feat(Carta Capital): Adiciona Spider e Scraper para a Carta Capital

Makefile

@@ -82,7 +82,10 @@ scrape_uol:
 
 scrape_folha:
 	docker compose exec scraper python scrape_no_openai.py --platform folha.uol.com.br
-
+
+scrape_cartacapital:
+	docker compose exec scraper python scrape_no_openai.py --platform cartacapital.com.br
+
 # Crawler para todos os portais ou específicos
 crawl:
 	docker compose run --rm scraper python crawl.py
@@ -112,6 +115,9 @@ crawl_ig:
 crawl_folha:
 	docker compose run scraper python crawl.py folhaspider
 
+crawl_cartacapital:
+	docker compose run scraper python crawl.py cartacapitalspider
+
 # Workflow completo de coleta de URLs
 crawl_all_working:
 	@echo "Executando crawl de todos os portais funcionais..."
@@ -123,6 +129,7 @@ crawl_all_working:
 	@make crawl_aliadosBrasil
 	@make crawl_ig
 	@make crawl_folha
+	@make crawl_cartacapital
 	@echo "Crawl de todos os portais concluído!"
 
 # Workflow completo de scraping
@@ -136,6 +143,7 @@ scrape_all_working:
 	@make scrape_r7
 	@make scrape_uol
 	@make scrape_folha
+	@make scrape_cartacapital
 	@echo "Scraping de todos os portais concluído!"
 
 # Pipeline completo: crawl + scrape
@@ -182,6 +190,7 @@ help:
 	@echo "  make crawl_aliadosBrasil - Coleta URLs do portal AliadosBrasil"
 	@echo "  make crawl_ig          - Coleta URLs do portal IG"
 	@echo "  make crawl_folha       - Coleta URLs do portal Folha"
+	@echo "  make crawl_cartacapital - Coleta URLs do portal CartaCapital"
 	@echo ""
 	@echo "=== COMANDOS DE SCRAPING (Extração de Anúncios) ==="
 	@echo "  make scrape_all_working - Executa scraping de todos os portais funcionais"
@@ -193,6 +202,7 @@ help:
 	@echo "  make scrape_r7          - Scraping do portal R7"
 	@echo "  make scrape_uol         - Scraping do portal UOL"
 	@echo "  make scrape_folha       - Scraping do portal Folha"
+	@echo "  make scrape_cartacapital - Scraping do portal CartaCapital"
 	@echo ""
 	@echo "=== WORKFLOWS COMPLETOS ==="
 	@echo "  make pipeline_complete  - Executa crawl + scraping de todos os portais"

crawl.py

@@ -54,4 +54,4 @@ def run(spider_name=None):
     if spider_name:
         logger.info(f"Running spider: {spider_name}")
     run(spider_name)
-    logger.info("Done!")
+    logger.info("Done!")

create_db.py

@@ -42,6 +42,8 @@
         ("Gazeta do Povo", "https://www.gazetadopovo.com.br/", "gazetadopovo"),
         ("Mais Goiás", "https://www.maisgoias.com.br/", "maisgoias"),
         ("Aliados Brasil", "https://www.aliadosbrasiloficial.com.br/", "aliadosbrasil"),
+        ("Carta Capital", "https://www.cartacapital.com.br/", "cartacapital"),
+
     ]
     portals_to_add = []
     for portal in portals:

plays/__init__.py

@@ -11,6 +11,7 @@
 from plays.gazetaDoPovo import GazetaDoPovoPlay
 from plays.maisGoias import MaisGoias
 from plays.aliadosBrasil import AliadosBrasilPlay
+from plays.cartaCapital import CartaCapitalPlay
 
 __all__ = [
     ClicRBSPlay,
@@ -26,4 +27,5 @@
     GazetaDoPovoPlay,
     MaisGoias,
     AliadosBrasilPlay,
+    CartaCapitalPlay,
 ]

