From f7f18bb588023e869c347ddfd23bf1ed5d16e527 Mon Sep 17 00:00:00 2001 From: Bryan Newbold Date: Tue, 11 Sep 2018 22:09:58 -0700 Subject: improvements to fatcat-export output --- rust/Cargo.lock | 57 +++++++++++++++++++++++++++ rust/Cargo.toml | 1 + rust/README.export.md | 13 +++++++ rust/src/bin/fatcat-export.rs | 89 +++++++++++++++++++++++++++++++------------ 4 files changed, 136 insertions(+), 24 deletions(-) create mode 100644 rust/README.export.md diff --git a/rust/Cargo.lock b/rust/Cargo.lock index 2a7d6017..a69f3d04 100644 --- a/rust/Cargo.lock +++ b/rust/Cargo.lock @@ -320,6 +320,18 @@ name = "encoding_index_tests" version = "0.1.4" source = "registry+https://github.com/rust-lang/crates.io-index" +[[package]] +name = "env_logger" +version = "0.5.12" +source = "registry+https://github.com/rust-lang/crates.io-index" +dependencies = [ + "atty 0.2.10 (registry+https://github.com/rust-lang/crates.io-index)", + "humantime 1.1.1 (registry+https://github.com/rust-lang/crates.io-index)", + "log 0.4.1 (registry+https://github.com/rust-lang/crates.io-index)", + "regex 1.0.2 (registry+https://github.com/rust-lang/crates.io-index)", + "termcolor 1.0.3 (registry+https://github.com/rust-lang/crates.io-index)", +] + [[package]] name = "error" version = "0.1.9" @@ -356,6 +368,7 @@ dependencies = [ "diesel 1.3.2 (registry+https://github.com/rust-lang/crates.io-index)", "diesel_migrations 1.3.0 (registry+https://github.com/rust-lang/crates.io-index)", "dotenv 0.9.0 (registry+https://github.com/rust-lang/crates.io-index)", + "env_logger 0.5.12 (registry+https://github.com/rust-lang/crates.io-index)", "error-chain 0.12.0 (registry+https://github.com/rust-lang/crates.io-index)", "fatcat-api-spec 0.1.0", "futures 0.1.21 (registry+https://github.com/rust-lang/crates.io-index)", @@ -439,6 +452,14 @@ name = "httparse" version = "1.2.4" source = "registry+https://github.com/rust-lang/crates.io-index" +[[package]] +name = "humantime" +version = "1.1.1" +source = "registry+https://github.com/rust-lang/crates.io-index" +dependencies = [ + "quick-error 1.2.2 (registry+https://github.com/rust-lang/crates.io-index)", +] + [[package]] name = "hyper" version = "0.10.13" @@ -863,6 +884,11 @@ dependencies = [ "unicode-xid 0.1.0 (registry+https://github.com/rust-lang/crates.io-index)", ] +[[package]] +name = "quick-error" +version = "1.2.2" +source = "registry+https://github.com/rust-lang/crates.io-index" + [[package]] name = "quote" version = "0.3.15" @@ -1200,6 +1226,14 @@ dependencies = [ "winapi 0.3.4 (registry+https://github.com/rust-lang/crates.io-index)", ] +[[package]] +name = "termcolor" +version = "1.0.3" +source = "registry+https://github.com/rust-lang/crates.io-index" +dependencies = [ + "wincolor 1.0.1 (registry+https://github.com/rust-lang/crates.io-index)", +] + [[package]] name = "termion" version = "1.5.1" @@ -1431,11 +1465,28 @@ name = "winapi-i686-pc-windows-gnu" version = "0.4.0" source = "registry+https://github.com/rust-lang/crates.io-index" +[[package]] +name = "winapi-util" +version = "0.1.1" +source = "registry+https://github.com/rust-lang/crates.io-index" +dependencies = [ + "winapi 0.3.4 (registry+https://github.com/rust-lang/crates.io-index)", +] + [[package]] name = "winapi-x86_64-pc-windows-gnu" version = "0.4.0" source = "registry+https://github.com/rust-lang/crates.io-index" +[[package]] +name = "wincolor" +version = "1.0.1" +source = "registry+https://github.com/rust-lang/crates.io-index" +dependencies = [ + "winapi 0.3.4 (registry+https://github.com/rust-lang/crates.io-index)", + "winapi-util 0.1.1 (registry+https://github.com/rust-lang/crates.io-index)", +] + [metadata] "checksum aho-corasick 0.6.4 (registry+https://github.com/rust-lang/crates.io-index)" = "d6531d44de723825aa81398a6415283229725a00fa30713812ab9323faa82fc4" "checksum ansi_term 0.11.0 (registry+https://github.com/rust-lang/crates.io-index)" = "ee49baf6cb617b853aa8d93bf420db2383fab46d314482ca2803b40d5fde979b" @@ -1476,6 +1527,7 @@ source = "registry+https://github.com/rust-lang/crates.io-index" "checksum encoding-index-singlebyte 1.20141219.5 (registry+https://github.com/rust-lang/crates.io-index)" = "3351d5acffb224af9ca265f435b859c7c01537c0849754d3db3fdf2bfe2ae84a" "checksum encoding-index-tradchinese 1.20141219.5 (registry+https://github.com/rust-lang/crates.io-index)" = "fd0e20d5688ce3cab59eb3ef3a2083a5c77bf496cb798dc6fcdb75f323890c18" "checksum encoding_index_tests 0.1.4 (registry+https://github.com/rust-lang/crates.io-index)" = "a246d82be1c9d791c5dfde9a2bd045fc3cbba3fa2b11ad558f27d01712f00569" +"checksum env_logger 0.5.12 (registry+https://github.com/rust-lang/crates.io-index)" = "f4d7e69c283751083d53d01eac767407343b8b69c4bd70058e08adc2637cb257" "checksum error 0.1.9 (registry+https://github.com/rust-lang/crates.io-index)" = "a6e606f14042bb87cc02ef6a14db6c90ab92ed6f62d87e69377bc759fd7987cc" "checksum error-chain 0.11.0 (registry+https://github.com/rust-lang/crates.io-index)" = "ff511d5dc435d703f4971bc399647c9bc38e20cb41452e3b9feb4765419ed3f3" "checksum error-chain 0.12.0 (registry+https://github.com/rust-lang/crates.io-index)" = "07e791d3be96241c77c43846b665ef1384606da2cd2a48730abe606a12906e02" @@ -1485,6 +1537,7 @@ source = "registry+https://github.com/rust-lang/crates.io-index" "checksum fuchsia-zircon-sys 0.3.3 (registry+https://github.com/rust-lang/crates.io-index)" = "3dcaa9ae7725d12cdb85b3ad99a434db70b468c09ded17e012d86b5c1010f7a7" "checksum futures 0.1.21 (registry+https://github.com/rust-lang/crates.io-index)" = "1a70b146671de62ec8c8ed572219ca5d594d9b06c0b364d5e67b722fc559b48c" "checksum httparse 1.2.4 (registry+https://github.com/rust-lang/crates.io-index)" = "c2f407128745b78abc95c0ffbe4e5d37427fdc0d45470710cfef8c44522a2e37" +"checksum humantime 1.1.1 (registry+https://github.com/rust-lang/crates.io-index)" = "0484fda3e7007f2a4a0d9c3a703ca38c71c54c55602ce4660c419fd32e188c9e" "checksum hyper 0.10.13 (registry+https://github.com/rust-lang/crates.io-index)" = "368cb56b2740ebf4230520e2b90ebb0461e69034d85d1945febd9b3971426db2" "checksum hyper-openssl 0.2.7 (registry+https://github.com/rust-lang/crates.io-index)" = "5ecb3cd8e4d53f8abe7cb2227e66674bb63c1bd0ba60ca9ba7b74ea1e0054891" "checksum idna 0.1.4 (registry+https://github.com/rust-lang/crates.io-index)" = "014b298351066f1512874135335d62a789ffe78a9974f94b43ed5621951eaf7d" @@ -1533,6 +1586,7 @@ source = "registry+https://github.com/rust-lang/crates.io-index" "checksum plugin 0.2.6 (registry+https://github.com/rust-lang/crates.io-index)" = "1a6a0dc3910bc8db877ffed8e457763b317cf880df4ae19109b9f77d277cf6e0" "checksum pq-sys 0.4.5 (registry+https://github.com/rust-lang/crates.io-index)" = "9323a6ce484fc41174d40f80ba87af6247f86a7ba57856af68d3aa0c8642d2f0" "checksum proc-macro2 0.3.8 (registry+https://github.com/rust-lang/crates.io-index)" = "1b06e2f335f48d24442b35a19df506a835fb3547bc3c06ef27340da9acf5cae7" +"checksum