refactor: refactor to `spdlog-rs` instead of `tracing` #538

NotEvenANeko · 2025-09-05T07:17:07Z

Refactor tracing to spdlog-rs for logging.

This closes #486 .

eatradish · 2025-09-05T07:21:55Z

Initially it looks good, but what I'm wondering is whether we should just replace the tracing family directly with spdlog instead of optional?

NotEvenANeko · 2025-09-05T07:24:32Z

Yeah we should replace tracing with spdlog, current codes is just convenient for debugging. I'll delete the tracing codes after spdlog parts are good to go.

SpriteOvO

Just a brief look and most of it seems to be fine. I'll probably merge SpriteOvO/spdlog-rs#93 in the next couple days. In the meantime, you can try removing the use of tracing from oma entirely.

SpriteOvO · 2025-09-06T23:06:59Z

src/main.rs

+    let rotating_sink = if let Ok(log_file) = &log_file {
+        Some(
+            AsyncPoolSink::builder()
+                .sink(Arc::new(
+                    RotatingFileSink::builder()
+                        .base_path(&log_file)
+                        .formatter(formatter.clone())
+                        .rotation_policy(RotationPolicy::Hourly)
+                        .build()
+                        .unwrap(),
+                ))
+                .overflow_policy(spdlog::sink::OverflowPolicy::DropIncoming)
+                .build()
+                .unwrap(),
+        )
+    } else {
+        None
+    };


The RotatingFileSink should not rotate hourly as oma did not do so. Could try RotationPolicy::FileSize(u64::MAX) and set max_files(config.save_log_count()).

Fixed, file size is set to 10MB.

SpriteOvO · 2025-09-09T20:25:00Z

src/main.rs

+            .join("oma")
+    };
+
+    let log_file = create_log_file(&log_dir);


Creating variants for filenames has been done internally by spdlog-rs, so oma no longer needs to create names manually.

SpriteOvO · 2025-09-17T04:04:24Z

FYI, SpriteOvO/spdlog-rs#93 is merged.

NotEvenANeko · 2025-09-17T04:52:13Z

I'll work on this PR tomorrow.

NotEvenANeko · 2025-09-19T07:58:43Z

spdlog-rs is set to main-dev branch now.

SpriteOvO · 2025-09-21T16:34:28Z

oma-console/src/print.rs

+///     StdStreamSink::builder().formatter(Box::new(OmaLayer::default())).build().unwrap()
+/// )).build().unwrap();


It is preferred that unwrap() not be used inside an example, and some of the error handling components be hidden if they make the example too difficult to follow.

Quote from https://doc.rust-lang.org/rustdoc/write-documentation/what-to-include.html?highlight=unwrap#examples

SpriteOvO · 2025-09-21T16:39:47Z

oma-console/src/print.rs

+            dest.write_str(&console::strip_ansi_codes(record.payload()))?;
+        }
+
+        dest.write_char('\n')?;


Also, we should write out the key-values of the record here.

Where we place the KV data? Just append it to the end of the log? 🤔

And it looks messy if we output the KV data in stdout.

oma/oma-console/src/print.rs

Lines 210 to 269 in 24424a1

let mut visitor = OmaRecorder(BTreeMap::new());

event.record(&mut visitor);

for (k, v) in visitor.0 {

if k == "message" {

if self.with_ansi {

self.writer.writeln(&prefix, &v).ok();

} else {

self.writer

.writeln(&prefix, &console::strip_ansi_codes(&v))

.ok();

}

}

}

}

}

/// OmaRecorder

/// `OmaRecorder` is used for recording oma-style logs.

///

/// # Example:

/// ```ignore

/// let mut visitor = OmaRecorder(BTreeMap::new());

/// event.record(&mut visitor);

/// for (k, v) in visitor.0 {

/// if k == "message" {

/// self.writer.writeln(&prefix, &v).ok();

/// }

/// }

/// ```

struct OmaRecorder<'a>(BTreeMap<&'a str, String>);

impl tracing::field::Visit for OmaRecorder<'_> {

fn record_f64(&mut self, field: &Field, value: f64) {

self.0.insert(field.name(), value.to_string());

}

fn record_i64(&mut self, field: &Field, value: i64) {

self.0.insert(field.name(), value.to_string());

}

fn record_u64(&mut self, field: &Field, value: u64) {

self.0.insert(field.name(), value.to_string());

}

fn record_bool(&mut self, field: &Field, value: bool) {

self.0.insert(field.name(), value.to_string());

}

fn record_str(&mut self, field: &Field, value: &str) {

self.0.insert(field.name(), value.to_string());

}

fn record_error(&mut self, field: &Field, value: &(dyn std::error::Error + 'static)) {

self.0.insert(field.name(), format!("{value:#?}"));

}

fn record_debug(&mut self, field: &Field, value: &dyn std::fmt::Debug) {

self.0.insert(field.name(), format!("{value:#?}"));

}

}

Since before this PR, oma is already used OmaRecorder to write KV to the end of a log record, I think keeping the behavior should be fine.

