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

require client for parsing

Browse source
This commit is contained in:
Jan Lukas Gernert 2020-02-10 18:01:35 +01:00
parent a7c247549a
commit d2960d8539
2 changed files with 31 additions and 48 deletions
Download patch file Download diff file Expand all files Collapse all files

40
src/images/mod.rs
View file

@ -15,29 +15,18 @@ use url;
mod error; mod error;
pub struct ImageDownloader { pub struct ImageDownloader {
client: Client,
max_size: (u32, u32), max_size: (u32, u32),
} }
impl ImageDownloader { impl ImageDownloader {
pub fn new(max_size: (u32, u32)) -> Self { pub fn new(max_size: (u32, u32)) -> Self {
Self::new_with_client(max_size, Client::new()) ImageDownloader { max_size }
}
pub fn new_with_client(max_size: (u32, u32), client: Client) -> Self {
ImageDownloader {
client,
max_size,
}
}
pub fn set_client(&mut self, client: Client) {
self.client = client;
} }
pub async fn download_images_from_string( pub async fn download_images_from_string(
&self, &self,
html: &str, html: &str,
client: &Client,
) -> Result<String, ImageDownloadError> { ) -> Result<String, ImageDownloadError> {
let parser = Parser::default_html(); let parser = Parser::default_html();
let doc = parser.parse_string(html).map_err(|_| { let doc = parser.parse_string(html).map_err(|_| {
@ -50,7 +39,8 @@ impl ImageDownloader {
ImageDownloadErrorKind::HtmlParse ImageDownloadErrorKind::HtmlParse
})?; })?;
self.download_images_from_context(&xpath_ctx).await?; self.download_images_from_context(&xpath_ctx, client)
.await?;
let options = SaveOptions { let options = SaveOptions {
format: false, format: false,
@ -68,6 +58,7 @@ impl ImageDownloader {
pub async fn download_images_from_context( pub async fn download_images_from_context(
&self, &self,
context: &Context, context: &Context,
client: &Client,
) -> Result<(), ImageDownloadError> { ) -> Result<(), ImageDownloadError> {
let xpath = "//img"; let xpath = "//img";
let node_vec = ArticleScraper::evaluate_xpath(context, xpath, false) let node_vec = ArticleScraper::evaluate_xpath(context, xpath, false)
@ -76,13 +67,13 @@ impl ImageDownloader {
if let Some(url) = node.get_property("src") { if let Some(url) = node.get_property("src") {
if !url.starts_with("data:") { if !url.starts_with("data:") {
if let Ok(url) = url::Url::parse(&url) { if let Ok(url) = url::Url::parse(&url) {
let parent_url = match self.check_image_parent(&node, &url).await { let parent_url = match self.check_image_parent(&node, &url, client).await {
Ok(url) => Some(url), Ok(url) => Some(url),
Err(_) => None, Err(_) => None,
}; };
if let Ok((small_image, big_image)) = if let Ok((small_image, big_image)) =
self.save_image(&url, &parent_url).await self.save_image(&url, &parent_url, client).await
{ {
if let Err(_) = node.set_property("src", &small_image) { if let Err(_) = node.set_property("src", &small_image) {
return Err(ImageDownloadErrorKind::HtmlParse)?; return Err(ImageDownloadErrorKind::HtmlParse)?;
@ -105,9 +96,9 @@ impl ImageDownloader {
&self, &self,
image_url: &url::Url, image_url: &url::Url,
parent_url: &Option<url::Url>, parent_url: &Option<url::Url>,
client: &Client,
) -> Result<(String, Option<String>), ImageDownloadError> { ) -> Result<(String, Option<String>), ImageDownloadError> {
let response = self let response = client
.client
.get(image_url.clone()) .get(image_url.clone())
.send() .send()
.await .await
@ -133,8 +124,7 @@ impl ImageDownloader {
let mut big_image: Option<Vec<u8>> = None; let mut big_image: Option<Vec<u8>> = None;
if let Some(parent_url) = parent_url { if let Some(parent_url) = parent_url {
let response_big = self let response_big = client
.client
.get(parent_url.clone()) .get(parent_url.clone())
.send() .send()
.await .await
@ -271,22 +261,21 @@ impl ImageDownloader {
&self, &self,
node: &Node, node: &Node,
child_url: &url::Url, child_url: &url::Url,
client: &Client,
) -> Result<url::Url, ImageDownloadError> { ) -> Result<url::Url, ImageDownloadError> {
if let Some(parent) = node.get_parent() { if let Some(parent) = node.get_parent() {
if parent.get_name() == "a" { if parent.get_name() == "a" {
if let Some(url) = parent.get_property("href") { if let Some(url) = parent.get_property("href") {
let parent_url = let parent_url =
url::Url::parse(&url).context(ImageDownloadErrorKind::ParentDownload)?; url::Url::parse(&url).context(ImageDownloadErrorKind::ParentDownload)?;
let parent_response = self let parent_response = client
.client
.head(parent_url.clone()) .head(parent_url.clone())
.send() .send()
.await .await
.context(ImageDownloadErrorKind::ParentDownload)?; .context(ImageDownloadErrorKind::ParentDownload)?;
let _ = ImageDownloader::check_image_content_type(&parent_response) let _ = ImageDownloader::check_image_content_type(&parent_response)
.context(ImageDownloadErrorKind::ParentDownload)?; .context(ImageDownloadErrorKind::ParentDownload)?;
let child_response = self let child_response = client
.client
.get(child_url.clone()) .get(child_url.clone())
.send() .send()
.await .await
@ -326,6 +315,7 @@ impl ImageDownloader {
#[cfg(test)] #[cfg(test)]
mod tests { mod tests {
use super::*; use super::*;
use reqwest::Client;
use std::fs; use std::fs;
use std::io::Write; use std::io::Write;
@ -335,7 +325,7 @@ mod tests {
let hdyleaflet = fs::read_to_string(r"./resources/tests/planetGnome/fedora31.html") let hdyleaflet = fs::read_to_string(r"./resources/tests/planetGnome/fedora31.html")
.expect("Failed to read HTML"); .expect("Failed to read HTML");
let result = image_dowloader let result = image_dowloader
.download_images_from_string(&hdyleaflet) .download_images_from_string(&hdyleaflet, &Client::new())
.await .await
.expect("Failed to downalod images"); .expect("Failed to downalod images");
let mut file = let mut file =

