Remove reference to log file

.dockerignore

@@ -1,3 +0,0 @@
-.git
-.github
-/target

.github/CODEOWNERS

@@ -1,2 +0,0 @@
-# Automatically request reviews from the synapse-core team when a pull request comes in.
-* @matrix-org/synapse-core

Cargo.toml

@@ -1,5 +1,5 @@
 [workspace]
-members = ["synapse_auto_compressor", "compressor_integration_tests"]
+members = ["auto_compressor", "compressor_integration_tests"]
 
 [package]
 authors = ["Erik Johnston"]
@@ -9,7 +9,9 @@ version = "0.1.0"
 edition = "2018"
 
 [dependencies]
+clap = "2.33.0"
 indicatif = "0.16.0"
+jemallocator = "0.3.2"
 openssl = "0.10.32"
 postgres = "0.19.0"
 postgres-openssl = "0.5.0"
@@ -18,7 +20,7 @@ rayon = "1.3.0"
 string_cache = "0.8.0"
 env_logger = "0.9.0"
 log = "0.4.14"
-pyo3-log = "0.6.0"
+pyo3-log = "0.4.0"
 log-panics = "2.0.0"
 
 [dependencies.state-map]
@@ -28,19 +30,11 @@ git = "https://github.com/matrix-org/rust-matrix-state-map"
 [lib]
 crate-type = ["cdylib", "rlib"]
 
-[dependencies.clap]
-version = "3.1.14"
-features = ["cargo"]
-
 [dependencies.pyo3]
-version = "0.16.4"
+version = "0.14.1"
-features = ["extension-module"]
+features = ["extension-module","abi3-py36"]
-
-[dependencies.tikv-jemallocator]
-version = "0.5.0"
-optional = true
 
 [features]
 default = ["jemalloc"]
-jemalloc = ["tikv-jemallocator"]
+jemalloc = []
 no-progress-bars = []

Dockerfile

@@ -1,22 +0,0 @@
-FROM rust:alpine AS builder
-
-RUN apk add python3 musl-dev pkgconfig openssl-dev make
-
-ENV RUSTFLAGS="-C target-feature=-crt-static"
-
-WORKDIR /opt/synapse-compressor/
-
-COPY . .
-
-RUN cargo build
-
-WORKDIR /opt/synapse-compressor/synapse_auto_compressor/
-
-RUN cargo build
-
-FROM alpine
-
-RUN apk add --no-cache libgcc
-
-COPY --from=builder /opt/synapse-compressor/target/debug/synapse_compress_state /usr/local/bin/synapse_compress_state
-COPY --from=builder /opt/synapse-compressor/target/debug/synapse_auto_compressor /usr/local/bin/synapse_auto_compressor

README.md

@@ -3,7 +3,7 @@
 This workspace contains experimental tools that attempt to reduce the number of
 rows in the `state_groups_state` table inside of a Synapse Postgresql database.
 
-# Automated tool: synapse_auto_compressor
+# Automated tool: auto_compressor
 
 ## Introduction:
 
@@ -11,7 +11,7 @@ This tool is significantly more simple to use than the manual tool (described be
 It scans through all of the rows in the `state_groups` database table from the start. When
 it finds a group that hasn't been compressed, it runs the compressor for a while on that
 group's room, saving where it got up to. After compressing a number of these chunks it stops,
-saving where it got up to for the next run of the `synapse_auto_compressor`.
+saving where it got up to for the next run of the `auto_compressor`.
 
 It creates three extra tables in the database: `state_compressor_state` which stores the
 information needed to stop and start the compressor for each room, `state_compressor_progress`
@@ -26,15 +26,14 @@ periodically.
 This tool requires `cargo` to be installed. See https://www.rust-lang.org/tools/install
 for instructions on how to do this.
 
-To build `synapse_auto_compressor`, clone this repository and navigate to the
+To build `auto_compressor`, clone this repository and navigate to the `autocompressor/` 
-`synapse_auto_compressor/` subdirectory. Then execute `cargo build`.
+subdirectory. Then execute `cargo build`.
 
-This will create an executable and store it in
+This will create an executable and store it in `auto_compressor/target/debug/auto_compressor`.
-`synapse_auto_compressor/target/debug/synapse_auto_compressor`.
 
 ## Example usage
 ```
-$ synapse_auto_compressor -p postgresql://user:pass@localhost/synapse -c 500 -n 100
+$ auto_compressor -p postgresql://user:pass@localhost/synapse -c 500 -n 100
 ```
 ## Running Options
 
@@ -197,10 +196,10 @@ $ docker-compose down
 # Using the synapse_compress_state library
 
 If you want to use the compressor in another project, it is recomended that you
-use jemalloc `https://github.com/tikv/jemallocator`.
+use jemalloc `https://github.com/gnzlbg/jemallocator`. 
 
 To prevent the progress bars from being shown, use the `no-progress-bars` feature.
-(See `synapse_auto_compressor/Cargo.toml` for an example)
+(See `auto_compressor/Cargo.toml` for an example)
 
 # Troubleshooting
 
@@ -226,17 +225,17 @@ setting in [`postgresql.conf`](https://www.postgresql.org/docs/current/runtime-c
 The amount of output the tools produce can be altered by setting the RUST_LOG 
 environment variable to something. 
 