Some dep crates may completely use KV to recording variables instead of formatting into the payload.

~~Oops, seems I understand it incorrectly, oma only cares value with "message" as the key.~~

~~My bad, let's just keep the old behavior. You can use KeyValues::get method to get value with "message" as the key.~~

EDIT: NVM, tracing uses "message" KV to store payload, the current code works the same as before already, so looks good to me.

I'll leave the original implementation here, which is not to print KV

The with_kv field is added to formatter but unused if we need to print KV later.

SpriteOvO · 2025-09-21T16:45:06Z

src/main.rs

+// FIXME: `spdlog::error` is conflict with `mod error`
+use spdlog::{
+    Level, LevelFilter, Logger, debug, default_logger, error as error2, info, set_default_logger,


Yes, this is definitely a mistake in spdlog-rs, I haven't figure out the right way to fix it. For now, we can import the error! macro from spdlog::prelude::error.

use spdlog:prelude:error has the same issue

Hmm, so it's not a conflict between spdlog::error mod and spdlog::error! macro? So, if not, then it's probably should not consider as a mistake from spdlog-rs.

Okay, let's just keep the alias here. I'll suggest another name error as err.

SpriteOvO · 2025-09-21T16:48:33Z

src/main.rs

+    .to_string_lossy()
+    .to_string();


Hmm, since spdlog-rs accepts a path type instead of a string type, I think we should place the conversion to the end of the function.

SpriteOvO · 2025-09-21T16:51:25Z

src/main.rs

+        let level_filter = LevelFilter::MoreSevereEqual(Level::Info);

-    if !debug && !dry_run {
-        let no_i18n_embd: EnvFilter = "i18n_embed=off,info".parse().unwrap();


Looks like we have not setup the log crate proxy in this PR. The logs from log crate will only be available after we set it up via spdlog::init_log_crate_proxy().

And we really should migrate env filters, but spdlog-rs currently does not support filter logs from log crate, we will probably support this feature in the next minor release.

For now, is it acceptable for not migrate this at the moment? @eatradish

Looks like we have not setup the log crate proxy in this PR. The logs from log crate will only be available after we set it up via spdlog::init_log_crate_proxy().

And we really should migrate env filters, but spdlog-rs currently does not support filter logs from log crate, we will probably support this feature in the next minor release.

For now, is it acceptable for not migrate this at the moment? @eatradish

Yes, but subsequent releases may include this feature as a breaking

init_log_crate_proxy() added.

Env filter for log crate is added in spdlog-rs v0.5.2 (docs), so filter directive i18n_embed=off can now be migrated! :D

@NotEvenANeko Please also adding this in the PR.

SpriteOvO · 2025-10-02T20:33:37Z

FYI, we have just released spdlog-rs v0.5.0. You could target the dependency spdlog-rs in this PR to the crates.io registry now.

NotEvenANeko · 2025-10-02T21:37:42Z

Updated to v0.5.0

NotEvenANeko · 2025-10-05T17:39:39Z

Doc tests should be fixed now

NotEvenANeko · 2025-10-13T10:38:25Z

Ok I think clippy check is passed now...

src/main.rs

SpriteOvO

Basically LGTM, @eatradish is there anything you are still concerned about?

eatradish · 2025-10-15T10:06:16Z

Basically LGTM, @eatradish is there anything you are still concerned about?

LGTM, I'll test the functionality tomorrow

eatradish · 2025-10-16T08:21:48Z

80col limit feature is broken

SpriteOvO · 2025-11-01T17:23:25Z

oma-console/src/print.rs

-                        .writeln(&prefix, &console::strip_ansi_codes(&v))
-                        .ok();
+                    time
                }


Looks like the 80col limit feature was implemented in oma_console::Writer, the original Writer directly writes to the IO handle stdout or stderr, probably we should implement one that writes to a string buf for spdlog-rs.

