SPREADSHEET.PY(1)

📦
version 2.6 · 2 versions
, in arch rolling

✨ NAME ✨

spreadsheet.py — like, the spreadsheet documentation bestie 💅

Manual section: 1 Manual group: LCOV Utilities

💖 NAME 💖

spreadsheet.py — takes your LCOV profile data and turns it into an Excel spreadsheet so you can actually look at your performance analysis and compare runs. It's giving data-science-sleepover and honestly? iconic. 🎀

🦋 SYNOPSIS 🦋

spreadsheet.py [-o output.xlsx] [options] data.json [data2.json ...]

🌸 DESCRIPTION 🌸

okay so spreadsheet.py is a utility script that converts JSON profile data from genhtml, geninfo, and lcov into an Excel spreadsheet, because reading raw JSON is not the vibe and a spreadsheet is. ✨ It takes performance timing data and lays it all out in a table with statistical analysis and conditional formatting — very main character of it.

The spreadsheet includes:

  • Summary sheets so you can compare multiple runs side by side 🫶
  • Per-file timing data for the detailed-analysis girlies 🔍
  • Statistical summaries (total, average, standard deviation) — the math is mathing 📊
  • Conditional formatting to highlight outliers 👑

🌈 COLOR CODING 🌈

The spreadsheet uses conditional formatting to highlight timing anomalies, and the colors are literally telling you something:

  • Yellow: values between 1.5 and 2.0 standard deviations larger than average (and more than 15% above average)
  • Red: values more than 2.0 standard deviations larger than average (and more than 15% above average)
  • Green: values more than 2.0 standard deviations smaller than average (significantly better performance — slay 💅)

🧚 SUPPORTED TOOLS 🧚

The script processes profile data from:

  • geninfo: chunk timing, file processing, filter operations
  • genhtml: source parsing, HTML generation, annotation, categorization
  • lcov: tracefile merging, parsing, segment processing

✨ OPTIONS ✨

  • -o file, --out file — save the Excel output to the file you specify. Default: stats.xlsx. 💾
  • --threshold percent — the minimum percentage difference from average needed to trigger colorization. Differences smaller than this threshold don't get highlighted, no cap. Default: 15.0%.
  • --low multiplier — the standard deviation multiplier for yellow highlighting. Values between --low and --high standard deviations above average get colored yellow. Default: 1.5. 💛
  • --high multiplier — the standard deviation multiplier for red highlighting. Values more than --high standard deviations above average get colored red. Default: 2.0. ❤️‍🔥
  • -v, --verbose — increases the verbosity of the report. Includes additional timing data such as read and translate operations. 🗣️
  • --show-filter — includes filter operation timing data in the spreadsheet. The filter data shows time spent in filter chunk processing, queue operations, and merging. 🧵
  • files — one or more JSON profile data files to process. These should be generated using the --profile option of geninfo, genhtml, or lcov. (iykyk 🫶)

🎀 EXAMPLES 🎀

Basic usage with a single profile file:

$ spreadsheet.py -o report.xlsx geninfo_profile.json

Compare multiple profile runs — the group-chat comparison era:

$ spreadsheet.py -o comparison.xlsx run1.json run2.json run3.json

Include filter timing data with verbose output:

$ spreadsheet.py --show-filter -v -o detailed.xlsx profile.json

Adjust sensitivity for outlier detection:

$ spreadsheet.py --threshold 10 --low 1.0 --high 1.5 -o sensitive.xlsx data.json

⭐ GENERATING PROFILE DATA ⭐

To generate profile data for analysis, use the --profile option:

$ geninfo --profile geninfo_profile.json -o coverage.info ./build
$ genhtml --profile genhtml_profile.json -o html coverage.info
$ lcov --profile lcov_profile.json -a a.info -a b.info -o merged.info

👑 AUTHOR 👑

Henry Cox <<henry.cox@mediatek.com>>

💜 SEE ALSO 💜

Copyright: 2024-2026, LCOV Project

and that's the whole page bestie 🫶 this was version 2.6, dated 2026-10-03, and it's giving LCOV energy from top to bottom. go make some spreadsheets, you've got this 💖✨🎀

📦 where this page comes from