Oliver/internet_mapper
1
0
Fork 0
Code Issues 11 Pull Requests Actions Packages Projects Releases Wiki Activity

epic metrics

Browse Source
This commit is contained in:
Oliver
2025-03-19 04:59:50 +00:00
parent de80418c00
commit 9aa34b3eee
10 changed files with 493 additions and 61 deletions
Download Patch File Download Diff File Expand all files Collapse all files

350
Cargo.lock generated
View File

@@ -398,6 +398,29 @@ version = "1.4.0"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "ace50bade8e6234aa140d9a2f552bbee1db4d353f69b8217bc503490fc1a9f26"
[[package]]
name = "aws-lc-rs"
version = "1.12.6"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "dabb68eb3a7aa08b46fddfd59a3d55c978243557a90ab804769f7e20e67d2b01"
dependencies = [
"aws-lc-sys",
"zeroize",
]
[[package]]
name = "aws-lc-sys"
version = "0.27.1"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "77926887776171ced7d662120a75998e444d3750c951abfe07f90da130514b1f"
dependencies = [
"bindgen",
"cc",
"cmake",
"dunce",
"fs_extra",
]
[[package]]
name = "backtrace"
version = "0.3.74"
@@ -453,6 +476,29 @@ dependencies = [
"serde",
]
[[package]]
name = "bindgen"
version = "0.69.5"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "271383c67ccabffb7381723dea0672a673f292304fcb45c01cc648c7a8d58088"
dependencies = [
"bitflags",
"cexpr",
"clang-sys",
"itertools 0.11.0",
"lazy_static",
"lazycell",
"log",
"prettyplease",
"proc-macro2",
"quote",
"regex",
"rustc-hash 1.1.0",
"shlex",
"syn 2.0.100",
"which",
]
[[package]]
name = "bit-set"
version = "0.5.3"
@@ -614,6 +660,8 @@ version = "1.2.16"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "be714c154be609ec7f5dad223a33bf1482fff90472de28f7362806e6d4832b8c"
dependencies = [
"jobserver",
"libc",
"shlex",
]
@@ -674,6 +722,15 @@ dependencies = [
"unicode-security",
]
[[package]]
name = "cexpr"
version = "0.6.0"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "6fac387a98bb7c37292057cffc56d62ecb629900026402633ae9160df93a8766"
dependencies = [
"nom",
]
[[package]]
name = "cfg-if"
version = "1.0.0"
@@ -738,6 +795,26 @@ dependencies = [
"inout",
]
[[package]]
name = "clang-sys"
version = "1.8.1"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "0b023947811758c97c59bf9d1c188fd619ad4718dcaa767947df1cadb14f39f4"
dependencies = [
"glob",
"libc",
"libloading",
]
[[package]]
name = "cmake"
version = "0.1.54"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "e7caa3f9de89ddbe2c607f4101924c5abec803763ae9534e4f4d7d8f84aa81f0"
dependencies = [
"cc",
]
[[package]]
name = "colorchoice"
version = "1.0.3"
@@ -769,6 +846,16 @@ dependencies = [
"libc",
]
[[package]]
name = "core-foundation"
version = "0.10.0"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "b55271e5c8c478ad3f38ad24ef34923091e0548492a266d19b3c0b4d82574c63"
dependencies = [
"core-foundation-sys",
"libc",
]
[[package]]
name = "core-foundation-sys"
version = "0.8.7"
@@ -1002,6 +1089,12 @@ version = "0.3.3"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "fea41bba32d969b513997752735605054bc0dfa92b4c56bf1189f2e174be7a10"
[[package]]
name = "dunce"
version = "1.0.5"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "92773504d58c093f6de2459af4af33faa518c13451eb8f2b5698ed3d36e7c813"
[[package]]
name = "earcutr"
version = "0.4.3"
@@ -1166,6 +1259,12 @@ dependencies = [
"percent-encoding",
]
[[package]]
name = "fs_extra"
version = "1.3.0"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "42703706b716c37f96a77aea830392ad231f44c9e9a67872fa5548707e11b11c"
[[package]]
name = "fst"
version = "0.4.7"
@@ -1388,6 +1487,12 @@ version = "0.31.1"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "07e28edb80900c19c28f1072f2e8aeca7fa06b23cd4169cefe1af5aa3260783f"
[[package]]
name = "glob"
version = "0.3.2"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "a8d1add55171497b4705a648c6b583acafb01d58050a51727785f0b2c8e0a2b2"
[[package]]
name = "h2"
version = "0.4.8"
@@ -1614,6 +1719,7 @@ dependencies = [
"hyper",
"hyper-util",
"rustls",
"rustls-native-certs",
"rustls-pki-types",
"tokio",
"tokio-rustls",
@@ -1924,6 +2030,15 @@ dependencies = [
"syn 2.0.100",
]
[[package]]
name = "jobserver"
version = "0.1.32"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "48d1dbcbbeb6a7fec7e059840aa538bd62aaccf972c7346c4d9d2059312853d0"
dependencies = [
"libc",
]
[[package]]
name = "js-sys"
version = "0.3.77"
@@ -1986,6 +2101,12 @@ version = "1.5.0"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "bbd2bcb4c963f2ddae06a2efc7e9f3591312473c50c6685e1f298068316e66fe"
[[package]]
name = "lazycell"
version = "1.3.0"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "830d08ce1d1d941e6b30645f1a0eb5643013d835ce3779a5fc208261dbe10f55"
[[package]]
name = "lexicmp"
version = "0.1.0"
@@ -2001,6 +2122,16 @@ version = "0.2.171"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "c19937216e9d3aa9956d9bb8dfc0b0c8beb6058fc4f7a4dc4d850edf86a237d6"
[[package]]
name = "libloading"
version = "0.8.6"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "fc2f4eb4bc735547cfed7c0a4922cbd04a4655978c09b54f1f7b228750664c34"
dependencies = [
"cfg-if",
"windows-targets 0.52.6",
]
[[package]]
name = "libm"
version = "0.2.11"
@@ -2029,6 +2160,12 @@ dependencies = [
"thiserror 1.0.69",
]
[[package]]
name = "linux-raw-sys"
version = "0.4.15"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "d26c52dbd32dccf2d10cac7725f8eae5296885fb5703b261f7d0a0739ec807ab"
[[package]]
name = "linux-raw-sys"
version = "0.9.3"
@@ -2149,6 +2286,53 @@ version = "2.7.4"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "78ca9ab1a0babb1e7d5695e3530886289c18cf2f87ec19a575a0abdce112e3a3"
[[package]]
name = "metrics"
version = "0.24.1"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "7a7deb012b3b2767169ff203fadb4c6b0b82b947512e5eb9e0b78c2e186ad9e3"
dependencies = [
"ahash 0.8.11",
"portable-atomic",
]
[[package]]
name = "metrics-exporter-prometheus"
version = "0.16.2"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "dd7399781913e5393588a8d8c6a2867bf85fb38eaf2502fdce465aad2dc6f034"
dependencies = [
"base64 0.22.1",
"http-body-util",
"hyper",
"hyper-rustls",
"hyper-util",
"indexmap 2.8.0",
"ipnet",
"metrics",
"metrics-util",
"quanta",
"thiserror 1.0.69",
"tokio",
"tracing",
]
[[package]]
name = "metrics-util"
version = "0.19.0"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "dbd4884b1dd24f7d6628274a2f5ae22465c337c5ba065ec9b6edccddf8acc673"
dependencies = [
"crossbeam-epoch",
"crossbeam-utils",
"hashbrown 0.15.2",
"metrics",
"quanta",
"rand 0.8.5",
"rand_xoshiro",
"sketches-ddsketch",
]
[[package]]
name = "miette"
version = "5.10.0"
@@ -2188,6 +2372,12 @@ dependencies = [
"unicase",
]
[[package]]
name = "minimal-lexical"
version = "0.2.1"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "68354c5c6bd36d73ff3feceb05efa59b6acb7626617f4962be322a825e61f79a"
[[package]]
name = "minio"
version = "0.2.0-alpha"