@NotEvenANeko Could you make time to have a look at it and try to fix?

Should be fixed now.

NotEvenANeko · 2025-11-07T20:27:58Z

I'm so sorry for AFK for 3 weeks, I'm busy at study recently so I don't have much time for this, really sorry for this :(

I'm continue this PR now.

After reviewing the code I think that if we can migrate console.rs/src/writer.rs to spdlog-rs? Since the Writer directly use console crate to print the log.

We can migrate all the Writer functions and behavior to OmaFormatter and remove Writer.

Will it be a good idea? 🤔

SpriteOvO · 2025-11-08T06:18:28Z

@NotEvenANeko

We can migrate all the Writer functions and behavior to OmaFormatter and remove Writer.

Will it be a good idea? 🤔

I think putting the behavior "writing to console" into a formatter is not an ideal design. Why don't we just replace the field term: Term in Writer with buf: String, then return it from the formatter?

NotEvenANeko · 2025-11-13T23:00:16Z

Not the print to console behavior, I mean that we can move the formatting part into OmaFormatter, and the printing part is obviously moved to StdStreamSink. However we may need information against terminal, I think we can make an OmaStdoutFormatter to handle terminal-related formatting instead of leaving these parts mixed together in Writer.

SpriteOvO · 2025-11-16T15:46:20Z

Not the print to console behavior, I mean that we can move the formatting part into OmaFormatter, and the printing part is obviously moved to StdStreamSink. However we may need information against terminal, I think we can make an OmaStdoutFormatter to handle terminal-related formatting instead of leaving these parts mixed together in Writer.

I think it's acceptable.

NotEvenANeko · 2026-01-12T15:13:22Z

80col limit feature is broken

Should be fixed now. Column wrap logic is extract to oma_console::terminal::wrap_content to reuse.

NotEvenANeko · 2026-01-12T15:45:54Z

It should look like this. I don't know if this error is normal for Debian 12 in Docker.

SpriteOvO

I left some reviews, though I didn't test it locally, maybe @eatradish wanna check the output's appearance.

SpriteOvO · 2026-01-12T15:42:39Z

oma-console/src/print.rs

+                Level::Debug => console::style("DEBUG").dim().to_string(),
+                Level::Info => console::style("INFO").blue().bold().to_string(),
+                Level::Warn => console::style("WARNING").yellow().bold().to_string(),
+                Level::Error => console::style("ERROR").red().bold().to_string(),
+                Level::Trace => console::style("TRACE").dim().to_string(),
+                Level::Critical => console::style("CRITICAL").red().bright().bold().to_string(),


@eatradish Just saying, if you care about performance, here is a good optimization point, by caching these string into a immutable array. However, since the previous also used to_string, so I think this can be done in future PRs, not in this one.

Same thing for the to_string uses below.

SpriteOvO · 2026-01-12T15:55:27Z

src/main.rs

+        let level_filter = LevelFilter::MoreSevereEqual(Level::Info);

-    if !debug && !dry_run {
-        let no_i18n_embd: EnvFilter = "i18n_embed=off,info".parse().unwrap();


Env filter for log crate is added in spdlog-rs v0.5.2 (docs), so filter directive i18n_embed=off can now be migrated! :D

@NotEvenANeko Please also adding this in the PR.

SpriteOvO · 2026-01-12T15:56:18Z

src/main.rs

+        (level_filter, formatter)
    } else {
-        let filter = EnvFilter::try_from_default_env()
-            .unwrap_or_else(|_| "hyper=off,rustls=off,debug".parse().unwrap());


Same thing for hyper=off,rustls=off directives.

Added. First we look for RUST_LOG environment variable as what tracing_subscriber does, if RUST_LOG is empty we use the hyper=off,rustls=off,debug directive for debug or dry run.

For any case besides debug and dry run we use i18n_embed=off,info

SpriteOvO · 2026-01-12T16:00:34Z

src/main.rs

+                // 10 MB
+                .rotation_policy(RotationPolicy::FileSize(10 * 1024 * 1024))


Before this PR, oma does not limit the single log file size, is 10MB for a single log file makes sense to you? @eatradish

If not, we could simply change to u64::MAX here :)

Before this PR, oma does not limit the single log file size, is 10MB for a single log file makes sense to you? @eatradish

If not, we could simply change to u64::MAX here :)

Agree 10MiB

SpriteOvO · 2026-01-12T16:04:13Z

src/main.rs

+                .build()
+                .unwrap(),
+        ))
+        .overflow_policy(spdlog::sink::OverflowPolicy::DropIncoming)


tracing also uses the same policy, quoting from their docs:

The queue has a fixed capacity, and if it becomes full, any logs written to it will be dropped until capacity is once again available.

