diff --git a/src/wxsy.net/catalog.py b/src/wxsy.net/catalog.py
index 1fab35f..8830c8e 100644
--- a/src/wxsy.net/catalog.py
+++ b/src/wxsy.net/catalog.py
@@ -12,6 +12,7 @@ import sys
 import json
 import time
 import requests
+from logger import logger
 from bs4 import BeautifulSoup
 
 basicUrl = 'https://m.wxsy.net/novel/57104/all.html'
@@ -48,7 +49,7 @@ 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)
-        print('Page: %d' % pageIndex, file = sys.stderr)
+        logger.info('Page: %d' % pageIndex)
         pageUrl = '%s?sort=1&page=%d' % (basicUrl, pageIndex)
         catalog.append(analysePage(httpRequest(pageUrl)))
         time.sleep(1)  # avoid being blocked by the server
diff --git a/src/wxsy.net/extract.py b/src/wxsy.net/extract.py
new file mode 100644
index 0000000..daf5fd0
--- /dev/null
+++ b/src/wxsy.net/extract.py
@@ -0,0 +1,78 @@
+#!/usr/bin/env python3
+# -*- coding: utf-8 -*-
+
+"""
+Extract data from raw html content.
+
+    USAGE: python3 extract.py [CATALOG] [HTML_DIR] [OUTPUT_DIR]
+"""
+
+import re
+import sys
+import json
+from logger import logger
+from bs4 import BeautifulSoup
+
+
+def splitHtml(rawHtml: str) -> dict:  # extract from raw html content
+    body = BeautifulSoup(rawHtml, 'lxml').body
+    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"]')],
+        '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],
+        'nextId': re.search(r'window\.__NEXTPAGE = "(\d*)"', script)[1],
+        'myId': re.search(r'window\.chapterNum = (\d+)', script)[1],
+        'caption': re.search(r'window\.chapterName = \'(.+)\'', script)[1],
+    }
+    if not info['title'].startswith(info['caption']):  # chapter title should start with caption
+        logger.error('Title error -> %s' % info['caption'])
+    info['index'] = info['title'].replace(info['caption'], '')  # remove caption string
+    info.pop('title')
+    return info
+
+
+def combinePage(id: str) -> dict:  # combine sub pages
+    page_1 = splitHtml(open('%s/%s-1.html' % (sys.argv[2], id)).read())
+    page_2 = splitHtml(open('%s/%s-2.html' % (sys.argv[2], id)).read())
+
+    # page info check
+    if not page_1['index'] == '[1/2页]' or not page_2['index'] == '[2/2页]':
+        logger.error('Sub page error -> `%s` <-> `%s`' % (page_1['index'], page_2['index']))
+    if not page_1['caption'] == page_2['caption']:
+        logger.error('Caption error -> `%s` <-> `%s`' % (page_1['caption'], page_2['caption']))
+    if not page_1['myId'] == page_2['myId']:
+        logger.error('Page ID error -> `%s` <-> `%s`' % (page_1['myId'], page_2['myId']))
+    if not page_1['preId'] == page_2['preId']:
+        logger.error('Pre page ID error -> `%s` <-> `%s`' % (page_1['preId'], page_2['preId']))
+    if not page_1['nextId'] == page_2['nextId']:
+        logger.error('Next page ID error -> `%s` <-> `%s`' % (page_1['nextId'], page_2['nextId']))
+
+    # check by pre-page and next-page url
+    if not page_1['prePage'] == '/novel/57104/read_%s.html' % page_1['preId']:
+        logger.warning('Page-1 pre url -> `%s` (ID = %s)' % (page_1['prePage'], id))
+    if not page_1['nextPage'] == '/novel/57104/read_%s/2.html' % page_1['myId']:
+        logger.warning('Page-1 next url -> `%s` (ID = %s)' % (page_1['nextPage'], id))
+    if not page_2['prePage'] == '/novel/57104/read_%s.html' % page_2['myId']:
+        logger.warning('Page-2 pre url -> `%s` (ID = %s)' % (page_2['prePage'], id))
+    if not page_2['nextPage'] == '/novel/57104/read_%s.html' % page_2['nextId']:
+        logger.warning('Page-2 next url -> `%s` (ID = %s)' % (page_2['nextPage'], id))
+    # there are empty links in the first and last page (ignore it)
+
+    return {  # release chapter
+        'title': page_1['caption'],
+        'preId': page_1['preId'],
+        'myId': page_1['myId'],
+        'nextId': page_1['nextId'],
+        'contents': page_1['contents'] + page_2['contents']
+    }
+
+
+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:
+        fileObj.write(json.dumps(combinePage(chapterId)))
diff --git a/src/wxsy.net/fetch.py b/src/wxsy.net/fetch.py
index 9d4c825..4480aff 100644
--- a/src/wxsy.net/fetch.py
+++ b/src/wxsy.net/fetch.py
@@ -44,10 +44,10 @@ def httpRequest(url: str, fileName: str) -> bool:  # save html content
 
 catalog = json.loads(open(sys.argv[1]).read())  # load catalog
 
-for _, pageId in catalog.items():  # traverse all chapters
+for _, chapterId in catalog.items():  # traverse all chapters
     for subPage in [1, 2]:  # two sub pages in one chapter
-        pageUrl = '%s/read_%s/%d.html' % (basicUrl, pageId, subPage)
-        pageFile = '%s/%s-%d.html' % (sys.argv[2], pageId, subPage)
+        pageUrl = '%s/read_%s/%d.html' % (basicUrl, chapterId, subPage)
+        pageFile = '%s/%s-%d.html' % (sys.argv[2], chapterId, subPage)
         if httpRequest(pageUrl, pageFile):  # save html content
             logger.info('Page request success -> %s' % pageUrl)
         else: