sistema replicavel administracaopublica okfn-brasil#1247

nova spider ma_turilandia

correcao sufixo em seletor administracaopublica

Remoção de trechos e imports sobre tipagem

Corrige erros de digitação

Corrige lógica de geração de datas

Troca forma de reconhecimento de intervalos vazios

Resolve casos de redirecionamento

Corrige reconhecimento de layout do site

Corrige coleta de metadados

Reduz intervalo de recorrência de datas

data_collection/gazette/spiders/base/administracaopublica.py

@@ -0,0 +1,58 @@
+import re
+from datetime import datetime as dt
+
+from dateutil.rrule import DAILY, rrule
+from scrapy import Request
+
+from gazette.items import Gazette
+from gazette.spiders.base import BaseGazetteSpider
+
+
+class BaseAdministracaoPublicaSpider(BaseGazetteSpider):
+    """
+    Base spider for cities using the https://administracaopublica.com.br/diario-oficial?token= plataform.
+    Gazettes are also available in http://www.transparenciadministrativa.com.br/diario/diariov2.xhtml?token=.
+    """
+
+    allowed_domains = ["administracaopublica.com.br"]
+
+    def start_requests(self):
+        dates = list(
+            rrule(freq=DAILY, interval=10, dtstart=self.start_date, until=self.end_date)
+        )
+        dt_end_date = dt(self.end_date.year, self.end_date.month, self.end_date.day)
+        if dt_end_date not in dates:
+            dates.append(dt_end_date)
+
+        for i in range(len(dates) - 1):
+            start = dates[i].strftime("%Y-%m-%d")
+            end = dates[i + 1].strftime("%Y-%m-%d")
+            yield Request(
+                f"https://www.administracaopublica.com.br/diario-oficial?token={self.token}&de={start}&ate={end}"
+            )
+
+    def parse(self, response):
+        gazettes = response.css('[class*="diario_item_diario__"]')
+        for gazette in gazettes:
+            if "Nenhum resultado encontrado" not in response.text:
+                href = gazette.css(
+                    '[class*="generics_button_baixar__"]::attr(href)'
+                ).get()
+                pattern = gazette.css("::text").getall()
+
+                match pattern:
+                    case [edition, _, power, date, _]:
+                        pass
+                    case [edition, _, date, _]:
+                        power = ""
+                power_dict = {
+                    "EXECUTIVO": "executive",
+                    "LEGISLATIVO": "legislative",
+                }
+                yield Gazette(
+                    edition_number=re.search(r"(\d+)/", edition).group(1),
+                    date=dt.strptime(date, "%d/%m/%Y").date(),
+                    file_urls=[f"https://www.administracaopublica.com.br{href}"],
+                    is_extra_edition=power == "EXTRA",
+                    power=power_dict.get(power, "executive_legislative"),
+                )

data_collection/gazette/spiders/ma/ma_nova_iorque.py

@@ -0,0 +1,10 @@
+import datetime as dt
+
+from gazette.spiders.base.administracaopublica import BaseAdministracaoPublicaSpider
+
+
+class MaNovaIorqueSpider(BaseAdministracaoPublicaSpider):
+    TERRITORY_ID = "2107308"
+    name = "ma_nova_iorque"
+    start_date = dt.date(2017, 2, 15)
+    token = "4f1cf16edf5d73feaad4fec2a03c7c9e1cf536aa"

data_collection/gazette/spiders/ma/ma_peritoro.py

@@ -1,11 +1,10 @@
 import datetime as dt
 
-from gazette.spiders.base.aplus import BaseAplusSpider
+from gazette.spiders.base.administracaopublica import BaseAdministracaoPublicaSpider
 
 
-class MaPeritoroSpider(BaseAplusSpider):
+class MaPeritoroSpider(BaseAdministracaoPublicaSpider):
     TERRITORY_ID = "2108454"
     name = "ma_peritoro"
-    start_date = dt.date(2020, 1, 4)
-    allowed_domains = ["peritoro.ma.gov.br"]
-    url_base = "https://www.peritoro.ma.gov.br/diario/"
+    start_date = dt.date(2017, 1, 2)
+    token = "9de645b503b922df799865ffcb07a6ec7b9cb53e"

data_collection/gazette/spiders/ma/ma_turilandia.py

@@ -0,0 +1,10 @@
+import datetime as dt
+
+from gazette.spiders.base.administracaopublica import BaseAdministracaoPublicaSpider
+
+
+class MaTurilandiaSpider(BaseAdministracaoPublicaSpider):
+    TERRITORY_ID = "2112456"
+    name = "ma_turilandia"
+    start_date = dt.date(2021, 3, 11)
+    token = "9664abfc624b73571a05e874f98fd6d114834924"