So from the perspective of migration, this is fine. 👍

SpriteOvO · 2026-01-12T16:08:46Z

src/main.rs

+    logger_builder
+        .level_filter(level_filter)
+        .sink(Arc::new(stream_sink))
+        .sink(Arc::new(rotating_sink));


Since we have wrapped RotatingFileSink with AsyncPoolSink, writing logs to file will not block the current thread, we can set the flush level filter for the logger to All, for flushing the buffer in time.

SpriteOvO · 2026-01-12T16:17:17Z

src/subcommand/search.rs

+                    "{}",
+                    color_formatter().color_str(body.trim(), Action::Secondary)
+                )
+                // Keep original behavior (?


I think this is fine as long as we can display the same results as before.

SpriteOvO · 2026-01-12T16:22:21Z

oma-console/src/print.rs

-/// `OmaLayer` is used for outputting oma-style logs to `tracing`
+
+// '1970-01-01T00:00:00Z' costs 24 chars, and a space costs 1.
+const TIME_RFC3339_LEN: u16 = 24 + 1;


So it should be 20 + 1?

Or we could play some fancy feature here.

Suggested change

const TIME_RFC3339_LEN: u16 = 24 + 1;

const TIME_RFC3339_LEN: usize = "1970-01-01T00:00:00Z ".len();

It is a const variable, so it computes at compile-time.

Oh I forgot to count the milliseconds part...

eatradish · 2026-01-13T08:33:19Z

LGTM output style:

SpriteOvO

LGTM. If @eatradish has no further concerns, I feel this PR can be merged.

EDIT: Oh, looks like CI clippy failed, just one more tiny fix and a rebase needed.

Signed-off-by: Neko <[email protected]>

1

eatradish · 2026-01-13T10:51:55Z

Thanks!

SpriteOvO · 2026-01-13T11:00:17Z

@eatradish Thanks for your reviews and @NotEvenANeko's hard work on this.

If you experience any issues while using spdlog-rs in oma in the future, feel free to CC me and @NotEvenANeko :)

SpriteOvO reviewed Sep 9, 2025

View reviewed changes

NotEvenANeko force-pushed the refactor/spdlog branch from 3ef6995 to 95052c9 Compare September 19, 2025 08:13

SpriteOvO reviewed Sep 21, 2025

View reviewed changes

NotEvenANeko force-pushed the refactor/spdlog branch from 95052c9 to c0266fb Compare October 2, 2025 21:35

NotEvenANeko force-pushed the refactor/spdlog branch 4 times, most recently from 125bbbf to ecaf9a1 Compare October 5, 2025 17:39

NotEvenANeko force-pushed the refactor/spdlog branch from ecaf9a1 to 55e6fc9 Compare October 13, 2025 10:37

eatradish previously requested changes Oct 14, 2025

View reviewed changes

src/main.rs Outdated Show resolved Hide resolved

SpriteOvO approved these changes Oct 15, 2025

View reviewed changes

SpriteOvO reviewed Nov 1, 2025

View reviewed changes

TimXiedada approved these changes Nov 11, 2025

View reviewed changes

NotEvenANeko force-pushed the refactor/spdlog branch 2 times, most recently from 232f031 to 1ca1cb5 Compare January 12, 2026 15:09

SpriteOvO requested changes Jan 12, 2026

View reviewed changes

SpriteOvO approved these changes Jan 13, 2026

View reviewed changes

NotEvenANeko and others added 7 commits January 13, 2026 18:18

refactor: migrate to spdlog

cd1d9d6

Signed-off-by: Neko <[email protected]>

refactor: extract common logic between Writer and Formatter

2220560

fix: fix time string length

d73aaac

fix: count millisecond in rfc3339 format

4a23cc2

refactor: add env filter

928195a

docs: update comment

cf3c0b5

clippy

ba8e4fb

eatradish force-pushed the refactor/spdlog branch from b478406 to ba8e4fb Compare January 13, 2026 10:23

eatradish added 3 commits January 13, 2026 18:24

fmt

09e8339

clippy

6359bf1

--debug / --dry-run do not wrap output

4f96964

eatradish merged commit 61b3f5f into AOSC-Dev:master Jan 13, 2026
3 checks passed

