update: enhance crawler of `108shu.com`

src/crawler/108shu.com/catalog.py

@@ -8,31 +8,17 @@ Fetch catalog and output as JSON format.
 """
 
 import re
+import sys
 import json
-import requests
-from logger import logger
+sys.path.append('..')
+from utils import logger
+from utils import httpRequest
 from bs4 import BeautifulSoup
 
-userAgent = (  # default user agent
-    'Mozilla/5.0 (Windows NT 10.0; Win64; x64) '
-    'AppleWebKit/537.36 (KHTML, like Gecko) '
-    'Chrome/106.0.0.0 Safari/537.36 Edg/106.0.1370.47'
-)
 
-
-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 extractCatalog(rawHtml: str) -> dict:  # extract catalog from html content
+def extractCatalog(rawHtml: bytes) -> dict:  # extract catalog from html content
     catalog = {}
-    html = BeautifulSoup(rawHtml, 'lxml')
+    html = BeautifulSoup(str(rawHtml, encoding = 'utf-8'), 'lxml')
     items = html.select('div[class="section-box"]')[1]
     for item in items.select('a'):
         name = re.search(r'^(第\d+章)(.*)$', item.text)
@@ -45,9 +31,10 @@ def fetchCatalog(pageNum: int) -> dict:  # fetch all catalog
     catalog = {}
     for pageId in range(1, pageNum + 1):  # traverse all pages
         pageUrl = 'http://www.108shu.com/book/54247/index_%d.html' % pageId
-        logger.info('Page: %d -> `%s`' % (pageId, pageUrl))
+        logger.info('Catalog page -> %d' % pageId)
         catalog.update(extractCatalog(httpRequest(pageUrl)))
     return catalog
 
 
+logger.warning('Fetch catalog of `108shu.com`')
 print(json.dumps(fetchCatalog(45)))

src/crawler/108shu.com/check.sh

@@ -0,0 +1,6 @@
+#!/usr/bin/env bash
+
+cd "$(dirname "$0")"
+
+diff <(xz -cdk ./archive/catalog.json.xz | jq .) <(jq . ./data/catalog.json)
+diff <(xz -cdk ./archive/xxrs.json.xz | jq .) <(jq . ./data/xxrs.json)

src/crawler/108shu.com/crawler.sh

@@ -1,17 +1,11 @@
 #!/usr/bin/env bash
 
-cd `dirname $0`
+cd "$(dirname "$0")"
 mkdir -p ./data/html/
 
-python3 catalog.py > ./data/catalog.json
-python3 fetch.py ./data/catalog.json ./data/html/
-python3 extract.py ./data/catalog.json ./data/html/ > ./data/xxrs.json
-
-cd ./data/
-xz -k9 catalog.json
-tar cJf html.tar.xz html/
-xz -k9 xxrs.json
+[ -z "${DELAY}" ] && DELAY=1
+[ -z "${THREAD}" ] && THREAD=1
 
-mkdir -p ../archive/
-mv *.xz ../archive/
-cd ../
+python3 catalog.py "" > ./data/catalog.json
+python3 fetch.py ./data/catalog.json ./data/html/ "${PROXY}" ${THREAD} ${DELAY}
+python3 extract.py ./data/catalog.json ./data/html/ > ./data/xxrs.json

src/crawler/108shu.com/extract.py

@@ -11,13 +11,13 @@ import os
 import re
 import sys
 import json
-from logger import logger
+sys.path.append('..')
+from utils import logger
 from bs4 import BeautifulSoup
 
 
 def splitHtml(rawHtml: str) -> dict:  # extract from raw html content
     html = BeautifulSoup(rawHtml, 'lxml')
-    content = [x.text.strip() for x in html.select('div[class="content"]')[0].select('p')]
     title = re.search(r'^(第\d+章)(.*)$', html.select('h1')[0].text)
     return {
         'title': '%s %s' % (title[1], title[2].strip()),
@@ -25,9 +25,9 @@ def splitHtml(rawHtml: str) -> dict:  # extract from raw html content
     }
 
 
-def combinePage(chapterId: str) -> dict:  # combine sub pages
-    page_1 = splitHtml(open(os.path.join(sys.argv[2], '%s-1.html' % chapterId)).read())
-    page_2 = splitHtml(open(os.path.join(sys.argv[2], '%s-2.html' % chapterId)).read())
+def combinePage(pageId: str) -> dict:  # combine sub pages
+    page_1 = splitHtml(open(os.path.join(sys.argv[2], '%s-1.html' % pageId)).read())
+    page_2 = splitHtml(open(os.path.join(sys.argv[2], '%s-2.html' % pageId)).read())
     if page_1['title'] != page_2['title']:
         logger.error('Title error -> `%s`' % page_1['title'])
     return {
@@ -37,13 +37,12 @@ def combinePage(chapterId: str) -> dict:  # combine sub pages
 
 
 result = {}
+logger.warning('Extract info of `108shu.com`')
 catalog = json.loads(open(sys.argv[1]).read())  # load catalog
-
 for chapterName, chapterId in catalog.items():  # traverse all chapters
     logger.info('Analyse chapter `%s`' % chapterId)
     info = combinePage(chapterId)
     if chapterName != info['title']:
         logger.error('Title error -> %s' % info['title'])
     result[chapterName] = info['content']
-
 print(json.dumps(result))

src/crawler/108shu.com/fetch.py

@@ -4,51 +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
-import requests
-from logger import logger
-
-userAgent = (  # default user agent
-    'Mozilla/5.0 (Windows NT 10.0; Win64; x64) '
-    'AppleWebKit/537.36 (KHTML, like Gecko) '
-    'Chrome/106.0.0.0 Safari/537.36 Edg/106.0.1370.47'
-)
+sys.path.append('..')
+from utils import logger
+from utils import htmlFetch
 
 
-def httpRequest(fileUrl: str, fileName: str) -> bool:  # save html content
-    try:
-        logger.debug('Http request `%s` -> `%s`' % (fileUrl, fileName))
-        request = requests.get(fileUrl, timeout = 30,  # timeout -> 30s
-            headers = {
-                'user-agent': userAgent,  # with fake user-agent
+def loadChapter():
+    catalog = json.loads(open(sys.argv[1]).read())  # load catalog
+    for _, chapterId in catalog.items():  # traverse all chapters
+        for subPage in [1, 2]:  # two sub-pages in one chapter
+            yield {
+                'url': 'http://www.108shu.com/book/54247/%s_%d.html' % (chapterId, subPage),
+                'file': os.path.join(sys.argv[2], '%s-%d.html' % (chapterId, subPage)),
             }
-        )
-        if request.status_code not in range(200, 300):  # http status code 2xx
-            logger.warning('Http request failed -> `%s`' % fileUrl)
-            return False
-        logger.debug('Http request success -> `%s`' % fileUrl)
-        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
-    for subPage in [1, 2]:
-        pageUrl = 'http://www.108shu.com/book/54247/%s_%d.html' % (chapterId, subPage)
-        pageFile = os.path.join(sys.argv[2], '%s-%d.html' % (chapterId, subPage))
-        if httpRequest(pageUrl, pageFile):  # save html content
-            logger.info('Page request success -> %s' % pageUrl)
-        else:
-            logger.error('Page request failed -> %s' % pageUrl)
-        time.sleep(1)  # avoid being blocked by the server
+logger.warning('Fetch html of `108shu.com`')
+htmlFetch(
+    loadChapter(),
+    proxy = sys.argv[3],
+    thread = int(sys.argv[4]),
+    delay = float(sys.argv[5]),
+)