soupvschompjs
soup is a Go library for parsing and querying HTML documents.
It provides a simple and intuitive interface for extracting information from HTML pages. It's inspired by popular Python web scraping
library BeautifulSoup and shares similar use API implementing functions like Find and FindAll.
soup can also use go's built-in http client to download HTML content.
Note that unlike beautifulsoup, soup does not support CSS selectors or XPath.
chompjs can be used in web scrapping for turning JavaScript objects embedded in pages into valid Python dictionaries.
In web scraping this is particularly useful for parsing Javascript variables like:
import chompjs
js = """
var myObj = {
myMethod: function(params) {
// ...
},
myValue: 100
}
"""
chompjs.parse_js_object(js, json_params={'strict': False})
{'myMethod': 'function(params) {\n // ...\n }', 'myValue': 100}
In practice this can be used to extract hidden JSON data like data from <script id=__NEXT_DATA__> elements
from nextjs (and similar) websites. Unlike json.loads command chompjs can ingest json documents that contain
javascript natives like functions making it a super easy way to scrape hidden web data objects.
Example Use
package main
import (
"fmt"
"log"
"github.com/anaskhan96/soup"
)
func main() {
url := "https://www.bing.com/search?q=weather+Toronto"
# soup has basic HTTP client though it's not recommended for scraping:
resp, err := soup.Get(url)
if err != nil {
log.Fatal(err)
}
# create soup object from HTML
doc := soup.HTMLParse(resp)
# html elements can be found using Find or FindStrict methods:
# in this case find <div> elements where "class" attribute matches some values:
grid := doc.FindStrict("div", "class", "b_antiTopBleed b_antiSideBleed b_antiBottomBleed")
# note: to find all elements FindAll() method can be used the same way
# elements can be further searched for descendents:
heading := grid.Find("div", "class", "wtr_titleCtrn").Find("div").Text()
conditions := grid.Find("div", "class", "wtr_condition")
primaryCondition := conditions.Find("div")
secondaryCondition := primaryCondition.FindNextElementSibling()
temp := primaryCondition.Find("div", "class", "wtr_condiTemp").Find("div").Text()
others := primaryCondition.Find("div", "class", "wtr_condiAttribs").FindAll("div")
caption := secondaryCondition.Find("div").Text()
fmt.Println("City Name : " + heading)
fmt.Println("Temperature : " + temp + "˚C")
for _, i := range others {
fmt.Println(i.Text())
}
fmt.Println(caption)
}
# basic use
import chompjs
js = """
var myObj = {
myMethod: function(params) {
// ...
},
myValue: 100
}
"""
chompjs.parse_js_object(js, json_params={'strict': False})
{'myMethod': 'function(params) {\n // ...\n }', 'myValue': 100}
# example how to use with hidden data parsing:
import httpx
import chompjs
from parsel import Selector
response = httpx.get("http://example.com")
hidden_script = Selector(response.text).css("script#__NEXT_DATA__::text").get()
data = chompjs.parse_js_object(hidden_script)
print(data['props'])