Skip to content

data(scenarii): gpt-5.5 ES/AR/FA/ZH translations (5344 cells, 100%) - #365

Merged
jsboige merged 1 commit into
masterfrom
data/scenarii-es-gpt55
May 28, 2026
Merged

data(scenarii): gpt-5.5 ES/AR/FA/ZH translations (5344 cells, 100%)#365
jsboige merged 1 commit into
masterfrom
data/scenarii-es-gpt55

Conversation

@jsboige

@jsboige jsboige commented May 28, 2026

Copy link
Copy Markdown
Contributor

Summary

  • Scenarii dataset fully translated to 4 new languages via DatasetUpdater with gpt-5.5 function calling
  • ES: 1336/1336 cells (100%) — Spanish
  • AR: 1336/1336 cells (100%) — Arabic
  • FA: 1336/1336 cells (100%) — Persian/Farsi
  • ZH: 1336/1336 cells (100%) — Chinese Simplified
  • 167 records × 8 fields per language = 5344 total cells filled
  • Adds 8 prompt files (user/assistant per language) + DatasetUpdater task configs
  • TargetPath fix: .\Target\Datasets\ instead of ..\..\Cards\

Config changes

  • 4 new DatasetUpdater task configs for Scenarii ES/AR/FA/ZH
  • All tasks Enabled = false (ready for review, not active)
  • Mode reverted to WebBasedImageGeneration | QuestPdfGeneration

Test plan

  • Verify CSV cell counts: ES/AR/FA/ZH each 1336/1336
  • Spot-check translations quality (2-3 records per language)
  • Build passes: dotnet build with 0 errors

🤖 Generated with Claude Code

…ls, 100%)

Scenarii dataset fully translated to 4 new languages via DatasetUpdater
with gpt-5.5 function calling (167 records × 8 fields × 4 languages).

- ES: 1336/1336 cells (100%) — Spanish
- AR: 1336/1336 cells (100%) — Arabic
- FA: 1336/1336 cells (100%) — Persian/Farsi
- ZH: 1336/1336 cells (100%) — Chinese Simplified

Adds 8 prompt files (user/assistant per language) and DatasetUpdater task
configs with TargetPath fix (.\Target\Datasets\ instead of ..\..\Cards\).

All tasks disabled, Mode reverted to pipeline defaults.

Co-Authored-By: Claude Opus 4.7 <noreply@anthropic.com>
@jsboige
jsboige merged commit f7eb226 into master May 28, 2026
3 checks passed

@clusterManager-Myia clusterManager-Myia left a comment

Copy link
Copy Markdown
Collaborator

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

[Hermes] — APPROVED

PR: data(scenarii): gpt-5.5 ES/AR/FA/ZH translations (5344 cells, 100%)
SHA: 686d5fe

Checklist

Check Status
Security scan ✅ CLEAN (0 real credential matches — "secret" appears in fallacy scenario content, not credentials)
Cross-repo impact ✅ N/A (Argumentum repo, data-only)
CI ✅ GitGuardian pass + Debug/Release builds pass
Second reviewer ✅ Non requis (data + config + prompts, no logic)

Verification

  • 8 prompt templates (Es/Ar/Fa/Zh × Assistant/User) — consistent pattern: translate _fr → target language via UpdateRecord parallel calls. Same template structure as existing RU/PT translations.
  • C# DatasetUpdaterRootConfig: +1709/-1449 — registration of 4 new translation resource files + string entries. No executable logic change.
  • CSV Cards: +168/-168 — data alignment for new language columns.
  • GitGuardian security scan passed.

Verdict

Data-only translation PR following established repo pattern. No logic, no credentials, CI green. Safe to merge.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

2 participants