@@ -2295,7 +2485,7 @@ dependencies = [
"openssl-probe",
"openssl-sys",
"schannel",
"security-framework",
"security-framework 2.11.1",
"security-framework-sys",
"tempfile",
]
@@ -2353,6 +2543,16 @@ dependencies = [
"num-traits",
]
[[package]]
name = "nom"
version = "7.1.3"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "d273983c5a657a70a3e8f2a01329822f3b8c8172b73826411a55751e404a0a4a"
dependencies = [
"memchr",
"minimal-lexical",
]
[[package]]
name = "ntapi"
version = "0.4.1"
@@ -2757,6 +2957,16 @@ version = "0.1.1"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "925383efa346730478fb4838dbe9137d2a47675ad789c546d150a6e1dd4ab31c"
[[package]]
name = "prettyplease"
version = "0.2.31"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "5316f57387668042f561aae71480de936257848f9c43ce528e311d89a07cadeb"
dependencies = [
"proc-macro2",
"syn 2.0.100",
]
[[package]]
name = "proc-macro-crate"
version = "3.3.0"
@@ -2810,6 +3020,21 @@ dependencies = [
"syn 1.0.109",
]
[[package]]
name = "quanta"
version = "0.12.5"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "3bd1fe6824cea6538803de3ff1bc0cf3949024db3d43c9643024bfb33a807c0e"
dependencies = [
"crossbeam-utils",
"libc",
"once_cell",
"raw-cpuid",
"wasi 0.11.0+wasi-snapshot-preview1",
"web-sys",
"winapi",
]
[[package]]
name = "quick_cache"
version = "0.5.2"
@@ -2833,7 +3058,7 @@ dependencies = [
"pin-project-lite",
"quinn-proto",
"quinn-udp",
"rustc-hash",
"rustc-hash 2.1.1",
"rustls",
"socket2",
"thiserror 2.0.12",
@@ -2852,7 +3077,7 @@ dependencies = [
"getrandom 0.3.2",
"rand 0.9.0",
"ring",
"rustc-hash",
"rustc-hash 2.1.1",
"rustls",
"rustls-pki-types",
"slab",
@@ -2968,6 +3193,24 @@ dependencies = [
"getrandom 0.3.2",
]
[[package]]
name = "rand_xoshiro"
version = "0.6.0"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "6f97cdb2a36ed4183de61b2f824cc45c9f1037f28afe0a322e9fff4c108b5aaa"
dependencies = [
"rand_core 0.6.4",
]
[[package]]
name = "raw-cpuid"
version = "11.5.0"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "c6df7ab838ed27997ba19a4664507e6f82b41fe6e20be42929332156e5e85146"
dependencies = [
"bitflags",
]
[[package]]
name = "rawpointer"
version = "0.2.1"
@@ -3296,6 +3539,12 @@ version = "0.1.24"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "719b953e2095829ee67db738b3bfa9fa368c94900df327b3f07fe6e794d2fe1f"
[[package]]
name = "rustc-hash"
version = "1.1.0"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "08d43f7aa6b08d49f382cde6a7982047c3426db949b1424bc4b7ec9ae12c6ce2"
[[package]]
name = "rustc-hash"
version = "2.1.1"
@@ -3320,6 +3569,19 @@ dependencies = [
"semver",
]
[[package]]
name = "rustix"
version = "0.38.44"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "fdb5bc1ae2baa591800df16c9ca78619bf65c0488b41b96ccec5d11220d8c154"
dependencies = [
"bitflags",
"errno",
"libc",
"linux-raw-sys 0.4.15",
"windows-sys 0.59.0",
]
[[package]]
name = "rustix"
version = "1.0.3"
@@ -3329,7 +3591,7 @@ dependencies = [
"bitflags",
"errno",
"libc",
"linux-raw-sys",
"linux-raw-sys 0.9.3",
"windows-sys 0.59.0",
]
@@ -3339,6 +3601,7 @@ version = "0.23.25"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "822ee9188ac4ec04a2f0531e55d035fb2de73f18b41a63c70c2712503b6fb13c"
dependencies = [
"aws-lc-rs",
"log",
"once_cell",
"ring",
@@ -3348,6 +3611,18 @@ dependencies = [
"zeroize",
]
[[package]]
name = "rustls-native-certs"
version = "0.8.1"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "7fcff2dd52b58a8d98a70243663a0d234c4e2b79235637849d15913394a247d3"
