Mostly working download loop done

Still need to fix a panic that happens on some .rar archive files.
This commit is contained in:
Tyler Hallada 2021-06-13 22:30:40 -04:00
parent b132a94c64
commit d6b8f4e74a
3 changed files with 350 additions and 282 deletions

4
Cargo.lock generated
View File

@ -1626,9 +1626,9 @@ checksum = "cbce6d4507c7e4a3962091436e56e95290cb71fa302d0d270e32130b75fbff27"
[[package]] [[package]]
name = "skyrim-cell-dump" name = "skyrim-cell-dump"
version = "0.1.2" version = "0.1.3"
source = "registry+https://github.com/rust-lang/crates.io-index" source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "372b96816596c25ba82afdc4819aae92e3750c9f4d965aa99d46f25fe53bbc3f" checksum = "b8ff27163eeca52326be9a89a4adc15dd7ed3d7c0c44dd981aa2bbacff10aede"
dependencies = [ dependencies = [
"anyhow", "anyhow",
"bitflags", "bitflags",

View File

@ -22,7 +22,7 @@ seahash = "4.1"
serde = { version = "1.0", features = ["derive"] } serde = { version = "1.0", features = ["derive"] }
serde_json = "1.0" serde_json = "1.0"
sqlx = { version = "0.5", features = ["runtime-tokio-native-tls", "postgres", "migrate", "chrono"] } sqlx = { version = "0.5", features = ["runtime-tokio-native-tls", "postgres", "migrate", "chrono"] }
skyrim-cell-dump = "0.1.2" skyrim-cell-dump = "0.1.3"
tempfile = "3.2" tempfile = "3.2"
tokio = { version = "1.5.0", features = ["full"] } tokio = { version = "1.5.0", features = ["full"] }
tokio-util = { version = "0.6", features = ["compat"] } tokio-util = { version = "0.6", features = ["compat"] }