-To get more logs when running the synapse_auto_compressor tool try the following:
+To get more logs when running the auto_compressor tool try the following:
 
 ```
-$ RUST_LOG=debug synapse_auto_compressor -p postgresql://user:pass@localhost/synapse -c 50 -n 100
+$ RUST_LOG=debug auto_compressor -p postgresql://user:pass@localhost/synapse -c 50 -n 100
 ```
 
 If you want to suppress all the debugging info you are getting from the 
 Postgres client then try:
 
 ```
-RUST_LOG=synapse_auto_compressor=debug,synapse_compress_state=debug synapse_auto_compressor [etc.]
+RUST_LOG=auto_compressor=debug,synapse_compress_state=debug auto_compressor [etc.]
 ```
 
 This will only print the debugging information from those two packages. For more info see 
@@ -266,7 +265,7 @@ be a large problem.
 
 ## Compressor is trying to increase the number of rows
 
-Backfilling can lead to issues with compression. The synapse_auto_compressor will
+Backfilling can lead to issues with compression. The auto_compressor will
 skip chunks it can't reduce the size of and so this should help jump over the backfilled 
 state_groups. Lots of state resolution might also impact the ability to use the compressor.
 

auto_compressor/Cargo.toml

@@ -0,0 +1,30 @@
+[package]
+name = "auto_compressor"
+authors = ["William Ashton"]
+version = "0.1.0"
+edition = "2018"
+
+# See more keys and their definitions at https://doc.rust-lang.org/cargo/reference/manifest.html
+
+[dependencies]
+clap = "2.33.0"
+openssl = "0.10.32"
+postgres = "0.19.0"
+postgres-openssl = "0.5.0"
+jemallocator = "0.3.2"
+rand = "0.8.0"
+serial_test = "0.5.1"
+synapse_compress_state = { path = "../", features = ["no-progress-bars"] }
+env_logger = "0.9.0"
+log = "0.4.14"
+log-panics = "2.0.0"
+anyhow = "1.0.42"
+pyo3-log = "0.4.0"
+
+# Needed for pyo3 support
+[lib]
+crate-type = ["cdylib", "rlib"]
+
+[dependencies.pyo3]
+version = "0.14.1"
+features = ["extension-module","abi3-py36"]

auto_compressor/src/lib.rs

@@ -57,16 +57,15 @@ impl FromStr for LevelInfo {
 
 // PyO3 INTERFACE STARTS HERE
 #[pymodule]
-fn synapse_auto_compressor(_py: Python, m: &PyModule) -> PyResult<()> {
+fn auto_compressor(_py: Python, m: &PyModule) -> PyResult<()> {
     let _ = pyo3_log::Logger::default()
         // don't send out anything lower than a warning from other crates
         .filter(LevelFilter::Warn)
-        // don't log warnings from synapse_compress_state, the
+        // don't log warnings from synapse_compress_state, the auto_compressor handles these
-        // synapse_auto_compressor handles these situations and provides better
+        // situations and provides better log messages
-        // log messages
         .filter_target("synapse_compress_state".to_owned(), LevelFilter::Error)
-        // log info and above for the synapse_auto_compressor
+        // log info and above for the auto_compressor
-        .filter_target("synapse_auto_compressor".to_owned(), LevelFilter::Debug)
+        .filter_target("auto_compressor".to_owned(), LevelFilter::Debug)
         .install();
     // ensure any panics produce error messages in the log
     log_panics::init();
@@ -93,7 +92,7 @@ fn synapse_auto_compressor(_py: Python, m: &PyModule) -> PyResult<()> {
         number_of_chunks: i64,
     ) -> PyResult<()> {
         // Announce the start of the program to the logs
-        log::info!("synapse_auto_compressor started");
+        log::info!("auto_compressor started");
 
         // Parse the default_level string into a LevelInfo struct
         let default_levels: LevelInfo = match default_levels.parse() {
@@ -121,7 +120,7 @@ fn synapse_auto_compressor(_py: Python, m: &PyModule) -> PyResult<()> {
             return Err(PyErr::new::<PyRuntimeError, _>(format!("{:?}", e)));
         }
 
-        log::info!("synapse_auto_compressor finished");
+        log::info!("auto_compressor finished");
         Ok(())
     }
     Ok(())

auto_compressor/src/main.rs

@@ -17,46 +17,53 @@
 //! continue from where it left off.
 
