Spaces:
Running
Running
Update app.py
Browse files
app.py
CHANGED
|
@@ -239,23 +239,24 @@ def get_japan_originator(ing_ja, page):
|
|
| 239 |
log, brands, companies = [], set(), set()
|
| 240 |
try:
|
| 241 |
log.append(f"1. 前往 PMDA (搜尋: {ing_ja})...")
|
| 242 |
-
page.goto("https://www.pmda.go.jp/PmdaSearch/iyakuSearch/", timeout=
|
| 243 |
|
| 244 |
-
log.append("2.
|
| 245 |
-
|
| 246 |
-
search_input
|
| 247 |
-
search_input.
|
|
|
|
| 248 |
|
| 249 |
-
log.append("3. 監聽新
|
| 250 |
with page.expect_popup() as popup_info:
|
| 251 |
search_input.press("Enter")
|
| 252 |
popup = popup_info.value
|
| 253 |
|
| 254 |
-
log.append("4. 等待新視窗表格並掃描
|
| 255 |
current_page = 1
|
| 256 |
|
| 257 |
-
# 💡 新增:自動翻頁迴圈
|
| 258 |
while True:
|
|
|
|
| 259 |
popup.wait_for_selector('table#ResultList, .errormsg, .non-result', timeout=15000)
|
| 260 |
soup = BeautifulSoup(popup.content(), 'html.parser')
|
| 261 |
table = soup.find('table', id='ResultList')
|
|
@@ -267,6 +268,7 @@ def get_japan_originator(ing_ja, page):
|
|
| 267 |
tds = tr.find_all('td')
|
| 268 |
if len(tds) >= 3:
|
| 269 |
title = tds[1].get_text(strip=True)
|
|
|
|
| 270 |
if not is_generic(title, "", ing_ja):
|
| 271 |
brands.add(clean_brand_name(title))
|
| 272 |
companies.add(tds[2].get_text(separator=" ", strip=True).replace('製造販売元/', ''))
|
|
@@ -274,15 +276,14 @@ def get_japan_originator(ing_ja, page):
|
|
| 274 |
log.append(f"❌ 第 {current_page} 頁未出現 ResultList。")
|
| 275 |
break
|
| 276 |
|
| 277 |
-
#
|
| 278 |
current_page += 1
|
| 279 |
next_page_link = popup.locator(f'a[href="javascript:changePg({current_page});"]')
|
| 280 |
|
| 281 |
if next_page_link.count() > 0:
|
| 282 |
-
log.append(f" -> 發現第 {current_page} 頁,
|
| 283 |
-
next_page_link.first.click()
|
| 284 |
-
popup.
|
| 285 |
-
popup.wait_for_timeout(1500) # 給予表格重新渲染的緩衝時間
|
| 286 |
else:
|
| 287 |
log.append(" -> 已無下一頁,結束掃描。")
|
| 288 |
break
|
|
@@ -298,30 +299,24 @@ def get_switzerland_originator(ing_de, page):
|
|
| 298 |
log, brands, companies = [], set(), set()
|
| 299 |
try:
|
| 300 |
log.append(f"1. 前往 Swissmedicinfo 搜尋頁面 (搜尋: {ing_de})...")
|
| 301 |
-
page.goto("https://swissmedicinfo.ch/SearchPage", timeout=
|
| 302 |
|
| 303 |
-
|
| 304 |
-
|
| 305 |
-
|
| 306 |
-
label.wait_for(state="visible", timeout=15000)
|
| 307 |
|
| 308 |
-
|
| 309 |
-
|
| 310 |
-
|
| 311 |
-
ing_input = page.locator(f'#{input_id}')
|
| 312 |
-
else:
|
| 313 |
-
log.append(" -> 找不到 for 屬性,啟動備用相鄰節點定位...")
|
| 314 |
-
ing_input = page.locator('input[type="text"]').nth(1)
|
| 315 |
-
|
| 316 |
-
ing_input.fill(ing_de)
|
| 317 |
|
| 318 |
-
log.append("3. 點擊
|
| 319 |
-
search_btn = page.locator('button'
|
| 320 |
-
search_btn.click()
|
| 321 |
|
| 322 |
-
log.append("4. 等待
|
| 323 |
-
page.
|
| 324 |
-
|
|
|
|
| 325 |
|
| 326 |
soup = BeautifulSoup(page.content(), 'html.parser')
|
| 327 |
cards = soup.find_all('div', class_=re.compile('medicament-card'))
|
|
|
|
| 239 |
log, brands, companies = [], set(), set()
|
| 240 |
try:
|
| 241 |
log.append(f"1. 前往 PMDA (搜尋: {ing_ja})...")
