html-to-markdown 3.4.0 → 3.6.0.pre.rc.23

data/vendor/html-to-markdown-rs/src/metadata/extraction.rs

@@ -1,398 +0,0 @@
-//! Metadata extraction utilities and helpers.
-
-use super::types::{DocumentMetadata, StructuredData, StructuredDataType, TextDirection};
-use std::collections::BTreeMap;
-
-/// Extract document metadata from collected head metadata.
-///
-/// Parses head metadata into structured document metadata,
-/// handling special cases like Open Graph, Twitter Card, keywords, etc.
-pub(crate) fn extract_document_metadata(
-    head_metadata: BTreeMap<String, String>,
-    lang: Option<String>,
-    dir: Option<String>,
-) -> DocumentMetadata {
-    let mut doc = DocumentMetadata::default();
-
-    for (raw_key, value) in head_metadata {
-        let mut key = raw_key.as_str();
-        let mut replaced_key: Option<String> = None;
-
-        if let Some(stripped) = key.strip_prefix("meta-") {
-            key = stripped;
-        }
-
-        if key.as_bytes().contains(&b':') {
-            replaced_key = Some(key.replace(':', "-"));
-            key = replaced_key.as_deref().unwrap_or(key);
-        }
-
-        // Normalize to lowercase for case-insensitive matching (per HTML spec,
-        // meta name attributes are compared ASCII case-insensitively).
-        let lower_key = key.to_ascii_lowercase();
-
-        match lower_key.as_str() {
-            "title" => doc.title = Some(value),
-            "description" => doc.description = Some(value),
-            "author" | "creator" | "publisher" => {
-                if doc.author.is_none() {
-                    doc.author = Some(value);
-                }
-            }
-            "canonical" => doc.canonical_url = Some(value),
-            "base" | "base-href" => doc.base_href = Some(value),
-            k if k.starts_with("og-") => {
-                let og_key = k.trim_start_matches("og-").replace('-', "_");
-                doc.open_graph.insert(og_key, value);
-            }
-            k if k.starts_with("twitter-") => {
-                let tw_key = k.trim_start_matches("twitter-").replace('-', "_");
-                doc.twitter_card.insert(tw_key, value);
-            }
-            // Dublin Core: DC.* and DCTERMS.* prefixes (dot becomes part of key after meta- strip).
-            // Map DC/DCTERMS fields to dedicated struct fields where applicable.
-            k if k.starts_with("dc.") || k.starts_with("dc-") => {
-                let dc_field = k.trim_start_matches("dc.").trim_start_matches("dc-");
-                match dc_field {
-                    "title" => {
-                        if doc.title.is_none() {
-                            doc.title = Some(value);
-                        }
-                    }
-                    "description" => {
-                        if doc.description.is_none() {
-                            doc.description = Some(value);
-                        }
-                    }
-                    "creator" | "contributor" | "publisher" => {
-                        if doc.author.is_none() {
-                            doc.author = Some(value);
-                        }
-                    }
-                    "subject" | "keywords" => {
-                        if doc.keywords.is_empty() {
-                            doc.keywords = split_keywords(&value);
-                        }
-                    }
-                    _ => {
-                        let meta_key = format!("dc_{}", dc_field.replace('-', "_"));
-                        doc.meta_tags.insert(meta_key, value);
-                    }
-                }
-            }
-            k if k.starts_with("dcterms.") || k.starts_with("dcterms-") => {
-                let dc_field = k.trim_start_matches("dcterms.").trim_start_matches("dcterms-");
-                match dc_field {
