fix: Resolve playwright function deprecations and io/ioutil function deprecations.

README.md

@@ -4,16 +4,18 @@ Rollup aggregates the contents of text-based files and webpages into a markdown
 
 ## Features
 
-- File type filtering
+- File type filtering for targeted content aggregation
-- Ignore patterns for excluding files
+- Ignore patterns for excluding specific files or directories
-- Support for code-generated file detection
+- Support for code-generated file detection and exclusion
-- Advanced web scraping functionality
+- Advanced web scraping functionality with depth control
-- Verbose logging option for detailed output
+- Verbose logging option for detailed operation insights
-- Exclusionary CSS selectors for web scraping
+- Exclusionary CSS selectors for precise web content extraction
-- Support for multiple URLs in web scraping
+- Support for multiple URLs in web scraping operations
 - Configurable output format for web scraping (single file or separate files)
-- Configuration file support (YAML)
+- Flexible configuration file support (YAML)
-- Generation of default configuration file
+- Automatic generation of default configuration file
+- Custom output file naming
+- Concurrent processing for improved performance
 
 ## Installation
 
@@ -74,14 +76,27 @@ ignore:
 code_generated:
   - **/generated/**
 scrape:
-  urls:
+  sites:
-    - url: https://example.com
+    - base_url: https://example.com
       css_locator: .content
       exclude_selectors:
         - .ads
         - .navigation
+      max_depth: 2
+      allowed_paths:
+        - /blog
+        - /docs
+      exclude_paths:
+        - /admin
       output_alias: example
+      path_overrides:
+        - path: /special-page
+          css_locator: .special-content
+          exclude_selectors:
+            - .special-ads
   output_type: single
+  requests_per_second: 1.0
+  burst_limit: 3
 ```
 
 ## Examples
@@ -92,10 +107,10 @@ scrape:
    rollup files
    ```
 
-2. Web scraping with multiple URLs:
+2. Web scraping with multiple URLs and increased concurrency:
 
    ```bash
-   rollup web --urls=https://example.com,https://another-example.com
+   rollup web --urls=https://example.com,https://another-example.com --concurrent=8
    ```
 
 3. Generate a default configuration file:
@@ -104,15 +119,25 @@ scrape:
    rollup generate
    ```
 
-4. Use a custom configuration file:
+4. Use a custom configuration file and specify output:
 
    ```bash
-   rollup files --config=my-config.yml
+   rollup files --config=my-config.yml --output=project_summary.md
    ```
 
-5. Web scraping with separate output files:
+5. Web scraping with separate output files and custom timeout:
    ```bash
-   rollup web --urls=https://example.com,https://another-example.com --output=separate
+   rollup web --urls=https://example.com,https://another-example.com --output=separate --timeout=60
+   ```
+
+6. Rollup files with specific types and ignore patterns:
+   ```bash
+   rollup files --types=.go,.md --ignore=vendor/**,*_test.go
+   ```
+
+7. Web scraping with depth and CSS selector:
+   ```bash
+   rollup web --urls=https://example.com --depth=2 --css=.main-content
    ```
 
 ## Contributing

cmd/files.go

@@ -8,8 +8,11 @@ import (
 	"time"
 
 	"github.com/spf13/cobra"
+	"github.com/tnypxl/rollup/internal/config"
 )
 
+var cfg *config.Config
+
 var (
 	path            string
 	fileTypes       string
@@ -24,7 +27,7 @@ var filesCmd = &cobra.Command{
 in a given project, current path or a custom path, to a single timestamped markdown file
 whose name is <project-directory-name>-rollup-<timestamp>.md.`,
 	RunE: func(cmd *cobra.Command, args []string) error {
-		return runRollup()
+		return runRollup(cfg)
 	},
 }
 
@@ -87,16 +90,23 @@ func isIgnored(filePath string, patterns []string) bool {
 				return true
 			}
 		} else {
-			matched, err := filepath.Match(pattern, filepath.Base(filePath))
+			// Check if the pattern matches the full path or any part of it
-			if err == nil && matched {
+			if matched, _ := filepath.Match(pattern, filePath); matched {
 				return true
 			}
+			pathParts := strings.Split(filePath, string(os.PathSeparator))
+			for i := range pathParts {
+				partialPath := filepath.Join(pathParts[:i+1]...)
