifying the locations of a particular test character within a data string and for producing decompression instructions based upon the search results;

• wherein said character searcher is characterized by finding a range which is the distance in the data string between a last position of said test character and an initial test position of said test character by using the last position of said test character minus said initial test position of said test character plus one;
• and finding a test interval which is the distance in a data string between an initial test position of said test character and a next sequential position of said test character by using said next sequential position of said test character minus said initial test position of said test character, and best position count is the greatest number of times said test character is found at a particular interval from an initial test position;
• and best interval is the particular interval that provides said best position count;
• said character searcher consisting:
• a) means to produce a temporary test string that is the same length as said data string and said temporary test string is filled with a test background character such as character code 00X or character code 01X;
• b) means to sequentially search said data string for said test character and place said test character on said test string at corresponding positions;
• c) means to analyze said test string for said test character by sequentially locating initial test position of said test character, calculating the range by last position of said test character minus said initial test position plus one, setting the change count to one, best position count to zero, best interval to zero, and test interval to zero, and if said range is greater than one, then search said test string for next sequential position of said test character, and calculating said test interval as said next sequential position minus said initial test position, setting change count to two, setting best position count to one, setting best interval to the value of test interval, and while said test interval is less than said range divided by said change count, then set test count to two and position count to one, and count said test character at multiples of said test interval, if character equals test character then incrementing said test count and said position count, if character equals overwrite character code FFX or code FEX then incrementing said position count, and stop counting when said test background character is encountered, and if said test count is greater than said change count then setting said change count to said test count, said best position count to said position count, said best interval to said test interval, then find the next sequential test character at next position after said test interval, setting said test interval to said next position minus said initial test position, ending the search when said test interval is greater than said range divided by said change count;
• d) means to produce a multiple position decompression instruction using said initial test position, said best position count, and said best interval, and decrement said test character count by said change count;
• e) means to execute said decompression instruction on said test string and place said overwrite character code FFX or code FEX at positions specified by said decompression instruction, and also append said decompression instruction to instruction string;
• f) means to repeat the processes above starting at “c) means to analyze said test string for said test character”, while said test character count in said test string is greater than zero.

A method of compressing data by using a decompression language and producing decompression instructions that replace said data;

• wherein a compression level is how many times a string has been compressed, from zero compression to a maximum of 31;
• and a string at said compression level zero contains the uncompressed original data string, and a string at said compression level 1 contains the compressed said compression level zero string, and a string at said compression level 2 contains the compressed said compression level 1 string, continuing to a string at said compression level 31 that contains the compressed said compression level 30 string;
• the decompression instructions are characterized by:
• counting an occurrence of the number of times each character appears in a string of characters;
• filling a string of data with background character which is the character that has the highest number of occurrences and is also the character with the greatest count;
• overwriting one or more said background character at required positions within the string of data of said background character with subsequent current character, whose number of occurrences in the string of characters is greater than zero, and is positioned based on its positions within said string of characters and the decompression instructions;
• and best position count is the greatest number of times said current character is found at a particular interval from an initial test position;
• and best interval is the particular interval that provides said best position count;
• and previous length is the string length of the previous set of decompression instructions at the current compression level minus 1;
• said method of compressing data consisting:
• a) means to determine original data length of said data and whether said data are in bits or bytes, and produce an initial segment in a new data string by issuing a compressed segment decompression instruction with the compression level equal to zero, and an original data segment decompression instruction that contains said original data, then increment the compression level by one and produce a compressed segment instruction, then determine whether the new data string must be split into a plurality of sections based upon the length of said new data string and available memory;
• b) means to input and process said new data string, or each section of said new data string consecutively by issuing a segment instruction for each section of said new data string at the current compression level, and count the occurrence of the possible unique characters within said new data string or section of said new data string, and produce a decompression instruction for the character with the greatest count which is the background character to fill a string whose length is equal to said new data string or section of said new data string, and reset its count to zero, and means for each said unique character with a count greater than zero to become the current character, for a character searcher to determine each said current characters positions within said string and issue optimum decompression instructions based on test position, best position count and best interval until each said characters positions have been identified, and repeat process above until all sections have been analyzed consecutively;
• c) means to determine if current said decompression instructions have a total length less than the previous length, and if no then using the previous set of decompression instructions and write said previous decompression instructions to an output file or string, otherwise determine if a greater level of compression is desired and the current compression level is less than 31, and if no then using said current decompression instructions and write said current decompression instructions to an output file or string, otherwise discarding said previous decompression instructions at current compression level minus one, and using current set of said decompression instructions at said current compression level, as input for repeated processing at an incremented said compression level, until said previous length of decompression instructions at said current compression level minus one is less than said current set of decompression instructions at said current compression level, and write said previous decompression instructions at said current compression level minus one to an output file or string.

A method of decompressing decompression instructions based upon a decompression language, wherein original data is the original uncompressed input data from a file that was compressed;

• the decompression instructions are characterized by and were developed by:
• counting an occurrence of the number of times each character appeared in a string of characters;
• filling a string of data with background character which was the character that had the highest number of occurrences;
• overwriting one or more said background character at required positions within the string of data of said background character with subsequent current character, whose number of occurrences in the string of characters was greater than zero, and was positioned based on its positions within said string of characters and the decompression instructions;
• and said method of decompressing decompression instructions to retrieve the said original data, consisting:
• a) means to identify segment decompression instruction and compression level;
• b) means to determine if said compression level is greater than zero, and if it is, then extract data from said segment by decoding said segment to determine if there are sub-segments, number of subsegments and, compression level, length, start and end of said segment or each sub-segment, wherein any said sub-segments must be at the same compression level as said segment;
• c) means for said segment or each said sub-segment to decode a decompression instruction to produce a temporary string with specified background character, and to decode each successive decompression instruction and execute said instructions to change the current character, and to position said current character on said temporary string, and when finished with said segment or each said sub-segment to append said temporary string to output string;
• d) means to repeat steps a, b, and c above, while said compression level is greater than zero;
• e) means to extract said original data when said compression level equals zero by decoding the decompression instruction for said original data and writing the output to a file or output string.