KCG Text Splitter User Guide

Quick Start

1. Select your Input File.
2. Select an Output Folder.
3. Choose a Split Method.
4. Set Base File Name and Start Sequence if needed.
5. Click Split.

Progress and messages appear in the status/log area during processing.

Main Split Settings

• Input File: Source text file to split.
• Output Folder: Where chunk files are written.
• Base File Name: Prefix used for each output file.
• Start Sequence: Starting number for chunk numbering.
• Output Naming:
  • Sequence Number After Name -> MyFile - 001.txt
  • Sequence Number Before Name -> 001 - MyFile.txt

Split Methods

1) Split By Size

Splits by target size in kilobytes.

Target Split Size: Enter size in KB.

2) Split By Character Count

Splits by maximum character count per chunk.

Target Character Count: Enter size in number of characters.

3) Split By Paragraph

Splits text by paragraph boundaries.

• Paragraph Breaks:
  • One or more newlines
  • Two or more newlines
• Paragraph Max Size (KB) (optional): Groups paragraphs up to a limit.

4) Split By Keyword

Starts a new chunk when a keyword is found.

• Keyword: Required.
• Keep keyword as first line of next chunk: Keeps delimiter text with the following chunk.
• Keyword Mode:
  • Exact
  • Case Insensitive

5) Split By Chapter

Starts a new chunk when a chapter header is detected.

• Chapter Rule:
  • Lines starting with Chapter
  • Markdown headings (#, ##, ###)
  • Regex pattern
• Keep chapter header in chunk: Keeps heading text at the start of its chunk.

Split Boundary (Used by Size and Character Count)

Boundary mode determines where the split lands near the target limit.

• Nearest Sentence: Tries to split at sentence endings.
• Nearest Paragraph (One or more newlines)
• Nearest Paragraph (Two or more newlines)

If no suitable boundary is found, the split uses the hard limit.

Combine Mode

1. Enable Combine mode.
2. Select Input Folder containing chunk files.
3. Select Output File for the combined text.
4. Choose optional combine settings and click Combine.

• Sort numerically first: Prefer numeric sequence ordering when available.
• Add separator blank line between files: Inserts a blank line between merged files.

Overwrite Handling

If an output file already exists, choose one of these options:

• Overwrite: Replace the existing file.
• Skip: Leave the existing file and continue.
• Cancel: Stop the current job.

Troubleshooting

• No output files created: Verify input path, output folder, and method-specific required fields.
• Unexpected split points: Change the split boundary mode.
• Keyword split not triggering: Check keyword spelling and keyword mode.
• Chapter split not triggering: Confirm the selected chapter rule matches your text format.
• Regex rule errors: Use a valid pattern for chapter detection.

Best Practices

• Use a dedicated output folder for each run.
• Choose a base name that identifies the source text and run purpose.
• For AI prompt chunking, start with Split By Character Count at 7000 and adjust as needed.
• Run a small test first when using keyword/chapter splitting rules.