misc-mirrors/article_scraper
1
0
Fork 0
mirror of https://gitlab.com/news-flash/article_scraper.git synced 2025-07-08 08:30:00 +02:00
Code Releases Activity

wip: test libxml XML_SAVE_NO_EMPTY option

Browse source
This commit is contained in:
Jan Lukas Gernert 2019-09-24 18:45:06 +02:00
parent a44ac3663c
commit a99b8dec47
2 changed files with 64 additions and 0 deletions
Download patch file Download diff file Expand all files Collapse all files

21
src/images/mod.rs
View file

@ -46,6 +46,7 @@ impl ImageDownloader {
self.download_images_from_context(&xpath_ctx)?;
Ok(doc.to_string(/*format:*/ false))
//Ok(doc.to_string_custom())
}
pub fn download_images_from_context(&self, context: &Context) -> Result<(), ImageDownloadError> {
@ -236,4 +237,24 @@ impl ImageDownloader {
}
Err(ImageDownloadErrorKind::ContentLenght)?
}
}
#[cfg(test)]
mod tests {
use super::*;
use std::fs;
use std::io::Write;
#[test]
pub fn close_tags() {
let image_dowloader = ImageDownloader::new((2048, 2048));
let hdyleaflet = fs::read_to_string(r"./resources/tests/planetGnome/fedora31.html")
.expect("Failed to read HTML");
let result = image_dowloader.download_images_from_string(&hdyleaflet)
.expect("Failed to downalod images");
let mut file = fs::File::create(r"./resources/tests/planetGnome/fedora31_images_downloaded.html")
.expect("Failed to create output file");
file.write_all(result.as_bytes()).expect("Failed to write result to file");
}
}