dependencies = [
"openssl-probe",
"rustls-pki-types",
"schannel",
"security-framework 3.2.0",
]
[[package]]
name = "rustls-pemfile"
version = "2.2.0"
@@ -3372,6 +3647,7 @@ version = "0.103.0"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "0aa4eeac2588ffff23e9d7a7e9b3f971c5fb5b7ebc9452745e0c232c64f83b2f"
dependencies = [
"aws-lc-rs",
"ring",
"rustls-pki-types",
"untrusted",
@@ -3447,7 +3723,20 @@ source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "897b2245f0b511c87893af39b033e5ca9cce68824c4d7e7630b5a1d339658d02"
dependencies = [
"bitflags",
"core-foundation",
"core-foundation 0.9.4",
"core-foundation-sys",
"libc",
"security-framework-sys",
]
[[package]]
name = "security-framework"
version = "3.2.0"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "271720403f46ca04f7ba6f55d438f8bd878d6b8ca0a1046e8228c4145bcbb316"
dependencies = [
"bitflags",
"core-foundation 0.10.0",
"core-foundation-sys",
"libc",
"security-framework-sys",
@@ -3520,6 +3809,15 @@ dependencies = [
"serde",
]
[[package]]
name = "serde_spanned"
version = "0.6.8"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "87607cb1398ed59d48732e575a4c28a7a8ebf2454b964fe3f224f2afc07909e1"
dependencies = [
"serde",
]
[[package]]
name = "serde_urlencoded"
version = "0.7.1"
@@ -3632,6 +3930,12 @@ version = "1.0.1"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "56199f7ddabf13fe5074ce809e7d3f42b42ae711800501b5b16ea82ad029c39d"
[[package]]
name = "sketches-ddsketch"
version = "0.3.0"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "c1e9a774a6c28142ac54bb25d25562e6bcf957493a184f15ad4eebccb23e410a"
[[package]]
name = "slab"
version = "0.4.9"
@@ -3814,11 +4118,14 @@ version = "0.1.0"
dependencies = [
"base64 0.22.1",
"html5ever 0.29.1",
"metrics",
"metrics-exporter-prometheus",
"minio",
"reqwest",
"serde",
"surrealdb",
"tokio",
"toml",
"tracing",
"tracing-subscriber",
"url",
@@ -4015,7 +4322,7 @@ source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "3c879d448e9d986b661742763247d3693ed13609438cf3d006f51f5368a5ba6b"
dependencies = [
"bitflags",
"core-foundation",
"core-foundation 0.9.4",
"system-configuration-sys",
]
@@ -4044,7 +4351,7 @@ dependencies = [
"fastrand",
"getrandom 0.3.2",
"once_cell",
"rustix",
"rustix 1.0.3",
"windows-sys 0.59.0",
]
@@ -4277,11 +4584,26 @@ dependencies = [
"tokio",
]
[[package]]
name = "toml"
version = "0.8.20"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "cd87a5cdd6ffab733b2f74bc4fd7ee5fff6634124999ac278c35fc78c6120148"
dependencies = [
"serde",
"serde_spanned",
"toml_datetime",
"toml_edit",
]
[[package]]
name = "toml_datetime"
version = "0.6.8"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "0dd7358ecb8fc2f8d014bf86f6f638ce72ba252a2c3a2572f2a795f1d23efb41"
dependencies = [
"serde",
]
[[package]]
name = "toml_edit"
@@ -4290,6 +4612,8 @@ source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "17b4795ff5edd201c7cd6dca065ae59972ce77d1b80fa0a84d94950ece7d1474"
dependencies = [
"indexmap 2.8.0",
"serde",
"serde_spanned",
"toml_datetime",
"winnow",
]
@@ -4750,6 +5074,18 @@ dependencies = [
"rustls-pki-types",
]
[[package]]
name = "which"
version = "4.4.2"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "87ba24419a2078cd2b0f2ede2691b6c66d8e47836da3b6db8265ebad47afbfc7"
dependencies = [
"either",
"home",
"once_cell",
"rustix 0.38.44",
]
[[package]]
name = "winapi"
version = "0.3.9"

3
Cargo.toml
View File

@@ -6,12 +6,15 @@ edition = "2021"
[dependencies]
base64 = "0.22.1"
html5ever = "0.29"
metrics = "0.24.1"
metrics-exporter-prometheus = { version = "0.16.2", features=["http-listener"]}
# minio = "0.1.0"
minio = {git="https://github.com/minio/minio-rs.git", rev = "c28f576"}
reqwest = { version = "0.12", features = ["gzip", "default", "rustls-tls"] }
serde = { version = "1.0", features = ["derive"] }
surrealdb = "2.2"
tokio = { version="1.41.0", features = ["full"] }
toml = "0.8.20"
tracing = "0.1"
tracing-subscriber = { version = "0.3", features = ["env-filter", "local-time", "json"] }
url = { version = "2.5", features = ["serde"] }

16
Crawler.toml Normal file
View File

@@ -0,0 +1,16 @@
# Surreal config
surreal_url = "localhost:8000"
surreal_username = "root"
surreal_password = "root"
surreal_ns = "test"
surreal_db = "v1.12"
# Minio config
s3_bucket = "v1.12"
s3_url = "http://localhost:9000"
s3_access_key = "jLDPKGuu513VENc8kJwX"
s3_secret_key = "4T1nymEzsGYOlKSAb1WX7V3stnQn9a5ZoTQjDfcL"
# Crawler config
crawl_filter = "en.wikipedia.com"
budget = 200

12
docker/compose.yml
View File

@@ -50,6 +50,17 @@ services:
volumes:
- ./loki.yaml:/etc/loki/local-config.yaml
# Metrics collector
prometheus:
image: prom/prometheus:latest
expose:
- 9090
volumes:
- ./prometheus.yaml:/etc/prometheus/prometheus.yml
# persist data
- prometheus_storage:/prometheus
command: --web.enable-lifecycle --config.file=/etc/prometheus/prometheus.yml
# Everything viewer
grafana:
image: grafana/grafana:latest
@@ -65,6 +76,7 @@ services:
- 3000:3000
volumes:
prometheus_storage:
grafana_storage:
alloy_storage:
surrealdb_storage:

12
docker/grafana.yaml
View File

@@ -10,3 +10,15 @@ datasources:
isDefault: true
version: 1
editable: false
- name: Prometheus
type: prometheus
uid: prometheus
access: proxy
orgId: 1
url: http://prometheus:9090
basicAuth: false
isDefault: false
version: 1
editable: false
jsonData:
httpMethod: GET

