update: enhance crawler of `m.wxsy.net`

src/crawler/m.wxsy.net/catalog.py

@@ -4,37 +4,17 @@
 """
 Fetch catalog and output as JSON format.
 
-    USAGE: python3 catalog.py
+    USAGE: python3 catalog.py [PROXY]
 """
 
-import sys
-sys.path.append('..')
-
 import re
+import sys
 import json
-import time
+sys.path.append('..')
-import requests
+from utils import logger
-from logger import logger
+from utils import httpRequest
 from bs4 import BeautifulSoup
 
-basicUrl = 'https://m.wxsy.net/novel/57104/all.html'
-
-userAgent = (  # default user-agent
-    'Mozilla/5.0 (Linux; Android 10; moto g(7) play) '
-    'AppleWebKit/537.36 (KHTML, like Gecko) '
-    'Chrome/100.0.4896.79 Mobile Safari/537.36'
-)
-
-
-def httpRequest(url: str) -> str:  # fetch raw html content
-    request = requests.get(url, headers = {
-        'user-agent': userAgent,  # with fake user-agent
-        'accept-encoding': 'gzip, deflate',  # allow content compress
-    })
-    if request.status_code not in range(200, 300):  # http status code 2xx
-        raise RuntimeError('Http request failed')
-    return request.text
-
 
 def analysePage(rawHtml: str) -> list:  # extract catalog from html content
     analyseRet = []
@@ -51,10 +31,13 @@ def analysePage(rawHtml: str) -> list:  # extract catalog from html content
 def fetchCatalog(pageNum: int) -> list:  # fetch raw catalog
     catalog = []
     for pageIndex in range(1, pageNum + 1):  # traverse all pages (1 ~ pageNum)
-        logger.info('Page: %d' % pageIndex)
+        logger.info('Catalog page -> %d' % pageIndex)
-        pageUrl = '%s?sort=1&page=%d' % (basicUrl, pageIndex)
+        catalog.append(analysePage(
-        catalog.append(analysePage(httpRequest(pageUrl)))
+            httpRequest(
-        time.sleep(1)  # avoid being blocked by the server
+                'https://m.wxsy.net/novel/57104/all.html?sort=1&page=%d' % pageIndex,
+                proxy = sys.argv[1]
+            )
+        ))
     return catalog
 
 
@@ -70,5 +53,6 @@ def formatCatalog(rawCatalog: list) -> dict:
     return {x[0]: x[1] for x in catalog}  # formatted output
 
 
+logger.warning('Fetch catalog of `m.wxsy.net`')
 release = formatCatalog(fetchCatalog(18))  # 18 pages in total
 print(json.dumps(release))  # output as JSON format

src/crawler/m.wxsy.net/crawler.sh

@@ -4,7 +4,10 @@ cd `dirname $0`
 mkdir -p ./data/html/
 mkdir -p ./data/json/
 
-python3 catalog.py > ./data/catalog.json
+[ -z ${DELAY} ] && DELAY=1
-python3 fetch.py ./data/catalog.json ./data/html/
+[ -z ${THREAD} ] && THREAD=1
+
+python3 catalog.py "${PROXY}" > ./data/catalog.json
+python3 fetch.py ./data/catalog.json ./data/html/ "${PROXY}" ${THREAD} ${DELAY}
 python3 extract.py ./data/catalog.json ./data/html/ ./data/json
 python3 release.py ./data/catalog.json ./data/json/ > ./data/xxrs.json

src/crawler/m.wxsy.net/extract.py

@@ -7,10 +7,12 @@ Extract data from raw html content.
     USAGE: python3 extract.py [CATALOG] [HTML_DIR] [OUTPUT_DIR]
 """
 
+import os
 import re
 import sys
 import json
-from logger import logger
+sys.path.append('..')
+from utils import logger
 from bs4 import BeautifulSoup
 
 
@@ -19,7 +21,7 @@ def splitHtml(rawHtml: str) -> dict:  # extract from raw html content
     script = body.select('script')[5].text  # js code with chapter info
     info = {
         'title': body.select('div[class="size18 w100 text-center lh100 pt30 pb15"]')[0].text.strip(),
-        'contents': [x.text.strip() for x in body.select('p[class="content_detail"]')],
+        'content': [x.text.strip() for x in body.select('p[class="content_detail"]')],
         'prePage': body.select('div[class="pt-prechapter"]')[0].a.attrs['href'],
         'nextPage': body.select('div[class="pt-nextchapter"]')[0].a.attrs['href'],
         'preId': re.search(r'window\.__PREVPAGE = "(\d*)"', script)[1],
@@ -35,8 +37,8 @@ def splitHtml(rawHtml: str) -> dict:  # extract from raw html content
 
 
 def combinePage(id: str) -> dict:  # combine sub pages
-    page_1 = splitHtml(open('%s/%s-1.html' % (sys.argv[2], id)).read())
+    page_1 = splitHtml(open(os.path.join(sys.argv[2], '%s-1.html' % id)).read())
-    page_2 = splitHtml(open('%s/%s-2.html' % (sys.argv[2], id)).read())
+    page_2 = splitHtml(open(os.path.join(sys.argv[2], '%s-2.html' % id)).read())
 
