Last active
May 5, 2017 18:36
-
-
Save akiraak/cec2e0acc7a127220b602c91515b18e1 to your computer and use it in GitHub Desktop.
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
package main | |
import ( | |
"fmt" | |
"github.com/PuerkitoBio/goquery" | |
"io/ioutil" | |
"strconv" | |
"strings" | |
"time" | |
) | |
type ImageInfo struct { | |
url string | |
width int | |
height int | |
} | |
type BookInfo struct { | |
asin string | |
title string | |
binding string | |
author string | |
publisher string | |
publicationDate string | |
images map[string]ImageInfo | |
} | |
func parseXmls(xmls []string) []BookInfo { | |
bookInfos := []BookInfo{} | |
for _, xml := range xmls { | |
dom, _ := goquery.NewDocumentFromReader(strings.NewReader(xml)) | |
dom.Find("Item").Each(func(_ int, item *goquery.Selection) { | |
bookInfo := BookInfo{} | |
bookInfo.asin = item.Find("ASIN").Text() | |
attributes := item.Find("ItemAttributes").First() | |
if attributes.Length() > 0 { | |
bookInfo.title = attributes.Find("Title").Text() | |
bookInfo.binding = attributes.Find("Binding").Text() | |
bookInfo.author = attributes.Find("Author").Text() | |
bookInfo.publisher = attributes.Find("Publisher").Text() | |
bookInfo.publicationDate = attributes.Find("PublicationDate").Text() | |
} | |
imageLabels := []string{ | |
"SmallImage", | |
"MediumImage", | |
"LargeImage", | |
} | |
images := map[string]ImageInfo{} | |
for _, imageLabel := range imageLabels { | |
xml := item.Find(imageLabel).First() | |
url := xml.Find("URL").Text() | |
width, _ := strconv.Atoi(xml.Find("Height").Text()) | |
height, _ := strconv.Atoi(xml.Find("Width").Text()) | |
image := ImageInfo{url, width, height} | |
images[imageLabel] = image | |
} | |
bookInfo.images = images | |
bookInfos = append(bookInfos, bookInfo) | |
}) | |
} | |
return bookInfos | |
} | |
func getXmls() []string { | |
xmls := []string{} | |
for i := 0; i <= 1440; i += 10 { | |
path := fmt.Sprintf("xmls/%d.xml", i) | |
xml, _ := ioutil.ReadFile(path) | |
xmls = append(xmls, string(xml)) | |
} | |
return xmls | |
} | |
func main() { | |
xmls := getXmls() | |
start := time.Now() | |
bookInfos := parseXmls(xmls) | |
end := time.Now() | |
fmt.Printf("xml数: %d\n", len(xmls)) | |
fmt.Printf("book数: %d\n", len(bookInfos)) | |
fmt.Printf("parse時間: %f秒\n", (end.Sub(start)).Seconds()) | |
} |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment