diff --git a/.github/workflows/ci-cd.yml b/.github/workflows/ci-cd.yml
index ebbe7bc..65a0ef4 100755
--- a/.github/workflows/ci-cd.yml
+++ b/.github/workflows/ci-cd.yml
@@ -32,13 +32,15 @@ jobs:
poetry install --with dev
- name: Ruff
- run: poetry run ruff check app/ integrations/ utils.py tests/test_alo_source_parser.py
+ run: poetry run ruff check app/ integrations/ utils.py tests/test_alo_source_parser.py tests/test_imoti_source_parser.py
- name: Ruff format check
- run: poetry run ruff format --check app/ integrations/ utils.py tests/test_alo_source_parser.py
+ run: poetry run ruff format --check app/ integrations/ utils.py tests/test_alo_source_parser.py tests/test_imoti_source_parser.py
- - name: ALO source parser regression test
- run: poetry run python -m unittest discover -s tests -p "test_alo_source_parser.py"
+ - name: Source parser regression tests
+ run: |
+ poetry run python -m unittest discover -s tests -p "test_alo_source_parser.py"
+ poetry run python -m unittest discover -s tests -p "test_imoti_source_parser.py"
typing:
runs-on: ubuntu-latest
diff --git a/app/scraper/sources/__init__.py b/app/scraper/sources/__init__.py
index 27e2d5c..f8de65c 100644
--- a/app/scraper/sources/__init__.py
+++ b/app/scraper/sources/__init__.py
@@ -1,3 +1,4 @@
from app.scraper.sources.alo import AloSourceParser
+from app.scraper.sources.imoti import ImotiSourceParser
-__all__ = ["AloSourceParser"]
+__all__ = ["AloSourceParser", "ImotiSourceParser"]
diff --git a/app/scraper/sources/imoti.py b/app/scraper/sources/imoti.py
new file mode 100644
index 0000000..5fe394a
--- /dev/null
+++ b/app/scraper/sources/imoti.py
@@ -0,0 +1,348 @@
+from __future__ import annotations
+
+import re
+from typing import Callable, Protocol
+
+from bs4 import BeautifulSoup, Tag
+
+from app.scraper.models import ScrapedListing
+
+extract_names: Callable[[str], list[str]] | None
+looks_like_person_name: Callable[[str], bool] | None
+
+try:
+ from utils import (
+ extract_names as _extract_names,
+ looks_like_person_name as _looks_like_person_name,
+ )
+
+ extract_names = _extract_names
+ looks_like_person_name = _looks_like_person_name
+except ImportError: # pragma: no cover - fallback path for isolated runtimes
+ extract_names = None
+ looks_like_person_name = None
+
+EMAIL_RE = re.compile(r"[A-Z0-9._%+-]+@[A-Z0-9.-]+\.[A-Z]{2,}", flags=re.I)
+PROPERTY_NAME_TOKENS: tuple[str, ...] = (
+ "апартамент",
+ "апартаменти",
+ "едностаен",
+ "двустаен",
+ "тристаен",
+ "четиристаен",
+ "многостаен",
+ "квартира",
+ "квартири",
+ "аренда",
+ "наем",
+ "оренда",
+ "жилье",
+ "житло",
+ "apartment",
+ "flat",
+ "rent",
+)
+
+
+class ImotiParserContext(Protocol):
+ @staticmethod
+ def _attr_str(tag: Tag | None, name: str, default: str = "") -> str: ...
+
+ def _normalize_link(self, base_url: str, href: str | None) -> str: ...
+
+ def _clean_text(self, value: str) -> str: ...
+
+ def _is_listing_candidate(self, title: str) -> bool: ...
+
+ def _pick_card_container(self, anchor: Tag) -> Tag: ...
+
+ def _extract_ad_id(self, link: str) -> str: ...
+
+ def _extract_price(self, text: str) -> str: ...
+
+ def _extract_location(self, text: str) -> str: ...
+
+ def _extract_size(self, text: str) -> str: ...
+
+ def _extract_image(self, card: Tag | None, base_url: str) -> str: ...