-                    "title" | "alternative" => {
-                        if doc.title.is_none() {
-                            doc.title = Some(value);
-                        }
-                    }
-                    "description" | "abstract" => {
-                        if doc.description.is_none() {
-                            doc.description = Some(value);
-                        }
-                    }
-                    "creator" | "contributor" | "publisher" => {
-                        if doc.author.is_none() {
-                            doc.author = Some(value);
-                        }
-                    }
-                    "subject" | "keywords" => {
-                        if doc.keywords.is_empty() {
-                            doc.keywords = split_keywords(&value);
-                        }
-                    }
-                    _ => {
-                        let meta_key = format!("dcterms_{}", dc_field.replace('-', "_"));
-                        doc.meta_tags.insert(meta_key, value);
-                    }
-                }
-            }
-            // All keyword-bearing meta tag variants
-            "keywords" | "news_keywords" | "citation_keywords" | "subject" | "topic" | "category"
-            | "classification" => {
-                if doc.keywords.is_empty() {
-                    doc.keywords = split_keywords(&value);
-                }
-            }
-            _ => {
-                let meta_key = if key.as_ptr() == raw_key.as_ptr() && key.len() == raw_key.len() {
-                    raw_key
-                } else if let Some(replaced) = replaced_key {
-                    replaced
-                } else {
-                    key.to_string()
-                };
-                doc.meta_tags.insert(meta_key, value);
-            }
-        }
-    }
-
-    if let Some(lang) = lang {
-        doc.language = Some(lang);
-    }
-
-    if let Some(dir) = dir {
-        if let Some(parsed_dir) = TextDirection::parse(&dir) {
-            doc.text_direction = Some(parsed_dir);
-        }
-    }
-
-    doc
-}
-
-/// Split a comma-separated keywords string into a `Vec<String>`.
-fn split_keywords(value: &str) -> Vec<String> {
-    value
-        .split(',')
-        .map(|s| s.trim().to_string())
-        .filter(|s| !s.is_empty())
-        .collect()
-}
-
-/// Extract structured data blocks into `StructuredData` items.
-pub(crate) fn extract_structured_data(json_ld: Vec<String>) -> Vec<StructuredData> {
-    let mut result = Vec::with_capacity(json_ld.len());
-
-    for json_str in json_ld {
-        let schema_type = scan_schema_type(&json_str)
-            .or_else(|| {
-                if json_str.contains("\"@type\"") {
-                    serde_json::from_str::<serde_json::Value>(&json_str).ok().and_then(|v| {
-                        v.get("@type")
-                            .and_then(|t| t.as_str().map(std::string::ToString::to_string))
-                    })
-                } else {
-                    None
-                }
-            })
-            .or_else(|| {
-                if !json_str.contains("\"@graph\"") {
-                    return None;
-                }
-
-                let value = serde_json::from_str::<serde_json::Value>(&json_str).ok()?;
-                let graph = value.get("@graph")?;
-                let items = graph.as_array()?;
-                items.iter().find_map(|item| {
-                    item.get("@type")
-                        .and_then(|t| t.as_str().map(std::string::ToString::to_string))
-                })
-            });
-
-        result.push(StructuredData {
-            data_type: StructuredDataType::JsonLd,
-            raw_json: json_str,
-            schema_type,
-        });
-    }
-
-    result
-}
-
-/// Scan for @type in JSON string without full parsing.