plays/cartaCapital.py

@@ -0,0 +1,199 @@
+from playwright.sync_api import sync_playwright
+
+from plays.base import BasePlay
+from plays.items import EntryItem
+from plog import logger
+
+
+class CartaCapitalPlay(BasePlay):
+    name = "cartacapital"
+
+    @classmethod
+    def match(cls, url: str) -> bool:
+        return "cartacapital.com.br" in url
+
+    def run(self) -> EntryItem:
+        """
+        Abre a página da notícia, extrai título, descrição, corpo e tags
+        e devolve um EntryItem.
+        """
+        with sync_playwright() as p:
+            browser = self.launch_browser(
+                p,
+                viewport={"width": 1366, "height": 768},
+            )
+            page = browser.new_page()
+
+            logger.info(f"[{self.name}] Opening URL '{self.url}'...")
+            page.goto(self.url, timeout=180_000, wait_until="networkidle")
+
+            title = self._extract_title(page)
+            description = self._extract_description(page)
+            body = self._extract_body(page)
+            tags = self._extract_tags(page)
+
+            browser.close()
+
+            return EntryItem(
+                title=title,
+                url=self.url,
+                description=description,
+                body=body,
+                tags=tags,
+            )
+
+    def _extract_title(self, page) -> str:
+        """
+        CartaCapital tem título principal em <h1>.
+        """
+        selectors = [
+            "article h1",
+            "main h1",
+            "h1.entry-title",
+            "h1",
+        ]
+
+        for selector in selectors:
+            try:
+                locator = page.locator(selector)
+                if locator.count() > 0:
+                    text = locator.first.inner_text().strip()
+                    if text:
+                        logger.info(f"[{self.name}] Title extracted: {text[:80]}...")
+                        return text
+            except Exception as exc:
+                logger.warning(
+                    f"[{self.name}] Failed to extract title with '{selector}': {exc}"
+                )
+
+        logger.error(f"[{self.name}] Failed to extract title")
+        return ""
+
+    def _extract_description(self, page) -> str:
+        """
+        Subtítulo / linha de apoio (se existir).
+        Se não achar, cai no primeiro parágrafo do corpo.
+        """
+        selectors = [
+            ".subtitle",
+            ".sub-title",
+            ".lead",
+            ".summary",
+            ".excerpt",
+        ]
+
+        for selector in selectors:
+            try:
+                locator = page.locator(selector)
+                if locator.count() > 0:
+                    text = locator.first.inner_text().strip()
+                    if text and len(text) > 20:
+                        logger.info(
+                            f"[{self.name}] Description extracted: {text[:80]}..."
+                        )
+                        return text
+            except Exception as exc:
+                logger.warning(
+                    f"[{self.name}] Failed to extract description with '{selector}': {exc}"
+                )
+
+        first_paragraph = self._first_paragraph(page)
+        if first_paragraph:
+            logger.info(
+                f"[{self.name}] Description fallback from first paragraph: "
+                f"{first_paragraph[:80]}..."
+            )
+        return first_paragraph
+
+    def _first_paragraph(self, page) -> str:
+        """
+        Pega o primeiro parágrafo "decente" do artigo.
+        """
+        try:
+            locator = page.locator("article p")
+            if locator.count() == 0:
+                locator = page.locator("main article p")
+            if locator.count() == 0:
+                locator = page.locator("main p")
+
+            for i in range(locator.count()):
+                text = locator.nth(i).inner_text().strip()
+                if text and len(text) > 40:
+                    return text
+        except Exception:
+            pass
+        return ""
+
+    def _extract_body(self, page) -> str:
+        """
+        Extrai o texto completo do corpo da matéria.
+        """
+        selectors = [
+            "article .entry-content",
+            "article .post-content",
+            "article .content",
+            "article",
+            "main article",
+        ]
+
+        for selector in selectors:
+            try:
+                locator = page.locator(selector)
+                if locator.count() == 0:
+                    continue
+
+                root = locator.first
+
+                try:
+                    root.locator(
+                        "script, style, .ad, .advertisement, nav, footer"
+                    ).evaluate_all("els => els.forEach(e => e.remove())")
+                except Exception:
+                    pass
+
+                text = root.inner_text().strip()
+                if text and len(text) > 200:
+                    logger.info(
+                        f"[{self.name}] Body extracted ({len(text)} chars)"
+                    )
+                    return text
+            except Exception as exc:
+                logger.warning(
+                    f"[{self.name}] Failed to extract body with '{selector}': {exc}"
+                )
+
+        logger.error(f"[{self.name}] Failed to extract body")
+        return ""
+
+    def _extract_tags(self, page) -> list[str]:
+        """
+        Extrai tags/tópicos se houver bloco de tags.
+        Na CartaCapital geralmente aparecem como links em blocos de 'ENTENDA MAIS SOBRE', etc.
+        """
+        selectors = [
+            ".tags a",
+            ".post-tags a",
+            ".article-tags a",
+            ".c-tags a",
+            "[rel='tag']",
+        ]
+
+        tags: list[str] = []
+
+        for selector in selectors:
+            try:
+                loc = page.locator(selector)
+                count = loc.count()
+                if count == 0:
+                    continue
+
+                for i in range(count):
+                    tag_text = loc.nth(i).inner_text().strip()
+                    if tag_text and tag_text not in tags:
+                        tags.append(tag_text)
+            except Exception as exc:
+                logger.warning(
+                    f"[{self.name}] Failed to extract tags with '{selector}': {exc}"
+                )
+
+        return tags

spiders/__init__.py

@@ -12,6 +12,7 @@
 from spiders.gazetaDoPovo import GazetaDoPovoSpider
 from spiders.maisGoias import MaisGoiasSpider
 from spiders.aliadosBrasil import AliadosBrasilSpider
+from spiders.cartaCapital import CartaCapitalSpider
 
 __all__ = [
     BaseSpider,
@@ -28,4 +29,5 @@
     GazetaDoPovoSpider,
     MaisGoiasSpider,
     AliadosBrasilSpider,
+    CartaCapitalSpider,
 ]