shimun/htmlextract
1
0
Fork 0
Code Issues Pull Requests Releases 1 Wiki Activity

serialize to html instead of text [CI SKIP]

Browse Source
This commit is contained in:
Drone CI 2019-03-25 14:12:08 +01:00
parent f07983db53
commit 5b1c58c659
3 changed files with 15 additions and 0 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

2
Cargo.lock generated
View File

@ -147,7 +147,9 @@ dependencies = [
name = "htmlextract"
version = "0.1.0"
dependencies = [
"html5ever 0.22.5 (registry+https://github.com/rust-lang/crates.io-index)",
"kuchiki 0.7.2 (registry+https://github.com/rust-lang/crates.io-index)",
"markup5ever 0.7.5 (registry+https://github.com/rust-lang/crates.io-index)",
"structopt 0.2.15 (registry+https://github.com/rust-lang/crates.io-index)",
]

2
Cargo.toml
View File

@ -13,3 +13,5 @@ opt-level = 'z'
[dependencies]
kuchiki = "0.7.2"
structopt = "0.2.15"
html5ever = "0.22.5"
markup5ever = "0.7.5"

11
src/main.rs
View File

@ -4,6 +4,9 @@ use structopt::StructOpt;
mod opts;
use html5ever::serialize::{serialize, SerializeOpts};
use kuchiki::NodeRef;
use markup5ever::serialize::Serialize as MSerialize;
use opts::Opts;
use std::error::Error;
use std::fs::File;
@ -20,6 +23,13 @@ fn read_html<I: Read>(source: &mut I) -> io::Result<String> {
Ok(html)
}
fn serialize_node<N: MSerialize>(node: &N) -> Result<String, Box<dyn Error>> {
let mut bytes = Vec::new();
serialize(&mut bytes, node, SerializeOpts::default())?;
let s = String::from_utf8(bytes)?;
Ok(s)
}
fn main() -> Result<(), Box<Error>> {
let stdin = stdin();
let opt = Opts::from_args();
@ -58,6 +68,7 @@ fn main() -> Result<(), Box<Error>> {
//let text = text_node.as_text().unwrap().borrow();
if let Some(child) = as_node.first_child() {
dbg!(serialize_node(&child));
//TODO: Convert Nodes to String, as this only works for Nodes containing plain text
if let Some(text) = child.as_text() {
// Prints: