"hq read" renamed to "hq output-log"

CHANGELOG.md

@@ -16,7 +16,7 @@
 ## Removed
 
 * Because worker streaming fully replaces original streaming, the original server streaming was removed.
-  For most cases, you can rename `--log` to `--stream` and `hq log` to `hq read`. See the docs for more details.
+  For most cases, you can rename `--log` to `--stream` and `hq log` to `hq output-log`. See the docs for more details.
 
 ## Fixes
 
@@ -373,7 +373,7 @@ an old client/worker to a new server (Connecting a new client/worker to an old s
 
 ### Stream log
 
-* Basic export of stream log into JSON (`hq read <log_file> export`)
+* Basic export of stream log into JSON (`hq output-log <log_file> export`)
 
 ### Server
 

crates/hyperqueue/src/bin/hq.rs

@@ -14,7 +14,7 @@ use hyperqueue::client::commands::job::{
     output_job_summary, JobCancelOpts, JobCatOpts, JobCloseOpts, JobForgetOpts, JobInfoOpts,
     JobListOpts, JobTaskIdsOpts,
 };
-use hyperqueue::client::commands::reader::command_reader;
+use hyperqueue::client::commands::outputlog::command_reader;
 use hyperqueue::client::commands::server::command_server;
 use hyperqueue::client::commands::submit::command::{open_job, SubmitJobConfOpts};
 use hyperqueue::client::commands::submit::{
@@ -474,7 +474,7 @@ async fn main() -> hyperqueue::Result<()> {
         }) => command_task_info(&gsettings, opts).await,
         #[cfg(feature = "dashboard")]
         SubCommand::Dashboard(opts) => command_dashboard_start(&gsettings, opts).await,
-        SubCommand::Read(opts) => command_reader(&gsettings, opts),
+        SubCommand::OutputLog(opts) => command_reader(&gsettings, opts),
         SubCommand::AutoAlloc(opts) => command_autoalloc(&gsettings, opts).await,
         SubCommand::EventLog(opts) => command_event_log(&gsettings, opts).await,
         SubCommand::GenerateCompletion(opts) => generate_completion(opts),

crates/hyperqueue/src/client/commands/mod.rs

@@ -1,7 +1,7 @@
 pub mod autoalloc;
 pub mod event;
 pub mod job;
-pub mod reader;
+pub mod outputlog;
 pub mod server;
 pub mod submit;
 pub mod wait;

crates/hyperqueue/src/client/commands/reader.rs → crates/hyperqueue/src/client/commands/outputlog.rs

@@ -6,7 +6,7 @@ use clap::Parser;
 use std::path::PathBuf;
 
 #[derive(Parser)]
-pub struct ReadOpts {
+pub struct OutputLogOpts {
     /// Path of log file
     path: PathBuf,
 
@@ -85,7 +85,7 @@ pub enum Channel {
     Stderr,
 }
 
-pub fn command_reader(gsettings: &GlobalSettings, opts: ReadOpts) -> anyhow::Result<()> {
+pub fn command_reader(gsettings: &GlobalSettings, opts: OutputLogOpts) -> anyhow::Result<()> {
     let mut stream_dir = StreamDir::open(&opts.path, opts.server_uid.as_deref())?;
     match opts.command {
         StreamCommand::Summary(_) => {

crates/hyperqueue/src/common/cli.rs

@@ -12,7 +12,7 @@ use crate::client::commands::job::{
     JobCancelOpts, JobCatOpts, JobCloseOpts, JobForgetOpts, JobInfoOpts, JobListOpts,
     JobTaskIdsOpts,
 };
-use crate::client::commands::reader::ReadOpts;
+use crate::client::commands::outputlog::OutputLogOpts;
 use crate::client::commands::server::ServerOpts;
 use crate::client::commands::submit::command::SubmitJobConfOpts;
 use crate::client::commands::submit::{JobSubmitFileOpts, JobSubmitOpts};
@@ -199,7 +199,7 @@ pub enum SubCommand {
     /// Commands for controlling HyperQueue workers
     Worker(WorkerOpts),
     /// Operations with log
-    Read(ReadOpts),
+    OutputLog(OutputLogOpts),
     /// Automatic allocation management
     #[command(name = "alloc")]
     AutoAlloc(AutoAllocOpts),

crates/hyperqueue/src/stream/reader/streamdir.rs

@@ -1,4 +1,4 @@
-use crate::client::commands::reader::{CatOpts, Channel, ExportOpts, ShowOpts};
+use crate::client::commands::outputlog::{CatOpts, Channel, ExportOpts, ShowOpts};
 use crate::common::arraydef::IntArray;
 use crate::common::error::HqError;
 use crate::server::event::bincode_config;

docs/jobs/streaming.md

@@ -77,23 +77,23 @@ HyperQueue lets you inspect the data stored inside the stream file using various
 the following structure:
 
 ```bash
-$ hq read <stream-path> <subcommand> <subcommand-args>
+$ hq output-log <stream-path> <subcommand> <subcommand-args>
 ```
 
 ### Stream summary
 
 You can display a summary of a log file using the `summary` subcommand:
 
 ```bash
-$ hq read <stream-path> summary
+$ hq output-log <stream-path> summary
 ```
 
 ### Stream jobs
 
 To print all job IDs that streaming in the stream path, you can run the following command:
 
 ```bash
-$ hq read <stream-path> jobs
+$ hq output-log <stream-path> jobs
 ```
 
 ### Printing stream content
@@ -102,7 +102,7 @@ If you want to simply print the (textual) content of the log file, without any a
 `cat` subcommand:
 
 ```bash
-$ hq read <stream-path> cat <job-id> <stdout/stderr>
+$ hq output-log <stream-path> cat <job-id> <stdout/stderr>
 ```
 
 It will print the raw content of either `stdout` or `stderr`, ordered by task id. All outputs will be concatenated one
@@ -119,7 +119,7 @@ If you want to inspect the contents of the log, along with its inner metadata th
 has produced which part of the data, you can use the `show` subcommand:
 
 ```commandline
-$ hq read <log-file-path> show
+$ hq output-log <log-file-path> show
 ```
 
 The output will have the form `J.T:C> DATA` where `J` is a job id, `T` is a task id and `C` is `0` for `stdout` channel
@@ -132,7 +132,7 @@ You can filter a specific channel with the `--channel=stdout/stderr` flag.
 Log can be exported into JSON by the following command:
 
 ```commandline
-$ hq read <log-file-path> export
+$ hq output-log <log-file-path> export
 ```
 
 This prints the log file into a JSON format on standard output.
@@ -145,14 +145,14 @@ hence HyperQueue streaming is able to avoid mixing
 outputs from different executions of the same task, when a task is restarted.
 
 HyperQueue automatically marks all output from previous instance of a task except the last instance as *superseded*.
-You can see statistics about superseded data via `hq read <stream-path> summary` command.
+You can see statistics about superseded data via `hq output-log <stream-path> summary` command.
 In the current version, superseded data is ignored by all other commands.
 
 ## More server instances
 
 HyperQueue supports writing streams from the different server instances into the same directory.
-If you run `hq read` commands over such directory then it will detect the situation and prints all server uids
-that writes into the directory. You have to specify the server instance via `hq read --server-uid=<SERVER_UID> ...`
+If you run `hq output-log` commands over such directory then it will detect the situation and prints all server uids
+that writes into the directory. You have to specify the server instance via `hq output-log --server-uid=<SERVER_UID> ...`
 when working with such a stream directory.
 
 !!! note
@@ -163,4 +163,4 @@ when working with such a stream directory.
 ## Working with non-shared file system
 
 You do not need to have a shared file system when working with streaming path. It is just your responsibility to
-collect all generated files into one directory before using `hq read` commands.
+collect all generated files into one directory before using `hq output-log` commands.

tests/test_placeholders.py

@@ -138,7 +138,7 @@ def test_stream_submit_placeholder(hq_env: HqEnv):
     hq_env.start_workers(1)
     wait_for_job_state(hq_env, 1, "FINISHED")
 
-    lines = set(hq_env.command(["read", "log-1", "show"], as_lines=True))
+    lines = set(hq_env.command(["output-log", "log-1", "show"], as_lines=True))
     assert "1.0:0> Hello" in lines