Switch to tracing for console logging

Cargo.toml

@@ -10,17 +10,24 @@ homepage = "https://murroughfoley.com"
 repository = "https://github.com/Murrough-Foley/rs-trafilatura"
 documentation = "https://docs.rs/rs-trafilatura"
 readme = "README.md"
-keywords = ["html", "content-extraction", "web-scraping", "trafilatura", "boilerplate-removal"]
+keywords = [
+    "html",
+    "content-extraction",
+    "web-scraping",
+    "trafilatura",
+    "boilerplate-removal",
+]
 categories = ["parser-implementations", "web-programming", "text-processing"]
 
 [dependencies]
+tracing = "0.1"
 html-cleaning = "0.3"
 quick_html2md = "0.2"
 web-page-classifier = "0.1"
-dom_query = "0.24"         # Primary DOM manipulation
-tendril = "0.4"            # Zero-copy text operations via StrTendril
+dom_query = "0.24"                                                             # Primary DOM manipulation
+tendril = "0.4"                                                                # Zero-copy text operations via StrTendril
 thiserror = "2.0"
-regex = "1.11"
+regex = "1.12"
 chrono = "0.4"
 encoding_rs = "0.8"
 serde = { version = "1.0", features = ["derive"] }
@@ -57,12 +64,12 @@ pedantic = { level = "warn", priority = -1 }
 unwrap_used = "deny"
 expect_used = "deny"
 # Allow these pedantic lints - they're too strict or low-value for this codebase
-needless_raw_string_hashes = "allow"    # Stylistic preference
-similar_names = "allow"                  # matched/matches, text_len/text_length are clear
-cast_precision_loss = "allow"            # Intentional in scoring calculations
-items_after_statements = "allow"         # Valid Rust style for inline helpers
-doc_markdown = "allow"                   # Too strict about backticks
-float_cmp = "allow"                      # Approximate comparisons are intentional
-too_many_lines = "allow"                 # Some complex functions need to be long
-missing_errors_doc = "allow"             # Error types are self-explanatory
-missing_panics_doc = "allow"             # We don't panic in production code
+needless_raw_string_hashes = "allow" # Stylistic preference
+similar_names = "allow"              # matched/matches, text_len/text_length are clear
+cast_precision_loss = "allow"        # Intentional in scoring calculations
+items_after_statements = "allow"     # Valid Rust style for inline helpers
+doc_markdown = "allow"               # Too strict about backticks
+float_cmp = "allow"                  # Approximate comparisons are intentional
+too_many_lines = "allow"             # Some complex functions need to be long
+missing_errors_doc = "allow"         # Error types are self-explanatory
+missing_panics_doc = "allow"         # We don't panic in production code

src/bin/batch_markdown.rs

@@ -11,15 +11,15 @@ use std::path::PathBuf;
 fn main() {
     let args: Vec<String> = std::env::args().collect();
     if args.len() < 3 {
-        eprintln!("Usage: batch_markdown <input_dir> <output_dir>");
+        tracing::error!("Usage: batch_markdown <input_dir> <output_dir>");
         std::process::exit(1);
     }
 
     let input_dir = PathBuf::from(&args[1]);
     let output_dir = PathBuf::from(&args[2]);
 
     if !input_dir.is_dir() {
-        eprintln!("Input directory does not exist: {}", input_dir.display());
+        tracing::error!("Input directory does not exist: {}", input_dir.display());
         std::process::exit(1);
     }
 
