Modularize scraping and api requests

This commit is contained in:
2021-07-09 00:37:08 -04:00
parent 19350081c3
commit 22757bc475
5 changed files with 416 additions and 292 deletions

View File

@@ -0,0 +1,70 @@
use anyhow::{anyhow, Result};
use futures::TryStreamExt;
use reqwest::Client;
use serde_json::Value;
use std::{env, time::Duration};
use tempfile::tempfile;
use tokio::fs::File;
use tokio_util::compat::FuturesAsyncReadCompatExt;
use super::{rate_limit_wait_duration, GAME_NAME, USER_AGENT};
pub struct DownloadLinkResponse {
pub wait: Option<Duration>,
json: Value,
}
pub async fn get(client: &Client, mod_id: i32, file_id: i64) -> Result<DownloadLinkResponse> {
let res = client
.get(format!(
"https://api.nexusmods.com/v1/games/{}/mods/{}/files/{}/download_link.json",
GAME_NAME, mod_id, file_id
))
.header("accept", "application/json")
.header("apikey", env::var("NEXUS_API_KEY")?)
.header("user-agent", USER_AGENT)
.send()
.await?
.error_for_status()?;
let wait = rate_limit_wait_duration(&res)?;
let json = res.json::<Value>().await?;
Ok(DownloadLinkResponse { wait, json })
}
impl DownloadLinkResponse {
pub fn link<'a>(&'a self) -> Result<&'a str> {
let link = self
.json
.get(0)
.ok_or_else(|| anyhow!("Links array in API response is missing first element"))?
.get("URI")
.ok_or_else(|| anyhow!("Missing URI key in link in API response"))?
.as_str()
.ok_or_else(|| anyhow!("URI value in API response link is not a string"))?;
Ok(link)
}
pub async fn download_file(&self, client: &Client) -> Result<File> {
let mut tokio_file = File::from_std(tempfile()?);
let res = client
.get(self.link()?)
.header("apikey", env::var("NEXUS_API_KEY")?)
.header("user-agent", USER_AGENT)
.send()
.await?
.error_for_status()?;
// See: https://github.com/benkay86/async-applied/blob/master/reqwest-tokio-compat/src/main.rs
let mut byte_stream = res
.bytes_stream()
.map_err(|e| futures::io::Error::new(futures::io::ErrorKind::Other, e))
.into_async_read()
.compat();
tokio::io::copy(&mut byte_stream, &mut tokio_file).await?;
return Ok(tokio_file);
}
}

107
src/nexus_api/files.rs Normal file
View File

@@ -0,0 +1,107 @@
use anyhow::{anyhow, Result};
use chrono::NaiveDateTime;
use reqwest::Client;
use serde_json::Value;
use std::{env, time::Duration};
use super::{rate_limit_wait_duration, GAME_NAME, USER_AGENT};
pub struct FilesResponse {
pub wait: Option<Duration>,
json: Value,
}
pub struct ApiFile<'a> {
pub file_id: i64,
pub name: &'a str,
pub file_name: &'a str,
pub category: Option<&'a str>,
pub version: Option<&'a str>,
pub mod_version: Option<&'a str>,
pub uploaded_at: NaiveDateTime,
}
pub async fn get(client: &Client, nexus_mod_id: i32) -> Result<FilesResponse> {
let res = client
.get(format!(
"https://api.nexusmods.com/v1/games/{}/mods/{}/files.json",
GAME_NAME, nexus_mod_id
))
.header("accept", "application/json")
.header("apikey", env::var("NEXUS_API_KEY")?)
.header("user-agent", USER_AGENT)
.send()
.await?
.error_for_status()?;
let wait = rate_limit_wait_duration(&res)?;
let json = res.json::<Value>().await?;
Ok(FilesResponse { wait, json })
}
impl FilesResponse {
pub fn files<'a>(&'a self) -> Result<Vec<ApiFile<'a>>> {
let files = self
.json
.get("files")
.ok_or_else(|| anyhow!("Missing files key in API response"))?
.as_array()
.ok_or_else(|| anyhow!("files value in API response is not an array"))?;
files
.into_iter()
.map(|file| {
let file_id = file
.get("file_id")
.ok_or_else(|| anyhow!("Missing file_id key in file in API response"))?
.as_i64()
.ok_or_else(|| anyhow!("file_id value in API response file is not a number"))?;
dbg!(file_id);
let name = file
.get("name")
.ok_or_else(|| anyhow!("Missing name key in file in API response"))?
.as_str()
.ok_or_else(|| anyhow!("name value in API response file is not a string"))?;
let file_name = file
.get("file_name")
.ok_or_else(|| anyhow!("Missing file_name key in file in API response"))?
.as_str()
.ok_or_else(|| {
anyhow!("file_name value in API response file is not a string")
})?;
let category = file
.get("category_name")
.ok_or_else(|| anyhow!("Missing category key in file in API response"))?
.as_str();
let version = file
.get("version")
.ok_or_else(|| anyhow!("Missing version key in file in API response"))?
.as_str();
let mod_version = file
.get("mod_version")
.ok_or_else(|| anyhow!("Missing mod_version key in file in API response"))?
.as_str();
let uploaded_timestamp = file
.get("uploaded_timestamp")
.ok_or_else(|| {
anyhow!("Missing uploaded_timestamp key in file in API response")
})?
.as_i64()
.ok_or_else(|| {
anyhow!("uploaded_timestamp value in API response file is not a number")
})?;
let uploaded_at = NaiveDateTime::from_timestamp(uploaded_timestamp, 0);
Ok(ApiFile {
file_id,
name,
file_name,
category,
version,
mod_version,
uploaded_at,
})
})
.collect()
}
}

44
src/nexus_api/mod.rs Normal file
View File

@@ -0,0 +1,44 @@
use anyhow::Result;
use chrono::DateTime;
use chrono::Duration;
use chrono::Utc;
use reqwest::Response;
pub mod download_link;
pub mod files;
pub static GAME_NAME: &str = "skyrimspecialedition";
pub const GAME_ID: u32 = 1704;
pub static USER_AGENT: &str = "mod-mapper/0.1";
pub fn rate_limit_wait_duration(res: &Response) -> Result<Option<std::time::Duration>> {
let daily_remaining = res
.headers()
.get("x-rl-daily-remaining")
.expect("No daily remaining in response headers");
let hourly_remaining = res
.headers()
.get("x-rl-hourly-remaining")
.expect("No hourly limit in response headers");
let hourly_reset = res
.headers()
.get("x-rl-hourly-reset")
.expect("No hourly reset in response headers");
dbg!(daily_remaining);
dbg!(hourly_remaining);
if hourly_remaining == "0" {
let hourly_reset = hourly_reset.to_str()?.trim();
let hourly_reset: DateTime<Utc> =
(DateTime::parse_from_str(hourly_reset, "%Y-%m-%d %H:%M:%S %z")?
+ Duration::seconds(5))
.into();
dbg!(hourly_reset);
let duration = (hourly_reset - Utc::now()).to_std()?;
dbg!(duration);
return Ok(Some(duration));
}
Ok(None)
}