		/// StdStreamSink::builder().formatter(Box::new(OmaLayer::default())).build().unwrap()
		/// )).build().unwrap();

	let mut visitor = OmaRecorder(BTreeMap::new());
	event.record(&mut visitor);

	for (k, v) in visitor.0 {
	if k == "message" {
	if self.with_ansi {
	self.writer.writeln(&prefix, &v).ok();
	} else {
	self.writer
	.writeln(&prefix, &console::strip_ansi_codes(&v))
	.ok();
	}
	}
	}
	}
	}
	/// OmaRecorder
	/// `OmaRecorder` is used for recording oma-style logs.
	///
	/// # Example:
	/// ```ignore
	/// let mut visitor = OmaRecorder(BTreeMap::new());
	/// event.record(&mut visitor);
	/// for (k, v) in visitor.0 {
	/// if k == "message" {
	/// self.writer.writeln(&prefix, &v).ok();
	/// }
	/// }
	/// ```
	struct OmaRecorder<'a>(BTreeMap<&'a str, String>);

	impl tracing::field::Visit for OmaRecorder<'_> {
	fn record_f64(&mut self, field: &Field, value: f64) {
	self.0.insert(field.name(), value.to_string());
	}

	fn record_i64(&mut self, field: &Field, value: i64) {
	self.0.insert(field.name(), value.to_string());
	}

	fn record_u64(&mut self, field: &Field, value: u64) {
	self.0.insert(field.name(), value.to_string());
	}

	fn record_bool(&mut self, field: &Field, value: bool) {
	self.0.insert(field.name(), value.to_string());
	}

	fn record_str(&mut self, field: &Field, value: &str) {
	self.0.insert(field.name(), value.to_string());
	}

	fn record_error(&mut self, field: &Field, value: &(dyn std::error::Error + 'static)) {
	self.0.insert(field.name(), format!("{value:#?}"));
	}

	fn record_debug(&mut self, field: &Field, value: &dyn std::fmt::Debug) {
	self.0.insert(field.name(), format!("{value:#?}"));
	}
	}

		// 10 MB
		.rotation_policy(RotationPolicy::FileSize(10 * 1024 * 1024))

	const TIME_RFC3339_LEN: u16 = 24 + 1;
	const TIME_RFC3339_LEN: usize = "1970-01-01T00:00:00Z ".len();

refactor: refactor to spdlog-rs instead of tracing #538

refactor: refactor to spdlog-rs instead of tracing #538

Uh oh!

Conversation

NotEvenANeko commented Sep 5, 2025 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

eatradish commented Sep 5, 2025 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

NotEvenANeko commented Sep 5, 2025

Uh oh!

SpriteOvO left a comment

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

SpriteOvO commented Sep 17, 2025

Uh oh!

NotEvenANeko commented Sep 17, 2025

Uh oh!

NotEvenANeko commented Sep 19, 2025

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

SpriteOvO Oct 2, 2025 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Choose a reason for hiding this comment

Uh oh!

SpriteOvO Oct 2, 2025 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

SpriteOvO Oct 2, 2025 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

SpriteOvO commented Oct 2, 2025

Uh oh!

NotEvenANeko commented Oct 2, 2025

Uh oh!

NotEvenANeko commented Oct 5, 2025

Uh oh!

NotEvenANeko commented Oct 13, 2025

Uh oh!

Uh oh!

SpriteOvO left a comment

Choose a reason for hiding this comment

Uh oh!

eatradish commented Oct 15, 2025

refactor: refactor to `spdlog-rs` instead of `tracing` #538

refactor: refactor to `spdlog-rs` instead of `tracing` #538

NotEvenANeko commented Sep 5, 2025 •

edited

Loading

eatradish commented Sep 5, 2025 •

edited

Loading

SpriteOvO Oct 2, 2025 •

edited

Loading

SpriteOvO Oct 2, 2025 •

edited

Loading

SpriteOvO Oct 2, 2025 •

edited

Loading

SpriteOvO Nov 1, 2025 •

edited

Loading

SpriteOvO commented Nov 8, 2025 •

edited

Loading

SpriteOvO left a comment •

edited

Loading

SpriteOvO Jan 12, 2026 •

edited

Loading

eatradish Jan 13, 2026 •

edited

Loading

SpriteOvO Jan 12, 2026 •

edited

Loading