-fn scan_schema_type(json_str: &str) -> Option<String> {
-    let needle = "\"@type\"";
-    let start = json_str.find(needle)? + needle.len();
-    let bytes = json_str.as_bytes();
-    let mut i = start;
-
-    while i < bytes.len() && bytes[i].is_ascii_whitespace() {
-        i += 1;
-    }
-    if i >= bytes.len() || bytes[i] != b':' {
-        return None;
-    }
-    i += 1;
-    while i < bytes.len() && bytes[i].is_ascii_whitespace() {
-        i += 1;
-    }
-    if i >= bytes.len() {
-        return None;
-    }
-
-    if bytes[i] == b'[' {
-        i += 1;
-        while i < bytes.len() && bytes[i].is_ascii_whitespace() {
-            i += 1;
-        }
-        if i >= bytes.len() || bytes[i] != b'"' {
-            return None;
-        }
-    } else if bytes[i] != b'"' {
-        return None;
-    }
-
-    let start_quote = i;
-    i += 1;
-    let mut escaped = false;
-    while i < bytes.len() {
-        let byte = bytes[i];
-        if escaped {
-            escaped = false;
-            i += 1;
-            continue;
-        }
-        if byte == b'\\' {
-            escaped = true;
-            i += 1;
-            continue;
-        }
-        if byte == b'"' {
-            let end_quote = i;
-            let slice = &json_str[start_quote..=end_quote];
-            return serde_json::from_str::<String>(slice).ok();
-        }
-        i += 1;
-    }
-
-    None
-}
-
-#[cfg(test)]
-mod tests {
-    use super::*;
-
-    #[test]
-    fn test_extract_document_metadata() {
-        let mut head_metadata = BTreeMap::new();
-        head_metadata.insert("title".to_string(), "Test Title".to_string());
-        head_metadata.insert("description".to_string(), "Test Description".to_string());
-        head_metadata.insert("keywords".to_string(), "rust, testing".to_string());
-
-        let doc = extract_document_metadata(head_metadata, Some("en".to_string()), Some("ltr".to_string()));
-
-        assert_eq!(doc.title, Some("Test Title".to_string()));
-        assert_eq!(doc.description, Some("Test Description".to_string()));
-        assert_eq!(doc.keywords, vec!["rust", "testing"]);
-        assert_eq!(doc.language, Some("en".to_string()));
-    }
-
-    #[test]
-    fn test_extract_document_metadata_with_og() {
-        let mut head_metadata = BTreeMap::new();
-        head_metadata.insert("og-title".to_string(), "OG Title".to_string());
-        head_metadata.insert("og-description".to_string(), "OG Description".to_string());
-
-        let doc = extract_document_metadata(head_metadata, None, None);
-
-        assert_eq!(doc.open_graph.get("title"), Some(&"OG Title".to_string()));
-        assert_eq!(doc.open_graph.get("description"), Some(&"OG Description".to_string()));
-    }
-
-    #[test]
-    fn test_keywords_case_insensitive() {
-        let mut head_metadata = BTreeMap::new();
-        head_metadata.insert("meta-Keywords".to_string(), "Rust, HTML, Markdown".to_string());
-
-        let doc = extract_document_metadata(head_metadata, None, None);
-        assert_eq!(doc.keywords, vec!["Rust", "HTML", "Markdown"]);
-    }
-
-    #[test]
-    fn test_keywords_dc_subject() {
-        let mut head_metadata = BTreeMap::new();
-        head_metadata.insert("meta-DC.subject".to_string(), "weather, forecast".to_string());
-
-        let doc = extract_document_metadata(head_metadata, None, None);
-        assert_eq!(doc.keywords, vec!["weather", "forecast"]);
-    }
-
-    #[test]
-    fn test_keywords_dc_keywords() {
-        let mut head_metadata = BTreeMap::new();
-        head_metadata.insert("meta-DC.keywords".to_string(), "climate, data".to_string());
-
-        let doc = extract_document_metadata(head_metadata, None, None);
-        assert_eq!(doc.keywords, vec!["climate", "data"]);
-    }
-
-    #[test]
-    fn test_keywords_dcterms_subject() {
-        let mut head_metadata = BTreeMap::new();
-        head_metadata.insert("meta-DCTERMS.subject".to_string(), "science, research".to_string());
-
-        let doc = extract_document_metadata(head_metadata, None, None);