+				if matched, _ := filepath.Match(pattern, partialPath); matched {
+					return true
+				}
+			}
 		}
 	}
 	return false
 }
 
-func runRollup() error {
+func runRollup(cfg *config.Config) error {
 	// Use config if available, otherwise use command-line flags
 	var types, codeGenList, ignoreList []string
 	if cfg != nil && len(cfg.FileTypes) > 0 {

cmd/files_test.go

@@ -21,6 +21,9 @@ func TestMatchGlob(t *testing.T) {
 		{"**/*.go", "dir/subdir/file.go", true},
 		{"dir/*.go", "dir/file.go", true},
 		{"dir/*.go", "otherdir/file.go", false},
+		{"**/test_*.go", "internal/test_helper.go", true},
+		{"docs/**/*.md", "docs/api/endpoints.md", true},
+		{"docs/**/*.md", "src/docs/readme.md", false},
 	}
 
 	for _, test := range tests {
@@ -32,7 +35,7 @@ func TestMatchGlob(t *testing.T) {
 }
 
 func TestIsCodeGenerated(t *testing.T) {
-	patterns := []string{"generated_*.go", "**/auto_*.go"}
+	patterns := []string{"generated_*.go", "**/auto_*.go", "**/*_gen.go"}
 	tests := []struct {
 		path     string
 		expected bool
@@ -41,6 +44,8 @@ func TestIsCodeGenerated(t *testing.T) {
 		{"normal_file.go", false},
 		{"subdir/auto_file.go", true},
 		{"subdir/normal_file.go", false},
+		{"pkg/models_gen.go", true},
+		{"pkg/handler.go", false},
 	}
 
 	for _, test := range tests {
@@ -52,7 +57,7 @@ func TestIsCodeGenerated(t *testing.T) {
 }
 
 func TestIsIgnored(t *testing.T) {
-	patterns := []string{"*.tmp", "**/*.log"}
+	patterns := []string{"*.tmp", "**/*.log", ".git/**", "vendor/**"}
 	tests := []struct {
 		path     string
 		expected bool
@@ -61,6 +66,10 @@ func TestIsIgnored(t *testing.T) {
 		{"file.go", false},
 		{"subdir/file.log", true},
 		{"subdir/file.txt", false},
+		{".git/config", true},
+		{"src/.git/config", false},
+		{"vendor/package/file.go", true},
+		{"internal/vendor/file.go", false},
 	}
 
 	for _, test := range tests {
@@ -85,45 +94,52 @@ func TestRunRollup(t *testing.T) {
 		"file2.txt":            "This is a text file.\n",
 		"subdir/file3.go":      "package subdir\n\nfunc Func() {}\n",
 		"subdir/file4.json":    "{\"key\": \"value\"}\n",
+		"generated_model.go":   "// Code generated DO NOT EDIT.\n\npackage model\n",
+		"docs/api/readme.md":   "# API Documentation\n",
+		".git/config":          "[core]\n\trepositoryformatversion = 0\n",
+		"vendor/lib/helper.go": "package lib\n\nfunc Helper() {}\n",
 	}
 
 	for name, content := range files {
 		path := filepath.Join(tempDir, name)
-		err := os.MkdirAll(filepath.Dir(path), 0755)
+		if err := os.MkdirAll(filepath.Dir(path), 0o755); err != nil {
-		if err != nil {
 			t.Fatalf("Failed to create directory: %v", err)
 		}
-		err = os.WriteFile(path, []byte(content), 0644)
+		if err := os.WriteFile(path, []byte(content), 0o644); err != nil {
-		if err != nil {
 			t.Fatalf("Failed to write file: %v", err)
 		}
 	}
 
 	// Set up test configuration
 	cfg = &config.Config{
-		FileTypes:     []string{"go", "txt"},
+		FileTypes:     []string{"go", "txt", "md"},
-		Ignore:        []string{"*.json"},