 #[global_allocator]
-static GLOBAL: tikv_jemallocator::Jemalloc = tikv_jemallocator::Jemalloc;
+static GLOBAL: jemallocator::Jemalloc = jemallocator::Jemalloc;
 
-use clap::{crate_authors, crate_description, crate_name, crate_version, Arg, Command};
+use auto_compressor::{manager, state_saving, LevelInfo};
+use clap::{crate_authors, crate_description, crate_name, crate_version, value_t, App, Arg};
 use log::LevelFilter;
-use std::env;
+use std::{env, fs::OpenOptions};
-use synapse_auto_compressor::{manager, state_saving, LevelInfo};
 
 /// Execution starts here
 fn main() {
-    // setup the logger for the synapse_auto_compressor
+    // setup the logger for the auto_compressor
     // The default can be overwritten with RUST_LOG
     // see the README for more information
+    let log_file = OpenOptions::new()
+        .append(true)
+        .create(true)
+        .open("auto_compressor.log")
+        .unwrap_or_else(|e| panic!("Error occured while opening the log file: {}", e));
+
     if env::var("RUST_LOG").is_err() {
         let mut log_builder = env_logger::builder();
         // Ensure panics still come through
         log_builder.filter_module("panic", LevelFilter::Error);
         // Only output errors from the synapse_compress state library
         log_builder.filter_module("synapse_compress_state", LevelFilter::Error);
-        // Output log levels info and above from synapse_auto_compressor
+        // Output log levels info and above from auto_compressor
-        log_builder.filter_module("synapse_auto_compressor", LevelFilter::Info);
+        log_builder.filter_module("auto_compressor", LevelFilter::Info);
         log_builder.init();
     } else {
         // If RUST_LOG was set then use that
         let mut log_builder = env_logger::Builder::from_env("RUST_LOG");
+        log_builder.target(env_logger::Target::Pipe(Box::new(log_file)));
         // Ensure panics still come through
         log_builder.filter_module("panic", LevelFilter::Error);
         log_builder.init();
     }
     log_panics::init();
     // Announce the start of the program to the logs
-    log::info!("synapse_auto_compressor started");
+    log::info!("auto_compressor started");
 
     // parse the command line arguments using the clap crate
-    let arguments = Command::new(crate_name!())
+    let arguments = App::new(crate_name!())
         .version(crate_version!())
         .author(crate_authors!("\n"))
         .about(crate_description!())
         .arg(
-            Arg::new("postgres-url")
+            Arg::with_name("postgres-url")
-                .short('p')
+                .short("p")
                 .value_name("POSTGRES_LOCATION")
                 .help("The configruation for connecting to the postgres database.")
                 .long_help(concat!(
@@ -69,8 +76,8 @@ fn main() {
                 .takes_value(true)
                 .required(true),
         ).arg(
-            Arg::new("chunk_size")
+            Arg::with_name("chunk_size")
-                .short('c')
+                .short("c")
                 .value_name("COUNT")
                 .help("The maximum number of state groups to load into memroy at once")
                 .long_help(concat!(
@@ -85,8 +92,8 @@ fn main() {
                 .takes_value(true)
                 .required(true),
         ).arg(
-            Arg::new("default_levels")
+            Arg::with_name("default_levels")
-                .short('l')
+                .short("l")
                 .value_name("LEVELS")
                 .help("Sizes of each new level in the compression algorithm, as a comma separated list.")
                 .long_help(concat!(
@@ -103,8 +110,8 @@ fn main() {
                 .takes_value(true)
                 .required(false),
         ).arg(
-            Arg::new("number_of_chunks")
+            Arg::with_name("number_of_chunks")
-                .short('n')
+                .short("n")
                 .value_name("CHUNKS_TO_COMPRESS")
                 .help("The number of chunks to compress") 
                 .long_help(concat!(
@@ -127,8 +134,7 @@ fn main() {
         .expect("A chunk size is required");
 
     // The default structure to use when compressing
-    let default_levels = arguments
+    let default_levels = value_t!(arguments, "default_levels", LevelInfo)
-        .value_of_t::<LevelInfo>("default_levels")
         .unwrap_or_else(|e| panic!("Unable to parse default levels: {}", e));
 
     // The number of rooms to compress with this tool
@@ -149,5 +155,5 @@ fn main() {
     manager::compress_chunks_of_database(db_url, chunk_size, &default_levels.0, number_of_chunks)
         .unwrap();
 
-    log::info!("synapse_auto_compressor finished");
+    log::info!("auto_compressor finished");
 }

compressor_integration_tests/Cargo.toml

@@ -13,7 +13,7 @@ postgres = "0.19.0"
 postgres-openssl = "0.5.0"
 rand = "0.8.0"
 synapse_compress_state = { path = "../", features = ["no-progress-bars"] }
-synapse_auto_compressor = { path = "../synapse_auto_compressor/" }
+auto_compressor = { path = "../auto_compressor/" }
 env_logger = "0.9.0"
 log = "0.4.14"
 

compressor_integration_tests/src/lib.rs

@@ -356,7 +356,7 @@ fn functions_are_self_consistent() {
 }
 
 pub fn setup_logger() {
-    // setup the logger for the synapse_auto_compressor
+    // setup the logger for the auto_compressor
     // The default can be overwritten with RUST_LOG
     // see the README for more information
     if env::var("RUST_LOG").is_err() {
@@ -366,7 +366,7 @@ pub fn setup_logger() {
         // default to printing the debug information for both packages being tested
         // (Note that just setting the global level to debug will log every sql transaction)
         log_builder.filter_module("synapse_compress_state", LevelFilter::Debug);
-        log_builder.filter_module("synapse_auto_compressor", LevelFilter::Debug);
+        log_builder.filter_module("auto_compressor", LevelFilter::Debug);
         // use try_init() incase the logger has been setup by some previous test
         let _ = log_builder.try_init();
     } else {

compressor_integration_tests/tests/auto_compressor_manager_tests.rs

@@ -1,5 +1,9 @@
 use std::collections::BTreeMap;
 