+
+ def _extract_seller_name(self, card: Tag | None) -> str: ...
+
+ def _detect_ad_type(self, seller_name: str) -> str: ...
+
+ def _passes_filters(self, listing: ScrapedListing) -> bool: ...
+
+ def _extract_phone_from_text(self, text: str) -> str: ...
+
+
+class ImotiSourceParser:
+ def __init__(
+ self,
+ context: ImotiParserContext,
+ source_site: str,
+ selectors: dict[str, str],
+ ) -> None:
+ self._context = context
+ self._source_site = source_site
+ self._selectors = selectors
+
+ def parse_listing_page(self, html: str, base_url: str) -> list[ScrapedListing]:
+ soup = BeautifulSoup(html, "html.parser")
+ exact_cards = soup.select("article.product-classic")
+ if exact_cards:
+ exact_results: list[ScrapedListing] = []
+ seen_ids: set[str] = set()
+ for article in exact_cards:
+ listing = self._parse_exact_card(article, base_url)
+ if listing is None or listing.ad_id in seen_ids:
+ continue
+ seen_ids.add(listing.ad_id)
+ if self._context._passes_filters(listing):
+ exact_results.append(listing)
+ return exact_results
+
+ soup = BeautifulSoup(html, "lxml")
+ links = soup.select(self._selectors.get("link", "a[href*='/наеми/']"))
+ results: list[ScrapedListing] = []
+ seen_links: set[str] = set()
+
+ for link_el in links:
+ link = self._context._normalize_link(
+ base_url,
+ self._context._attr_str(link_el, "href"),
+ )
+ if not link or link in seen_links:
+ continue
+ seen_links.add(link)
+
+ title = self._context._clean_text(link_el.get_text(" ", strip=True))
+ if not self._context._is_listing_candidate(title):
+ continue
+
+ card = self._context._pick_card_container(link_el)
+ card_text = card.get_text("\n", strip=True) if card else title
+ seller_name = self._context._extract_seller_name(card)
+ listing = ScrapedListing(
+ ad_id=self._context._extract_ad_id(link),
+ title=title,
+ price=self._context._extract_price(card_text),
+ location=self._context._extract_location(card_text),
+ size=self._context._extract_size(card_text),
+ link=link,
+ image_url=self._context._extract_image(card, base_url),
+ seller_name=seller_name,
+ ad_type=self._context._detect_ad_type(seller_name),
+ )
+ if self._context._passes_filters(listing):
+ results.append(listing)
+
+ return results
+
+ def _parse_exact_card(self, article: Tag, base_url: str) -> ScrapedListing | None:
+ context = self._context
+ title_anchor = article.select_one("h4.product-classic-title a")
+ if title_anchor is None:
+ return None
+
+ link = context._normalize_link(base_url, context._attr_str(title_anchor, "href"))
+ if not link:
+ return None
+
+ ad_id = context._extract_ad_id(link)
+ if not ad_id:
+ return None
+
+ title = context._clean_text(title_anchor.get_text(strip=True))
+ if not context._is_listing_candidate(title):
+ return None
+
+ price_el = article.select_one(".product-classic-price")
+ if price_el:
+ price_lines = [
+ line.strip() for line in price_el.get_text("\n").splitlines() if line.strip()
+ ]
+ price = price_lines[0] if price_lines else ""
+ else:
+ price = ""
+
+ location_el = article.select_one(".btext")
+ location = context._clean_text(location_el.get_text(strip=True) if location_el else "")
+
+ size = ""
+ for li in article.select(".product-classic-list li"):
+ text = context._clean_text(li.get_text(strip=True))
+ if "кв.м." in text or "м²" in text:
+ size = text.replace("м2", "м²")
+ break
+
+ seller_name = self._extract_seller_name(article)
+ phone = self._extract_phone(article)
+
+ return ScrapedListing(
+ ad_id=ad_id,
+ title=title,
+ price=price,
+ location=location,
+ size=size,
+ link=link,
+ image_url=context._extract_image(article, base_url),
