feat(monitoring): attribute SaaS citations by registrable domain

Persist host/registrable_domain/site_key on monitoring article URL aliases, populate them on desktop-publish completion, and use them in the citation pipeline to count how often SaaS-published content appears as a citation source. Expose a /citation-summary endpoint with a 7/30-day window switch and surface the new source-share metrics in the tracking dashboard. Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
2026-04-30 17:12:09 +08:00
parent 6e6e19cccb
commit fb166fa9fe
15 changed files with 836 additions and 143 deletions
@@ -0,0 +1,108 @@
+package app
+
+import (
+	"context"
+	"net/url"
+	"strings"
+)
+
+func (s *DesktopTaskService) syncMonitoringArticleAlias(ctx context.Context, input *monitoringArticleAliasInput) error {
+	if s == nil || s.monitoringPool == nil || input == nil {
+		return nil
+	}
+
+	originalURL, confidence := monitoringAliasOriginalURL(input)
+	if originalURL == "" {
+		return nil
+	}
+
+	normalizedURL := normalizeCitationURL(originalURL)
+	if normalizedURL == "" {
+		return nil
+	}
+	host, registrableDomain, _, _ := deriveCitationURLParts(originalURL, normalizedURL)
+	if registrableDomain == "" {
+		registrableDomain = host
+	}
+	siteKey := host
+	if siteKey == "" {
+		siteKey = registrableDomain
+	}
+	lastPathSegment := monitoringAliasLastPathSegment(normalizedURL)
+
+	_, err := s.monitoringPool.Exec(ctx, `
+		INSERT INTO monitoring_article_url_aliases (
+			tenant_id, article_id, publish_record_id, platform_id, publish_platform_name_snapshot,
+			external_article_id, article_title_snapshot, original_url, normalized_url,
+			host, registrable_domain, site_key, last_path_segment, confidence
+		)
+		VALUES (
+			$1, $2, $3, $4, $5,
+			$6, $7, $8, $9,
+			$10, $11, $12, $13, $14
+		)
+		ON CONFLICT (tenant_id, normalized_url)
+		DO UPDATE SET
+			article_id = EXCLUDED.article_id,
+			publish_record_id = EXCLUDED.publish_record_id,
+			platform_id = EXCLUDED.platform_id,
+			publish_platform_name_snapshot = EXCLUDED.publish_platform_name_snapshot,
+			external_article_id = EXCLUDED.external_article_id,
+			article_title_snapshot = EXCLUDED.article_title_snapshot,
+			original_url = EXCLUDED.original_url,
+			host = EXCLUDED.host,
+			registrable_domain = EXCLUDED.registrable_domain,
+			site_key = EXCLUDED.site_key,
+			last_path_segment = EXCLUDED.last_path_segment,
+			confidence = EXCLUDED.confidence,
+			updated_at = NOW()
+	`,
+		input.TenantID,
+		input.ArticleID,
+		input.PublishRecordID,
+		strings.TrimSpace(input.PlatformID),
+		strings.TrimSpace(input.PlatformName),
+		normalizeStringPointer(input.ExternalArticleID),
+		strings.TrimSpace(input.ArticleTitle),
+		originalURL,
+		normalizedURL,
+		host,
+		registrableDomain,
+		siteKey,
+		nullableString(lastPathSegment),
+		confidence,
+	)
+	return err
+}
+
+func monitoringAliasOriginalURL(input *monitoringArticleAliasInput) (string, string) {
+	if input == nil {
+		return "", ""
+	}
+	if strings.TrimSpace(input.PlatformID) == "baijiahao" {
+		if articleID := strings.TrimSpace(monitoringStringValue(input.ExternalArticleID)); articleID != "" {
+			return baijiahaoPublishedArticleURL(articleID), "high"
+		}
+	}
+	if value := strings.TrimSpace(monitoringStringValue(input.ExternalArticleURL)); value != "" {
+		return value, "high"
+	}
+	if value := strings.TrimSpace(monitoringStringValue(input.ExternalManageURL)); value != "" {
+		return value, "medium"
+	}
+	return "", ""
+}
+
+func monitoringAliasLastPathSegment(rawURL string) *string {
+	parsed, err := url.Parse(strings.TrimSpace(rawURL))
+	if err != nil {
+		return nil
+	}
+	segments := strings.Split(strings.Trim(parsed.Path, "/"), "/")
+	for index := len(segments) - 1; index >= 0; index-- {
+		if segment := strings.TrimSpace(segments[index]); segment != "" {
+			return &segment
+		}
+	}
+	return nil
+}