+use auto_compressor::{
+    manager::{compress_chunks_of_database, run_compressor_on_room_chunk},
+    state_saving::{connect_to_database, create_tables_if_needed},
+};
 use compressor_integration_tests::{
     add_contents_to_database, clear_compressor_state, database_collapsed_states_match_map,
     database_structure_matches_map, empty_database,
@@ -10,10 +14,6 @@ use compressor_integration_tests::{
     setup_logger, DB_URL,
 };
 use serial_test::serial;
-use synapse_auto_compressor::{
-    manager::{compress_chunks_of_database, run_compressor_on_room_chunk},
-    state_saving::{connect_to_database, create_tables_if_needed},
-};
 use synapse_compress_state::Level;
 
 #[test]

compressor_integration_tests/tests/auto_compressor_state_saving_tests.rs

@@ -1,9 +1,9 @@
-use compressor_integration_tests::{clear_compressor_state, setup_logger, DB_URL};
+use auto_compressor::state_saving::{
-use serial_test::serial;
-use synapse_auto_compressor::state_saving::{
     connect_to_database, create_tables_if_needed, read_room_compressor_state,
     write_room_compressor_state,
 };
+use compressor_integration_tests::{clear_compressor_state, setup_logger, DB_URL};
+use serial_test::serial;
 use synapse_compress_state::Level;
 
 #[test]

compressor_integration_tests/tests/compressor_config_tests.rs

@@ -46,11 +46,10 @@ fn run_succeeds_without_crashing() {
     let transactions = true;
     let graphs = false;
     let commit_changes = false;
-    let verify = true;
 
     let config = Config::new(
-        db_url,
+        db_url.clone(),
-        room_id,
+        room_id.clone(),
         output_file,
         min_state_group,
         groups_to_compress,
@@ -60,7 +59,6 @@ fn run_succeeds_without_crashing() {
         transactions,
         graphs,
         commit_changes,
-        verify,
     )
     .unwrap();
 
@@ -96,7 +94,6 @@ fn changes_commited_if_no_min_saved_rows() {
     let transactions = true;
     let graphs = false;
     let commit_changes = true;
-    let verify = true;
 
     let config = Config::new(
         db_url,
@@ -110,7 +107,6 @@ fn changes_commited_if_no_min_saved_rows() {
         transactions,
         graphs,
         commit_changes,
-        verify,
     )
     .unwrap();
 
@@ -164,7 +160,6 @@ fn changes_commited_if_min_saved_rows_exceeded() {
     let transactions = true;
     let graphs = false;
     let commit_changes = true;
-    let verify = true;
 
     let config = Config::new(
         db_url,
@@ -178,7 +173,6 @@ fn changes_commited_if_min_saved_rows_exceeded() {
         transactions,
         graphs,
         commit_changes,
-        verify,
     )
     .unwrap();
 
@@ -233,7 +227,6 @@ fn changes_not_commited_if_fewer_than_min_saved_rows() {
     let transactions = true;
     let graphs = false;
     let commit_changes = true;
-    let verify = true;
 
     let config = Config::new(
         db_url,
@@ -247,7 +240,6 @@ fn changes_not_commited_if_fewer_than_min_saved_rows() {
         transactions,
         graphs,
         commit_changes,
-        verify,
     )
     .unwrap();
 
@@ -288,7 +280,6 @@ fn run_panics_if_invalid_db_url() {
     let transactions = true;
     let graphs = false;
     let commit_changes = true;
-    let verify = true;
 
     let config = Config::new(
         db_url,
@@ -302,7 +293,6 @@ fn run_panics_if_invalid_db_url() {
         transactions,
         graphs,
         commit_changes,
-        verify,
     )
     .unwrap();
 
@@ -346,7 +336,6 @@ fn run_only_affects_given_room_id() {
     let transactions = true;
     let graphs = false;
     let commit_changes = true;
-    let verify = true;
 
     let config = Config::new(
         db_url,
@@ -360,7 +349,6 @@ fn run_only_affects_given_room_id() {
         transactions,
         graphs,
         commit_changes,
-        verify,
     )
     .unwrap();
 
@@ -418,7 +406,6 @@ fn run_respects_groups_to_compress() {
     let transactions = true;
     let graphs = false;
     let commit_changes = true;
-    let verify = true;
 
     let config = Config::new(
         db_url,
@@ -432,7 +419,6 @@ fn run_respects_groups_to_compress() {
         transactions,
         graphs,
         commit_changes,
-        verify,
     )
     .unwrap();
 
@@ -506,7 +492,6 @@ fn run_is_idempotent_when_run_on_whole_room() {
     let transactions = true;
     let graphs = false;
     let commit_changes = true;
-    let verify = true;
 
     let config1 = Config::new(
         db_url.clone(),
@@ -520,23 +505,21 @@ fn run_is_idempotent_when_run_on_whole_room() {
         transactions,
         graphs,
         commit_changes,
-        verify,
     )
     .unwrap();
 
     let config2 = Config::new(
-        db_url,
+        db_url.clone(),
-        room_id,
+        room_id.clone(),
         output_file2,
         min_state_group,
         groups_to_compress,
         min_saved_rows,
         max_state_group,
-        level_sizes,
+        level_sizes.clone(),
         transactions,
         graphs,
         commit_changes,
-        verify,
     )
     .unwrap();
 

compressor_integration_tests/tests/compressor_continue_run_tests.rs

@@ -56,7 +56,7 @@ fn continue_run_called_twice_same_as_run() {
 
     let start = Some(6);
     let chunk_size = 7;
-    let level_info = chunk_stats_1.new_level_info;
+    let level_info = chunk_stats_1.new_level_info.clone();
 
     // Run the compressor with those settings
     let chunk_stats_2 = continue_run(start, chunk_size, &db_url, &room_id, &level_info).unwrap();

docs/python.md

@@ -18,7 +18,7 @@ the compressor is run.
 
 3. Navigate to the correct location  
 For the automatic tool:  
-`$ cd /home/synapse/rust-synapse-compress-state/synpase_auto_compressor`   
+`$ cd /home/synapse/rust-synapse-compress-state/auto_compressor`   
 For the manual tool:  
 `$ cd /home/synapse/rust-synapse-compress-state`   
 