-        assert_eq!(doc.keywords, vec!["science", "research"]);
-    }
-
-    #[test]
-    fn test_keywords_news_keywords() {
-        let mut head_metadata = BTreeMap::new();
-        head_metadata.insert("meta-news_keywords".to_string(), "breaking, world".to_string());
-
-        let doc = extract_document_metadata(head_metadata, None, None);
-        assert_eq!(doc.keywords, vec!["breaking", "world"]);
-    }
-
-    #[test]
-    fn test_keywords_citation_keywords() {
-        let mut head_metadata = BTreeMap::new();
-        head_metadata.insert("meta-citation_keywords".to_string(), "biology, genetics".to_string());
-
-        let doc = extract_document_metadata(head_metadata, None, None);
-        assert_eq!(doc.keywords, vec!["biology", "genetics"]);
-    }
-
-    #[test]
-    fn test_dc_title_and_description_fallback() {
-        let mut head_metadata = BTreeMap::new();
-        head_metadata.insert("meta-DC.title".to_string(), "DC Title".to_string());
-        head_metadata.insert("meta-DC.description".to_string(), "DC Description".to_string());
-        head_metadata.insert("meta-DC.creator".to_string(), "DC Author".to_string());
-
-        let doc = extract_document_metadata(head_metadata, None, None);
-        assert_eq!(doc.title, Some("DC Title".to_string()));
-        assert_eq!(doc.description, Some("DC Description".to_string()));
-        assert_eq!(doc.author, Some("DC Author".to_string()));
-    }
-
-    #[test]
-    fn test_dc_does_not_override_standard_fields() {
-        let mut head_metadata = BTreeMap::new();
-        // Standard fields come first alphabetically in BTreeMap
-        head_metadata.insert("description".to_string(), "Standard Description".to_string());
-        head_metadata.insert("meta-DC.description".to_string(), "DC Description".to_string());
-        head_metadata.insert("title".to_string(), "Standard Title".to_string());
-        head_metadata.insert("meta-DC.title".to_string(), "DC Title".to_string());
-
-        let doc = extract_document_metadata(head_metadata, None, None);
-        assert_eq!(doc.title, Some("Standard Title".to_string()));
-        assert_eq!(doc.description, Some("Standard Description".to_string()));
-    }
-
-    #[test]
-    fn test_case_insensitive_title_description_author() {
-        let mut head_metadata = BTreeMap::new();
-        head_metadata.insert("meta-Title".to_string(), "My Title".to_string());
-        head_metadata.insert("meta-Description".to_string(), "My Desc".to_string());
-        head_metadata.insert("meta-Author".to_string(), "My Author".to_string());
-
-        let doc = extract_document_metadata(head_metadata, None, None);
-        assert_eq!(doc.title, Some("My Title".to_string()));
-        assert_eq!(doc.description, Some("My Desc".to_string()));
-        assert_eq!(doc.author, Some("My Author".to_string()));
-    }
-
-    #[test]
-    fn test_dcterms_remaining_go_to_meta_tags() {
-        let mut head_metadata = BTreeMap::new();
-        head_metadata.insert("meta-DCTERMS.license".to_string(), "MIT".to_string());
-
-        let doc = extract_document_metadata(head_metadata, None, None);
-        assert_eq!(doc.meta_tags.get("dcterms_license"), Some(&"MIT".to_string()));
-    }
-
-    #[test]
-    fn test_scan_schema_type() {
-        let json = r#"{"@type":"Article","title":"Test"}"#;
-        assert_eq!(scan_schema_type(json), Some("Article".to_string()));
-
-        let json_array = r#"{"@type":["Article","NewsArticle"]}"#;
-        assert_eq!(scan_schema_type(json_array), Some("Article".to_string()));
-
-        let no_type = r#"{"title":"Test"}"#;
-        assert_eq!(scan_schema_type(no_type), None);
-    }
-}

data/vendor/html-to-markdown-rs/src/metadata/mod.rs

@@ -1,288 +0,0 @@
-#![allow(clippy::cast_precision_loss, clippy::cast_sign_loss, clippy::unused_self)]
-//! Metadata extraction for HTML to Markdown conversion.
-//!
-//! This module provides comprehensive, type-safe metadata extraction during HTML-to-Markdown
-//! conversion, enabling content analysis, SEO optimization, and document indexing workflows.
