Specify (at least a major) version when adding this project as dependency. Whenever the API has incompatible changes the major version changes!
- no new functionality.
- adapted to dart2. This library is no longer a codec!
parseNumbers
has been renamed toshouldParseNumbers
FirstOccurenceSettingsDetector
has been renamed toFirstOccurrenceSettingsDetector
- Speed improvements.
- added option
delimitAllFields
- fix issue #5 (endDelimiter was not always taken correctly from delimiter)
- fix issue #10 (SDK version was overlay restrictive)
A dart csv to list converter.
If you have a String
of all rows with RFC conform separators and delimiters,
simply convert them with:
List<List<dynamic>> rowsAsListOfValues = const CsvToListConverter().convert(yourString);
To convert to a Csv string your values must be in a List<List<dynamic>>
representing a List of Rows where every Row is a List of values.
You can then convert with:
String csv = const ListToCsvConverter().convert(yourListOfLists);
The default (RFC conform) configuration is:
- , as field separator
- " as text delimiter and
- \r\n as eol.
See below if you need other settings, or want to autodetect them.
This converter may be used as transformer for streams:
final stream = Stream.fromIterable([['a', 'b'], [1, 2]]);
final csvRowStream = stream.transform(ListToCsvConverter());
Or the decoder side:
final input = File('a/csv/file.txt').openRead();
final fields = await input.transform(utf8.decoder).transform(CsvToListConverter()).toList();
The converter is highly customizable and even allows multiple characters as delimiters or separators.
Every csv row is converted to a list of values. Unquoted strings looking like
numbers (integers and doubles) are by default converted to int
s or double
s.
The input must be a List
of List
s. Every inner list is converted to one
output csv row. The string representation of values is obtained by calling
toString
.
This converter follows the rules of rfc4180.
This means that text fields containing any delimiter or an eol are quoted.
The default configuration is:
- , as field separator
- " as text delimiter and
- \r\n as eol.
This parser will accept eol and text-delimiters inside unquoted text and not throw an error.
In addition, this converter supports multiple characters for all delimiters
and eol. Also, the start text delimiter and end text delimiter may be
different. This means the following text can be parsed:
«abc«d»*|*«xy»»z»*|*123
And (if configured correctly) will return ['abc«d', 'xy»z', 123]
If the default values are fine, simply instantiate ListToCsvConverter
and
call convert
:
final res = const ListToCsvConverter().convert([[',b', 3.1, 42], ['n\n']]);
assert(res == '",b",3.1,42\r\n"n\n"');
Consider using the returnString = false
option to work around a performance bug.
There are 2 interesting things to note:
- Not all rows have to be the same length.
- The default eol is
'\r\n'
and'\n'
is also quoted. The appearance of only one character is enough for the string to be quoted.
The converter takes the following configurations either in the constructor or
the convert
function:
fieldDelimiter
: the separator between fields. By default','
but another common value is';'
.textDelimiter
: the quotation string. By default'"'
.textEndDelimiter
: the end quotation string. By default, equalstextDelimiter
. The string used to end a quoted string.eol
: The new line string. By default'\r\n'
. Another common value:'\n'
convertNullTo
: If a value isnull
, use this value instead. To get an empty entry instead of stringnull
, use''
.
All configuration values may be multiple characters!:
const conv = const ListToCsvConverter(fieldDelimiter: '|*|',
textDelimiter: '<<',
textEndDelimiter: '>>',
eol: '**\n');
final res = conv.convert([['a','>'], ['<<', '>>'], [1, 2]]);
assert(res == 'a|*|<<>>>**\n<<<<>>|*|<<>>>>>>**\n1|*|2');
final res2 = const ListToCsvConverter()
.convert([['a','>'], ['<<', '>>'], [1, 2]],
fieldDelimiter: '|*|',
textDelimiter: '<<',
textEndDelimiter: '>>',
eol: '**\n',
convertNullTo: '');
assert(res == res2);
Note that:
'>'
is quoted'<<'
is quoted as well, but because it is "only" a start text delimiter it is not doubled. (See rule 7. below).'>>'
is quoted. Only the end-quote string is doubled!
If the default values are fine, simply instantiate CsvToListConverter
and
call convert
:
final res = const CsvToListConverter().convert('",b",3.1,42\r\n"n\n"');
assert(res.toString() == [[',b', 3.1, 42], ['n\n']].toString());
Again please note that depending on the input not all rows have the same number of values.
The CsvToListConverter
takes the same arguments as the ListToCsvConverter
(except for convertNullTo
)
plus
-
shouldParseNumbers
: by default true. If you want the output to beString
s only set this to false. -
allowInvalid
: by default true. The converter will by default never throw an exception. Even iffieldDelimiter
,textDelimiter
,... don't make sense or the csv-String is invalid. This may for instance happen if the csv-String ends with a quoted String without the end-quote (textEndDelimiter
) string. -
csvSettingsDetector
: must be an object which extends fromCsvSettingsDetector
. There implementation simply selects the first occurrence of a list of possible values as value. -
convertEmptyTo
: If a field is empty (and not quoted!) use this value instead.
convertEmptyTo: [1, 2, 3]
would insert the list[1, 2, 3]
whenever a field is empty.
To convert empty fields tonull
, use the enum:EmptyValue.NULL
:var csv = CsvToListConverter(convertEmptyTo: EmptyValue.NULL).convert(input);
var d = FirstOccurrenceSettingsDetector(eols: ['\r\n', '\n'], textDelimiters: ['"', "'"]); CsvToListConverter(csvSettingsDetector: d);
In this case eol
will either be '\r\n'
or '\n'
depending on which of
those 2 comes first in the csv string. Note that the
FirstOccurrenceSettingsDetector
doesn't parse the csv string! For instance
if eol should be '\r\n'
but there is a field with a correctly quoted '\n'
in the first row, '\n'
is used instead.
If you csv String contains a (simple) header row, or all eols are equal this is good enough.
Feel free to submit something more intelligent.
To check your configuration values there is CsvToListConverter.verifySettings
and verifyCurrentSettings
. Both return an empty list if all settings are valid,
or a list of errors. If the optional throwError
is true an error is thrown in
case the settings are invalid.
All settings must be set, i.e. not be null, and delimiters, separators and eols must be distinguishable, i.e. they may not be the start of another settings.
Ad rule 3: removed as it is not relevant for this converter.
-
Each record is located on a separate line, delimited by a line break (CRLF). For example: aaa,bbb,ccc CRLF zzz,yyy,xxx CRLF
-
The last record in the file may or may not have an ending line break. For example:
aaa,bbb,ccc CRLF
zzz,yyy,xxx -
... (Header-lines)
-
Within the header and each record, there may be one or more fields, separated by commas. Each line should contain the same number of fields throughout the file. Spaces are considered part of a field and should not be ignored. The last field in the record must not be followed by a comma. For example:
aaa,bbb,ccc
-
Each field may or may not be enclosed in double quotes (however some programs, such as Microsoft Excel, do not use double quotes at all). If fields are not enclosed with double quotes, then double quotes may not appear inside the fields. For example:
"aaa","bbb","ccc" CRLF zzz,yyy,xxx
-
Fields containing line breaks (CRLF), double quotes, and commas should be enclosed in double-quotes. For example:
"aaa","b CRLF bb","ccc" CRLF zzz,yyy,xxx
-
If double-quotes are used to enclose fields, then a double-quote appearing inside a field must be escaped by preceding it with another double quote. For example:
"aaa","b""bb","ccc"