@@ -30,9 +30,9 @@ This will install the relevant compressor tool into the activated virtual enviro
 ## Automatic tool example:
 
 ```python
-import synapse_auto_compressor
+import auto_compressor
 
-synapse_auto_compressor.compress_state_events_table(
+auto_compressor.compress_state_events_table(
   db_url="postgresql://localhost/synapse",
   chunk_size=500,
   default_levels="100,50,25",

src/compressor.rs

@@ -181,11 +181,12 @@ impl<'a> Compressor<'a> {
             panic!("Can only call `create_new_tree` once");
         }
 
-        let pb = if cfg!(feature = "no-progress-bars") {
+        let pb: ProgressBar;
-            ProgressBar::hidden()
+        if cfg!(feature = "no-progress-bars") {
+            pb = ProgressBar::hidden();
         } else {
-            ProgressBar::new(self.original_state_map.len() as u64)
+            pb = ProgressBar::new(self.original_state_map.len() as u64);
-        };
+        }
         pb.set_style(
             ProgressStyle::default_bar().template("[{elapsed_precise}] {bar} {pos}/{len} {msg}"),
         );

src/compressor/compressor_tests.rs

@@ -96,7 +96,11 @@ fn create_new_tree_does_nothing_if_already_compressed() {
         let pred_group = initial_edges.get(&i);
 
         // Need Option<i64> not Option<&i64>
-        let prev = pred_group.copied();
+        let prev;
+        match pred_group {
+            Some(i) => prev = Some(*i),
+            None => prev = None,
+        }
 
         // insert that edge into the initial map
         initial.insert(

src/compressor/level_tests.rs

@@ -54,7 +54,7 @@ fn get_head_returns_head() {
 #[test]
 fn has_space_returns_true_if_empty() {
     let l = Level::new(15);
-    assert!(l.has_space());
+    assert_eq!(l.has_space(), true);
 }
 
 #[test]
@@ -65,7 +65,7 @@ fn has_space_returns_true_if_part_full() {
     l.update(1, true);
     l.update(143, true);
     l.update(15, true);
-    assert!(l.has_space());
+    assert_eq!(l.has_space(), true);
 }
 
 #[test]
@@ -76,5 +76,5 @@ fn has_space_returns_false_if_full() {
     l.update(3, true);
     l.update(4, true);
     l.update(5, true);
-    assert!(!l.has_space());
+    assert_eq!(l.has_space(), false);
 }

src/compressor/stats_tests.rs

@@ -145,7 +145,11 @@ fn stats_correct_if_no_changes() {
         let pred_group = initial_edges.get(&i);
 
         // Need Option<i64> not Option<&i64>
-        let prev = pred_group.copied();
+        let prev;
+        match pred_group {
+            Some(i) => prev = Some(*i),
+            None => prev = None,
+        }
 
         // insert that edge into the initial map
         initial.insert(

src/database.rs

@@ -372,11 +372,12 @@ fn get_initial_data_from_db(
     // Copy the data from the database into a map
     let mut state_group_map: BTreeMap<i64, StateGroupEntry> = BTreeMap::new();
 
-    let pb = if cfg!(feature = "no-progress-bars") {
+    let pb: ProgressBar;
-        ProgressBar::hidden()
+    if cfg!(feature = "no-progress-bars") {
+        pb = ProgressBar::hidden();
     } else {
-        ProgressBar::new_spinner()
+        pb = ProgressBar::new_spinner();
-    };
+    }
     pb.set_style(
         ProgressStyle::default_spinner().template("{spinner} [{elapsed}] {pos} rows retrieved"),
     );
@@ -536,11 +537,12 @@ pub fn send_changes_to_db(
     debug!("Writing changes...");
 
     // setup the progress bar
-    let pb = if cfg!(feature = "no-progress-bars") {
+    let pb: ProgressBar;
-        ProgressBar::hidden()
+    if cfg!(feature = "no-progress-bars") {
+        pb = ProgressBar::hidden();
     } else {
-        ProgressBar::new(old_map.len() as u64)
+        pb = ProgressBar::new(old_map.len() as u64);
-    };
+    }
     pb.set_style(
         ProgressStyle::default_bar().template("[{elapsed_precise}] {bar} {pos}/{len} {msg}"),
     );

src/lib.rs

@@ -23,11 +23,11 @@
 use log::{info, warn, LevelFilter};
 use pyo3::{exceptions, prelude::*};
 
-use clap::{crate_authors, crate_description, crate_name, crate_version, Arg, Command};
+use clap::{crate_authors, crate_description, crate_name, crate_version, value_t, App, Arg};
 use indicatif::{ProgressBar, ProgressStyle};
 use rayon::prelude::*;
 use state_map::StateMap;
-use std::{collections::BTreeMap, convert::TryInto, fs::File, io::Write, str::FromStr};
+use std::{collections::BTreeMap, fs::File, io::Write, str::FromStr};
 use string_cache::DefaultAtom as Atom;
 
 mod compressor;
@@ -109,21 +109,18 @@ pub struct Config {
     // Whether or not to commit changes to the database automatically
     // N.B. currently assumes transactions is true (to be on the safe side)
     commit_changes: bool,
-    // Whether to verify the correctness of the compressed state groups by
-    // comparing them to the original groups
-    verify: bool,
 }
 
 impl Config {
     /// Build up config from command line arguments
     pub fn parse_arguments() -> Config {
-        let matches = Command::new(crate_name!())
+        let matches = App::new(crate_name!())
         .version(crate_version!())
         .author(crate_authors!("\n"))
         .about(crate_description!())
         .arg(
-            Arg::new("postgres-url")
+            Arg::with_name("postgres-url")
-                .short('p')
+                .short("p")
                 .value_name("POSTGRES_LOCATION")
                 .help("The configruation for connecting to the postgres database.")
                 .long_help(concat!(
@@ -136,8 +133,8 @@ impl Config {
                 .takes_value(true)
                 .required(true),
         ).arg(
-            Arg::new("room_id")
+            Arg::with_name("room_id")
-                .short('r')
+                .short("r")
                 .value_name("ROOM_ID")
                 .help("The room to process")
                 .long_help(concat!(
@@ -147,23 +144,23 @@ impl Config {
                 .takes_value(true)
                 .required(true),
         ).arg(
-            Arg::new("min_state_group")
+            Arg::with_name("min_state_group")
-                .short('b')
+                .short("b")
                 .value_name("MIN_STATE_GROUP")
                 .help("The state group to start processing from (non inclusive)")