16
docker/prometheus.yaml Normal file
View File

@@ -0,0 +1,16 @@
global:
scrape_interval: 5s
query_log_file: /etc/prometheus/query.log
scrape_configs:
- job_name: crawler
static_configs:
# change this your machine's ip, localhost won't work
# because localhost refers to the docker container.
- targets: ['192.168.8.209:2500']
- job_name: loki
static_configs:
- targets: ['loki:3100']
- job_name: prometheus
static_configs:
- targets: ['localhost:9090']

26
src/db.rs
View File

@@ -1,4 +1,5 @@
use std::fmt::Debug;
use metrics::counter;
use serde::{Deserialize, Serialize};
use surrealdb::{
engine::remote::ws::{Client, Ws}, error::Db, opt::auth::Root, sql::Thing, Response, Surreal
@@ -8,6 +9,10 @@ use url::Url;
use crate::{Config, Timer};
const ROUND_TRIP_METRIC: &'static str = "surql_trips";
const STORE: &'static str = "surql_store_calls";
const LINK: &'static str = "surql_link_calls";
#[derive(Serialize, Deserialize, Clone)]
pub struct Website {
/// The url that this data is found at
@@ -47,6 +52,7 @@ impl Website {
#[instrument(skip_all)]
pub async fn links_to(&self, other: Vec<Thing>, db: &Surreal<Client>) {
let len = other.len();
if len == 0 {return}
@@ -57,6 +63,8 @@ impl Website {
let timer = Timer::start(&msg);
// prevent the timer from being dropped instantly.
let _ = timer;
counter!(ROUND_TRIP_METRIC).increment(1);
counter!(LINK).increment(1);
match db
.query("COUNT(RELATE (SELECT id FROM website WHERE site = $in) -> links_to -> $out)")
.bind(("in", from))
@@ -89,8 +97,12 @@ impl Website {
#[instrument(name = "surql_store", skip_all)]
pub async fn store(&self, db: &Surreal<Client>) -> Option<Thing> {
counter!(STORE).increment(1);
let counter = counter!(ROUND_TRIP_METRIC);
let t = Timer::start("Stored link");
let _ = t;
counter.increment(1);
// check if it's been gone thru before
let mut response = db
.query("SELECT * FROM ONLY website WHERE site = $site LIMIT 1")
@@ -105,6 +117,7 @@ impl Website {
let mut new = self.clone();
new.crawled = old.crawled | new.crawled;
counter.increment(1);
// update the record
match db.upsert((id.tb, id.id.to_string())).content(new).await {
Ok(e) => {
@@ -130,6 +143,7 @@ impl Website {
};
}
} else {
counter.increment(1);
// sites hasn't existed yet
match db.create("website").content(self.clone()).await {
Ok(e) => {
@@ -165,23 +179,23 @@ pub struct Record {
}
#[instrument(skip_all, name = "SurrealDB")]
pub async fn connect(config: &Config<'_>) -> surrealdb::Result<Surreal<Client>> {
pub async fn connect(config: &Config) -> surrealdb::Result<Surreal<Client>> {
trace!("Establishing connection to surreal...");
// Connect to the server
let db = Surreal::new::<Ws>(config.surreal_url).await?;
let db = Surreal::new::<Ws>(&config.surreal_url).await?;
trace!("Logging in...");
// Signin as a namespace, database, or root user
db.signin(Root {
username: config.surreal_username,
password: config.surreal_password,
username: &config.surreal_username,
password: &config.surreal_password,
})
.await?;
// Select a specific namespace / database
db
.use_ns(config.surreal_ns)
.use_db(config.surreal_db)
.use_ns(&config.surreal_ns)
.use_db(&config.surreal_db)
.await?;
let setup = include_bytes!("setup.surql");

