Csvbeans: parse CSV in Java

1. Introduction to CSV and csvbeans
CSV: comma seperated values; Description: A row represents a record, and a record has multiple fields (attributes). Each attribute is separated by commas (or other symbols.

Csvbeans open-source project: an open-source tool that converts each line of CSV data into a JavaBean;

Download URL: http://sourceforge.net/projects/csvbeans/files/

Of course, parsing CSV is not only a tool, but also opencsv of Apache;

Dependency packages should also be added in the lib directory, including: commons-resources.jarcommons-logging.jarcommons-lang-2.1.jarcommons-digester-1.7.jarcommons-codec-1.3.jarcommons-beanutils-core.jarcommons-beanutils-bean-collections.jarcommons-beanutils.jar
2. Solving project defects
First, I have found two small problems in this project: (1) CSV file data cannot be separated by \ t; (2) starttag must be included when writing CSV files;

Starttag must exist, which means:Persons----> This is starttagaaa, 20aaa, 20aaa, 20. The starttag cannot be absent, which means AAA, 20aaa, 20aaa, 20 If starttag is set to null, a problem occurs;

Solution:
(1) For the first problem, if a CSV file is separated by \ t, I wrote a helper class that can be used to rewrite the CSV file to a CSV file separated by commas;

Package com.xiazdong.csv; import Java. io. bufferedreader; import Java. io. file; import Java. io. fileinputstream; import Java. io. inputstreamreader; import Java. io. printwriter; import Java. util. UUID; public class csvutils {/*** this function is used to assist csvbeans. Because this open-source package has a defect, it cannot parse CSV files separated by tab, therefore, this function is used to convert a tab to a comma and overwrite the file to be converted from the original file * @ Param fromfile * @ throws exception */public static void parsefromtabtocomma (File fromfile) throws exc Eption {bufferedreader reader = new bufferedreader (New inputstreamreader (New fileinputstream (fromfile); file tmpfile = new file (UUID. randomuuid (). tostring (); // generate a temporary file, but no printwriter writer = new printwriter (tmpfile); string line = NULL; while (line = reader. readline ())! = NULL) {string ss [] = line. split ("\ t"); // separated by tab (INT I = 0; I <ss. length; I ++) {writer. write (ss [I]); if (I <(ss. length-1) {writer. write (",") ;}} writer. write ("\ n");} writer. close (); reader. close (); If (fromfile. exists () {fromfile. delete (); tmpfile. renameto (fromfile) ;}} public static void main (string [] ARGs) throws exception {parsefromtabtocomma (new file ("src/persons.csv "));}}

(2) For the second problem, you must modify the source code;
Source code download: http://sourceforge.net/projects/csvbeans/files/csvbeans/csvbeans-0.7.1/
Csvbeans-0.7.1 \ SRC \ Java \ org \ csvbeans \ builders \ csvbuilder. Java AddbeansFunction changed to: Public void addbeans (string tag, list beans ){
If (beans! = NULL) {This. Beans. Add (new records (TAG, beans ));}}
WritebeansThe function starts with try {If (tag! = NULL) writer. writeline (TAG); recordspecification record = getrecordspecification (TAG );
After modifying the source code, ant package also needs to be compiled. The ant package automatically generates csvbeans. jar;

Iii. Preparations

1. Write Javabean to indicate a CSV record;
2. write an XML file with the following content: (1) attributes of a CSV record, including the name and maxlen of each attribute; (2) separators between each attribute; (3) classes of JavaBean; (4) parser class and builder class used;

4. Programming

Directory structure:


First, it indicates that both the an and mapping. XML for reading and writing CSV are the same;
1. Read CSV

Person. Java
Mapping. xml

<? XML version = "1.0"?> <Csvbeans> <strategy> <parser classname = "org.csv beans. parsers. csvparser "/> <builder classname =" org.csv beans. builders. csvbuilder "/> </strategy> <property name =" separator "value =", "/> <! -- Delimiter --> <property name = "nostarttag" value = "true"/> <! -- No starttag --> <record classname = "com.xiazdong.csv. domain. Person"> <! -- One record --> <field name = "name" maxlen = "100" required = "true"/> <field name = "Age" maxlen = "10" required = "true "/> </record> </csvbeans>

Read the CSV program:

Private Static void read () throws exception, specificationsfileexception, filenotfoundexception, parsingexception {file F = new file ("src/persons.csv"); csvutils. parsefromtabtocomma (f); // If CSV files are separated by tabs, use this function specificationsfileparser specsparser = new specificationsfileparser (); specificationsfile specs = specsparser. parse (New fileinputstream ("src/mapping. XML "); parsingstrategy parser = specs. getparsingstrategy (); // obtain mapping. csvparserparser in XML. parse (New inputstreamlinesreader (New fileinputstream (F); list products = parser. getbeans (null); // No header for (iterator it = products. iterator (); it. hasnext ();) {person Product = (person) it. next (); system. out. println (product );}}

2. Write CSV

Private Static void write () throws specificationsfileexception, filenotfoundexception, generationexception {list <person> Persons = new arraylist <person> (); persons. add (new person ("I", 20, 20); persons. add (new person ("AAA", 20, 20); persons. add (new person ("AAA", 20, 20); specificationsfileparser specsparser = new specificationsfileparser (); specificationsfile specs = specsparser. parse (New fileinputstream ("src/mapping. XML "); buildingstrategy builder = specs. getbuildingstrategy (); // obtain mapping. in XML, csvbuildercsvbuilder csvbuilder = (csvbuilder) builder; csvbuilder. addbeans (null, persons); csvbuilder. build (New outputstreamlineswriter (New fileoutputstream ("src/output.csv ")));}