Advanced Parsing

Advanced Parsing & Text Extraction

RenLocalizer uses a sophisticated multi-stage pipeline to find and extract text from Ren'Py games without breaking the underlying code.

1. Traditional Regex Parsing

The first layer of scanning uses highly optimized Regular Expressions to find standard Ren'Py dialogue and UI strings:

• Dialogue: character_name "Dialogue text"
• Direct strings: _("Text") or "Text"
• Menu items: menu: blocks

2. AST (Abstract Syntax Tree) Scanning

When simple regex isn't enough, RenLocalizer uses Python's ast module to analyze scripts. This allows us to:

• Find strings inside complex nested functions.
• Identify translatable text in init python blocks.
• Smartly distinguish between technical code and game content.

3. RPYC & RPYMC Readers

Ren'Py compiles human-readable .rpy files into binary .rpyc files. Many "obfuscated" games hide their source code.

• RPYC Reader: RenLocalizer can "unpickle" binary RPYC files to extract the original Abstract Syntax Tree. This allows translation even when the source .rpy file is missing.
• RPYMC Reader: Specifically handles screen cache files, ensuring even complex UI elements are localized.

4. Deep Scan Technology

Enable Deep Scan in settings to trigger a recursive AST analysis of the entire project.

• What it finds: Variable assignments, dictionary keys, and list items that are used as game text but don't follow standard _() localization markers.
• Safety: Deep Scan uses a heuristic "Technical String Filter" to ensure it doesn't accidentally translate internal Ren'Py paths or variable names (e.g., it will translate "Health" but skip "renpy.dissolve").

5. Normalization & Encoding

RenLocalizer automatically detects and normalizes file encodings to UTF-8 with BOM. This prevents "Mojibake" (broken characters) in languages like Russian, Chinese, or Japanese.