106
src/main.rs
View File

@@ -1,32 +1,43 @@
#![feature(type_alias_impl_trait)]
#![feature(const_async_blocks)]
#![feature(ip_from)]
extern crate html5ever;
use std::time::Instant;
use std::{fs::File, io::Read, net::{IpAddr, Ipv4Addr}, time::Instant};
use db::{connect, Website};
use metrics::{counter, gauge};
use metrics_exporter_prometheus::PrometheusBuilder;
use s3::S3;
use serde::Deserialize;
use surrealdb::{engine::remote::ws::Client, Surreal};
use tokio::{task::JoinSet};
use tokio::task::JoinSet;
use tracing::{debug, info, instrument, trace, trace_span, warn};
use tracing_subscriber::{fmt::{self, time::LocalTime}, layer::{Filter, SubscriberExt}, EnvFilter, Layer, Registry};
use tracing_subscriber::{fmt, layer::SubscriberExt, EnvFilter, Layer, Registry};
mod db;
mod parser;
mod s3;
struct Config<'a> {
surreal_ns: &'a str,
surreal_db: &'a str,
surreal_url: &'a str,
surreal_username: &'a str,
surreal_password: &'a str,
const GET_METRIC: &'static str = "total_gets";
const GET_IN_FLIGHT: &'static str = "gets_in_flight";
const SITES_CRAWLED: &'static str = "pages_crawled";
const BEING_PROCESSED: &'static str = "pages_being_processed";
s3_url: &'a str,
s3_bucket: &'a str,
s3_access_key: &'a str,
s3_secret_key: &'a str,
#[derive(Deserialize)]
struct Config {
surreal_ns: String,
surreal_db: String,
surreal_url: String,
surreal_username: String,
surreal_password: String,
s3_url: String,
s3_bucket: String,
s3_access_key: String,
s3_secret_key: String,
crawl_filter: String,
budget: usize,
}
#[tokio::main]
@@ -48,38 +59,33 @@ async fn main() {
// .with_timer(LocalTime::rfc_3339()) // Loki or alloy does this automatically
.json()
.with_writer(writer)
// .with_filter(EnvFilter::from_default_env())
);
tracing::subscriber::set_global_default(registry).expect("Failed to set default subscriber");
// tracing_subscriber::fmt()
// .with_env_filter(EnvFilter::from_default_env())
// .with_line_number(true)
// .with_thread_ids(true)
// .with_file(true)
// .with_timer(LocalTime::rfc_3339())
// .init();
let builder = PrometheusBuilder::new();
builder.with_http_listener(
std::net::SocketAddr::new(IpAddr::V4(Ipv4Addr::from_octets([0,0,0,0])), 2500)
)
.install()
.expect("failed to install recorder/exporter");
debug!("Starting...");
// Would probably take these in as parameters from a cli
// Would probably take these in as parameters from a cli
let starting_url = "https://en.wikipedia.org/";
// When getting uncrawled pages, name must contain this variable. "" will effectively get ignored.
let crawl_filter = "en.wikipedia.org/";
let budget = 50;
// let crawl_filter = "en.wikipedia.org/";
// let budget = 50;
let mut crawled = 0;
let config = Config {
surreal_url: "localhost:8000",
surreal_username: "root",
surreal_password: "root",
surreal_ns: "test",
surreal_db: "v1.12",
s3_bucket: "v1.12",
s3_url: "http://localhost:9000",
s3_access_key: "p8gXIZEO2FnWqWBiJYwo",
s3_secret_key: "1mRO0EYA2YAQ0xsKrlbkIIz4AT8KNXy6QIQPtxUu",
};
let mut file = File::open("./Crawler.toml").expect("Failed to read Crawler.toml");
let mut buf = String::new();
let _ = file.read_to_string(&mut buf);
let config: Config = toml::from_str(&buf).expect("Failed to parse Crawler.toml");
let db = connect(&config)
.await
@@ -106,14 +112,14 @@ async fn main() {
let span = trace_span!("Loop");
let span = span.enter();
while crawled < budget {
let get_num = if budget - crawled < 100 {
budget - crawled
while crawled < config.budget {
let get_num = if config.budget - crawled < 100 {
config.budget - crawled
} else {
100
};
let uncrawled = get_uncrawled_links(&db, get_num, crawl_filter.to_string()).await;
let uncrawled = get_uncrawled_links(&db, get_num, config.crawl_filter.clone()).await;
if uncrawled.len() == 0 {
info!("Had more budget but finished crawling everything.");
return;
@@ -126,16 +132,20 @@ async fn main() {
{
let mut futures = JoinSet::new();
for site in uncrawled {
gauge!(BEING_PROCESSED).increment(1);
futures.spawn(get(site, db.clone(), reqwest.clone(), s3.clone()));
// technically the site hasn't be crawled *yet*, but the future
// where it is crawled has been set up.
crawled += 1;
// let percent = format!("{:.2}%", (crawled as f32 / budget as f32) * 100f32);
// info!("Crawled {crawled} out of {budget} pages. ({percent})");
}
debug!("Joining {} futures...", futures.len());
// join all the gets together
let _ = futures.join_all().await;
let c = counter!(SITES_CRAWLED);
// As futures complete runs code in while block
while let Some(_) = futures.join_next().await {
c.increment(1);
gauge!(BEING_PROCESSED).decrement(1);
crawled += 1;
}
}
}
drop(span);
@@ -154,9 +164,15 @@ async fn get(mut site: Website, db: Surreal<Client>, reqwest: reqwest::Client, s
let request_builder = reqwest.get(site.to_string());
timer.stop();
let g = gauge!(GET_IN_FLIGHT);
g.increment(1);
let timer = Timer::start("Got page");
if let Ok(response) = request_builder.send().await {
timer.stop();
g.decrement(1);
counter!(GET_METRIC).increment(1);
debug!("Getting body...");
// Get body

