gcfg.v1: gopkg.in/gcfg.v1 Index | Examples | Files | Directories

package gcfg

import "gopkg.in/gcfg.v1"

Package gcfg reads "INI-style" text-based configuration files with "name=value" pairs grouped into sections (gcfg files).

This package is still a work in progress; see the sections below for planned changes.

Syntax

The syntax is based on that used by git config: http://git-scm.com/docs/git-config#_syntax . There are some (planned) differences compared to the git config format:

- improve data portability:
  - must be encoded in UTF-8 (for now) and must not contain the 0 byte
  - include and "path" type is not supported
    (path type may be implementable as a user-defined type)
- internationalization
  - section and variable names can contain unicode letters, unicode digits
    (as defined in http://golang.org/ref/spec#Characters ) and hyphens
    (U+002D), starting with a unicode letter
- disallow potentially ambiguous or misleading definitions:
  - `[sec.sub]` format is not allowed (deprecated in gitconfig)
  - `[sec ""]` is not allowed
    - use `[sec]` for section name "sec" and empty subsection name
  - (planned) within a single file, definitions must be contiguous for each:
    - section: '[secA]' -> '[secB]' -> '[secA]' is an error
    - subsection: '[sec "A"]' -> '[sec "B"]' -> '[sec "A"]' is an error
    - multivalued variable: 'multi=a' -> 'other=x' -> 'multi=b' is an error

Data structure

The functions in this package read values into a user-defined struct. Each section corresponds to a struct field in the config struct, and each variable in a section corresponds to a data field in the section struct. The mapping of each section or variable name to fields is done either based on the "gcfg" struct tag or by matching the name of the section or variable, ignoring case. In the latter case, hyphens '-' in section and variable names correspond to underscores '_' in field names. Fields must be exported; to use a section or variable name starting with a letter that is neither upper- or lower-case, prefix the field name with 'X'. (See https://code.google.com/p/go/issues/detail?id=5763#c4 .)

For sections with subsections, the corresponding field in config must be a map, rather than a struct, with string keys and pointer-to-struct values. Values for subsection variables are stored in the map with the subsection name used as the map key. (Note that unlike section and variable names, subsection names are case sensitive.) When using a map, and there is a section with the same section name but without a subsection name, its values are stored with the empty string used as the key. It is possible to provide default values for subsections in the section "default-<sectionname>" (or by setting values in the corresponding struct field "Default_<sectionname>").

The functions in this package panic if config is not a pointer to a struct, or when a field is not of a suitable type (either a struct or a map with string keys and pointer-to-struct values).

Parsing of values

The section structs in the config struct may contain single-valued or multi-valued variables. Variables of unnamed slice type (that is, a type starting with `[]`) are treated as multi-value; all others (including named slice types) are treated as single-valued variables.

Single-valued variables are handled based on the type as follows. Unnamed pointer types (that is, types starting with `*`) are dereferenced, and if necessary, a new instance is allocated.

For types implementing the encoding.TextUnmarshaler interface, the UnmarshalText method is used to set the value. Implementing this method is the recommended way for parsing user-defined types.

For fields of string kind, the value string is assigned to the field, after unquoting and unescaping as needed. For fields of bool kind, the field is set to true if the value is "true", "yes", "on" or "1", and set to false if the value is "false", "no", "off" or "0", ignoring case. In addition, single-valued bool fields can be specified with a "blank" value (variable name without equals sign and value); in such case the value is set to true.

Predefined integer types [u]int(|8|16|32|64) and big.Int are parsed as decimal or hexadecimal (if having '0x' prefix). (This is to prevent unintuitively handling zero-padded numbers as octal.) Other types having [u]int* as the underlying type, such as os.FileMode and uintptr allow decimal, hexadecimal, or octal values. Parsing mode for integer types can be overridden using the struct tag option ",int=mode" where mode is a combination of the 'd', 'h', and 'o' characters (each standing for decimal, hexadecimal, and octal, respectively.)

All other types are parsed using fmt.Sscanf with the "%v" verb.

For multi-valued variables, each individual value is parsed as above and appended to the slice. If the first value is specified as a "blank" value (variable name without equals sign and value), a new slice is allocated; that is any values previously set in the slice will be ignored.

The types subpackage for provides helpers for parsing "enum-like" and integer types.

Error handling

There are 3 types of errors:

- programmer errors / panics:
  - invalid configuration structure
- data errors:
  - fatal errors:
    - invalid configuration syntax
  - warnings:
    - data that doesn't belong to any part of the config structure

Programmer errors trigger panics. These are should be fixed by the programmer before releasing code that uses gcfg.

Data errors cause gcfg to return a non-nil error value. This includes the case when there are extra unknown key-value definitions in the configuration data (extra data). However, in some occasions it is desirable to be able to proceed in situations when the only data error is that of extra data. These errors are handled at a different (warning) priority and can be filtered out programmatically. To ignore extra data warnings, wrap the gcfg.Read*Into invocation into a call to gcfg.FatalOnly.