quick-error 1.2.2 (registry+https://github.com/rust-lang/crates.io-index)" = "9274b940887ce9addde99c4eee6b5c44cc494b182b97e73dc8ffdcb3397fd3f0" "checksum quote 0.3.15 (registry+https://github.com/rust-lang/crates.io-index)" = "7a6e920b65c65f10b2ae65c831a81a073a89edd28c7cce89475bff467ab4167a" "checksum quote 0.5.2 (registry+https://github.com/rust-lang/crates.io-index)" = "9949cfe66888ffe1d53e6ec9d9f3b70714083854be20fd5e271b232a017401e8" "checksum r2d2 0.8.2 (registry+https://github.com/rust-lang/crates.io-index)" = "f9078ca6a8a5568ed142083bb2f7dc9295b69d16f867ddcc9849e51b17d8db46" @@ -1576,6 +1630,7 @@ source = "registry+https://github.com/rust-lang/crates.io-index" "checksum take_mut 0.2.2 (registry+https://github.com/rust-lang/crates.io-index)" = "f764005d11ee5f36500a149ace24e00e3da98b0158b3e2d53a7495660d3f4d60" "checksum tempdir 0.3.7 (registry+https://github.com/rust-lang/crates.io-index)" = "15f2b5fb00ccdf689e0149d1b1b3c03fead81c2b37735d812fa8bddbbf41b6d8" "checksum term 0.5.1 (registry+https://github.com/rust-lang/crates.io-index)" = "5e6b677dd1e8214ea1ef4297f85dbcbed8e8cdddb561040cc998ca2551c37561" +"checksum termcolor 1.0.3 (registry+https://github.com/rust-lang/crates.io-index)" = "ff3bac0e465b59f194e7037ed404b0326e56ff234d767edc4c5cc9cd49e7a2c7" "checksum termion 1.5.1 (registry+https://github.com/rust-lang/crates.io-index)" = "689a3bdfaab439fd92bc87df5c4c78417d3cbe537487274e9b0b2dce76e92096" "checksum textwrap 0.9.0 (registry+https://github.com/rust-lang/crates.io-index)" = "c0b59b6b4b44d867f1370ef1bd91bfb262bf07bf0ae65c202ea2fbc16153b693" "checksum thread_local 0.3.5 (registry+https://github.com/rust-lang/crates.io-index)" = "279ef31c19ededf577bfd12dfae728040a21f635b06a24cd670ff510edd38963" @@ -1608,4 +1663,6 @@ source = "registry+https://github.com/rust-lang/crates.io-index" "checksum winapi 0.3.4 (registry+https://github.com/rust-lang/crates.io-index)" = "04e3bd221fcbe8a271359c04f21a76db7d0c6028862d1bb5512d85e1e2eb5bb3" "checksum winapi-build 0.1.1 (registry+https://github.com/rust-lang/crates.io-index)" = "2d315eee3b34aca4797b2da6b13ed88266e6d612562a0c46390af8299fc699bc" "checksum winapi-i686-pc-windows-gnu 0.4.0 (registry+https://github.com/rust-lang/crates.io-index)" = "ac3b87c63620426dd9b991e5ce0329eff545bccbbb34f3be09ff6fb6ab51b7b6" +"checksum winapi-util 0.1.1 (registry+https://github.com/rust-lang/crates.io-index)" = "afc5508759c5bf4285e61feb862b6083c8480aec864fa17a81fdec6f69b461ab" "checksum winapi-x86_64-pc-windows-gnu 0.4.0 (registry+https://github.com/rust-lang/crates.io-index)" = "712e227841d057c1ee1cd2fb22fa7e5a5461ae8e48fa2ca79ec42cfc1931183f" +"checksum wincolor 1.0.1 (registry+https://github.com/rust-lang/crates.io-index)" = "561ed901ae465d6185fa7864d63fbd5720d0ef718366c9a4dc83cf6170d7e9ba" diff --git a/rust/Cargo.toml b/rust/Cargo.toml index 7c6695d5..f9a439db 100644 --- a/rust/Cargo.toml +++ b/rust/Cargo.toml @@ -38,6 +38,7 @@ serde_json = "1.0" # Command-line tools crossbeam-channel = "0.2" num_cpus = "1" +env_logger = "*" # Unused (hyper 0.11 and https) #hyper-openssl = {version = "0.2", optional = true} diff --git a/rust/README.export.md b/rust/README.export.md new file mode 100644 index 00000000..7301a4db --- /dev/null +++ b/rust/README.export.md @@ -0,0 +1,13 @@ + +First create ident files: + + psql fatcat < ../extra/quick_dump.sql + +Then dump: + + cat /tmp/fatcat_ident_releases.tsv | ./target/debug/fatcat-export releases + +Or, perhaps, in production: + + cat /tmp/fatcat_ident_releases.tsv | ./target/release/fatcat-export release --expand files,contaner -j8 | pv | gzip > all_releases.json.gz + diff --git a/rust/src/bin/fatcat-export.rs b/rust/src/bin/fatcat-export.rs index 6f9b92cf..1af321ba 100644 --- a/rust/src/bin/fatcat-export.rs +++ b/rust/src/bin/fatcat-export.rs @@ -9,16 +9,14 @@ extern crate error_chain; extern crate fatcat; extern crate fatcat_api_spec; #[macro_use] -extern crate slog; -extern crate slog_async; -extern crate slog_term; +extern crate log; +extern crate env_logger; extern crate uuid; extern crate crossbeam_channel; extern crate serde_json; extern crate num_cpus; use clap::{App, Arg}; -use slog::{Drain, Logger}; use dotenv::dotenv; use std::env; @@ -32,6 +30,7 @@ use fatcat::api_helpers::*; use fatcat::api_entity_crud::*; use fatcat::errors::*; +use error_chain::ChainedError; use std::thread; //use std::io::{Stdout,StdoutLock}; use crossbeam_channel as channel; @@ -71,18 +70,26 @@ pub fn database_worker_pool() -> Result { macro_rules! generic_loop_work { ($fn_name:ident, $entity_model:ident) => { - fn $fn_name(row_receiver: channel::Receiver, output_sender: channel::Sender, db_conn: &DbConn, expand: Option) -> Result<()> { - for row in row_receiver { - let mut entity = $entity_model::db_get_rev(db_conn, row.rev_id.expect("valid, non-deleted row"))?; - //let mut entity = ReleaseEntity::db_get_rev(db_conn, row.rev_id.expect("valid, non-deleted row"))?; - entity.state = Some("active".to_string()); // XXX - entity.ident = Some(row.ident_id.to_string()); - if let Some(expand) = expand { - entity.db_expand(db_conn, expand)?; + fn $fn_name(row_receiver: channel::Receiver, output_sender: channel::Sender, db_conn: &DbConn, expand: Option) { + let result: Result<()> = (|| { + for row in row_receiver { + let mut entity = $entity_model::db_get_rev(db_conn, row.rev_id.expect("valid, non-deleted row")) + .chain_err(|| "reading entity from database")?; + //let mut entity = ReleaseEntity::db_get_rev(db_conn, row.rev_id.expect("valid, non-deleted row"))?; + entity.state = Some("active".to_string()); // XXX + entity.ident = Some(row.ident_id.to_string()); + if let Some(expand) = expand { + entity.db_expand(db_conn, expand) + .chain_err(|| "expanding sub-entities from database")?; + } + output_sender.send(serde_json::to_string(&entity)?); } - output_sender.send(serde_json::to_string(&entity)?); + Ok(()) + })(); + if let Err(ref e) = result { + error!("{}", e.display_chain()) } - Ok(()) + result.unwrap() } } } @@ -114,14 +121,23 @@ fn parse_line(s: String) -> Result { } Ok(IdentRow { ident_id: FatCatId::from_uuid(&Uuid::from_str(&fields[0])?), - rev_id: Some(Uuid::from_str(&fields[1])?), - redirect_id: None, + rev_id: match fields[1].as_ref() { + "\\N" => None, + val => Some(Uuid::from_str(&val)?), + }, + redirect_id: match fields[2].as_ref() { + "\\N" => None, + val => Some(FatCatId::from_uuid(&Uuid::from_str(&val)?)), + }, }) } #[test] fn test_parse_line() { - assert!(false) + assert!(parse_line("00000000-0000-0000-3333-000000000001\t00000000-0000-0000-3333-fff000000001\t\\N").is_ok()); + assert!(parse_line("00000-0000-0000-3333-000000000001\t00000000-0000-0000-3333-fff000000001\t\\N").is_err()); + assert!(parse_line("00000-0000-0000-3333-000000000001\t00000000-0000-0000-3333-fff000000001").is_err()); + assert!