+		Ignore:        []string{"*.json", ".git/**", "vendor/**"},
-		CodeGenerated: []string{},
+		CodeGenerated: []string{"generated_*.go"},
 	}
-	path = tempDir
+
+	// Change working directory to the temp directory
+	originalWd, _ := os.Getwd()
+	os.Chdir(tempDir)
+	defer os.Chdir(originalWd)
 
 	// Run the rollup
-	err = runRollup()
+	if err := runRollup(cfg); err != nil {
-	if err != nil {
 		t.Fatalf("runRollup() failed: %v", err)
 	}
 
 	// Check if the output file was created
-	outputFiles, err := filepath.Glob(filepath.Join(tempDir, "*.rollup.md"))
+	outputFiles, err := filepath.Glob("*.rollup.md")
 	if err != nil {
-		t.Fatalf("Failed to glob output files: %v", err)
+		t.Fatalf("Error globbing for output file: %v", err)
 	}
-	if len(outputFiles) != 1 {
+	if len(outputFiles) == 0 {
-		t.Fatalf("Expected 1 output file, got %d", len(outputFiles))
+		allFiles, _ := filepath.Glob("*")
+		t.Fatalf("No rollup.md file found. Files in directory: %v", allFiles)
 	}
+	outputFile := outputFiles[0]
 
 	// Read the content of the output file
-	content, err := os.ReadFile(outputFiles[0])
+	content, err := os.ReadFile(outputFile)
 	if err != nil {
 		t.Fatalf("Failed to read output file: %v", err)
 	}
@@ -133,6 +149,8 @@ func TestRunRollup(t *testing.T) {
 		"# File: file1.go",
 		"# File: file2.txt",
 		"# File: subdir/file3.go",
+		"# File: docs/api/readme.md",
+		"# File: generated_model.go (Code-generated, Read-only)",
 	}
 	for _, expected := range expectedContent {
 		if !strings.Contains(string(content), expected) {
@@ -140,8 +158,15 @@ func TestRunRollup(t *testing.T) {
 		}
 	}
 
-	// Check if the ignored file is not included
+	// Check if the ignored files are not included
-	if strings.Contains(string(content), "file4.json") {
+	ignoredContent := []string{
-		t.Errorf("Output file contains ignored file: file4.json")
+		"file4.json",
+		".git/config",
+		"vendor/lib/helper.go",
+	}
+	for _, ignored := range ignoredContent {
+		if strings.Contains(string(content), ignored) {
+			t.Errorf("Output file contains ignored file: %s", ignored)
+		}
 	}
 }

cmd/root.go

@@ -7,7 +7,6 @@ import (
 
 var (
 	configFile string
-	cfg        *config.Config
 	verbose    bool
 )
 
@@ -19,10 +18,10 @@ Use subcommands to perform specific operations.`,
 }
 
 func Execute(conf *config.Config) error {
-	cfg = conf
+	if conf == nil {
-	if cfg == nil {
+		conf = &config.Config{} // Use an empty config if none is provided
-		cfg = &config.Config{} // Use an empty config if none is provided
 	}
+	cfg = conf // Set the cfg variable in cmd/files.go
 	return rootCmd.Execute()
 }
 

cmd/web.go

@@ -2,7 +2,7 @@ package cmd
 
 import (
 	"fmt"
-	"io/ioutil"
+	"io"
 	"log"
 	"net/url"
 	"os"
@@ -44,7 +44,7 @@ func runWeb(cmd *cobra.Command, args []string) error {
 	scraper.SetupLogger(verbose)
 	logger := log.New(os.Stdout, "WEB: ", log.LstdFlags)
 	if !verbose {
-        logger.SetOutput(ioutil.Discard)
+		logger.SetOutput(io.Discard)
 	}