@@ -59,7 +59,7 @@ fn main() {
         let html = match fs::read_to_string(&path) {
             Ok(h) => h,
             Err(e) => {
-                eprintln!("[{}/{}] ERROR reading {}: {}", i + 1, total, stem, e);
+                tracing::error!("[{}/{}] ERROR reading {}: {}", i + 1, total, stem, e);
                 failed += 1;
                 continue;
             }
@@ -68,12 +68,10 @@ fn main() {
         match extract_with_options(&html, &options) {
             Ok(result) => {
                 // Prefer markdown, fall back to plain text
-                let content = result
-                    .content_markdown
-                    .unwrap_or(result.content_text);
+                let content = result.content_markdown.unwrap_or(result.content_text);
 
                 if content.trim().is_empty() {
-                    eprintln!(
+                    tracing::warn!(
                         "[{}/{}] EMPTY: {} (confidence: {:.2})",
                         i + 1,
                         total,
@@ -96,7 +94,10 @@ fn main() {
                 if let Some(ref date) = result.metadata.date {
                     md.push_str(&format!("date: \"{}\"\n", date.to_rfc3339()));
                 }
-                md.push_str(&format!("source_file: \"{}\"\n", path.file_name().unwrap().to_string_lossy()));
+                md.push_str(&format!(
+                    "source_file: \"{}\"\n",
+                    path.file_name().unwrap().to_string_lossy()
+                ));
                 md.push_str(&format!("confidence: {:.2}\n", result.extraction_quality));
                 if let Some(ref pt) = result.metadata.page_type {
                     md.push_str(&format!("page_type: \"{}\"\n", pt));
@@ -116,7 +117,7 @@ fn main() {
                 success += 1;
             }
             Err(e) => {
-                eprintln!("[{}/{}] EXTRACT ERROR {}: {}", i + 1, total, stem, e);
+                tracing::error!("[{}/{}] EXTRACT ERROR {}: {}", i + 1, total, stem, e);
                 failed += 1;
             }
         }

src/bin/extract_stdin.rs

@@ -1,8 +1,8 @@
 //! Simple CLI that reads HTML from stdin and outputs JSON to stdout.
 //! Used by the text-extraction-benchmark Python wrapper.
 
-use rs_trafilatura::{extract_with_options, Options};
 use rs_trafilatura::page_type::PageType;
+use rs_trafilatura::{extract_with_options, Options};
 use serde::Serialize;
 use std::io::{self, Read};
 
@@ -42,7 +42,7 @@ fn main() {
                     url = Some(args[i + 1].clone());
                     i += 2;
                 } else {
-                    eprintln!("--url requires a value");
+                    tracing::error!("--url requires a value");
                     std::process::exit(1);
                 }
             }
@@ -51,13 +51,13 @@ fn main() {
                     match args[i + 1].parse::<PageType>() {
                         Ok(pt) => page_type_override = Some(pt),
                         Err(e) => {
-                            eprintln!("Invalid --page-type: {e}");
+                            tracing::error!("Invalid --page-type: {e}");
                             std::process::exit(1);
                         }
                     }
                     i += 2;
                 } else {
-                    eprintln!("--page-type requires a value");
+                    tracing::error!("--page-type requires a value");
                     std::process::exit(1);
                 }
             }
@@ -69,14 +69,16 @@ fn main() {
                 markdown = true;
                 i += 1;
             }
-            _ => { i += 1; }
+            _ => {
+                i += 1;
+            }
         }
     }
 
     // Read HTML from stdin
     let mut html = String::new();
     if io::stdin().read_to_string(&mut html).is_err() {
-        eprintln!("Failed to read from stdin");
+        tracing::error!("Failed to read from stdin");
         std::process::exit(1);
     }
 
@@ -85,9 +87,21 @@ fn main() {
         url,
         page_type: page_type_override,
         output_markdown: markdown,
-        include_tables: if markdown { true } else { Options::default().include_tables },
-        include_links: if markdown { true } else { Options::default().include_links },
-        include_formatting: if markdown { true } else { Options::default().include_formatting },
+        include_tables: if markdown {
+            true
+        } else {
+            Options::default().include_tables
+        },
+        include_links: if markdown {
+            true
+        } else {
+            Options::default().include_links
+        },
+        include_formatting: if markdown {
+            true
+        } else {
+            Options::default().include_formatting
+        },
         ..Options::default()
     };