1
src/parser.rs
View File

@@ -97,6 +97,7 @@ pub async fn parse(db: &Surreal<Client>, site: &mut Website, data: &str) {
let mut links_to = Vec::with_capacity(other_sites.len());
for a in other_sites {
let other = a.store(db).await;
if let Some(o) = other {
links_to.push(o);

8
src/s3.rs
View File

@@ -1,4 +1,5 @@
use base64::{alphabet, engine::{self, general_purpose}, Engine};
use metrics::counter;
use minio::s3::{
args::{BucketExistsArgs, MakeBucketArgs},
client::ClientBuilder,
@@ -14,6 +15,8 @@ use crate::{Config, Timer};
const CUSTOM_ENGINE: engine::GeneralPurpose = engine::GeneralPurpose::new(&alphabet::URL_SAFE, general_purpose::NO_PAD);
const ROUND_TRIP_METRIC: &'static str = "s3_trips";
#[derive(Clone)]
pub struct S3 {
bucket_name: String,
@@ -22,7 +25,7 @@ pub struct S3 {
impl S3 {
#[instrument(skip_all, name = "S3")]
pub async fn connect(config: &Config<'_>) -> Result<Self, Error> {
pub async fn connect(config: &Config) -> Result<Self, Error> {
let base_url = config
.s3_url
.parse::<BaseUrl>()
@@ -62,6 +65,7 @@ impl S3 {
#[instrument(name = "s3_store", skip_all)]
pub async fn store(&self, data: &str, url: &Url) {
let counter = counter!(ROUND_TRIP_METRIC);
let t = Timer::start("Stored page");
let _ = t; // prevent compiler drop
if let Some(domain) = url.domain() {
@@ -69,6 +73,7 @@ impl S3 {
trace!("Created filename: {filename} from raw: {}", url.to_string());
counter.increment(1);
let _ = match &self
.client
.put_object_content(&self.bucket_name, &filename, data.to_owned())
@@ -83,6 +88,7 @@ impl S3 {
let filename: String = domain.to_owned() + &CUSTOM_ENGINE.encode(url.path());
counter.increment(1);
let _ = &self
.client
.put_object_content(&self.bucket_name, &filename, data.to_owned())