     # page info check
     if not page_1['index'] == '[1/2页]' or not page_2['index'] == '[2/2页]':
@@ -66,13 +68,13 @@ def combinePage(id: str) -> dict:  # combine sub pages
         'preId': page_1['preId'],
         'myId': page_1['myId'],
         'nextId': page_1['nextId'],
-        'contents': page_1['contents'] + page_2['contents']
+        'content': page_1['content'] + page_2['content']
     }
 
 
+logger.warning('Extract info of `m.wxsy.net`')
 catalog = json.loads(open(sys.argv[1]).read())  # load catalog
-
 for _, chapterId in catalog.items():  # traverse all chapters
     logger.info('Analyse chapter `%s`' % chapterId)
-    with open('%s/%s.json' % (sys.argv[3], chapterId), 'w') as fileObj:
+    with open(os.path.join(sys.argv[3], '%s.json' % chapterId), 'w') as fileObj:
         fileObj.write(json.dumps(combinePage(chapterId)))

src/crawler/m.wxsy.net/fetch.py

@@ -4,52 +4,31 @@
 """
 Download raw html content as `.html` files.
 
-    USAGE: python3 fetch.py [CATALOG] [OUTPUT_DIR]
+    USAGE: python3 fetch.py [CATALOG] [OUTPUT_DIR] [PROXY] [THREAD] [DELAY]
 """
 
+import os
 import sys
 import json
-import time
+sys.path.append('..')
-import requests
+from utils import logger
-from logger import logger
+from utils import htmlFetch
-
+
-basicUrl = 'https://m.wxsy.net/novel/57104'
+
-
+def loadChapter():
-userAgent = (  # default user-agent
+    catalog = json.loads(open(sys.argv[1]).read())  # load catalog
-    'Mozilla/5.0 (Linux; Android 10; moto g(7) play) '
+    for _, chapterId in catalog.items():  # traverse all chapters
-    'AppleWebKit/537.36 (KHTML, like Gecko) '
+        for subPage in [1, 2]:  # two sub pages in one chapter
-    'Chrome/100.0.4896.79 Mobile Safari/537.36'
+            yield {
-)
+                'url': 'https://m.wxsy.net/novel/57104/read_%s/%d.html' % (chapterId, subPage),
-
+                'file': os.path.join(sys.argv[2], '%s-%d.html' % (chapterId, subPage)),
-
-def httpRequest(url: str, fileName: str) -> bool:  # save html content
-    try:
-        logger.debug('Http request `%s` -> %s' % (url, fileName))
-        request = requests.get(url, timeout = 30,  # timeout -> 30s
-            headers = {
-                'user-agent': userAgent,  # with fake user-agent
             }
-        )
-        if request.status_code not in range(200, 300):  # http status code 2xx
-            logger.warning('Http request failed -> %s' % url)
-            return False
-        logger.debug('Http request success -> %s' % url)
-        with open(fileName, 'w') as fileObj:  # save html content
-            fileObj.write(request.text)
-        logger.debug('File save success -> %s' % fileName)
-    except:
-        return False
-    return True
-
 
-catalog = json.loads(open(sys.argv[1]).read())  # load catalog
 
-for _, chapterId in catalog.items():  # traverse all chapters
+logger.warning('Fetch html of `m.wxsy.net`')
-    for subPage in [1, 2]:  # two sub pages in one chapter
+htmlFetch(
-        pageUrl = '%s/read_%s/%d.html' % (basicUrl, chapterId, subPage)
+    loadChapter(),
-        pageFile = '%s/%s-%d.html' % (sys.argv[2], chapterId, subPage)
+    proxy = sys.argv[3],
-        if httpRequest(pageUrl, pageFile):  # save html content
+    thread = int(sys.argv[4]),
-            logger.info('Page request success -> %s' % pageUrl)
+    delay = float(sys.argv[5]),
-        else:
+)
-            logger.error('Page request failed -> %s' % pageUrl)
-        time.sleep(1)  # avoid being blocked by the server

src/crawler/m.wxsy.net/release.py

@@ -7,16 +7,18 @@ Combine all chapters from json files.
     USAGE: python3 release.py [CATALOG] [JSON_DIR]
 """
 
+import os
 import sys
 import json
-from logger import logger
+sys.path.append('..')
+from utils import logger
 
 
 def loadData(catalog: dict) -> dict:  # load data from json files
     data = {}
     for _, chapterId in catalog.items():
         data[chapterId] = json.loads(
-            open('%s/%s.json' % (sys.argv[2], chapterId)).read()  # read json content
+            open(os.path.join(sys.argv[2], '%s.json' % chapterId)).read()  # read json content
         )
     return data
 
@@ -59,8 +61,9 @@ def combine() -> dict:  # combine all chapters
 
     result = {}
     for _, info in data.items():  # combine contents
-        result[info['title']] = info['contents']
+        result[info['title']] = info['content']
     return result
 
 
+logger.warning('Release info of `m.wxsy.net`')
 print(json.dumps(combine()))