+ source_site=self._source_site,
+ phone=phone,
+ seller_name=seller_name,
+ ad_type=context._detect_ad_type(seller_name),
+ )
+
+ def enrich_detail(self, soup: BeautifulSoup, listing: ScrapedListing) -> None:
+ context = self._context
+
+ if not self._looks_like_real_seller_name(listing.seller_name):
+ listing.seller_name = ""
+
+ for block in soup.select("div.block-person-link"):
+ icon = block.select_one("span.icon")
+ icon_classes = context._attr_str(icon, "class")
+ block_text = context._clean_text(block.get_text(" ", strip=True))
+
+ if "mdi-account" in icon_classes and block_text:
+ listing.seller_name = block_text
+ if not listing.contact_name or listing.contact_name == "-":
+ listing.contact_name = block_text
+ continue
+
+ if "mdi-phone" in icon_classes and not listing.phone:
+ tel = block.select_one("a[href^='tel:']")
+ phone_source = context._attr_str(tel, "href") if tel is not None else block_text
+ phone = context._extract_phone_from_text(phone_source)
+ if phone:
+ listing.phone = phone
+ continue
+
+ if "mdi-email" in icon_classes and (
+ not listing.contact_email or listing.contact_email == "-"
+ ):
+ email_anchor = block.select_one("a[href]")
+ email_text = context._clean_text(
+ email_anchor.get_text(" ", strip=True)
+ if email_anchor is not None
+ else block_text
+ )
+ email_match = EMAIL_RE.search(email_text)
+ if email_match:
+ listing.contact_email = email_match.group(0)
+
+ if not listing.seller_name:
+ for selector in (
+ "h1 a",
+ ".product-title a",
+ ".property-title",
+ "[class*='owner']",
+ "[class*='seller']",
+ "[class*='agency']",
+ ):
+ element = soup.select_one(selector)
+ if element is None:
+ continue
+ candidate = context._clean_text(element.get_text(" ", strip=True))
+ if self._looks_like_real_seller_name(candidate):
+ listing.seller_name = candidate
+ break
+
+ if not listing.contact_name or listing.contact_name == "-":
+ listing.contact_name = self._extract_contact_name(soup)
+
+ if listing.seller_name:
+ listing.ad_type = context._detect_ad_type(listing.seller_name)
+
+ def _extract_phone(self, article: Tag) -> str:
+ context = self._context
+ tel_link = article.select_one('a[href^="tel:"]')
+ if tel_link:
+ phone = context._extract_phone_from_text(context._attr_str(tel_link, "href"))
+ if phone:
+ return phone
+
+ for selector in (
+ "[class*='phone']",
+ "[class*='tel']",
+ ".contact-phone",
+ ".phone-number",
+ "span.phone",
+ ):
+ element = article.select_one(selector)
+ if element is not None:
+ phone = context._extract_phone_from_text(element.get_text(" ", strip=True))
+ if phone:
+ return phone
+
+ return context._extract_phone_from_text(article.get_text(" ", strip=True))
+
+ def _extract_seller_name(self, article: Tag) -> str:
+ context = self._context
+ for selector in (
+ ".product-classic-agency",
+ ".agency-name",
+ ".seller-name",
+ "[class*='agency']",
+ "[class*='seller']",
+ ".block-info h3",
+ ):
+ element = article.select_one(selector)
+ if element is None:
+ continue
+ name = context._clean_text(element.get_text(strip=True))
+ if self._looks_like_real_seller_name(name):
+ return name
+
+ text_content = article.get_text(" ", strip=True)
+ if extract_names is not None:
+ names = extract_names(text_content)
+ if names:
+ return names[0]
+
+ if looks_like_person_name is not None:
+ chunks = [
+ context._clean_text(chunk) for chunk in text_content.split(" ") if chunk.strip()
+ ]
+ for chunk in chunks:
+ if looks_like_person_name(chunk):
+ return chunk
+ return ""
+
+ def _extract_contact_name(self, soup: BeautifulSoup) -> str:
+ context = self._context