TODO

The following is a list of changes under consideration:

- documentation
  - self-contained syntax documentation
  - more practical examples
  - move TODOs to issue tracker (eventually)
- syntax
  - reconsider valid escape sequences
    (gitconfig doesn't support \r in value, \t in subsection name, etc.)
- reading / parsing gcfg files
  - define internal representation structure
  - support multiple inputs (readers, strings, files)
  - support declaring encoding (?)
  - support varying fields sets for subsections (?)
- writing gcfg files
- error handling
  - make error context accessible programmatically?
  - limit input size?

Index

Examples

Package Files

doc.go errors.go read.go set.go

func FatalOnly Uses

func FatalOnly(err error) error

FatalOnly filters the results of a Read*Into invocation and returns only fatal errors. That is, errors (warnings) indicating data for unknown sections / variables is ignored. Example invocation:

err := gcfg.FatalOnly(gcfg.ReadFileInto(&cfg, configFile))
if err != nil {
    ...

func ReadFileInto Uses

func ReadFileInto(config interface{}, filename string) error

ReadFileInto reads gcfg formatted data from the file filename and sets the values into the corresponding fields in config.

For compatibility with files created on Windows, the ReadFileInto skips a single leading UTF8 BOM sequence if it exists.

func ReadInto Uses

func ReadInto(config interface{}, reader io.Reader) error

ReadInto reads gcfg formatted data from reader and sets the values into the corresponding fields in config.

func ReadStringInto Uses

func ReadStringInto(config interface{}, str string) error

ReadStringInto reads gcfg formatted data from str and sets the values into the corresponding fields in config.

Code:

cfgStr := `; Comment line
[section]
name=value # comment`
cfg := struct {
    Section struct {
        Name string
    }
}{}
err := gcfg.ReadStringInto(&cfg, cfgStr)
if err != nil {
    log.Fatalf("Failed to parse gcfg data: %s", err)
}
fmt.Println(cfg.Section.Name)

Output:

value

Code:

cfgStr := `; Comment line
[section]
switch=on`
cfg := struct {
    Section struct {
        Switch bool
    }
}{}
err := gcfg.ReadStringInto(&cfg, cfgStr)
if err != nil {
    log.Fatalf("Failed to parse gcfg data: %s", err)
}
fmt.Println(cfg.Section.Switch)

Output:

true

Code:

cfgStr := `; Comment line
[section-name]
variable-name=value # comment`
cfg := struct {
    Section_Name struct {
        Variable_Name string
    }
}{}
err := gcfg.ReadStringInto(&cfg, cfgStr)
if err != nil {
    log.Fatalf("Failed to parse gcfg data: %s", err)
}
fmt.Println(cfg.Section_Name.Variable_Name)

Output:

value

Code:

cfgStr := `; Comment line
[section]
multi=value1
multi=value2`
cfg := struct {
    Section struct {
        Multi []string
    }
}{}
err := gcfg.ReadStringInto(&cfg, cfgStr)
if err != nil {
    log.Fatalf("Failed to parse gcfg data: %s", err)
}
fmt.Println(cfg.Section.Multi)

Output:

[value1 value2]

Code:

cfgStr := `; Comment line
[profile "A"]
color = white

[profile "B"]
color = black
`
cfg := struct {
    Profile map[string]*struct {
        Color string
    }
}{}
err := gcfg.ReadStringInto(&cfg, cfgStr)
if err != nil {
    log.Fatalf("Failed to parse gcfg data: %s", err)
}
fmt.Printf("%s %s\n", cfg.Profile["A"].Color, cfg.Profile["B"].Color)

Output:

white black

Code:

cfgStr := `; Comment line
[section]
var-name=value # comment`
cfg := struct {
    Section struct {
        FieldName string `gcfg:"var-name"`
    }
}{}
err := gcfg.ReadStringInto(&cfg, cfgStr)
if err != nil {
    log.Fatalf("Failed to parse gcfg data: %s", err)
}
fmt.Println(cfg.Section.FieldName)

Output:

value

Code:

cfgStr := `; Comment line
[甲]
乙=丙 # comment`
cfg := struct {
    X甲 struct {
        X乙 string
    }
}{}
err := gcfg.ReadStringInto(&cfg, cfgStr)
if err != nil {
    log.Fatalf("Failed to parse gcfg data: %s", err)
}
fmt.Println(cfg.X甲.X乙)

Output:

Directories

PathSynopsis
scannerPackage scanner implements a scanner for gcfg configuration text.
tokenPackage token defines constants representing the lexical tokens of the gcfg configuration syntax and basic operations on tokens (printing, predicates).
typesPackage types defines helpers for type conversions.

Package gcfg imports 16 packages (graph) and is imported by 1065 packages. Updated 2019-11-15. Refresh now. Tools for package owners.