WARNING

This page is machine-translated by Gemini.

Continuous Translation

In professional translation, tasks rarely end with a single delivery. You often receive additional files midway through a project or need to reuse previous project data as Translation Memories (TM) or Termbases (TB).

Built on plain text and open JSON standards, SheepWeave excels at continuous translation and agile asset reuse.

---

Adding Source Files Midway

If the client supplies new files (Word, Excel, XLIFF, etc.) during an ongoing project:

1. Place New Files: Copy the new source files into Working/02_SOURCE.
2. Update Project (Flow Tab): In the SheepWeave panel Flow tab under On Working, click 🧶 Add Source Files.
3. Automatic Editor Sync: The new segments are extracted and appended to the end of Target.shwvt without affecting your confirmed translations or project.json.

---

Utilizing Reference Files

Reusing project.json as a TM

Any project.json from completed SheepWeave projects functions directly as a full TM. Copy project.json into the new project's Data/Ref/TM folder. Matches and similarity ratios will be automatically calculated in the Translate tab.

Leveraging Generated Assets

• phrase.jsonl (IntelliSense Phrases): Store frequent expressions in Working/01_REF/phrase.jsonl for instant Ctrl + Space completion.
• auto_replace_log.jsonl (Direct TB): Replacements made with F2 or Ctrl + Shift + Z are automatically logged and recognized as live termbase candidates.

Standard Format Compatibility

Place standard TMX, TBX, Excel (.xlsx), or CSV files into Data/Ref/TM or Data/Ref/TB, then execute 📈 Reanalyze Project Files & Add Reference Files in the Flow tab.

---

Future Roadmap

ManticoreSearch Indexing

Because SheepWeave data is structured JSON, indexing hundreds of thousands of past segments into the lightning-fast ManticoreSearch engine enables sub-millisecond concordance lookups and rich RAG context for LLMs.

Git Version Control

All SheepWeave files (.shwvt, .shwvs, project.json) are 100% plain text. Using Git allows precise diff tracking per line, branch-based collaborative reviews, and effortless cloud synchronization.