39
src/lib.rs
View file

@ -27,15 +27,10 @@ use url;
pub struct ArticleScraper { pub struct ArticleScraper {
pub image_downloader: ImageDownloader, pub image_downloader: ImageDownloader,
config_files: Arc<RwLock<Option<ConfigCollection>>>, config_files: Arc<RwLock<Option<ConfigCollection>>>,
client: Client,
} }
impl ArticleScraper { impl ArticleScraper {
pub fn new(config_path: PathBuf) -> Self { pub fn new(config_path: PathBuf) -> Self {
Self::new_with_client(config_path, Client::new())
}
pub fn new_with_client(config_path: PathBuf, client: Client) -> Self {
let config_files = Arc::new(RwLock::new(None)); let config_files = Arc::new(RwLock::new(None));
let locked_config_files = config_files.clone(); let locked_config_files = config_files.clone();
@ -54,25 +49,19 @@ impl ArticleScraper {
}); });
ArticleScraper { ArticleScraper {
image_downloader: ImageDownloader::new_with_client((2048, 2048), client.clone()), image_downloader: ImageDownloader::new((2048, 2048)),
config_files, config_files,
client,
} }
} }
pub fn set_client(&mut self, client: Client) {
self.client = client.clone();
self.image_downloader.set_client(client);
}
pub async fn parse( pub async fn parse(
&self, &self,
url: url::Url, url: url::Url,
download_images: bool, download_images: bool,
client: &Client,
) -> Result<Article, ScraperError> { ) -> Result<Article, ScraperError> {
info!("Scraping article: '{}'", url.as_str()); info!("Scraping article: '{}'", url.as_str());
let response = self let response = client
.client
.head(url.clone()) .head(url.clone())
.send() .send()
.await .await
@ -117,7 +106,7 @@ impl ArticleScraper {
ArticleScraper::generate_head(&mut root, &document)?; ArticleScraper::generate_head(&mut root, &document)?;
self.parse_pages(&mut article, &url, &mut root, &config) self.parse_pages(&mut article, &url, &mut root, &config, client)
.await?; .await?;
let context = Context::new(&document).map_err(|()| { let context = Context::new(&document).map_err(|()| {
@ -138,7 +127,7 @@ impl ArticleScraper {
if download_images { if download_images {
if let Err(error) = self if let Err(error) = self
.image_downloader .image_downloader
.download_images_from_context(&context) .download_images_from_context(&context, client)
.await .await
{ {
error!("Downloading images failed: '{}'", error); error!("Downloading images failed: '{}'", error);
@ -168,8 +157,9 @@ impl ArticleScraper {
url: &url::Url, url: &url::Url,
root: &mut Node, root: &mut Node,
config: &GrabberConfig, config: &GrabberConfig,
client: &Client,
) -> Result<(), ScraperError> { ) -> Result<(), ScraperError> {
let html = ArticleScraper::download(&url, &self.client).await?; let html = ArticleScraper::download(&url, client).await?;
let mut document = Self::parse_html(html, config)?; let mut document = Self::parse_html(html, config)?;
let mut xpath_ctx = Self::get_xpath_ctx(&document)?; let mut xpath_ctx = Self::get_xpath_ctx(&document)?;
@ -183,9 +173,10 @@ impl ArticleScraper {
if !result.trim().is_empty() { if !result.trim().is_empty() {