+ for block in soup.select("div.block-person-link"):
+ text = context._clean_text(block.get_text(" ", strip=True))
+ if not text:
+ continue
+ if looks_like_person_name is not None and looks_like_person_name(text):
+ return text
+
+ for selector in (
+ ".contact-name",
+ ".agent-name",
+ ".block-agent-contact .name",
+ ".contact-person",
+ "[class*='contact-name']",
+ ):
+ element = soup.select_one(selector)
+ if element is None:
+ continue
+ candidate = context._clean_text(element.get_text(" ", strip=True))
+ if not candidate:
+ continue
+ if looks_like_person_name is None or looks_like_person_name(candidate):
+ return candidate
+
+ return "-"
+
+ def _looks_like_real_seller_name(self, value: str) -> bool:
+ normalized = self._context._clean_text(value).lower()
+ if not normalized:
+ return False
+ if "*" in normalized:
+ return False
+ if any(word in normalized for word in PROPERTY_NAME_TOKENS):
+ return False
+ if any(token in normalized for token in ("кв.м", "месец", "eur", "лв", "€", "$")):
+ return False
+ return True
diff --git a/app/services/async_scraper.py b/app/services/async_scraper.py
index c822b2f..18d61e5 100755
--- a/app/services/async_scraper.py
+++ b/app/services/async_scraper.py
@@ -14,28 +14,21 @@
from app.core.logging import get_logger
from app.scraper.models import ScrapedListing
from app.scraper.sources.alo import AloSourceParser
+from app.scraper.sources.imoti import ImotiSourceParser
-extract_names: Callable[[str], list[str]] | None
extract_phone_numbers: Callable[[str], list[str]] | None
-looks_like_person_name: Callable[[str], bool] | None
normalize_phone_number: Callable[[str], str] | None
try:
from utils import (
- extract_names as _extract_names,
extract_phone_numbers as _extract_phone_numbers,
- looks_like_person_name as _looks_like_person_name,
normalize_phone_number as _normalize_phone_number,
)
- extract_names = _extract_names
extract_phone_numbers = _extract_phone_numbers
- looks_like_person_name = _looks_like_person_name
normalize_phone_number = _normalize_phone_number
except ImportError: # pragma: no cover - fallback path for isolated runtimes
- extract_names = None
extract_phone_numbers = None
- looks_like_person_name = None
normalize_phone_number = None
logger = get_logger("async_scraper")
@@ -107,6 +100,11 @@ def __init__(self, site_config: SiteConfig, settings: Settings) -> None:
)
self._proxy_index = 0
self._alo_parser = AloSourceParser(self, site_config.name)
+ self._imoti_parser = ImotiSourceParser(
+ self,
+ site_config.name,
+ site_config.selectors,
+ )
def _get_client_kwargs(self) -> dict[str, Any]:
headers = {
@@ -290,7 +288,7 @@ def _build_page_url(self, page: int) -> str:
def _parse_listing_page(self, html: str, base_url: str) -> list[ScrapedListing]:
parsers = {
- "imoti.bg": self._parse_imoti_listing_page,
+ "imoti.bg": self._imoti_parser.parse_listing_page,
"alo.bg": self._alo_parser.parse_listing_page,
"dom.ria.com": self._parse_generic_anchor_page,
"olx.ua": self._parse_generic_anchor_page,
@@ -299,107 +297,6 @@ def _parse_listing_page(self, html: str, base_url: str) -> list[ScrapedListing]:
parser = parsers.get(self.site_config.name, self._parse_generic_cards)
return parser(html=html, base_url=base_url)
- def _parse_imoti_listing_page(self, html: str, base_url: str) -> list[ScrapedListing]:
- soup = BeautifulSoup(html, "html.parser")
- exact_cards = soup.select("article.product-classic")
- if exact_cards:
- exact_results: list[ScrapedListing] = []
- seen_ids: set[str] = set()
- for article in exact_cards:
- listing = self._parse_imoti_card_exact(article, base_url)