 	logger.Printf("Starting web scraping process with verbose mode: %v", verbose)
 	scraperConfig.Verbose = verbose
@@ -131,7 +131,7 @@ func runWeb(cmd *cobra.Command, args []string) error {
 }
 
 func writeSingleFile(content map[string]string) error {
-	outputFile := generateDefaultFilename(urls)
+	outputFile := generateDefaultFilename()
 	file, err := os.Create(outputFile)
 	if err != nil {
 		return fmt.Errorf("error creating output file: %v", err)
@@ -139,7 +139,7 @@ func writeSingleFile(content map[string]string) error {
 	defer file.Close()
 
 	for url, c := range content {
-		_, err = file.WriteString(fmt.Sprintf("# Content from %s\n\n%s\n\n---\n\n", url, c))
+		_, err = fmt.Fprintf(file, "# ::: Content from %s\n\n%s\n\n---\n\n", url, c)
 		if err != nil {
 			return fmt.Errorf("error writing content to file: %v", err)
 		}
@@ -161,7 +161,7 @@ func writeMultipleFiles(content map[string]string) error {
 			return fmt.Errorf("error creating output file %s: %v", filename, err)
 		}
 
-		_, err = file.WriteString(fmt.Sprintf("# Content from %s\n\n%s\n", url, c))
+		_, err = file.WriteString(fmt.Sprintf("# ::: Content from %s\n\n%s\n", url, c))
 		if err != nil {
 			file.Close()
 			return fmt.Errorf("error writing content to file %s: %v", filename, err)
@@ -174,7 +174,7 @@ func writeMultipleFiles(content map[string]string) error {
 	return nil
 }
 
-func generateDefaultFilename(urls []string) string {
+func generateDefaultFilename() string {
 	timestamp := time.Now().Format("20060102-150405")
 	return fmt.Sprintf("web-%s.rollup.md", timestamp)
 }
@@ -215,8 +215,10 @@ func scrapeURL(urlStr string, depth int, visited map[string]bool) (string, error
 	return content, nil
 }
 
-var testExtractAndConvertContent = extractAndConvertContent
+var (
-var testExtractLinks = scraper.ExtractLinks
+	testExtractAndConvertContent = extractAndConvertContent
+	testExtractLinks             = scraper.ExtractLinks
+)
 
 func extractAndConvertContent(urlStr string) (string, error) {
 	content, err := scraper.FetchWebpageContent(urlStr)
@@ -240,7 +242,7 @@ func extractAndConvertContent(urlStr string) (string, error) {
 	if err != nil {
 		return "", fmt.Errorf("error parsing URL: %v", err)
 	}
-	header := fmt.Sprintf("# Content from %s\n\n", parsedURL.String())
+	header := fmt.Sprintf("# ::: Content from %s\n\n", parsedURL.String())
 
 	return header + markdown + "\n\n", nil
 }

cmd/web_test.go

@@ -103,7 +103,7 @@ func mockExtractAndConvertContent(urlStr string) (string, error) {
 	return "Mocked content for " + urlStr, nil
 }
 
-func mockExtractLinks(urlStr string) ([]string, error) {
+func mockExtractLinks() ([]string, error) {
 	return []string{"http://example.com/link1", "http://example.com/link2"}, nil
 }
 

docs/CHANGELOG.md

@@ -0,0 +1,21 @@
+# Changelog
+
+All notable changes to this project will be documented in this file.
+
+The format is based on [Keep a Changelog](https://keepachangelog.com/en/1.0.0/),
+and this project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0.html).