// parse again with single page url // parse again with single page url
debug!("Single page link found '{}'", result); debug!("Single page link found '{}'", result);
let single_page_url = url::Url::parse(&result).context(ScraperErrorKind::Url)?; let single_page_url =
url::Url::parse(&result).context(ScraperErrorKind::Url)?;
return self return self
.parse_single_page(article, &single_page_url, root, config) .parse_single_page(article, &single_page_url, root, config, client)
.await; .await;
} }
} }
@ -197,7 +188,7 @@ impl ArticleScraper {
loop { loop {
if let Some(url) = self.check_for_next_page(&xpath_ctx, config) { if let Some(url) = self.check_for_next_page(&xpath_ctx, config) {
let html = ArticleScraper::download(&url, &self.client).await?; let html = ArticleScraper::download(&url, client).await?;
document = Self::parse_html(html, config)?; document = Self::parse_html(html, config)?;
xpath_ctx = Self::get_xpath_ctx(&document)?; xpath_ctx = Self::get_xpath_ctx(&document)?;
ArticleScraper::strip_junk(&xpath_ctx, config, &url); ArticleScraper::strip_junk(&xpath_ctx, config, &url);
@ -261,8 +252,9 @@ impl ArticleScraper {
url: &url::Url, url: &url::Url,
root: &mut Node, root: &mut Node,
config: &GrabberConfig, config: &GrabberConfig,
client: &Client,
) -> Result<(), ScraperError> { ) -> Result<(), ScraperError> {
let html = ArticleScraper::download(&url, &self.client).await?; let html = ArticleScraper::download(&url, client).await?;
let document = Self::parse_html(html, config)?; let document = Self::parse_html(html, config)?;
let xpath_ctx = Self::get_xpath_ctx(&document)?; let xpath_ctx = Self::get_xpath_ctx(&document)?;
ArticleScraper::extract_metadata(&xpath_ctx, config, article); ArticleScraper::extract_metadata(&xpath_ctx, config, article);
@ -811,6 +803,7 @@ impl ArticleScraper {
#[cfg(test)] #[cfg(test)]
mod tests { mod tests {
use crate::*; use crate::*;
use reqwest::Client;
#[tokio::test(basic_scheduler)] #[tokio::test(basic_scheduler)]
async fn golem() { async fn golem() {
@ -819,7 +812,7 @@ mod tests {
let url = url::Url::parse("https://www.golem.de/news/http-error-418-fehlercode-ich-bin-eine-teekanne-darf-bleiben-1708-129460.html").unwrap(); let url = url::Url::parse("https://www.golem.de/news/http-error-418-fehlercode-ich-bin-eine-teekanne-darf-bleiben-1708-129460.html").unwrap();
let grabber = ArticleScraper::new(config_path); let grabber = ArticleScraper::new(config_path);
let article = grabber.parse(url, true).await.unwrap(); let article = grabber.parse(url, true, &Client::new()).await.unwrap();
article.save_html(&out_path).unwrap(); article.save_html(&out_path).unwrap();
assert_eq!( assert_eq!(
@ -841,7 +834,7 @@ mod tests {
.unwrap(); .unwrap();
let grabber = ArticleScraper::new(config_path); let grabber = ArticleScraper::new(config_path);
let article = grabber.parse(url, true).await.unwrap(); let article = grabber.parse(url, true, &Client::new()).await.unwrap();
article.save_html(&out_path).unwrap(); article.save_html(&out_path).unwrap();
assert_eq!( assert_eq!(