|
| 242 |
+
page.goto("https://www.pmda.go.jp/PmdaSearch/iyakuSearch/", timeout=45000, wait_until="domcontentloaded")
|
| 243 |
|
| 244 |
+
log.append("2. 準確定位置首頁搜尋框 (id='txtName')...")
|
| 245 |
+
# 依照您提供的 HTML 結構,直接鎖定 txtName
|
| 246 |
+
search_input = page.locator('input#txtName')
|
| 247 |
+
search_input.wait_for(state="attached", timeout=15000)
|
| 248 |
+
search_input.fill(ing_ja, force=True)
|
| 249 |
|
| 250 |
+
log.append("3. 觸發搜尋並監聽彈出新視窗...")
|
| 251 |
with page.expect_popup() as popup_info:
|
| 252 |
search_input.press("Enter")
|
| 253 |
popup = popup_info.value
|
| 254 |
|
| 255 |
+
log.append("4. 等待新視窗表格並啟動自動翻頁掃描...")
|
| 256 |
current_page = 1
|
| 257 |
|
|
|
|
| 258 |
while True:
|
| 259 |
+
# 等待表格出現
|
| 260 |
popup.wait_for_selector('table#ResultList, .errormsg, .non-result', timeout=15000)
|
| 261 |
soup = BeautifulSoup(popup.content(), 'html.parser')
|
| 262 |
table = soup.find('table', id='ResultList')
|
|
|
|
| 268 |
tds = tr.find_all('td')
|
| 269 |
if len(tds) >= 3:
|
| 270 |
title = tds[1].get_text(strip=True)
|
| 271 |
+
# 這裡會自動把帶有「括號」或廠商名稱的學名藥濾掉
|
| 272 |
if not is_generic(title, "", ing_ja):
|
| 273 |
brands.add(clean_brand_name(title))
|
| 274 |
companies.add(tds[2].get_text(separator=" ", strip=True).replace('製造販売元/', ''))
|
|
|
|
| 276 |
log.append(f"❌ 第 {current_page} 頁未出現 ResultList。")
|
| 277 |
break
|
| 278 |
|
| 279 |
+
# 尋找下一頁的按鈕 (例如 javascript:changePg(2);)
|
| 280 |
current_page += 1
|
| 281 |
next_page_link = popup.locator(f'a[href="javascript:changePg({current_page});"]')
|
| 282 |
|
| 283 |
if next_page_link.count() > 0:
|
| 284 |
+
log.append(f" -> 發現第 {current_page} 頁,執行翻頁...")
|
| 285 |
+
next_page_link.first.click(force=True)
|
| 286 |
+
popup.wait_for_timeout(2000) # 給予表格重新渲染的緩衝時間
|
|
|
|
| 287 |
else:
|
| 288 |
log.append(" -> 已無下一頁,結束掃描。")
|
| 289 |
break
|
|
|
|
| 299 |
log, brands, companies = [], set(), set()
|
| 300 |
try:
|
| 301 |
log.append(f"1. 前往 Swissmedicinfo 搜尋頁面 (搜尋: {ing_de})...")
|
| 302 |
+
page.goto("https://swissmedicinfo.ch/SearchPage", timeout=45000, wait_until="domcontentloaded")
|
| 303 |
|
| 304 |
+
log.append("2. 透過父層容器定位 Wirkstoff 欄位...")
|
| 305 |
+
# 絕對定位:尋找包含 Wirkstoff 的 Vue 欄位容器,抓取內部的 input,無視任何 ID
|
| 306 |
+
search_input = page.locator('div.v-field__field').filter(has_text="Wirkstoff").locator('input[type="text"]').first
|
|
|
|
| 307 |
|
| 308 |
+
# 💡 關鍵修正:改為等待 attached (存在於DOM),並用 force=True 無視 UI 動畫遮罩強制填寫
|
| 309 |
+
search_input.wait_for(state="attached", timeout=15000)
|
| 310 |
+
search_input.fill(ing_de, force=True)
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 311 |
|
| 312 |
+
log.append("3. 強制點擊搜尋按鈕...")
|
| 313 |
+
search_btn = page.locator('button').filter(has_text='Packungsbeilage suchen').first
|
| 314 |
+
search_btn.click(force=True)
|
| 315 |
|
| 316 |
+
log.append("4. 等待卡片 (.medicament-card) 渲染...")
|
| 317 |
+
page.wait_for_timeout(3000) # 給予 Vue.js 充分時間生成卡片
|
| 318 |
+
# 確保畫面出現卡片,或者出現找不到資料的提示
|
| 319 |
+
page.wait_for_selector('.medicament-card, .v-alert', timeout=15000)
|
| 320 |
|
| 321 |
soup = BeautifulSoup(page.content(), 'html.parser')
|
| 322 |
cards = soup.find_all('div', class_=re.compile('medicament-card'))
|