-//! Metadata includes:
-//! - **Document metadata**: Title, description, author, language, canonical URL, Open Graph, Twitter Card
-//! - **Headers**: Heading elements (h1-h6) with hierarchy, IDs, and positions
-//! - **Links**: Hyperlinks with type classification (anchor, internal, external, email, phone)
-//! - **Images**: Image elements with source, alt text, dimensions, and type (data URI, external, etc.)
-//! - **Structured data**: JSON-LD, Microdata, and RDFa blocks
-//!
-//! The implementation follows a single-pass collector pattern for zero-overhead extraction
-//! when metadata features are disabled.
-//!
-//! # Architecture
-//!
-//! Metadata extraction uses the [`MetadataCollector`] pattern:
-//! - **Single-pass collection**: Metadata is gathered during the primary tree traversal without additional passes
-//! - **Zero overhead when disabled**: Entire module can be compiled out via feature flags
-//! - **Configurable granularity**: Use [`MetadataConfig`] to select which metadata types to extract
-//! - **Type-safe APIs**: All metadata types are enum-based with exhaustive matching
-//! - **Memory-bounded**: Size limits prevent memory exhaustion from adversarial documents
-//! - **Pre-allocated buffers**: Typical documents (32 headers, 64 links, 16 images) handled efficiently
-//!
-//! # Type Overview
-//!
-//! ## Enumerations
-//!
-//! - [`TextDirection`]: Document directionality (LTR, RTL, Auto)
-//! - [`LinkType`]: Link classification (Anchor, Internal, External, Email, Phone, Other)
-//! - [`ImageType`]: Image source type (DataUri, External, Relative, InlineSvg)
-//! - [`StructuredDataType`]: Structured data format (JsonLd, Microdata, RDFa)
-//!
-//! ## Structures
-//!
-//! - [`DocumentMetadata`]: Head-level metadata with maps for Open Graph and Twitter Card
-//! - [`HeaderMetadata`]: Heading element with level (1-6), text, ID, hierarchy depth, and position
-//! - [`LinkMetadata`]: Hyperlink with href, text, title, type, rel attributes, and custom attributes
-//! - [`ImageMetadata`]: Image element with src, alt, title, dimensions, type, and attributes
-//! - [`StructuredData`]: Structured data block with type and raw JSON
-//! - [`MetadataConfig`]: Configuration controlling extraction granularity and size limits
-//! - [`HtmlMetadata`]: Top-level result containing all extracted metadata
-//!
-//! # Examples
-//!
-//! ## Basic Usage with `convert()`
-//!
-//! ```text
-//! use html_to_markdown_rs::convert;
-//!
-//! let html = r#"
-//!   <html lang="en">
-//!     <head>
-//!       <title>My Article</title>
-//!       <meta name="description" content="An interesting read">
-//!     </head>
-//!     <body>
-//!       <h1 id="main">Title</h1>
-//!       <a href="https://example.com">External Link</a>
-//!       <img src="photo.jpg" alt="A photo">
-//!     </body>
-//!   </html>
-//! "#;
-//!
-//! let result = convert(html, None)?;
-//! let metadata = result.metadata.unwrap();
-//!
-//! // Access document metadata
-//! assert_eq!(metadata.document.title, Some("My Article".to_string()));
-//! assert_eq!(metadata.document.language, Some("en".to_string()));
-//!
-//! // Access headers
-//! assert_eq!(metadata.headers.len(), 1);
-//! assert_eq!(metadata.headers[0].level, 1);
-//! assert_eq!(metadata.headers[0].id, Some("main".to_string()));
-//!
-//! // Access links
-//! assert_eq!(metadata.links.len(), 1);
-//! assert_eq!(metadata.links[0].link_type, LinkType::External);
-//!
