Skip to content

Commit

Permalink
add ParseReader
Browse files Browse the repository at this point in the history
  • Loading branch information
foolin committed Apr 20, 2020
1 parent 708158d commit 11010a6
Show file tree
Hide file tree
Showing 4 changed files with 75 additions and 6 deletions.
9 changes: 3 additions & 6 deletions _examples/basic/main.go
Original file line number Diff line number Diff line change
Expand Up @@ -2,8 +2,8 @@ package main

import (
"encoding/json"
"fmt"
"github.com/foolin/pagser"
"log"
)

const rawPageHtml = `
Expand Down Expand Up @@ -43,20 +43,17 @@ type ExampPage struct {
func main() {
//New default config
p := pagser.New()

//data parser model
var page ExampPage

//parse html data
err := p.Parse(&page, rawPageHtml)

//check error
if err != nil {
panic(err)
log.Fatal(err)
}

//print data
fmt.Printf("Page data json: \n-------------\n%v\n-------------\n", toJson(page))
log.Printf("Page data json: \n-------------\n%v\n-------------\n", toJson(page))
}

func toJson(v interface{}) string {
Expand Down
45 changes: 45 additions & 0 deletions _examples/http/main.go
Original file line number Diff line number Diff line change
@@ -0,0 +1,45 @@
package main

import (
"encoding/json"
"github.com/foolin/pagser"
"log"
"net/http"
)

type ExampPage struct {
Title string `pagser:"title"`
H1 string `pagser:"h1"`
Navs []struct {
ID int `pagser:"->attrInt(id, -1)"`
Name string `pagser:"a"`
Url string `pagser:"a->attr(href)"`
} `pagser:".navlink li"`
}

func main() {
resp, err := http.Get("https://raw.githubusercontent.com/foolin/pagser/master/_examples/pages/demo.html")
if err != nil {
log.Fatal(err)
}
defer resp.Body.Close()

//New default config
p := pagser.New()
//data parser model
var page ExampPage
//parse html data
err = p.ParseReader(&page, resp.Body)
//check error
if err != nil {
panic(err)
}

//print data
log.Printf("Page data json: \n-------------\n%v\n-------------\n", toJson(page))
}

func toJson(v interface{}) string {
data, _ := json.MarshalIndent(v, "", "\t")
return string(data)
}
17 changes: 17 additions & 0 deletions pagser_test.go
Original file line number Diff line number Diff line change
Expand Up @@ -3,6 +3,7 @@ package pagser
import (
"fmt"
"github.com/PuerkitoBio/goquery"
"net/http"
"strings"
"testing"
)
Expand Down Expand Up @@ -76,6 +77,22 @@ func TestParse(t *testing.T) {
fmt.Printf("json: %v\n", prettyJson(data))
}

func TestPagser_ParseReader(t *testing.T) {
res, err := http.Get("https://raw.githubusercontent.com/foolin/pagser/master/_examples/pages/demo.html")
if err != nil {
t.Fatal(err)
}
defer res.Body.Close()

p := New()
var data HtmlPage
err = p.ParseReader(&data, res.Body)
if err != nil {
t.Fatal(err)
}
fmt.Printf("json: %v\n", prettyJson(data))
}

func TestParseFuncArgs(t *testing.T) {
//const str = `"Hello world, Andy", abc, 123, 1.2,3, -1, ',','abc\'x', -12.3, true, '1,2,3', false, "[1,2,3]",bool`
//const str = `Foo bar random "letters lol" stuff`
Expand Down
10 changes: 10 additions & 0 deletions parse.go
Original file line number Diff line number Diff line change
Expand Up @@ -4,6 +4,7 @@ import (
"fmt"
"github.com/PuerkitoBio/goquery"
"github.com/spf13/cast"
"io"
"reflect"
"strings"
)
Expand All @@ -17,6 +18,15 @@ func (p *Pagser) Parse(v interface{}, document string) (err error) {
return p.ParseDocument(v, reader)
}

// Parse parse html to struct
func (p *Pagser) ParseReader(v interface{}, reader io.Reader) (err error) {
doc, err := goquery.NewDocumentFromReader(reader)
if err != nil {
return err
}
return p.ParseDocument(v, doc)
}

// ParseDocument parse document to struct
func (p *Pagser) ParseDocument(v interface{}, document *goquery.Document) (err error) {
return p.ParseSelection(v, document.Selection)
Expand Down

0 comments on commit 11010a6

Please sign in to comment.