Trim, sanitize, and modify struct string fields in place, based on tags.
Now also works with embedded structs
Turns this...
typePersonstruct {
FirstNamestring`conform:"name"`LastNamestring`conform:"ucfirst,trim"`Emailstring`conform:"email"`CamelCasestring`conform:"camel"`UserNamestring`conform:"snake"`Slugstring`conform:"slug"`Blurbstring`conform:"title"`Leftstring`conform:"ltrim"`Rightstring`conform:"rtrim"`Skills []string`conform:"upper"`Examplesmap[string]string`conform:"!html"`
}
p1:=Person{
" LEE ",
" Benson",
" LEE@LEEbenson.com ",
"I love new york city",
"lee benson",
"LeeBensonWasHere",
"this is a little bit about me...",
" Left trim ",
" Right trim ",
[]string{"HtmL", "Yaml"},
map[string]string{"<best>":"<body><p>I know this & that.</p></body>"},
}Into this...
p2:=p1// <-- copy the Person struct into a new one, to see the differenceconform.Strings(&p2) // <-- this does the work/* p1 (left) vs. p2 (right) FirstName: ' LEE ' -> 'Lee' LastName: ' Benson' -> 'Benson' Email: ' LEE@LEEbenson.com ' -> 'lee@leebenson.com' CamelCase: 'I love new york city' -> 'ILoveNewYorkCity' UserName: 'lee benson' -> 'lee_benson' Slug: 'LeeBensonWasHere' -> 'lee-benson-was-here' Blurb: 'this is a little bit about me...' -> 'This Is A Little Bit About Me...' Left: ' Left trim ' -> 'Left trim ' Right: ' Right trim ' -> ' Right trim', Skills: { 'HtmL', 'Yaml' } -> { 'HTML', 'YAML' }, Examples: { '<best>': '<body><p>I know this & that.</p></body>' } -> { '<best>': '<body><p>I know this & that.</p></body>' }*/Note: No map keys are changed.
Conform helps you fix and format user strings quickly, without writing functions.
If you do form processing with Gorilla Schema or similar, you probably shuttle user data into structs using tags. Adding a conform tag to your string field gives you "first pass" clean up against user input.
Use it for names, e-mail addresses, URL slugs, or any other form field where formatting matters.
Conform doesn't attempt any kind of validation on your fields. Check out govalidator for a slew of common validation funcs, or validator which is an uber-flexible Swiss Army knife for validating pretty much any kind of data you can imagine. Both have struct tag syntax and can be used with conform.
Grab the package from the command line with:
go get github.com/leebenson/conform
And import in the usual way in your Go app:
import "github.com/leebenson/conform"
Add a conform tag to your structs, for all of the string fields that you want Conform to transform. Add the name of the transform (known as the "tag") in double quotes, and separate multiple tags with commas. Example: conform:"trim,lowercase"
To format in place, pass your struct pointer to conform.Strings.
Note: your struct will be edited in place. This will OVERWRITE any data that is already stored in your string fields.
Here's an example that formats e-mail addresses:
package main
import (
"fmt""github.com/leebenson/conform"
)
typeUserFormstruct {
Emailstring`conform:"email"`
}
funcmain() {
input:=UserForm{
Email: " POORLYFormaTTED@EXAMPlE.COM ",
}
conform.Strings(&input) // <-- pass in a pointer to your structfmt.Println(input.Email) // prints "poorlyformatted@example.com"
}Just add a conform tag along with your Gorilla schema tags:
// ...import (
"net/http""github.com/gorilla/schema""github.com/leebenson/conform"
)
// the struct that will be filled from the post request...typenewUserFormstruct {
FirstNamestring`schema:"firstName" conform:"name"`Emailstring`schema:"emailAddress" conform:"email"`Passwordstring`schema:"password"`// <-- no tag? no changeDob time.Time`schema:"dateOfBirth"`// <-- non-strings ignored by conform
}
// ProcessNewUser attempts to register a new userfuncProcessNewUser(r*http.Request) error {
form:=new(newUserForm)
schema.NewDecoder().Decode(form, r.PostForm) // <-- Gorilla Schemaconform.Strings(form) // <-- Conform. Pass in the same pointer that Schema used// ...
}
// HTTP handlers, etc...See the public API / exported methods on Godoc.
You can use multiple tags in the format of conform:"tag1,tag2"
Trims leading and trailing spaces. Example: " string " -> "string"
Trims leading spaces only. Example: " string " -> "string "
Trims trailing spaces only. Example: " string " -> " string"
Converts string to lowercase. Example: "STRING" -> "string"
Converts string to uppercase. Example: "string" -> "STRING"
Converts string to Title Case, e.g. "this is a sentence" -> "This Is A Sentence"
Converts to camel case via stringUp, Example provided by library: this is it => thisIsIt, this\_is\_it => thisIsIt, this-is-it => thisIsIt
Converts to snake_case. Example: "CamelCase" -> "camel_case", "regular string" -> "regular_string"
Special thanks to snaker for inspiration (credited in license)
Turns strings into slugs. Example: "CamelCase" -> "camel-case", "blog title here" -> "blog-title-here"
Uppercases first character. Example: "all lower" -> "All lower"
Trims, strips numbers and special characters (except dashes and spaces separating names), converts multiple spaces and dashes to single characters, title cases multiple names. Example: "3493€848Jo-s$%£@Ann " -> "Jo-Ann", " ~~ The Dude ~~" -> "The Dude", "**susan**" -> "Susan", " hugh fearnley-whittingstall" -> "Hugh Fearnley-Whittingstall"
Trims and lowercases the domain portion of the string. Example: "UNSIGHTLY-EMAIL@EXamPLE.com " -> "UNSIGHTLY-EMAIL@example.com"
Removes all non-numeric characters. Example: "the price is €30,38" -> "3038"
Note: The struct field will remain a string. No type conversion takes place.
Removes all numbers. Example "39472349D34a34v69e8932747" -> "Dave"
Removes non-alpha unicode characters. Example: "!@£$%^&'()Hello 1234567890 World+[];\" -> "HelloWorld"
Removes alpha unicode characters. Example: "Everything's here but the letters!" -> "' !"
Escapes HTML so that it is safe for display. Characters are substituted by their respective HTML codes. Internally uses template.HTMLEscapeString.
Example: "' " & < > \000" -> "' " & < > \uFFFD"
Escapes JavaScript. Internally uses template.JSEscapeString. Example: "\ ' " < > & =" -> "\\ \' \u003C \u003E \u0026 \u003D"