-//! // Access images
-//! assert_eq!(metadata.images.len(), 1);
-//! assert_eq!(metadata.images[0].image_type, ImageType::Relative);
-//! # Ok::<(), html_to_markdown_rs::ConversionError>(())
-//! ```
-//!
-//! ## Selective Extraction
-//!
-//! ```text
-//! use html_to_markdown_rs::{convert, ConversionOptions};
-//!
-//! let options = ConversionOptions {
-//!     extract_metadata: false,  // Disable metadata extraction
-//!     ..Default::default()
-//! };
-//!
-//! let result = convert(html, Some(options))?;
-//! assert!(result.metadata.is_none());  // Metadata not extracted
-//! # Ok::<(), html_to_markdown_rs::ConversionError>(())
-//! ```
-//!
-//! ## Analyzing Link Types
-//!
-//! ```text
-//! use html_to_markdown_rs::convert;
-//! use html_to_markdown_rs::metadata::LinkType;
-//!
-//! let result = convert(html, None)?;
-//! let metadata = result.metadata.unwrap();
-//!
-//! for link in &metadata.links {
-//!     match link.link_type {
-//!         LinkType::External => println!("External: {}", link.href),
-//!         LinkType::Internal => println!("Internal: {}", link.href),
-//!         LinkType::Anchor => println!("Anchor: {}", link.href),
-//!         LinkType::Email => println!("Email: {}", link.href),
-//!         _ => {}
-//!     }
-//! }
-//! # Ok::<(), html_to_markdown_rs::ConversionError>(())
-//! ```
-//!
-//! # Serialization
-//!
-//! All types in this module support serialization via `serde` when the `metadata` feature is enabled.
-//! This enables easy export to JSON, YAML, or other formats:
-//!
-//! ```text
-//! use html_to_markdown_rs::convert;
-//!
-//! let result = convert(html, None)?;
-//! if let Some(metadata) = &result.metadata {
-//!     let json = serde_json::to_string_pretty(metadata)?;
-//!     println!("{}", json);
-//! }
-//! # Ok::<(), Box<dyn std::error::Error>>(())
-//! ```
-
-pub mod collector;
-pub mod config;
-pub mod extraction;
-pub mod types;
-
-// Re-export public types
-pub use collector::MetadataCollector;
-pub use config::{DEFAULT_MAX_STRUCTURED_DATA_SIZE, MetadataConfig, MetadataConfigUpdate};
-pub use types::{
-    DocumentMetadata, HeaderMetadata, HtmlMetadata, ImageMetadata, ImageType, LinkMetadata, LinkType, StructuredData,
-    StructuredDataType, TextDirection,
-};
-
-// Internal handle type for shared mutable access during tree traversal
-use std::cell::RefCell;
-use std::rc::Rc;
-
-/// Handle to a metadata collector via reference-counted mutable cell.
-///
-/// Used internally for sharing collector state across the tree traversal.
