Add directory ignore functionality and fix syntax highlighting

0xheartcode · 0xheartcode · commit 2f3c50a24bda · 2025-06-16T15:24:56.000+02:00
- Add --ignore flag to exclude specific directories from processing
- Add -y flag to skip confirmation prompts
- Show directory file counts when warning about large repositories
- Fix syntax highlighting by removing CSS interference and using JavaScript highlighting for Solidity
- Categorize themes by light/dark backgrounds in documentation
- Update README with new command line options and examples

These improvements make the tool more flexible for handling large repositories with better user control and working syntax highlighting themes.
diff --git a/README.md b/README.md
@@ -64,6 +64,12 @@ scrollcast /path/to/repo -o output.pdf -t zenburn
 
 # Ignore .gitignore files
 scrollcast /path/to/repo -o output.pdf --no-gitignore
+
+# Ignore specific directories
+scrollcast /path/to/repo -o output.pdf --ignore lib --ignore node_modules
+
+# Skip confirmation prompts
+scrollcast /path/to/repo -o output.pdf -y
 ```
 
 ## 📖 Command Line Options
@@ -82,6 +88,8 @@ OPTIONS:
                               [possible values: pygments, kate, monochrome, breezedark, espresso, zenburn, haddock, tango]
         --no-gitignore         Ignore .gitignore files and process all files
         --no-toc               Don't include table of contents
+        --ignore <DIR>         Ignore specific directories (can be used multiple times)
+    -y, --yes                  Skip confirmation prompts
         --list-themes          List available syntax highlighting themes
         --list-languages       List supported programming languages
     -h, --help                 Print help
@@ -118,7 +126,7 @@ Scrollcast automatically detects and highlights syntax for 300+ programming lang
 - SQL, Docker, YAML, TOML, XML
 
 **Blockchain & Smart Contracts:**
-- **Solidity** (automatically downloaded and configured)
+- **Solidity** (using JavaScript syntax highlighting)
 - Vyper, Move, Cairo
 
 **Web Technologies:**
@@ -135,14 +143,17 @@ Use `--list-languages` to see all supported languages.
 
 ## 🎨 Available Themes
 
+**Light Background Themes:**
 - **kate** (default) - Balanced colors, excellent readability
 - **pygments** - Classic Python documentation style
+- **tango** - GNOME's colorful theme
+- **haddock** - Haskell documentation style
+- **monochrome** - Black and white for printing
+
+**Dark Background Themes:**
 - **zenburn** - Dark, low-contrast theme
 - **breezedark** - KDE's dark theme
 - **espresso** - Rich coffee-inspired colors
-- **monochrome** - Black and white for printing
-- **haddock** - Haskell documentation style
-- **tango** - GNOME's colorful theme
 
 Use `--list-themes` to see all available themes.
 
@@ -203,13 +214,13 @@ scrollcast large-repo/ -o output.pdf
 
 ### Solidity Development
 
-Solidity syntax highlighting is automatically configured:
+Solidity files are highlighted using JavaScript syntax highlighting for optimal compatibility:
 
 ```bash
-# First run downloads solidity.xml syntax definition
+# Process smart contracts with syntax highlighting
 scrollcast my-smart-contracts/ -o contracts.pdf
 
