Add an input URL list parameter

tests/test_ecommerce.py

@@ -395,6 +395,22 @@ def test_metadata():
                     "title": "URL",
                     "type": "string",
                 },
+                "urls": {
+                    "anyOf": [
+                        {"items": {"type": "string"}, "type": "array"},
+                        {"type": "null"},
+                    ],
+                    "default": None,
+                    "description": (
+                        "Initial URLs for the crawl, separated by new lines. Enter the "
+                        "full URL including http(s), you can copy and paste it from your "
+                        "browser. Example: https://toscrape.com/"
+                    ),
+                    "exclusiveRequired": True,
+                    "group": "inputs",
+                    "title": "URLs",
+                    "widget": "textarea",
+                },
                 "seed_url": {
                     "default": "",
                     "description": (
@@ -698,6 +714,18 @@ def test_input_multiple():
         EcommerceSpider.from_crawler(
             crawler,
             url="https://a.example",
+            urls=["https://b.example"],
+        )
+    with pytest.raises(ValueError):
+        EcommerceSpider.from_crawler(
+            crawler,
+            url="https://a.example",
+            seed_url="https://b.example",
+        )
+    with pytest.raises(ValueError):
+        EcommerceSpider.from_crawler(
+            crawler,
+            urls=["https://a.example"],
             seed_url="https://b.example",
         )
 
@@ -708,6 +736,47 @@ def test_url_invalid():
         EcommerceSpider.from_crawler(crawler, url="foo")
 
 
+def test_urls(caplog):
+    crawler = get_crawler()
+    url = "https://example.com"
+
+    spider = EcommerceSpider.from_crawler(crawler, urls=[url])
+    start_requests = list(spider.start_requests())
+    assert len(start_requests) == 1
+    assert start_requests[0].url == url
+    assert start_requests[0].callback == spider.parse_navigation
+
+    spider = EcommerceSpider.from_crawler(crawler, urls=url)
+    start_requests = list(spider.start_requests())
+    assert len(start_requests) == 1
+    assert start_requests[0].url == url
+    assert start_requests[0].callback == spider.parse_navigation
+
+    caplog.clear()
+    spider = EcommerceSpider.from_crawler(
+        crawler,
+        urls="https://a.example\n \nhttps://b.example\nhttps://c.example\nfoo\n\n",
+    )
+    assert "'foo', from the 'urls' spider argument, is not a valid URL" in caplog.text
+    start_requests = list(spider.start_requests())
+    assert len(start_requests) == 3
+    assert all(
+        request.callback == spider.parse_navigation for request in start_requests
+    )
+    assert start_requests[0].url == "https://a.example"
+    assert start_requests[1].url == "https://b.example"
+    assert start_requests[2].url == "https://c.example"
+
+    caplog.clear()
+    with pytest.raises(ValueError):
+        spider = EcommerceSpider.from_crawler(
+            crawler,
+            urls="foo\nbar",
+        )
+    assert "'foo', from the 'urls' spider argument, is not a valid URL" in caplog.text
+    assert "'bar', from the 'urls' spider argument, is not a valid URL" in caplog.text
+
+
 def test_seed_url():
     crawler = get_crawler()
     url = "https://example.com"

zyte_spider_templates/params.py

@@ -1,7 +1,9 @@
+import re
 from enum import Enum
-from typing import Optional
+from logging import getLogger
+from typing import List, Optional, Union
 
-from pydantic import BaseModel, Field
+from pydantic import BaseModel, Field, field_validator
 
 from zyte_spider_templates._geolocations import (
     GEOLOCATION_OPTIONS_WITH_CODE,
@@ -11,6 +13,8 @@
 
 from .utils import _URL_PATTERN
 
+logger = getLogger(__name__)
+
 
 @document_enum
 class ExtractFrom(str, Enum):
@@ -107,3 +111,51 @@ class UrlParam(BaseModel):
             "exclusiveRequired": True,
         },
     )
+
+
+class UrlsParam(BaseModel):
+    urls: Optional[List[str]] = Field(
+        title="URLs",
+        description=(
+            "Initial URLs for the crawl, separated by new lines. Enter the "
+            "full URL including http(s), you can copy and paste it from your "
+            "browser. Example: https://toscrape.com/"
+        ),
+        default=None,
+        json_schema_extra={
+            "group": "inputs",
+            "exclusiveRequired": True,
+            "widget": "textarea",
+        },
+    )
+
+    @field_validator("urls", mode="before")
+    @classmethod
+    def validate_url_list(cls, value: Union[List[str], str]) -> List[str]:
+        """Validate a list of URLs.
+
+        If a string is received as input, it is split into multiple strings
+        on new lines.
+
+        List items that do not match a URL pattern trigger a warning and are
+        removed from the list. If all URLs are invalid, validation fails.
+        """
+        if isinstance(value, str):
+            value = value.split("\n")
+        if not value:
+            return value
+        result = []
+        for v in value:
+            v = v.strip()
+            if not v:
+                continue
+            if not re.search(_URL_PATTERN, v):
+                logger.warning(
+                    f"{v!r}, from the 'urls' spider argument, is not a "
+                    f"valid URL and will be ignored."
+                )
+                continue
+            result.append(v)
+        if not result:
+            raise ValueError(f"No valid URL found in {value!r}")
+        return result

zyte_spider_templates/spiders/base.py

@@ -11,19 +11,21 @@
     MaxRequestsParam,
     SeedUrlParam,
     UrlParam,
+    UrlsParam,
 )
 
 # Higher priority than command-line-defined settings (40).
 ARG_SETTING_PRIORITY: int = 50
 
-_INPUT_FIELDS = ("url", "seed_url")
+_INPUT_FIELDS = ("url", "urls", "seed_url")
 
 
 class BaseSpiderParams(
     ExtractFromParam,
     MaxRequestsParam,
     GeolocationParam,
     SeedUrlParam,
+    UrlsParam,
     UrlParam,
     BaseModel,
 ):

zyte_spider_templates/spiders/ecommerce.py

@@ -105,6 +105,8 @@ def _init_input(self):
             urls = load_url_list(response.text)
             self.logger.info(f"Loaded {len(urls)} initial URLs from {seed_url}.")
             self.start_urls = urls
+        elif self.args.urls:
+            self.start_urls = self.args.urls
         else:
             self.start_urls = [self.args.url]
         self.allowed_domains = list(set(get_domain(url) for url in self.start_urls))