                 .takes_value(true)
                 .required(false),
         ).arg(
-            Arg::new("min_saved_rows")
+            Arg::with_name("min_saved_rows")
-                .short('m')
+                .short("m")
                 .value_name("COUNT")
                 .help("Abort if fewer than COUNT rows would be saved")
                 .long_help("If the compressor cannot save this many rows from the database then it will stop early")
                 .takes_value(true)
                 .required(false),
         ).arg(
-            Arg::new("groups_to_compress")
+            Arg::with_name("groups_to_compress")
-                .short('n')
+                .short("n")
                 .value_name("GROUPS_TO_COMPRESS")
                 .help("How many groups to load into memory to compress") 
                 .long_help(concat!(
@@ -172,14 +169,14 @@ impl Config {
                 .takes_value(true)
                 .required(false),
         ).arg(
-            Arg::new("output_file")
+            Arg::with_name("output_file")
-                .short('o')
+                .short("o")
                 .value_name("FILE")
                 .help("File to output the changes to in SQL")
                 .takes_value(true),
         ).arg(
-            Arg::new("max_state_group")
+            Arg::with_name("max_state_group")
-                .short('s')
+                .short("s")
                 .value_name("MAX_STATE_GROUP")
                 .help("The maximum state group to process up to")
                 .long_help(concat!(
@@ -188,8 +185,8 @@ impl Config {
                 .takes_value(true)
                 .required(false),
         ).arg(
-            Arg::new("level_sizes")
+            Arg::with_name("level_sizes")
-                .short('l')
+                .short("l")
                 .value_name("LEVELS")
                 .help("Sizes of each new level in the compression algorithm, as a comma separated list.")
                 .long_help(concat!(
@@ -205,34 +202,27 @@ impl Config {
                 .default_value("100,50,25")
                 .takes_value(true),
         ).arg(
-            Arg::new("transactions")
+            Arg::with_name("transactions")
-                .short('t')
+                .short("t")
                 .help("Whether to wrap each state group change in a transaction")
                 .long_help(concat!("If this flag is set then then each change to a particular",
                     " state group is wrapped in a transaction. This should be done if you wish to",
                     " apply the changes while synapse is still running."))
                 .requires("output_file"),
         ).arg(
-            Arg::new("graphs")
+            Arg::with_name("graphs")
-                .short('g')
+                .short("g")
                 .help("Output before and after graphs")
                 .long_help(concat!("If this flag is set then output the node and edge information for",
                     " the state_group directed graph built up from the predecessor state_group links.",
                     " These can be looked at in something like Gephi (https://gephi.org)")),
         ).arg(
-            Arg::new("commit_changes")
+            Arg::with_name("commit_changes")
-                .short('c')
+                .short("c")
                 .help("Commit changes to the database")
                 .long_help(concat!("If this flag is set then the changes the compressor makes will",
                     " be committed to the database. This should be safe to use while synapse is running",
                     " as it assumes by default that the transactions flag is set")),
-        ).arg(
-            Arg::new("no_verify")
-                .short('N')
-                .help("Do not double-check that the compression was performed correctly")
-                .long_help(concat!("If this flag is set then the verification of the compressed",
-                    " state groups, which compares them to the original groups, is skipped. This",
-                    " saves time at the cost of potentially generating mismatched state.")),
         ).get_matches();
 
         let db_url = matches
@@ -263,8 +253,7 @@ impl Config {
             .value_of("max_state_group")
             .map(|s| s.parse().expect("max_state_group must be an integer"));
 
-        let level_sizes = matches
+        let level_sizes = value_t!(matches, "level_sizes", LevelSizes)
-            .value_of_t::<LevelSizes>("level_sizes")
             .unwrap_or_else(|e| panic!("Unable to parse level_sizes: {}", e));
 
         let transactions = matches.is_present("transactions");
@@ -273,8 +262,6 @@ impl Config {
 
         let commit_changes = matches.is_present("commit_changes");
 
-        let verify = !matches.is_present("no_verify");
-
         Config {
             db_url: String::from(db_url),
             output_file,
@@ -287,7 +274,6 @@ impl Config {
             transactions,
             graphs,
             commit_changes,
-            verify,
         }
     }
 }
@@ -376,8 +362,8 @@ pub fn run(mut config: Config) {
     }
 
     if let Some(min) = config.min_saved_rows {
-        let saving = original_summed_size.saturating_sub(compressed_summed_size);
+        let saving = (original_summed_size - compressed_summed_size) as i32;
-        if saving < min.try_into().unwrap_or(0) {
+        if saving < min {
             warn!(
                 "Only {} rows would be saved by this compression. Skipping output.",
                 saving
@@ -386,9 +372,7 @@ pub fn run(mut config: Config) {
         }
     }
 
-    if config.verify {
+    check_that_maps_match(&state_group_map, new_state_group_map);
-        check_that_maps_match(&state_group_map, new_state_group_map);
-    }
 
     // If we are given an output file, we output the changes as SQL. If the
     // `transactions` argument is set we wrap each change to a state group in a
@@ -508,11 +492,12 @@ fn output_sql(
 
     info!("Writing changes...");
 
-    let pb = if cfg!(feature = "no-progress-bars") {
+    let pb: ProgressBar;
-        ProgressBar::hidden()
+    if cfg!(feature = "no-progress-bars") {