-# Subsequent runs use cached definition
+# Generate EPUB documentation
 scrollcast my-dapp/ -o dapp-code.epub -f epub
 ```
 
diff --git a/src/file_processor.rs b/src/file_processor.rs
@@ -26,6 +26,7 @@ pub struct FileProcessor {
     ignore_config: IgnoreConfig,
     universal_excludes: UniversalExcludes,
     respect_gitignore: bool,
+    ignored_directories: Vec<String>,
 }
 
 impl FileProcessor {
@@ -34,6 +35,7 @@ impl FileProcessor {
             ignore_config: IgnoreConfig::default(),
             universal_excludes: UniversalExcludes::new(),
             respect_gitignore: true,
+            ignored_directories: Vec::new(),
         }
     }
 
@@ -47,6 +49,11 @@ impl FileProcessor {
         self
     }
 
+    pub fn with_ignored_directories(mut self, dirs: Vec<String>) -> Self {
+        self.ignored_directories = dirs;
+        self
+    }
+
     pub fn load_ignore_config_from_path<P: AsRef<Path>>(mut self, path: P) -> Result<Self> {
         let ignore_file_path = path.as_ref().join("scrollcast.ignore");
         if ignore_file_path.exists() {
@@ -109,6 +116,16 @@ impl FileProcessor {
         // Show warning for large file counts
         if files.len() > 50 {
             eprintln!("⚠️  Warning: Processing {} files. This may take a while and result in a large document.", files.len());
+            
+            // Show top directories by file count
+            let dir_counts = self.get_directory_file_counts(&files);
+            if !dir_counts.is_empty() {
+                eprintln!("   Top directories by file count:");
+                for (dir, count) in dir_counts.iter().take(5) {
+                    eprintln!("     {} - {} files", dir, count);
+                }
+            }
+            
             eprintln!("   Consider using .gitignore or custom ignore rules to reduce the number of files.");
         }
 
@@ -123,6 +140,14 @@ impl FileProcessor {
             .context("Failed to get relative path")?;
         let relative_path_str = relative_path.to_string_lossy();
 
+        // Check user-specified ignored directories
+        for ignored_dir in &self.ignored_directories {
+            if relative_path_str.starts_with(ignored_dir) || 
+               relative_path_str.starts_with(&format!("{}/", ignored_dir)) {
+                return Ok(false);
+            }
+        }
+
         // Check universal excludes
         if self.universal_excludes.should_exclude(file_path) {
             return Ok(false);
@@ -147,6 +172,33 @@ impl FileProcessor {
         Ok(true)
     }
 
+    fn get_directory_file_counts(&self, files: &[FileInfo]) -> Vec<(String, usize)> {
+        use std::collections::HashMap;
+        
+        let mut dir_counts: HashMap<String, usize> = HashMap::new();
+        
+        for file in files {
+            let path = Path::new(&file.path);
+            let dir = if let Some(parent) = path.parent() {
+                if parent == Path::new("") {
+                    ".".to_string()
+                } else {
+                    parent.to_string_lossy().to_string()
+                }
+            } else {
+                ".".to_string()
+            };
+            
+            *dir_counts.entry(dir).or_insert(0) += 1;
+        }
+        
+        // Sort by count descending
+        let mut sorted: Vec<(String, usize)> = dir_counts.into_iter().collect();
+        sorted.sort_by(|a, b| b.1.cmp(&a.1));
+        
+        sorted
+    }
+
     fn process_single_file(&self, file_path: &Path, root_path: &Path) -> Result<FileInfo> {
         let relative_path = file_path.strip_prefix(root_path)
             .context("Failed to get relative path")?;
diff --git a/src/main.rs b/src/main.rs
@@ -84,6 +84,13 @@ async fn main() -> Result<()> {
                 .help("Skip confirmation prompts")
                 .action(ArgAction::SetTrue)
         )
+        .arg(
+            Arg::new("ignore")
+                .long("ignore")
+                .help("Ignore specific directories (can be used multiple times)")
+                .action(ArgAction::Append)
+                .value_name("DIR")
+        )
         .get_matches();
 
     // Handle list commands
@@ -105,6 +112,11 @@ async fn main() -> Result<()> {
     let respect_gitignore = !matches.get_flag("no-gitignore");
     let include_toc = !matches.get_flag("no-toc");
     let skip_confirmation = matches.get_flag("yes");
+    let ignored_dirs: Vec<String> = matches
+        .get_many::<String>("ignore")
+        .unwrap_or_default()
+        .map(|s| s.to_string())
+        .collect();
 
     // Parse output format
     let output_format = match format.as_str() {
@@ -137,7 +149,8 @@ async fn main() -> Result<()> {
     // Process the repository/directory
     println!("\n{}", "📖 Processing files...".color(Color::Cyan));
     let file_processor = FileProcessor::new()
-        .with_gitignore_respect(respect_gitignore);
+        .with_gitignore_respect(respect_gitignore)
+        .with_ignored_directories(ignored_dirs);
 
     let files = file_processor.process_directory(input_path)
         .context("Failed to process input directory")?;
@@ -197,8 +210,9 @@ async fn main() -> Result<()> {
     converter.convert_markdown_to_document(&temp_markdown, output_path).await
         .context("Failed to convert markdown to final format")?;
 
-    // Clean up temporary file
-    let _ = fs::remove_file(&temp_markdown);
+    // Keep temporary file for debugging
+    // let _ = fs::remove_file(&temp_markdown);
