Tutorial Casper IO

Casper datasets IO provides a set of classes for performing import and export to/from casper datasets. It includes functions/classes for:

• Import from a collection of POJO beans (CBuildFromCollection)

• Export to a collection of POJO beans (CExportBeans, CExportBeansCached)

• Import from CSV, XLS and XLSX files (CBuildFromFile)

• Export to CSV files (CExportCSV)

• Casper dataset file definitions, or in other words, specification of a casper dataset, its name, columns, column types, and PKs that is loaded from a file(CDataFileDef)

• Loading of casper dataset file definitions from files (CDataFileDefLoader)

• Create a casper dataset from a collection of beans

    // create a casper dataset called "people" from the
    // peopleColl collection using "id" for the primary key
    builderPK = new CBuildFromCollection("people", peopleColl,
                    Object.class, "id");
  
    // create a casper dataset called "people" from the
    // peopleColl collection without a primary key
    builderNoPK = new CBuildFromCollection("people", peopleColl,
                    Object.class, null);
  
    CDataCacheContainer container = new CDataCacheContainer(builderPK);
  

Export a casper dataset to a collection of beans

    // Exports a casper dataset to a collection of java beans using Apache commons
    // beanutils reflection code. Faster than CExportBeansCached for smaller
    // collections.

    CExportBeans<Person> cexportBeans = new CExportBeans<Person>(Person.class);
    container.export(cexportBeans);
    Collection<Person> beans = cexportBeans.getBeans();
    // Exports a casper dataset to a collection of java beans using a MethodCache
    // for reflection. Faster than CExportBeans for larger
    // collections.

    CExportBeansCached<Person> cexportBeans = new CExportBeansCached<Person>(Person.class);
    container.export(cexportBeans);
    Collection<Person> beans = cexportBeans.getBeans();

Load a casper dataset from a file

    // Load from an XLS file, reading the column header from the file 
    // and reading columns as whatever type is returned by the XLS reader
    CBuilder builder = new CBuildFromFile(new File("people.xls"));
    CDataCacheContainer container = new CDataCacheContainer(builder);

    // Load from a CSV file, reading the column header from the file 
    // and reading columns as String
    CBuilder builder = new CBuildFromFile(new File("people.csv"));
    CDataCacheContainer container = new CDataCacheContainer(builder);

    // Load CSV file, using the headers found in the file,
    // and converting each column to the narrowest possible
    // data type. Missing integers and doubles will be 
    // returned as CellReaders.MISSING_VALUE_INTEGER
    // or CellReaders.MISSING_VALUE_DOUBLE
    CBuilder builder = new CBuildNarrowedFile(new File("people.csv"))
                    .setConvertMissing(true);
    CDataCacheContainer container = new CDataCacheContainer(builder);

    // Load the specified columns, as the specified types (ie: CellReader)
    // from an CSV file
    String[] columnNames = new String[] {"crefnum",  
                    "sex", "refnum", "question", "age",  
                    "weight"};
    String[] PKs = new String[] {"refnum", "crefnum"};

    CellReader<?>[] columnOptionalReaders = new CellReader<?>[] {
                    CellReaders.OPTIONAL_INTEGER,
                    CellReaders.CHARACTER,
                    CellReaders.OPTIONAL_INTEGER,
                    CellReaders.OPTIONAL_INTEGER,
                    CellReaders.OPTIONAL_INTEGER,
                    CellReaders.OPTIONAL_DOUBLE};

    CBuilder builder = new CBuildFromFile(new File("people.csv"), "people dataset", 
            columnNames, columnOptionalReaders,PKs);
    CDataCacheContainer container = new CDataCacheContainer(builder);

Load a casper dataset from a file definition

    // Load the same file as the last example but using a CDataFileDef
    CDataFile cdf = new CDataFileDef("cdf people xls", columnNames,
            columnOptionalReaders, PKs);
    CDataCacheContainer container = cdf.loadDataset(new File("people.xls"));

    // Load the same file as the last example but from its definition supplied in people.json
    CDataFileDef builtCDF = loader.fromJsonFile(new File("people.json"));
    CDataCacheContainer container = builtCDF.loadDataset(new File("people.xls"));

    // people.json
    {
            "name":"cdf patients xls",
                "column_names":[
                "crefnum",
                "sex",
                "refnum",
                "question",
                "age",
                "weight"
            ],
            "column_types":[
                "optional integer",
                "character",
                "optional integer",
                "optional integer",
                "optional integer",
                "optional double"
            ],
            "primary_key":[
                "refnum",
                "crefnum"
            ]
    }

Export a casper dataset to a CSV file

    // Export complete dataset to a csv file
    CExportCSV csv = new CExportCSV(new File("export.csv"));
    container.export(csv);

    // Export only specified columns of dataset to a csv file
    CExportCSV csv = new CExportCSV(new File("export.csv"),"refnum,sex,crefnum,age");
    container.export(csv);