(parse_line("00000000-0000-0000-3333-000000000002\t00000000-0000-0000-3333-fff000000001\t00000000-0000-0000-3333-000000000001").is_ok()); } // Use num_cpus/2, or CLI arg for worker count @@ -135,7 +151,7 @@ fn test_parse_line() { // 4. read rows, pushing to row channel // => every N lines, log to stderr // 5. wait for all channels to finish, and quit -pub fn do_export(num_workers: usize, expand: Option, entity_type: ExportEntityType) -> Result<()> { +pub fn do_export(num_workers: usize, expand: Option, entity_type: ExportEntityType, redirects: bool) -> Result<()> { let db_pool = database_worker_pool()?; let buf_input = BufReader::new(std::io::stdin()); @@ -143,6 +159,8 @@ pub fn do_export(num_workers: usize, expand: Option, entity_type: E let (output_sender, output_receiver) = channel::bounded(CHANNEL_BUFFER_LEN); let (done_sender, done_receiver) = channel::bounded(0); + info!("Starting an export of {} entities", entity_type); + // Start row worker threads assert!(num_workers > 0); for _ in 0..num_workers { @@ -164,15 +182,25 @@ pub fn do_export(num_workers: usize, expand: Option, entity_type: E } drop(output_sender); // Start printer thread - thread::spawn(move || loop_printer(output_receiver, done_sender)); + thread::spawn(move || loop_printer(output_receiver, done_sender).expect("printing to stdout") ); + let mut count = 0; for line in buf_input.lines() { let line = line?; let row = parse_line(line)?; - row_sender.send(row); + match (row.rev_id, row.redirect_id, redirects) { + (None, _, _) => (), + (Some(_), Some(_), false) => (), + _ => row_sender.send(row), + } + count += 1; + if count % 1000 == 0 { + info!("processed {} lines...", count); + } } drop(row_sender); done_receiver.recv(); + info!("Done reading ({} lines), waiting for workers to exit...", count); Ok(()) } @@ -182,28 +210,41 @@ fn run() -> Result<()> { .version(env!("CARGO_PKG_VERSION")) .author("Bryan Newbold ") .about("Fast exports of database entities from an id list") + .arg(Arg::from_usage(" 'what entity type the idents correspond to'") + .possible_values(&ExportEntityType::variants()) + .case_insensitive(true)) .args_from_usage( - "-f --workers=[workers] 'number of threads (database connections) to use' - --expand=[expand] 'sub-entities to include in dump' - 'what entity type the idents correspond to'") + "-j --workers=[workers] 'number of threads (database connections) to use' + -q --quiet 'less status output to stderr' + --include-redirects 'include redirected idents (normally skipped)' + --expand=[expand] 'sub-entities to include in dump'") .after_help("Reads a ident table TSV dump from stdin (aka, ident_id, rev_id, redirect_id), \ and outputs JSON (one entity per line). Database connection info read from environment \ (DATABASE_URL, same as fatcatd).") .get_matches(); let num_workers: usize = match m.value_of("workers") { - Some(v) => value_t_or_exit!(m.value_of("workers"), usize), + Some(_) => value_t_or_exit!(m.value_of("workers"), usize), None => std::cmp::min(1, num_cpus::get() / 2) as usize, }; let expand = match m.value_of("expand") { Some(s) => Some(ExpandFlags::from_str(&s)?), None => None, }; + let log_level = if m.is_present("quiet") { + "warn" + } else { + "info" + }; + let env = env_logger::Env::default() + .filter_or(env_logger::DEFAULT_FILTER_ENV, log_level); + env_logger::Builder::from_env(env).init(); do_export( num_workers, expand, value_t_or_exit!(m.value_of("entity_type"), ExportEntityType), + m.is_present("include_redirects"), ) } -- cgit v1.2.3