- if listing is None or listing.ad_id in seen_ids:
- continue
- seen_ids.add(listing.ad_id)
- if self._passes_filters(listing):
- exact_results.append(listing)
- return exact_results
-
- soup = BeautifulSoup(html, "lxml")
- links = soup.select(self.site_config.selectors.get("link", "a[href*='/наеми/']"))
- results: list[ScrapedListing] = []
- seen_links: set[str] = set()
-
- for link_el in links:
- link = self._normalize_link(base_url, self._attr_str(link_el, "href"))
- if not link or link in seen_links:
- continue
- seen_links.add(link)
-
- title = self._clean_text(link_el.get_text(" ", strip=True))
- if not self._is_listing_candidate(title):
- continue
-
- card = self._pick_card_container(link_el)
- card_text = card.get_text("\n", strip=True) if card else title
- listing = ScrapedListing(
- ad_id=self._extract_ad_id(link),
- title=title,
- price=self._extract_price(card_text),
- location=self._extract_location(card_text),
- size=self._extract_size(card_text),
- link=link,
- image_url=self._extract_image(card, base_url),
- seller_name=self._extract_seller_name(card),
- ad_type=self._detect_ad_type(self._extract_seller_name(card)),
- )
- if self._passes_filters(listing):
- results.append(listing)
-
- return results
-
- def _parse_imoti_card_exact(self, article: Tag, base_url: str) -> ScrapedListing | None:
- title_anchor = article.select_one("h4.product-classic-title a")
- if title_anchor is None:
- return None
-
- link = self._normalize_link(base_url, self._attr_str(title_anchor, "href"))
- if not link:
- return None
-
- ad_id = self._extract_ad_id(link)
- if not ad_id:
- return None
-
- title = self._clean_text(title_anchor.get_text(strip=True))
- if not self._is_listing_candidate(title):
- return None
-
- price_el = article.select_one(".product-classic-price")
- if price_el:
- price_lines = [
- line.strip() for line in price_el.get_text("\n").splitlines() if line.strip()
- ]
- price = price_lines[0] if price_lines else ""
- else:
- price = ""
-
- location_el = article.select_one(".btext")
- location = self._clean_text(location_el.get_text(strip=True) if location_el else "")
-
- size = ""
- for li in article.select(".product-classic-list li"):
- text = self._clean_text(li.get_text(strip=True))
- if "кв.м." in text or "м²" in text:
- size = text.replace("м2", "м²")
- break
-
- seller_name = self._extract_seller_name_from_imoti_card(article)
- phone = self._extract_phone_from_imoti_card(article)
-
- return ScrapedListing(
- ad_id=ad_id,
- title=title,
- price=price,
- location=location,
- size=size,
- link=link,
- image_url=self._extract_image(article, base_url),
- source_site=self.site_config.name,
- phone=phone,
- seller_name=seller_name,
- ad_type=self._detect_ad_type(seller_name),
- )
-
def _parse_generic_cards(self, html: str, base_url: str) -> list[ScrapedListing]:
soup = BeautifulSoup(html, "lxml")
cards = soup.select(self.site_config.selectors.get("card", "article, li, div"))
@@ -517,7 +414,7 @@ async def _enrich_one(
soup = BeautifulSoup(response.text, "lxml")
if self.site_config.name == "imoti.bg":
- self._enrich_imoti_detail(soup, listing)
+ self._imoti_parser.enrich_detail(soup, listing)
elif self.site_config.name == "alo.bg":
self._alo_parser.enrich_detail(soup, listing)
else:
@@ -550,101 +447,6 @@ def _enrich_generic_detail(self, soup: BeautifulSoup, listing: ScrapedListing) -
if not listing.contact_name or listing.contact_name == "-":
listing.contact_name = listing.seller_name or self._guess_contact_name(text)
- def _enrich_imoti_detail(self, soup: BeautifulSoup, listing: ScrapedListing) -> None:
- if not self._looks_like_real_seller_name(listing.seller_name):
- listing.seller_name = ""
-
- for block in soup.select("div.block-person-link"):
- icon = block.select_one("span.icon")
- icon_classes = self._attr_str(icon, "class")
- block_text = self._clean_text(block.get_text(" ", strip=True))
-
- if "mdi-account" in icon_classes and block_text:
- listing.seller_name = block_text
- if not listing.contact_name or listing.contact_name == "-":
- listing.contact_name = block_text
- continue
-
- if "mdi-phone" in icon_classes and not listing.phone:
- tel = block.select_one("a[href^='tel:']")
- phone_source = self._attr_str(tel, "href") if tel is not None else block_text
- phone = self._extract_phone_from_text(phone_source)
- if phone:
- listing.phone = phone
- continue
-
- if "mdi-email" in icon_classes and (
- not listing.contact_email or listing.contact_email == "-"
- ):
- email_anchor = block.select_one("a[href]")
- email_text = self._clean_text(
- email_anchor.get_text(" ", strip=True)
- if email_anchor is not None
- else block_text
- )
- email_match = EMAIL_RE.search(email_text)
- if email_match:
- listing.contact_email = email_match.group(0)
-
- if not listing.seller_name:
- for sel in (
- "h1 a",
- ".product-title a",
- ".property-title",
- "[class*='owner']",
- "[class*='seller']",
- "[class*='agency']",
- ):
- el = soup.select_one(sel)
- if el is not None:
- candidate = self._clean_text(el.get_text(" ", strip=True))
- if self._looks_like_real_seller_name(candidate):
- listing.seller_name = candidate
- break
-
- if not listing.contact_name or listing.contact_name == "-":
- listing.contact_name = self._extract_contact_name_from_detail_soup(soup)
- if listing.seller_name:
- listing.ad_type = self._detect_ad_type(listing.seller_name)
-
- def _looks_like_real_seller_name(self, value: str) -> bool:
- normalized = self._clean_text(value).lower()
- if not normalized:
- return False
- if "*" in normalized:
- return False
- if any(word in normalized for word in APARTMENT_KEYWORDS):
- return False
- if any(token in normalized for token in ("кв.м", "месец", "eur", "лв", "€", "$")):
- return False
- return True
-
- def _extract_contact_name_from_detail_soup(self, soup: BeautifulSoup) -> str:
- for block in soup.select("div.block-person-link"):
- text = self._clean_text(block.get_text(" ", strip=True))
- if not text:
- continue
- if looks_like_person_name is not None and looks_like_person_name(text):
- return text
-
- for sel in (
- ".contact-name",
- ".agent-name",
- ".block-agent-contact .name",
- ".contact-person",
- "[class*='contact-name']",
- ):
- el = soup.select_one(sel)
- if el is None:
- continue
- candidate = self._clean_text(el.get_text(" ", strip=True))
- if not candidate:
- continue
- if looks_like_person_name is None or looks_like_person_name(candidate):
- return candidate
-
- return "-"
-
def _extract_contact_email_from_detail_soup(self, soup: BeautifulSoup) -> str:
for root in soup.select("div.block-person-link, div.block-info, div.block-agent"):
mail = root.select_one("a[href^='mailto:']")
@@ -846,59 +648,6 @@ def _guess_contact_name(self, text: str) -> str:
return sentence[:120]
return "-"
- def _extract_phone_from_imoti_card(self, article: Tag) -> str:
- tel_link = article.select_one('a[href^="tel:"]')
- if tel_link:
- phone = self._extract_phone_from_text(self._attr_str(tel_link, "href"))
- if phone:
- return phone
-
- for selector in (
- "[class*='phone']",
- "[class*='tel']",
- ".contact-phone",
- ".phone-number",
- "span.phone",
- ):
- el = article.select_one(selector)
- if el is not None:
