Statistics Tracking

[Xendergo]

Closes #50

MVP

• Rewrite my old statistics tracking code to only store data for the current day in memory; replacing the data whenever a new day is observed using atomic trickery (ce14147)
• Output the statistics data into a writer in CSV format (9262c9e)
• Research compression formats that supports being appended to and use one of them to write to a statistics file on disk

Important but maybe a future PR

• Allow people to download the data from the website
• Send a "weekly update" in the webring channel with statistics aggregated over the week

[codecov]

Codecov Report

❌ Patch coverage is 84.12698% with 20 lines in your changes missing coverage. Please review.
✅ Project coverage is 86.23%. Comparing base (01cebcf) to head (93d5271).
⚠️ Report is 1 commits behind head on master.
✅ All tests successful. No failed tests found.

Files with missing lines	Patch %	Lines
src/stats/mod.rs	84.29%	18 Missing and 1 partial ⚠️
src/webring.rs	75.00%	1 Missing ⚠️

Additional details and impacted files

@@            Coverage Diff             @@
##           master      #75      +/-   ##
==========================================
- Coverage   86.69%   86.23%   -0.46%     
==========================================
  Files           8        8              
  Lines        2533     2609      +76     
  Branches     2533     2609      +76     
==========================================
+ Hits         2196     2250      +54     
- Misses        296      317      +21     
- Partials       41       42       +1     

☔ View full report in Codecov by Sentry.
📢 Have feedback on the report? Share it here.

🚀 New features to boost your workflow:

• 📦 JS Bundle Analysis: Save yourself from yourself by tracking and limiting bundle sizes in JS merges.

[github-actions]

Test Results

46 tests   46 ✅  1s ⏱️
 1 suites   0 💤
 1 files     0 ❌

Results for commit 93d5271.

♻️ This comment has been updated with latest results.

[kdkasad]

Why not export metrics in OTel or Prometheus format so we can integrate with dashboards?

[Xendergo]

I don't know anything about either of those; could you point me to what I should read?

[kdkasad]

I don't know anything about either of those; could you point me to what I should read?

Both are systems for collecting/processing metrics from software.

Prometheus: https://prometheus.io/docs/introduction/overview/

OpenTelemetry (intro): https://opentelemetry.io/docs/what-is-opentelemetry/
OpenTelemetry (for devs): https://opentelemetry.io/docs/getting-started/dev/

Originally the plan was to add Prometheus metrics to collect standard performance/behavior metrics about the process, but it sounded like you wanted to collect other statistics (more like user behavior) so I left that up to you.

We did also discuss Prometheus a little in Discord: https://discord.com/channels/772576325897945119/1319140464812753009/1373470039478370385

Basically it boils down to a standard format for exporting metrics so we can use existing software to collect/process/display the data instead of having to re-implement all of that. There are also application-side libraries for Prometheus (and I assume OTel too) that make it easy to calculate and export more complex metrics like histograms of response latencies.

[Xendergo]

Ok sick, I will look into those, though I'm highkey locked in for spill so I may look at that over the winter