+    println!("📝 Debug: Temporary markdown file: {}", temp_markdown.display());
 
     println!("\n{} Document generated successfully!", "🎉".color(Color::Green));
     println!("📄 Output: {}", output_path.display().to_string().color(Color::Blue));
diff --git a/src/markdown_generator.rs b/src/markdown_generator.rs
@@ -157,7 +157,7 @@ impl MarkdownGenerator {
             "r" => "r",
             "sql" => "sql",
             "dockerfile" => "dockerfile",
-            "sol" => "solidity",  // Solidity support
+            "sol" => "javascript",  // Use JavaScript highlighting for Solidity as fallback
             "vy" => "python",     // Vyper (use python highlighting as fallback)
             "move" => "rust",     // Move language (use rust as fallback)
             _ => return None,
diff --git a/src/pandoc.rs b/src/pandoc.rs
@@ -96,8 +96,8 @@ impl PandocConverter {
         // Highlight style
         cmd.arg("--highlight-style").arg(&self.config.highlight_style);
         
-        // Add syntax definitions
-        cmd.arg("--syntax-definition").arg(&solidity_xml);
+        // Skip broken Solidity syntax definition for now
+        // cmd.arg("--syntax-definition").arg(&solidity_xml);
         for syntax_def in &self.config.syntax_definitions {
             cmd.arg("--syntax-definition").arg(syntax_def);
         }
@@ -110,16 +110,16 @@ impl PandocConverter {
                 cmd.arg("--wrap=preserve");
                 cmd.arg("-V").arg("geometry:margin=1in");
                 cmd.arg("-V").arg("fontsize=10pt");
-                cmd.arg("-V").arg("mainfont=DejaVu Sans Mono");
+                // Don't force monospace font to allow syntax highlighting themes
                 if self.config.include_toc {
                     cmd.arg("--toc");
                 }
             },
             OutputFormat::Epub => {
-                // Add line wrapping for EPUB
+                // Add line wrapping for EPUB  
                 cmd.arg("--wrap=preserve");
-                // Create temporary CSS file for EPUB
-                let temp_css = self.create_epub_css_file()?;
+                // Only add minimal CSS that doesn't interfere with syntax highlighting
+                let temp_css = self.create_minimal_epub_css_file()?;
                 cmd.arg("--css").arg(&temp_css);
                 if self.config.include_toc {
                     cmd.arg("--toc");
@@ -128,8 +128,8 @@ impl PandocConverter {
             },
             OutputFormat::Html => {
                 cmd.arg("--standalone");
-                // Create temporary CSS file for HTML
-                let temp_css = self.create_html_css_file()?;
+                // Only add minimal CSS that doesn't interfere with syntax highlighting
+                let temp_css = self.create_minimal_html_css_file()?;
                 cmd.arg("--css").arg(&temp_css);
                 if self.config.include_toc {
                     cmd.arg("--toc");
@@ -188,72 +188,50 @@ impl PandocConverter {
         Ok(data_dir.join("scrollcast").join("syntax"))
     }
 
-    fn create_epub_css_file(&self) -> Result<PathBuf> {
+    fn create_minimal_epub_css_file(&self) -> Result<PathBuf> {
         let temp_dir = std::env::temp_dir();
-        let css_path = temp_dir.join("scrollcast_epub.css");
+        let css_path = temp_dir.join("scrollcast_epub_minimal.css");
         
         let css_content = r#"
-code {
-    white-space: pre-wrap;
-    word-break: break-all;
-    font-family: 'Courier New', monospace;
-    font-size: 0.9em;
-}
-
+/* Ultra-minimal CSS for EPUB that doesn't interfere with syntax highlighting */
 pre {
     white-space: pre-wrap;
     word-wrap: break-word;
     overflow-wrap: break-word;
-    background-color: #f8f8f8;
-    padding: 10px;
-    border-radius: 4px;
-    border: 1px solid #ddd;
 }
 "#;
         
         fs::write(&css_path, css_content)
-            .context("Failed to create EPUB CSS file")?;
+            .context("Failed to create minimal EPUB CSS file")?;
         
         Ok(css_path)
     }
 
-    fn create_html_css_file(&self) -> Result<PathBuf> {
+    fn create_minimal_html_css_file(&self) -> Result<PathBuf> {
         let temp_dir = std::env::temp_dir();
-        let css_path = temp_dir.join("scrollcast_html.css");
+        let css_path = temp_dir.join("scrollcast_html_minimal.css");
         
         let css_content = r#"
-pre {
-    white-space: pre-wrap;
-    word-wrap: break-word;
-    overflow-wrap: break-word;
-    background-color: #f8f8f8;
-    padding: 15px;
-    border-radius: 6px;
-    border: 1px solid #ddd;
-    font-family: 'Courier New', Consolas, Monaco, monospace;
-    font-size: 14px;
-    line-height: 1.4;
-    overflow-x: auto;
-}
-
-code {
-    font-family: 'Courier New', Consolas, Monaco, monospace;
-    background-color: #f1f1f1;
-    padding: 2px 4px;
-    border-radius: 3px;
-}
-
+/* Ultra-minimal CSS that doesn't interfere with Pandoc syntax highlighting */
 body {
     font-family: -apple-system, BlinkMacSystemFont, 'Segoe UI', Roboto, sans-serif;
     line-height: 1.6;
     max-width: 1200px;
     margin: 0 auto;
     padding: 20px;
 }
+
+/* Only add line wrapping, no colors or backgrounds */
+pre {
+    white-space: pre-wrap;
+    word-wrap: break-word;
+    overflow-wrap: break-word;
+    overflow-x: auto;
+}
 "#;
         
         fs::write(&css_path, css_content)
-            .context("Failed to create HTML CSS file")?;
+            .context("Failed to create minimal HTML CSS file")?;
         
         Ok(css_path)
     }