+
+## [0.0.3] - 2024-09-22
+
+### Added
+- Implemented web scraping functionality using Playwright
+- Added support for CSS selectors to extract specific content
+- Introduced rate limiting for web requests
+- Created configuration options for scraping settings
+
+### Changed
+- Improved error handling and logging throughout the application
+- Enhanced URL parsing and validation
+
+### Fixed
+- Resolved issues with concurrent scraping operations

internal/config/config_test.go

@@ -45,10 +45,10 @@ scrape:
 	}
 	defer os.Remove(tmpfile.Name())
 
-	if _, err := tmpfile.Write(content); err != nil {
+	if _, err = tmpfile.Write(content); err != nil {
 		t.Fatalf("Failed to write to temp file: %v", err)
 	}
-	if err := tmpfile.Close(); err != nil {
+	if err = tmpfile.Close(); err != nil {
 		t.Fatalf("Failed to close temp file: %v", err)
 	}
 

internal/scraper/scraper.go

@@ -1,21 +1,21 @@
 package scraper
 
 import (
+	"context"
 	"fmt"
-	"io/ioutil"
+	"io"
 	"log"
 	"math/rand"
 	"net/url"
 	"os"
 	"regexp"
 	"strings"
-	"time"
 	"sync"
-	"context"
+	"time"
 
+	md "github.com/JohannesKaufmann/html-to-markdown"
 	"github.com/PuerkitoBio/goquery"
 	"github.com/playwright-community/playwright-go"
-	md "github.com/JohannesKaufmann/html-to-markdown"
 	"golang.org/x/time/rate"
 )
 
@@ -81,7 +81,7 @@ func ScrapeSites(config Config) (map[string]string, error) {
 				fullURL := site.BaseURL + path
 				totalURLs++
 				logger.Printf("Queueing URL for scraping: %s\n", fullURL)
-                scrapeSingleURL(fullURL, site, config, results, limiter)
+				scrapeSingleURL(fullURL, site, results, limiter)
 			}
 		}(site)
 	}
@@ -108,11 +108,12 @@ func ScrapeSites(config Config) (map[string]string, error) {
 	return scrapedContent, nil
 }
 
-func scrapeSingleURL(url string, site SiteConfig, config Config, results chan<- struct {
+func scrapeSingleURL(url string, site SiteConfig, results chan<- struct {
 	url     string
 	content string
 	err     error
-}, limiter *rate.Limiter) {
+}, limiter *rate.Limiter,
+) {
 	logger.Printf("Starting to scrape URL: %s\n", url)
 
 	// Wait for rate limiter before making the request
@@ -155,11 +156,12 @@ func scrapeSingleURL(url string, site SiteConfig, config Config, results chan<-
 	}{url, content, nil}
 }
 
-func scrapeSite(site SiteConfig, config Config, results chan<- struct {
+func scrapeSite(site SiteConfig, results chan<- struct {
 	url     string
 	content string
 	err     error
-}, limiter *rate.Limiter) {
+}, limiter *rate.Limiter,
+) {
 	visited := make(map[string]bool)
 	queue := []string{site.BaseURL}
 
@@ -296,7 +298,7 @@ func SetupLogger(verbose bool) {
 	if verbose {
 		logger = log.New(os.Stdout, "SCRAPER: ", log.LstdFlags)
 	} else {
-		logger = log.New(ioutil.Discard, "", 0)
+		logger = log.New(io.Discard, "", 0)
 	}
 }
 
@@ -339,6 +341,16 @@ func ClosePlaywright() {
 	}
 }
 
+// InitBrowser initializes the browser
+func InitBrowser() error {
+	return InitPlaywright()
+}
+
+// CloseBrowser closes the browser
+func CloseBrowser() {
+	ClosePlaywright()
+}
+
 // FetchWebpageContent retrieves the content of a webpage using Playwright
 func FetchWebpageContent(urlStr string) (string, error) {
 	logger.Printf("Fetching webpage content for URL: %s\n", urlStr)
@@ -377,7 +389,9 @@ func FetchWebpageContent(urlStr string) (string, error) {
 	}
 