View File

@ -1,9 +1,13 @@
use anyhow::{anyhow, Context, Result}; use anyhow::{anyhow, Context, Result};
use chrono::DateTime;
use chrono::Duration;
use chrono::NaiveDateTime; use chrono::NaiveDateTime;
use chrono::Utc;
use compress_tools::{list_archive_files, uncompress_archive_file}; use compress_tools::{list_archive_files, uncompress_archive_file};
use dotenv::dotenv; use dotenv::dotenv;
use futures::future::try_join_all; use futures::future::try_join_all;
use futures::stream::TryStreamExt; use futures::stream::TryStreamExt;
use reqwest::Response;
use scraper::{Html, Selector}; use scraper::{Html, Selector};
use serde::{Deserialize, Serialize}; use serde::{Deserialize, Serialize};
use serde_json::Value; use serde_json::Value;
@ -16,6 +20,7 @@ use std::io::Seek;
use std::io::SeekFrom; use std::io::SeekFrom;
use tempfile::tempfile; use tempfile::tempfile;
use tokio::io::{AsyncReadExt, AsyncSeekExt}; use tokio::io::{AsyncReadExt, AsyncSeekExt};
use tokio::time::sleep;
use tokio_util::compat::FuturesAsyncReadCompatExt; use tokio_util::compat::FuturesAsyncReadCompatExt;
use zip::write::{FileOptions, ZipWriter}; use zip::write::{FileOptions, ZipWriter};
@ -223,7 +228,9 @@ async fn insert_cell(
"INSERT INTO cells "INSERT INTO cells
(form_id, x, y, is_persistent, created_at, updated_at) (form_id, x, y, is_persistent, created_at, updated_at)
VALUES ($1, $2, $3, $4, now(), now()) VALUES ($1, $2, $3, $4, now(), now())
ON CONFLICT DO NOTHING ON CONFLICT (form_id) DO UPDATE
SET (x, y, is_persistent, updated_at) =
(EXCLUDED.x, EXCLUDED.y, EXCLUDED.is_persistent, now())
RETURNING *", RETURNING *",
form_id, form_id,
x, x,
@ -258,6 +265,38 @@ async fn insert_plugin_cell(
.context("Failed to insert cell") .context("Failed to insert cell")
} }
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 limit 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)
}
#[tokio::main] #[tokio::main]
pub async fn main() -> Result<()> { pub async fn main() -> Result<()> {
dotenv().ok(); dotenv().ok();
@ -268,23 +307,46 @@ pub async fn main() -> Result<()> {
let game = insert_game(&pool, GAME_NAME, GAME_ID as i32).await?; let game = insert_game(&pool, GAME_NAME, GAME_ID as i32).await?;
let client = reqwest::Client::new(); let client = reqwest::Client::new();
let mut page: i32 = 1;
let mut last_page: i32 = 1;
while page <= last_page {
let res = client let res = client
.get(format!( .get(format!(
"https://www.nexusmods.com/Core/Libs/Common/Widgets/ModList?RH_ModList=nav:true,home:false,type:0,user_id:0,game_id:{},advfilt:true,include_adult:true,page_size:80,show_game_filter:false,open:false,page:1,sort_by:OLD_u_downloads", "https://www.nexusmods.com/Core/Libs/Common/Widgets/ModList?RH_ModList=nav:true,home:false,type:0,user_id:0,game_id:{},advfilt:true,include_adult:true,page_size:80,show_game_filter:false,open:false,page:{},sort_by:OLD_u_downloads",
GAME_ID GAME_ID,
page
)) ))
.send() .send()
.await? .await?
.error_for_status()?; .error_for_status()?;
let html = res.text().await?; let html = res.text().await?;
let document = Html::parse_document(&html); let document = Html::parse_document(&html);
let mod_select = Selector::parse("div.mod-tile").expect("failed to parse CSS selector"); let mod_select = Selector::parse("li.mod-tile").expect("failed to parse CSS selector");
let left_select = Selector::parse("div.mod-tile-left").expect("failed to parse CSS selector"); let left_select =
let right_select = Selector::parse("div.mod-tile-right").expect("failed to parse CSS selector"); Selector::parse("div.mod-tile-left").expect("failed to parse CSS selector");
let right_select =
Selector::parse("div.mod-tile-right").expect("failed to parse CSS selector");
let name_select = Selector::parse("p.tile-name a").expect("failed to parse CSS selector"); let name_select = Selector::parse("p.tile-name a").expect("failed to parse CSS selector");
let category_select = Selector::parse("div.category a").expect("failed to parse CSS selector"); let category_select =
Selector::parse("div.category a").expect("failed to parse CSS selector");
let author_select = Selector::parse("div.author a").expect("failed to parse CSS selector"); let author_select = Selector::parse("div.author a").expect("failed to parse CSS selector");
let desc_select = Selector::parse("p.desc").expect("failed to parse CSS selector"); let desc_select = Selector::parse("p.desc").expect("failed to parse CSS selector");