+        pb = ProgressBar::hidden();
     } else {
-        ProgressBar::new(old_map.len() as u64)
+        pb = ProgressBar::new(old_map.len() as u64);
-    };
+    }
     pb.set_style(
         ProgressStyle::default_bar().template("[{elapsed_precise}] {bar} {pos}/{len} {msg}"),
     );
@@ -622,11 +607,12 @@ fn check_that_maps_match(
 ) {
     info!("Checking that state maps match...");
 
-    let pb = if cfg!(feature = "no-progress-bars") {
+    let pb: ProgressBar;
-        ProgressBar::hidden()
+    if cfg!(feature = "no-progress-bars") {
+        pb = ProgressBar::hidden();
     } else {
-        ProgressBar::new(old_map.len() as u64)
+        pb = ProgressBar::new(old_map.len() as u64);
-    };
+    }
     pb.set_style(
         ProgressStyle::default_bar().template("[{elapsed_precise}] {bar} {pos}/{len} {msg}"),
     );
@@ -709,7 +695,6 @@ impl Config {
         transactions: bool,
         graphs: bool,
         commit_changes: bool,
-        verify: bool,
     ) -> Result<Config, String> {
         let mut output: Option<File> = None;
         if let Some(file) = output_file {
@@ -737,7 +722,6 @@ impl Config {
             transactions,
             graphs,
             commit_changes,
-            verify,
         })
     }
 }
@@ -762,7 +746,6 @@ impl Config {
     transactions = true,
     graphs = false,
     commit_changes = false,
-    verify = true,
 )]
 fn run_compression(
     db_url: String,
@@ -776,7 +759,6 @@ fn run_compression(
     transactions: bool,
     graphs: bool,
     commit_changes: bool,
-    verify: bool,
 ) -> PyResult<()> {
     let config = Config::new(
         db_url,
@@ -790,7 +772,6 @@ fn run_compression(
         transactions,
         graphs,
         commit_changes,
-        verify,
     );
     match config {
         Err(e) => Err(PyErr::new::<exceptions::PyException, _>(e)),
@@ -807,7 +788,7 @@ fn synapse_compress_state(_py: Python, m: &PyModule) -> PyResult<()> {
     let _ = pyo3_log::Logger::default()
         // don't send out anything lower than a warning from other crates
         .filter(LevelFilter::Warn)
-        // don't log warnings from synapse_compress_state, the synapse_auto_compressor handles these
+        // don't log warnings from synapse_compress_state, the auto_compressor handles these
         // situations and provides better log messages
         .filter_target("synapse_compress_state".to_owned(), LevelFilter::Debug)
         .install();
@@ -974,6 +955,7 @@ mod lib_tests {
     #[test]
     fn check_that_maps_match_returns_if_both_empty() {
         check_that_maps_match(&BTreeMap::new(), &BTreeMap::new());
+        assert!(true);
     }
 
     #[test]
@@ -1006,6 +988,7 @@ mod lib_tests {
         }
 
         check_that_maps_match(&old_map, &BTreeMap::new());
+        assert!(true);
     }
 
     #[test]
@@ -1041,6 +1024,7 @@ mod lib_tests {
         }
 
         check_that_maps_match(&BTreeMap::new(), &new_map);
+        assert!(true);
     }
 
     #[test]
@@ -1072,6 +1056,7 @@ mod lib_tests {
         }
 
         check_that_maps_match(&BTreeMap::new(), &old_map.clone());
+        assert!(true);
     }
 
     #[test]
@@ -1134,6 +1119,7 @@ mod lib_tests {
         }
 
         check_that_maps_match(&old_map, &new_map);
+        assert!(true);
     }
 
     #[test]
@@ -1215,6 +1201,7 @@ mod lib_tests {
         );
 
         check_that_maps_match(&old_map, &new_map);
+        assert!(true);
     }
 
     //TODO: tests for correct SQL code produced by output_sql
@@ -1237,7 +1224,6 @@ mod pyo3_tests {
         let transactions = false;
         let graphs = false;
         let commit_changes = false;
-        let verify = true;
 
         let config = Config::new(
             db_url.clone(),
@@ -1251,7 +1237,6 @@ mod pyo3_tests {
             transactions,
             graphs,
             commit_changes,
-            verify,
         )
         .unwrap();
 
@@ -1285,7 +1270,6 @@ mod pyo3_tests {
         let transactions = true;
         let graphs = true;
         let commit_changes = true;
-        let verify = true;
 
         let config = Config::new(
             db_url.clone(),
@@ -1299,12 +1283,11 @@ mod pyo3_tests {
             transactions,
             graphs,
             commit_changes,
-            verify,
         )
         .unwrap();
 
         assert_eq!(config.db_url, db_url);
-        assert!(config.output_file.is_some());
+        assert!(!config.output_file.is_none());
         assert_eq!(config.room_id, room_id);
         assert_eq!(config.min_state_group, Some(3225));
         assert_eq!(config.groups_to_compress, Some(970));

src/main.rs

@@ -18,7 +18,7 @@
 
 #[cfg(feature = "jemalloc")]
 #[global_allocator]
-static GLOBAL: tikv_jemallocator::Jemalloc = tikv_jemallocator::Jemalloc;
+static GLOBAL: jemallocator::Jemalloc = jemallocator::Jemalloc;
 
 use log::LevelFilter;
 use std::env;

synapse_auto_compressor/Cargo.toml

@@ -1,39 +0,0 @@
-[package]
-name = "synapse_auto_compressor"
-authors = ["William Ashton"]
-version = "0.1.3"
-edition = "2018"
-
-[package.metadata.maturin]
-requires-python = ">=3.7"
-project-url = {Source = "https://github.com/matrix-org/rust-synapse-compress-state"}
-classifier = [
-    "Development Status :: 4 - Beta",
-    "Programming Language :: Rust",
-]
-
-[dependencies]
-openssl = "0.10.32"
-postgres = "0.19.0"
-postgres-openssl = "0.5.0"
-tikv-jemallocator = "0.5.0"
-rand = "0.8.0"
-serial_test = "0.5.1"
-synapse_compress_state = { path = "../", features = ["no-progress-bars"] }
-env_logger = "0.9.0"
-log = "0.4.14"
-log-panics = "2.0.0"
-anyhow = "1.0.42"
-pyo3-log = "0.6.0"
-
-# Needed for pyo3 support
-[lib]
-crate-type = ["cdylib", "rlib"]
-
-[dependencies.clap]
-version = "3.1.14"
-features = ["cargo"]
-
-[dependencies.pyo3]
-version = "0.16.4"
-features = ["extension-module"]

synapse_auto_compressor/README.md

@@ -1,12 +0,0 @@
-# Auto Compressor
-
-See the top level readme for information.
-
-
-## Publishing to PyPI
-
-Bump the version number and run from the root directory of the repo:
-
-```
-docker run -it --rm -v $(pwd):/io -e OPENSSL_STATIC=1 konstin2/maturin publish -m synapse_auto_compressor/Cargo.toml --cargo-extra-args "\--features='openssl/vendored'"
-```