-///
-/// # Examples
-///
-/// ```text
-/// let collector = MetadataCollector::new(MetadataConfig::default());
-/// let handle = Rc::new(RefCell::new(collector));
-///
-/// // In tree walk, can be passed and borrowed
-/// handle.borrow_mut().add_header(1, "Title".to_string(), None, 0, 100);
-///
-/// let metadata = handle.take().finish();
-/// ```
-#[allow(dead_code)]
-pub(crate) type MetadataCollectorHandle = Rc<RefCell<MetadataCollector>>;
-
-#[cfg(test)]
-mod tests {
-    use super::*;
-
-    #[test]
-    fn test_text_direction_parse() {
-        assert_eq!(TextDirection::parse("ltr"), Some(TextDirection::LeftToRight));
-        assert_eq!(TextDirection::parse("rtl"), Some(TextDirection::RightToLeft));
-        assert_eq!(TextDirection::parse("auto"), Some(TextDirection::Auto));
-        assert_eq!(TextDirection::parse("invalid"), None);
-        assert_eq!(TextDirection::parse("LTR"), Some(TextDirection::LeftToRight));
-    }
-
-    #[test]
-    fn test_text_direction_display() {
-        assert_eq!(TextDirection::LeftToRight.to_string(), "ltr");
-        assert_eq!(TextDirection::RightToLeft.to_string(), "rtl");
-        assert_eq!(TextDirection::Auto.to_string(), "auto");
-    }
-
-    #[test]
-    fn test_link_classification() {
-        assert_eq!(LinkMetadata::classify_link("#section"), LinkType::Anchor);
-        assert_eq!(LinkMetadata::classify_link("mailto:test@example.com"), LinkType::Email);
-        assert_eq!(LinkMetadata::classify_link("tel:+1234567890"), LinkType::Phone);
-        assert_eq!(LinkMetadata::classify_link("https://example.com"), LinkType::External);
-        assert_eq!(LinkMetadata::classify_link("http://example.com"), LinkType::External);
-        assert_eq!(LinkMetadata::classify_link("/path/to/page"), LinkType::Internal);
-        assert_eq!(LinkMetadata::classify_link("../relative"), LinkType::Internal);
-        assert_eq!(LinkMetadata::classify_link("./same"), LinkType::Internal);
-    }
-
-    #[test]
-    fn test_header_validation() {
-        let valid = HeaderMetadata {
-            level: 3,
-            text: "Title".to_string(),
-            id: None,
-            depth: 2,
-            html_offset: 100,
-        };
-        assert!(valid.is_valid());
-
-        let invalid_high = HeaderMetadata {
-            level: 7,
-            text: "Title".to_string(),
-            id: None,
-            depth: 2,
-            html_offset: 100,
-        };
-        assert!(!invalid_high.is_valid());
-
-        let invalid_low = HeaderMetadata {
-            level: 0,
-            text: "Title".to_string(),
-            id: None,
-            depth: 2,
-            html_offset: 100,
-        };
-        assert!(!invalid_low.is_valid());
-    }
-
-    #[test]
-    fn test_document_metadata_default() {
-        let doc = DocumentMetadata::default();
-
-        assert!(doc.title.is_none());
-        assert!(doc.description.is_none());
-        assert!(doc.keywords.is_empty());
-        assert!(doc.open_graph.is_empty());
-        assert!(doc.twitter_card.is_empty());
-        assert!(doc.meta_tags.is_empty());
-    }
-
-    #[test]
-    fn test_image_type_classification() {
-        let data_uri = ImageMetadata {
-            src: "data:image/png;base64,iVBORw0KG...".to_string(),
-            alt: None,
-            title: None,
-            dimensions: None,
-            image_type: ImageType::DataUri,
-            attributes: Default::default(),
-        };
-        assert_eq!(data_uri.image_type, ImageType::DataUri);
-
-        let external = ImageMetadata {
-            src: "https://example.com/image.jpg".to_string(),
-            alt: None,
-            title: None,
-            dimensions: None,
-            image_type: ImageType::External,
-            attributes: Default::default(),
-        };
-        assert_eq!(external.image_type, ImageType::External);
-    }
-
-    #[test]
-    fn test_link_type_display() {
-        assert_eq!(LinkType::Anchor.to_string(), "anchor");
-        assert_eq!(LinkType::Internal.to_string(), "internal");
-        assert_eq!(LinkType::External.to_string(), "external");
-        assert_eq!(LinkType::Email.to_string(), "email");
-        assert_eq!(LinkType::Phone.to_string(), "phone");
-        assert_eq!(LinkType::Other.to_string(), "other");
-    }
-
-    #[test]
-    fn test_structured_data_type_display() {
-        assert_eq!(StructuredDataType::JsonLd.to_string(), "json_ld");
-        assert_eq!(StructuredDataType::Microdata.to_string(), "microdata");
-        assert_eq!(StructuredDataType::RDFa.to_string(), "rdfa");
-    }
-}