- phone = self._extract_phone_from_text(el.get_text(" ", strip=True))
- if phone:
- return phone
-
- full_text = article.get_text(" ", strip=True)
- return self._extract_phone_from_text(full_text)
-
- def _extract_seller_name_from_imoti_card(self, article: Tag) -> str:
- for selector in (
- ".product-classic-agency",
- ".agency-name",
- ".seller-name",
- "[class*='agency']",
- "[class*='seller']",
- ".block-info h3",
- ):
- el = article.select_one(selector)
- if el is not None:
- name = self._clean_text(el.get_text(strip=True))
- if self._looks_like_real_seller_name(name):
- return name
-
- text_content = article.get_text(" ", strip=True)
- if extract_names is not None:
- names = extract_names(text_content)
- if names:
- return names[0]
-
- if looks_like_person_name is not None:
- chunks = [
- self._clean_text(chunk) for chunk in text_content.split(" ") if chunk.strip()
- ]
- for chunk in chunks:
- if looks_like_person_name(chunk):
- return chunk
- return ""
-
def _title_from_url(self, link: str) -> str:
path = urlparse(link).path.rstrip("/").split("/")[-1]
path = re.sub(r"[-_]+", " ", path)
diff --git a/tests/test_imoti_source_parser.py b/tests/test_imoti_source_parser.py
new file mode 100644
index 0000000..cb2eaa2
--- /dev/null
+++ b/tests/test_imoti_source_parser.py
@@ -0,0 +1,132 @@
+from __future__ import annotations
+
+import unittest
+
+from bs4 import BeautifulSoup
+
+from app.core.config import Settings, SiteConfig
+from app.services.async_scraper import BaseScraper
+
+
+class ImotiSourceParserTests(unittest.TestCase):
+ def setUp(self) -> None:
+ self.site = SiteConfig(
+ name="imoti.bg",
+ base_url="https://imoti.bg/наеми/page:{page}",
+ max_pages=1,
+ selectors={
+ "link": "h4 a[href*='/наеми/'], a[href*='/наеми/']",
+ "seller": "[class*='agency'], [class*='seller']",
+ },
+ allowed_domains=["imoti.bg"],
+ )
+ self.settings = Settings(
+ sites=[self.site],
+ city_filter=None,
+ scrape_detail_pages=False,
+ _env_file=None,
+ )
+ self.scraper = BaseScraper(self.site, self.settings)
+
+ def test_exact_card_parsing_is_delegated_to_imoti_parser(self) -> None:
+ html = """
+
+
+
+
+ 900 EUR
+ Допълнителен текст
+
+ София, Лозенец
+
+ Example Agency
+ +359 888 123 456
+ 
+
+
+
+
+ 900 EUR
+ София, Лозенец
+
+
+ """
+
+ listings = self.scraper._parse_listing_page(
+ html,
+ base_url="https://imoti.bg/наеми/",
+ )
+
+ self.assertEqual(len(listings), 1)
+ listing = listings[0]
+ self.assertEqual(listing.ad_id, "123456")
+ self.assertEqual(listing.title, "Двустаен апартамент под наем")
+ self.assertEqual(listing.price, "900 EUR")
+ self.assertEqual(listing.location, "София, Лозенец")
+ self.assertEqual(listing.size, "72 кв.м.")
+ self.assertEqual(listing.seller_name, "Example Agency")
+ self.assertEqual(listing.ad_type, "agency")
+ self.assertTrue(listing.phone)
+ self.assertEqual(listing.source_site, "imoti.bg")
+ self.assertEqual(
+ listing.link,
+ "https://imoti.bg/наеми/двустаен-апартамент-123456.htm",
+ )
+
+ def test_detail_enrichment_remains_source_specific(self) -> None:
+ listing = self.scraper._parse_listing_page(
+ """
+
+
+ 800 EUR
+ София
+
+ """,
+ base_url="https://imoti.bg/наеми/",
+ )[0]
+
+ soup = BeautifulSoup(
+ """
+
+
+
+ Example Agency
+
+
+
+
+ """,
+ "lxml",
+ )
+
+ self.scraper._imoti_parser.enrich_detail(soup, listing)
+
+ self.assertEqual(listing.seller_name, "Example Agency")
+ self.assertEqual(listing.contact_name, "Example Agency")
+ self.assertTrue(listing.phone)
+ self.assertEqual(listing.contact_email, "agent@example.com")
+ self.assertEqual(listing.ad_type, "agency")
+
+
+if __name__ == "__main__":
+ unittest.main()