penal-colony/kafka
2
0
Fork 0
Code Issues Pull requests Projects Releases Packages Wiki Activity Actions

refactor: clean up verbose and redundant comments
Some checks failed
Build and Push Docker Image / build-and-push (push) Failing after 7s
Details
Mirror to GitHub / mirror (push) Failing after 3s
Details
Tests / test (push) Successful in 25s
Details

Browse source 
Trim or remove comments that:
- State the obvious (function names already convey purpose)
- Repeat what the code clearly shows
- Are excessively long without adding value

Keep comments that explain *why*, not *what*.
This commit is contained in:
Franz Kafka 2026-03-22 11:10:50 +00:00
parent 805e7ffdc2
commit 5b942a5fd6
11 changed files with 16 additions and 102 deletions
Download patch file Download diff file Expand all files Collapse all files

26
internal/engines/google.go
View file

@ -57,7 +57,6 @@ func (e *GoogleEngine) Search(ctx context.Context, req contracts.SearchRequest)
start := (req.Pageno - 1) * 10
query := url.QueryEscape(req.Query)
// Build URL like SearXNG does.
u := fmt.Sprintf(
"https://www.google.com/search?q=%s&filter=0&start=%d&hl=%s&lr=%s&safe=%s",
query,
@ -118,7 +117,6 @@ func (e *GoogleEngine) Search(ctx context.Context, req contracts.SearchRequest)
}, nil
}
// detectGoogleSorry returns true if the response is a Google block/CAPTCHA page.
func detectGoogleSorry(resp *http.Response) bool {
if resp.Request != nil {
if resp.Request.URL.Host == "sorry.google.com" || strings.HasPrefix(resp.Request.URL.Path, "/sorry") {
@ -128,16 +126,9 @@ func detectGoogleSorry(resp *http.Response) bool {
return false
}
// parseGoogleResults extracts search results from Google's HTML.
// Uses the same selectors as SearXNG: div.MjjYud for result containers.
func parseGoogleResults(body, query string) []contracts.MainResult {
var results []contracts.MainResult
// SearXNG selector: .//div[contains(@class, "MjjYud")]
// Each result block contains a title link and snippet.
// We simulate the XPath matching with regex-based extraction.
// Find all MjjYud div blocks.
mjjPattern := regexp.MustCompile(`<div[^>]*class="[^"]*MjjYud[^"]*"[^>]*>(.*?)</div>\s*(?=<div[^>]*class="[^"]*MjjYud|$)`)
matches := mjjPattern.FindAllStringSubmatch(body, -1)
@ -147,15 +138,12 @@ func parseGoogleResults(body, query string) []contracts.MainResult {
}
block := match[1]
// Extract title and URL from the result link.
// Pattern: <a href="/url?q=ACTUAL_URL&amp;sa=..." ...>TITLE</a>
urlPattern := regexp.MustCompile(`<a[^>]+href="(/url\?q=[^"&]+)`)
urlMatch := urlPattern.FindStringSubmatch(block)
if len(urlMatch) < 2 {
continue
}
rawURL := urlMatch[1]
// Remove /url?q= prefix and decode.
actualURL := strings.TrimPrefix(rawURL, "/url?q=")
if amp := strings.Index(actualURL, "&amp;"); amp != -1 {
actualURL = actualURL[:amp]
@ -168,14 +156,12 @@ func parseGoogleResults(body, query string) []contracts.MainResult {
continue
}
// Extract title from the title tag.
titlePattern := regexp.MustCompile(`<span[^>]*class="[^"]*qrStP[^"]*"[^>]*>([^<]+)</span>`)
titleMatch := titlePattern.FindStringSubmatch(block)
title := query
if len(titleMatch) >= 2 {
title = stripTags(titleMatch[1])
} else {
// Fallback: extract visible text from an <a> with data-title or role="link"
linkTitlePattern := regexp.MustCompile(`<a[^>]+role="link"[^>]*>([^<]+)<`)
ltMatch := linkTitlePattern.FindStringSubmatch(block)
if len(ltMatch) >= 2 {
@ -183,7 +169,6 @@ func parseGoogleResults(body, query string) []contracts.MainResult {
}
}
// Extract snippet from data-sncf divs (SearXNG's approach).
snippet := extractGoogleSnippet(block)
urlPtr := actualURL
@ -202,10 +187,7 @@ func parseGoogleResults(body, query string) []contracts.MainResult {
return results
}
// extractGoogleSnippet extracts the snippet text from a Google result block.
func extractGoogleSnippet(block string) string {
// Google's snippets live in divs with data-sncf attribute.
// SearXNG looks for: .//div[contains(@data-sncf, "1")]
snippetPattern := regexp.MustCompile(`<div[^>]+data-sncf="1"[^>]*>(.*?)</div>`)
matches := snippetPattern.FindAllStringSubmatch(block, -1)
var parts []string
@ -221,10 +203,8 @@ func extractGoogleSnippet(block string) string {
return strings.Join(parts, " ")
}
// extractGoogleSuggestions extracts search suggestions from Google result cards.
func extractGoogleSuggestions(body string) []string {
var suggestions []string
// SearXNG xpath: //div[contains(@class, "ouy7Mc")]//a
suggestionPattern := regexp.MustCompile(`(?s)<div[^>]*class="[^"]*ouy7Mc[^"]*"[^>]*>.*?<a[^>]*>([^<]+)</a>`)
matches := suggestionPattern.FindAllStringSubmatch(body, -1)
seen := map[string]bool{}
@ -241,8 +221,6 @@ func extractGoogleSuggestions(body string) []string {
return suggestions
}
// googleHL maps SearXNG locale to Google hl (host language) parameter.
// e.g. "en-US" -> "en-US"
func googleHL(lang string) string {
lang = strings.ToLower(strings.TrimSpace(lang))
if lang == "" || lang == "auto" {
@ -251,8 +229,6 @@ func googleHL(lang string) string {
return lang
}
// googleUILanguage maps SearXNG language to Google lr (language restrict) parameter.
// e.g. "en" -> "lang_en", "de" -> "lang_de"
func googleUILanguage(lang string) string {
lang = strings.ToLower(strings.Split(lang, "-")[0])
if lang == "" || lang == "auto" {
@ -261,7 +237,6 @@ func googleUILanguage(lang string) string {
return "lang_" + lang
}
// googleSafeSearchLevel maps safesearch (0-2) to Google's safe parameter.
func googleSafeSearchLevel(safesearch int) string {
switch safesearch {
case 0:
@ -275,7 +250,6 @@ func googleSafeSearchLevel(safesearch int) string {
}
}
// stripTags removes HTML tags from a string.
func stripTags(s string) string {
stripper := regexp.MustCompile(`<[^>]*>`)
s = stripper.ReplaceAllString(s, "")