Data compressing apparatus, reconstructing apparatus, and its method

A data compressing apparatus for compressing non-compression data formed by a character code of a language having a word structure which is not separated by spaces, comprising:

• a first coding unit for comparing a registration character train which has been registered in a dictionary and serves as a processing unit when compressing with a partial character train in said non-compression data, thereby detecting the partial character train which coincides with said registration character train, and for allocating a predetermined character train code every said detected partial character train and outputting as an intermediate code;
• and
• a second coding unit for inputting an intermediate code train compressed by said first coding unit and compressing it again.

A data reconstructing apparatus for inputting compression data in which a coding at a first stage for detecting a registration character train which has been registered in a dictionary and serves as a processing unit upon compression for non-compression data, as a target, formed by character codes of a language having a word structure that is not separated by spaces and a coincident partial character train in said non-compression data and for outputting a predetermined character train code as an intermediate code and a coding at a second stage for inputting said intermediate code train and again coding have been executed and reconstructing the original non-compression data, comprising:

• a first decoding unit for inputting said compression data and reconstructing said intermediate code train;
• and
• a second decoding unit for inputting the intermediate code train reconstructed by said first decoding unit and reconstructing the original character train.

A data compressing method of compressing non-compression data formed by character codes of a language having a word structure which is not separated by spaces, comprising:

• a character train comparing step of comparing a registration character train in a character train dictionary storing unit in which a dictionary in which a character train serving as a processing unit upon compression was registered has been stored with a partial character train in said non-compression data, thereby detecting the partial character train which coincides with said registration character train;
• and
• a code output step of outputting a predetermined character train code every said partial character train detected by said character train comparing step.

A data reconstructing method of comparing a registration character train which has been registered in a dictionary and serves as a processing unit upon compression for non-compression data, as a target, formed by character codes of a language having a word structure which is not separated by spaces with a partial character train in said non-compression data, thereby detecting the partial character train which coincides with said registration character train, and inputting compression data to which a predetermined character train code has been allocated every said detected partial character train, and reconstructing the original non-compression data, comprising:

• a code train separating step of separating a character train code serving as a reconstructing unit from the compression data;
• and
• a character train reconstructing step of reconstructing the original character train with reference to the dictionary in which a reconstruction character train corresponding to the character train code serving as a processing unit upon reconstruction has been registered by the character train code separated in said code train separating step.

A data compressing method of compressing non-compression data formed by character codes of a language having a word structure which is not separated by spaces, comprising:

• a first coding step of comparing a registration character train which has been registered in a dictionary and serves as a processing unit upon compression with a partial character train in said non-compression data, detecting the partial character train which coincides with said registration character train, and allocating a predetermined character train code every said detected partial character train, and outputting as an intermediate code;
• and
• a second coding step of inputting the intermediate code train compressed by said first coding step and again compressing it.

A data reconstructing method of inputting compression data in which a coding at a first stage such that a registration character train which has been registered in a dictionary and serves as a processing unit upon compression for non-compression data, as a target, formed by character codes of a language having a word structure which is not separated by spaces and a coincident partial character train in said non-compression data are detected and a predetermined character train code is allocated and an intermediate code is outputted and a coding at a second stage such that said intermediate code train is inputted and is again coded have been performed and reconstructing the original non-compression data, comprising:

• a first decoding step of inputting said compression data and reconstructing said intermediate code train;
• and
• a second decoding step of inputting the intermediate code train decoded by said first decoding step and reconstructing the original character train.

A data compressing apparatus for compressing non-compression data formed by character codes, comprising:

• a first coding unit for comparing a registration character train which has been registered in a character train attribute dictionary and serves as a processing unit upon compression with a partial character train in said non-compression data, thereby detecting the partial character train which coincides with said registration character train, and allocating a set of a predetermined character train code and an attribute code every said coincidence detected partial character train as an intermediate code and outputting;
• and
• a second coding unit for inputting the intermediate code train compressed by said first coding unit and again compressing.

A data reconstructing apparatus for inputting compression data in which a coding at a first stage such that a registration character train which has been registered in a character train attribute dictionary and serves as a processing unit upon compression for non-compression data formed by character codes as a target and a coincident partial character train in said non-compression data are detected and a set of a predetermined character train code and an attribute code are allocated as an intermediate code and are outputted and a coding at a second stage for inputting said intermediate code train and coding again have been performed and reconstructing the original non-compression data, comprising:

• a first decoding unit for inputting said compression data and reconstructing said intermediate code train;
• and
• a second decoding unit for inputting the intermediate code train decoded by said first decoding unit and reconstructing the original character train.

A data compressing method of compressing non-compression data formed by character codes, comprising:

• a character train comparing step of comparing a registration character train in a dictionary in which character trains serving as a processing unit upon compression have been classified in accordance with attributes and divided into a plurality of attribute groups and registered with a partial character train in said non-compression data, thereby detecting the partial character train which coincides with said registration character train;
• and
• a code output step of allocating a set of a predetermined character train code and an attribute code showing said attribute group every said partial character train detected by said character train comparing step and outputting.

A data compressing method of compressing non-compression data formed by character codes, comprising:

• a first coding step of comparing a registration character train which has been registered in a character train attribute dictionary and serves as a processing unit upon compression with a partial character train in said non-compression data, thereby detecting the partial character train which coincides with said registration character train, and allocating a set of intermediate codes in which a predetermined character train code and an attribute code are coupled every said detected partial character train as an intermediate code, and outputting;
• and
• a second coding step of inputting the intermediate code train compressed by said first coding step and again compressing.

A data reconstructing method of inputting compression data in which a coding at a first stage such that a registration character train which has been registered in a character train attribute dictionary and serves as a processing unit upon compression for non-compression data formed by character codes as a target and a partial character train which coincides in said non-compression data are detected and a set of a predetermined character train code and an attribute code is allocated as an intermediate code and outputted and a coding at a second stage for inputting said intermediate code train and coding again have been executed and reconstructing the original non-compression data:

• a first decoding step of inputting said compression data and reconstructing said intermediate code train;
• and
• a second decoding step of inputting the intermediate code train reconstructed by said first decoding step and reconstructing the original character train.