README
¶
csvutil - Your CSV pocket-knife (golang)
#WARNING
I would advise against using this package. It was a language learning exercise from a time before "encoding/csv" existed. If encoding/csv doesn't fit your needs I would advise finding a different maintained package or writing a new one
Version
This is csvutil version 1.1_2
Synopsis
The csvutil package can be used to read CSV data from any io.Reader and, write CSV data to any io.Writer. It can automatically generate CSV rows from slices containing native Go types, other slices/arrays of native go types, or flat structs. It can also convert CSV rows and assign values to memory referenced by a slice of pointers.
package main
import "os"
import "github.com/bmatsuo/csvutil"
type Person struct {
Name string
Height float64
Weight float64
}
func main() {
writer := csvutil.NewWriter(os.Stdout, nil)
errdo := csvutil.DoFile(os.Args[1], func(r csvutil.Row) bool {
if r.HasError() {
panic(r.Error)
}
var person Person
if _, errc := r.Format(&person); errc != nil {
panic("Row is not a Person")
}
bmi := person.Weight / (person.Height * person.Height)
writer.WriteRow(csvutil.FormatRow(person, bmi).Fields...)
return true
})
if errdo != nil {
panic(errdo)
}
writer.Flush()
}
Given a CSV file 'in.csv' with contents
alice,1.4,50
bob,2,80
chris,1.6,67
When the above program is called and given the path 'in.csv', the following is printed to standard output.
alice,1.4,50,25.510204081632658
bob,2,80,20
chris,1.6,67,26.171874999999996
About
The csvutil package is written to make interacting with CSV data as easy as possible. Efficiency of its underlying functions and methods are slightly less important factors in its design. However, that being said, it is important. And, csvutil should be capable of handling both extremely large and fairly small streams of CSV data through input and output quite well in terms of speed and memory footprint. It should do this while not making your code bend over backwards (more than necessary). As libraries should never make you do.
Features
-
Slurping/spewing CSV data. That is, reading/writing whole files or data streams at once.
-
Iteration through individual rows of a CSV data stream for a smaller memory footprint.
-
Writing of individual writing fields/rows (along with batch writing).
-
Automated CSV row serialization and deserialization (formatting) for flat data structures and types.
Todo
- Enhance and clean the formatting API and allow better formatting of data.
Install
The easiest installation of csvutil is done through goinstall.
goinstall github.com/bmatsuo/csvutil
Documentation
The best way to read the current csvutil documentation is using godoc.
godoc github.com/bmatsuo/csvutil
Or better yet, you can run a godoc http server.
godoc -http=":6060"
Then go to the url http://localhost:6060/pkg/github.com/bmatsuo/csvutil/
Copyright & License
Copyright (c) 2011, Bryan Matsuo. All rights reserved.
Use of this source code is governed by a BSD-style license that can be found in the LICENSE file.
Documentation
¶
Overview ¶
Package csvutil is not maintained. It provides formatted reading and writing of CSV data.
Index ¶
- Variables
- func Do(r io.Reader, f func(r Row) bool)
- func DoFile(filename string, f func(r Row) bool) error
- func Read(r io.Reader) ([][]string, error)
- func ReadFile(filename string) ([][]string, error)
- func Write(w io.Writer, rows [][]string) (int, error)
- func WriteFile(filename string, perm os.FileMode, rows [][]string) (int, error)
- type Config
- type Reader
- func (csvr *Reader) Do(f func(Row) bool)
- func (csvr *Reader) DoN(n int, f func(Row) bool)
- func (csvr *Reader) LineNum() int
- func (csvr *Reader) ReadRow() Row
- func (csvr *Reader) ReadRows(rbuf [][]string) (int, error)
- func (csvr *Reader) RemainingRows() (rows [][]string, err error)
- func (csvr *Reader) RemainingRowsSize(size int) ([][]string, error)
- type Row
- type Writer
Constants ¶
This section is empty.
Variables ¶
var ( FloatFmt byte = 'g' FloatPrec int = -1 )
See strconv.Ftoa32()
var ( ErrorIndex = errors.New("Not enough fields to format") ErrorStruct = errors.New("Cannot format unreferenced structs") ErrorUnimplemented = errors.New("Unimplemented field type") ErrorFieldType = errors.New("Field type incompatible.") ErrorNonPointer = errors.New("Target is not a pointer.") ErrorCantSet = errors.New("Cannot set value.") )
var ( DefaultConfig = &Config{ Sep: ',', Trim: false, Cutset: " \t", CommentPrefix: "#", Comments: false, CommentsInBody: false} )
The default configuration is used for Readers and Writers when none is given.
Functions ¶
Types ¶
type Config ¶
type Config struct {
// General configuration
// Field seperator
Sep rune
// Trim leading/trailing whitespace in fields.
Trim bool
// Characters to trim from fields.
Cutset string
// Prefix for comment lines.
CommentPrefix string
// Reader specific config
// Are comments allowed in the input.
Comments bool
// Comments can appear in the body (Comments must be true).
CommentsInBody bool
}
A configuration structure that can be shared between a Reader and Writer.
func NewConfig ¶
func NewConfig() *Config
Return a freshly allocated Config that is initialized to DefaultConfig.
func (*Config) LooksLikeComment ¶
type Reader ¶
type Reader struct {
*Config
// contains filtered or unexported fields
}
A reader object for CSV data utilizing the bufio package.
func NewReaderSize ¶
Create a new reader with a buffer of a specified size.
func (*Reader) Do ¶
Iteratively read the remaining rows in the reader and call f on each of them. If f returns false, no more rows will be read.
func (*Reader) DoN ¶
Process rows from the reader like Do, but stop after processing n of them. If f returns false before n rows have been process, no more rows will be processed.
func (*Reader) ReadRow ¶
Attempt to read up to a new line, skipping any comment lines found in the process. Return a Row object containing the fields read and any error encountered.
func (*Reader) ReadRows ¶
Read rows into a preallocated buffer. Return the number of rows read, and any error encountered.
func (*Reader) RemainingRows ¶
Reads any remaining rows of CSV data in the underlying io.Reader.
type Row ¶
A simple row structure for rows read by a csvutil.Reader that encapsulates any read error enountered along with any data read prior to encountering an error.
func FormatRow ¶
func FormatRow(x ...interface{}) Row
Iteratively take values from the argument list and formats them (or their elements/fields) as a (list of) string(s). Returns a Row object that contains the formatted arguments, as well as any error that occured.
type Writer ¶
type Writer struct {
*Config
// contains filtered or unexported fields
}
A simple CSV file writer using the package bufio for effeciency. But, because of this, the method Flush() must be called to ensure data is written to any given io.Writer before it is closed.
func NewWriter ¶
Create a new CSV writer with the default field seperator and a buffer of a default size.
func NewWriterSize ¶
Create a new CSV writer using a buffer of at least n bytes.
See bufio.NewWriterSize(io.Writer, int) (*bufio.NewWriter).
func (*Writer) WriteComments ¶
Write a comment. Each comment string given will start on a new line. If the string is contains multiple lines, comment prefixes will be inserted at the beginning of each one.
func (*Writer) WriteFields ¶
Write a slice of field values with a trailing field seperator (no '\n'). Returns any error incurred from writing.
Source Files