 	logger.Println("Waiting for body element")
-	_, err = page.WaitForSelector("body", playwright.PageWaitForSelectorOptions{
+	
+	bodyElement := page.Locator("body")
+	err = bodyElement.WaitFor(playwright.LocatorWaitForOptions{
 		State: playwright.WaitForSelectorStateVisible,
 	})
 	if err != nil {
@@ -394,7 +408,7 @@ func FetchWebpageContent(urlStr string) (string, error) {
 
 	if content == "" {
 		logger.Println(" content is empty, falling back to body content")
-		content, err = page.InnerHTML("body")
+		content, err = bodyElement.InnerHTML()
 		if err != nil {
 			logger.Printf("Error getting body content: %v\n", err)
 			return "", fmt.Errorf("could not get body content: %v", err)
@@ -447,6 +461,8 @@ func scrollPage(page playwright.Page) error {
 		() => {
 			window.scrollTo(0, document.body.scrollHeight);
 			return document.body.scrollHeight;
+			// wait for 500 ms
+			new Promise(resolve => setTimeout(resolve, 500));
 		}
 	`
 
@@ -478,7 +494,9 @@ func scrollPage(page playwright.Page) error {
 
 		previousHeight = currentHeight
 
-		page.WaitForTimeout(500)
+		// Wait for a while before scrolling again
+		
+		
 	}
 
 	logger.Println("Scrolling back to top")
@@ -518,7 +536,9 @@ func ExtractLinks(urlStr string) ([]string, error) {
 
 	var result []string
 	for _, link := range links.([]interface{}) {
-		result = append(result, link.(string))
+		// Normalize URL by removing trailing slash
+		normalizedLink := strings.TrimRight(link.(string), "/")
+		result = append(result, normalizedLink)
 	}
 
 	logger.Printf("Extracted %d links\n", len(result))
@@ -552,6 +572,23 @@ func ExtractContentWithCSS(content, includeSelector string, excludeSelectors []s
 		return "", fmt.Errorf("error extracting content with CSS selector: %v", err)
 	}
 
+	// Trim leading and trailing whitespace
+	selectedContent = strings.TrimSpace(selectedContent)
+
+	// Normalize newlines
+	selectedContent = strings.ReplaceAll(selectedContent, "\r\n", "\n")
+	selectedContent = strings.ReplaceAll(selectedContent, "\r", "\n")
+
+	// Remove indentation while preserving structure
+	lines := strings.Split(selectedContent, "\n")
+	for i, line := range lines {
+		lines[i] = strings.TrimSpace(line)
+	}
+	selectedContent = strings.Join(lines, "\n")
+
+	// Remove any leading or trailing newlines
+	selectedContent = strings.Trim(selectedContent, "\n")
+
 	logger.Printf("Extracted content length: %d\n", len(selectedContent))
 	return selectedContent, nil
 }

internal/scraper/scraper_test.go

@@ -1,10 +1,13 @@
 package scraper
 
 import (
-	"testing"
+	"io"
+	"log"
 	"net/http"
 	"net/http/httptest"
+	"reflect"
 	"strings"
+	"testing"
 )
 
 func TestIsAllowedURL(t *testing.T) {
@@ -69,6 +72,9 @@ func TestGetOverrides(t *testing.T) {
 }
 
 func TestExtractContentWithCSS(t *testing.T) {
+	// Initialize logger for testing
+	logger = log.New(io.Discard, "", 0)
+
 	html := `
 		<html>
 			<body>
@@ -138,6 +144,12 @@ This is a **test** paragraph.
 }
 
 func TestExtractLinks(t *testing.T) {
+	// Initialize Playwright before running the test
+	if err := InitPlaywright(); err != nil {
+		t.Fatalf("Failed to initialize Playwright: %v", err)
+	}
+	defer ClosePlaywright()
+
 	server := httptest.NewServer(http.HandlerFunc(func(w http.ResponseWriter, r *http.Request) {
 		w.Header().Set("Content-Type", "text/html")
 		w.Write([]byte(`