let last_page_select =
Selector::parse("div.pagination li.extra a").expect("failed to parse CSS selector");
let last_page_elem = document
.select(&last_page_select)
.next()
.expect("Missing last page link");
last_page = last_page_elem
.text()
.next()
.expect("Missing last page text")
.trim()
.parse::<i32>()
.ok()
.expect("Failed to parse last page");
let mods = try_join_all(document.select(&mod_select).map(|element| { let mods = try_join_all(document.select(&mod_select).map(|element| {
let left = element let left = element
@ -328,18 +390,16 @@ pub async fn main() -> Result<()> {
.next() .next()
.expect("Missing desc elem for mod"); .expect("Missing desc elem for mod");
let desc = desc_elem.text().next(); let desc = desc_elem.text().next();
dbg!(name, nexus_mod_id, author, category, desc, game.id);
insert_mod(&pool, name, nexus_mod_id, author, category, desc, game.id) insert_mod(&pool, name, nexus_mod_id, author, category, desc, game.id)
})) }))
.await?; .await?;
dbg!(&mods);
for mod_obj in mods { for mod_obj in mods {
dbg!(mod_obj.id); dbg!(&mod_obj);
let res = client let res = client
.get(format!( .get(format!(
"https://api.nexusmods.com/v1/games/{}/mods/{}/files.json", "https://api.nexusmods.com/v1/games/{}/mods/{}/files.json",
GAME_NAME, mod_obj.id GAME_NAME, mod_obj.nexus_mod_id
)) ))
.header("accept", "application/json") .header("accept", "application/json")
.header("apikey", env::var("NEXUS_API_KEY")?) .header("apikey", env::var("NEXUS_API_KEY")?)
@ -347,6 +407,11 @@ pub async fn main() -> Result<()> {
.send() .send()
.await? .await?
.error_for_status()?; .error_for_status()?;
if let Some(duration) = rate_limit_wait_duration(&res)? {
sleep(duration).await;
}
let files = res.json::<Value>().await?; let files = res.json::<Value>().await?;
let files = files let files = files
.get("files") .get("files")
@ -378,9 +443,11 @@ pub async fn main() -> Result<()> {
.get("file_name") .get("file_name")
.ok_or_else(|| anyhow!("Missing file_name key in file in API response"))? .ok_or_else(|| anyhow!("Missing file_name key in file in API response"))?
.as_str() .as_str()
.ok_or_else(|| anyhow!("file_name value in API response file is not a string"))?; .ok_or_else(|| {
anyhow!("file_name value in API response file is not a string")
})?;
let category = file let category = file
.get("category") .get("category_name")
.ok_or_else(|| anyhow!("Missing category key in file in API response"))? .ok_or_else(|| anyhow!("Missing category key in file in API response"))?
.as_str(); .as_str();
let version = file let version = file
@ -393,13 +460,15 @@ pub async fn main() -> Result<()> {
.as_str(); .as_str();
let uploaded_timestamp = file let uploaded_timestamp = file
.get("uploaded_timestamp") .get("uploaded_timestamp")
.ok_or_else(|| anyhow!("Missing uploaded_timestamp key in file in API response"))? .ok_or_else(|| {
anyhow!("Missing uploaded_timestamp key in file in API response")
})?
.as_i64() .as_i64()
.ok_or_else(|| { .ok_or_else(|| {
anyhow!("uploaded_timestamp value in API response file is not a number") anyhow!("uploaded_timestamp value in API response file is not a number")
})?; })?;
let uploaded_at = NaiveDateTime::from_timestamp(uploaded_timestamp, 0); let uploaded_at = NaiveDateTime::from_timestamp(uploaded_timestamp, 0);
insert_file( let db_file = insert_file(
&pool, &pool,
name, name,
file_name, file_name,
@ -414,7 +483,7 @@ pub async fn main() -> Result<()> {
let res = client let res = client
.get(format!( .get(format!(
"https://api.nexusmods.com/v1/games/{}/mods/{}/files/{}/download_link.json", "https://api.nexusmods.com/v1/games/{}/mods/{}/files/{}/download_link.json",
GAME_NAME, mod_obj.id, file_id GAME_NAME, mod_obj.nexus_mod_id, file_id
)) ))
.header("accept", "application/json") .header("accept", "application/json")
.header("apikey", env::var("NEXUS_API_KEY")?) .header("apikey", env::var("NEXUS_API_KEY")?)
@ -422,6 +491,7 @@ pub async fn main() -> Result<()> {
.send() .send()
.await? .await?
.error_for_status()?; .error_for_status()?;
let links = res.json::<Value>().await?; let links = res.json::<Value>().await?;
let link = links let link = links
.get(0) .get(0)
@ -440,6 +510,8 @@ pub async fn main() -> Result<()> {
.await? .await?
.error_for_status()?; .error_for_status()?;
let duration = rate_limit_wait_duration(&res)?;
