Class CsvUtil

java.lang.Object

org.bgerp.plugin.inventoru.sync1c.service.CsvUtil

public class CsvUtil
extends Object

Minimal RFC4180 CSV reader/writer for the sync1c File protocol — no CSV library is otherwise used in this project, so a dedicated dependency isn't warranted for two simple flat formats. Real exports from 1C are not RFC4180-shaped: the de-facto industry format (e.g. the Cleverence "Склад 15" exchange) is semicolon-separated, because a comma is the decimal separator in the Russian locale, and the file often comes in Windows-1251 rather than UTF-8. Both the delimiter and the charset are therefore detected from the content, with an explicit override available per connector (csv.delimiter / csv.encoding).

Field Summary

Fields

Modifier and Type	Field	Description
static final Charset	WINDOWS_1251	Charset of most Cyrillic exports that are not UTF-8.

Method Summary

Modifier and Type	Method	Description
static Charset	detectCharset(byte[] data)	Detects the charset: an UTF-8 BOM decides it outright, otherwise the content is decoded as strict UTF-8 — a file that is not valid UTF-8 is assumed to be Windows-1251, which is the only other charset seen in practice for these exports.
static char	detectDelimiter(String text)	Detects the delimiter by counting candidates outside quoted fields in the first line: the header row is the most reliable sample, since data rows may legitimately contain any of them inside quotes.
static List<String[]>	parse(byte[] data)	Parses CSV with the delimiter and charset detected from the content.
static List<String[]>	parse(byte[] data, Character delimiter, Charset charset)	Parses CSV.
static List<String[]>	parse(InputStream in)	Deprecated. stream variant kept for callers that don't hold the whole content; it buffers the stream anyway, since both charset and delimiter detection need to look at the bytes before decoding.
static void	write(OutputStream out, List<String[]> rows)	Writes rows as UTF-8 CSV with a BOM, comma-separated.
static void	write(OutputStream out, List<String[]> rows, char delimiter, Charset charset)	Writes rows as CSV, quoting fields that contain the delimiter, a quote or a newline.

Methods inherited from class Object

clone, equals, finalize, getClass, hashCode, notify, notifyAll, toString, wait, wait, wait

Field Details

WINDOWS_1251

public static final Charset WINDOWS_1251

Charset of most Cyrillic exports that are not UTF-8.

Method Details

parse

public static List<String[]> parse(byte[] data)
                            throws IOException

Parses CSV with the delimiter and charset detected from the content.

Parameters:

data - whole file content

Throws:

IOException

parse

public static List<String[]> parse(byte[] data,
 Character delimiter,
 Charset charset)
                            throws IOException

Parses CSV.

Parameters:

data - whole file content

delimiter - explicit delimiter, null to detect

charset - explicit charset, null to detect

Throws:

IOException

parse

@Deprecated
public static List<String[]> parse(InputStream in)
                            throws IOException

Deprecated.

stream variant kept for callers that don't hold the whole content; it buffers the stream anyway, since both charset and delimiter detection need to look at the bytes before decoding. Prefer parse(byte[], Character, Charset).

Throws:

IOException

detectCharset

public static Charset detectCharset(byte[] data)

Detects the charset: an UTF-8 BOM decides it outright, otherwise the content is decoded as strict UTF-8 — a file that is not valid UTF-8 is assumed to be Windows-1251, which is the only other charset seen in practice for these exports. Pure ASCII decodes as UTF-8, and for ASCII the two charsets agree anyway.

detectDelimiter

public static char detectDelimiter(String text)

Detects the delimiter by counting candidates outside quoted fields in the first line: the header row is the most reliable sample, since data rows may legitimately contain any of them inside quotes. Falls back to a comma when the line holds no candidate at all.

write

public static void write(OutputStream out,
 List<String[]> rows)
                  throws IOException

Writes rows as UTF-8 CSV with a BOM, comma-separated.

Throws:

IOException

write

public static void write(OutputStream out,
 List<String[]> rows,
 char delimiter,
 Charset charset)
                  throws IOException

Writes rows as CSV, quoting fields that contain the delimiter, a quote or a newline. A BOM is prepended for UTF-8 so Excel opens Cyrillic text correctly.

Throws:

IOException