// See: https://github.com/benkay86/async-applied/blob/master/reqwest-tokio-compat/src/main.rs // See: https://github.com/benkay86/async-applied/blob/master/reqwest-tokio-compat/src/main.rs
let mut byte_stream = res let mut byte_stream = res
.bytes_stream() .bytes_stream()
@ -459,7 +531,7 @@ pub async fn main() -> Result<()> {
.open("plugins.zip")?, .open("plugins.zip")?,
); );
plugin_archive.add_directory( plugin_archive.add_directory(
format!("{}/{}/{}", GAME_NAME, mod_obj.id, file_id), format!("{}/{}/{}", GAME_NAME, mod_obj.nexus_mod_id, file_id),
FileOptions::default(), FileOptions::default(),
)?; )?;
plugin_archive.finish()?; plugin_archive.finish()?;
@ -473,7 +545,6 @@ pub async fn main() -> Result<()> {
let mut initial_bytes = [0; 8]; let mut initial_bytes = [0; 8];
tokio_file.seek(SeekFrom::Start(0)).await?; tokio_file.seek(SeekFrom::Start(0)).await?;
tokio_file.read_exact(&mut initial_bytes).await?; tokio_file.read_exact(&mut initial_bytes).await?;
dbg!(&initial_bytes);
let kind = infer::get(&initial_bytes).expect("unknown file type of file download"); let kind = infer::get(&initial_bytes).expect("unknown file type of file download");
match kind.mime_type() { match kind.mime_type() {
// "application/zip" => { // "application/zip" => {
@ -505,7 +576,6 @@ pub async fn main() -> Result<()> {
let mut file = tokio_file.into_std().await; let mut file = tokio_file.into_std().await;
let mut plugin_file_paths = Vec::new(); let mut plugin_file_paths = Vec::new();
for file_name in list_archive_files(&file)? { for file_name in list_archive_files(&file)? {
dbg!(&file_name);
if file_name.ends_with(".esp") if file_name.ends_with(".esp")
|| file_name.ends_with(".esm") || file_name.ends_with(".esm")
|| file_name.ends_with(".esl") || file_name.ends_with(".esl")
@ -513,21 +583,18 @@ pub async fn main() -> Result<()> {
plugin_file_paths.push(file_name); plugin_file_paths.push(file_name);
} }
} }
file.seek(SeekFrom::Start(0))?;
for file_name in plugin_file_paths.iter() { for file_name in plugin_file_paths.iter() {
file.seek(SeekFrom::Start(0))?;
dbg!(file_name); dbg!(file_name);
let mut buf = Vec::default(); let mut buf = Vec::default();
uncompress_archive_file(&mut file, &mut buf, file_name)?; uncompress_archive_file(&mut file, &mut buf, file_name)?;
let plugin = parse_plugin(&buf)?; let plugin = parse_plugin(&buf)?;
dbg!(&plugin);
let hash = seahash::hash(&buf); let hash = seahash::hash(&buf);
dbg!(&hash);
let plugin_row = insert_plugin( let plugin_row = insert_plugin(
&pool, &pool,
name, name,
// TODO: how to make i64 hash? hash as i64,
hash.try_into()?, db_file.id,
file_id as i32,
Some(plugin.header.version as f64), Some(plugin.header.version as f64),
plugin.header.author, plugin.header.author,
plugin.header.description, plugin.header.description,
@ -550,11 +617,19 @@ pub async fn main() -> Result<()> {
cell.is_persistent, cell.is_persistent,
) )
.await?; .await?;
insert_plugin_cell(&pool, plugin_row.id, cell_row.id, cell.editor_id) insert_plugin_cell(
&pool,
plugin_row.id,
cell_row.id,
cell.editor_id,
)
.await?; .await?;
} }
plugin_archive.start_file( plugin_archive.start_file(
format!("{}/{}/{}/{}", GAME_NAME, mod_obj.id, file_id, file_name), format!(
"{}/{}/{}/{}",
GAME_NAME, mod_obj.nexus_mod_id, file_id, file_name
),
FileOptions::default(), FileOptions::default(),
)?; )?;
std::io::copy(&mut buf.as_slice(), &mut plugin_archive)?; std::io::copy(&mut buf.as_slice(), &mut plugin_archive)?;
@ -563,21 +638,14 @@ pub async fn main() -> Result<()> {
}; };
plugin_archive.finish()?; plugin_archive.finish()?;
break; // temporarily just grabbing first file if let Some(duration) = duration {
sleep(duration).await;
} }
break; // temporarily just grabbing first mod
} }
// let mod_id = 4119; // hardcoded temporarily }
// let res = client
// .get("https://cf-files.nexusmods.com/cdn/1704/351/Kynesgrove-351-2-0-8-1602105523.7z?md5=hUgu4epNAuzlp8yTUMNPgQ&expires=1621585205&user_id=512579&rip=24.218.205.137") page += 1;
// .header("apikey", env::var("NEXUS_API_KEY")?) }
// .header("user-agent", USER_AGENT)
// .send()
// .await?;
// dbg!(&res);
// let bytes = res.bytes().await?;
// let mut bytes = read("C:\\Users\\tyler\\Downloads\\Crime Overhaul Expanded 1.1-19188-1-1.rar")?;
// let mut bytes = read("C:\\Users\\tyler\\Downloads\\YourMarketStall-15814-1-4-2.zip")?;
// let mut reader = std::io::Cursor::new(&bytes);
Ok(()) Ok(())
} }