readabilite: custom xpath for article detection

.drone.yml

@@ -0,0 +1,15 @@
+kind: pipeline
+name: default
+
+steps:
+- name: isort
+  image: python:alpine
+  commands:
+  - pip install isort
+  - isort --check-only --diff .
+- name: pylint
+  image: alpine
+  commands:
+  - apk add --no-cache python3 py3-lxml py3-pip py3-wheel py3-pylint py3-enchant hunspell-en
+  - pip3 install --no-cache-dir .[full]
+  - pylint morss --rcfile=.pylintrc --disable=C,R,W --fail-under=8

.pylintrc

@@ -0,0 +1,50 @@
+[MASTER]
+ignore=CVS
+suggestion-mode=yes
+extension-pkg-allow-list=lxml.etree
+
+[MESSAGES CONTROL]
+disable=missing-function-docstring,
+        missing-class-docstring,
+        missing-module-docstring,
+        wrong-spelling-in-comment,
+
+[REPORTS]
+reports=yes
+score=yes
+
+[SPELLING]
+spelling-dict=en_GB
+spelling-ignore-words=morss
+
+[STRING]
+check-quote-consistency=yes
+check-str-concat-over-line-jumps=yes
+
+[VARIABLES]
+allow-global-unused-variables=no
+init-import=no
+
+[FORMAT]
+expected-line-ending-format=LF
+indent-string='    '
+max-line-length=120
+max-module-lines=1000
+
+[BASIC]
+argument-naming-style=snake_case
+attr-naming-style=snake_case
+class-attribute-naming-style=snake_case
+class-const-naming-style=UPPER_CASE
+class-naming-style=PascalCase
+const-naming-style=UPPER_CASE
+function-naming-style=snake_case
+inlinevar-naming-style=snake_case
+method-naming-style=snake_case
+module-naming-style=snake_case
+variable-naming-style=snake_case
+
+include-naming-hint=yes
+
+bad-names=foo, bar
+good-names=i, j, k

Dockerfile

@@ -3,6 +3,6 @@ FROM alpine:latest
 RUN apk add --no-cache python3 py3-lxml py3-pip py3-wheel git
 
 ADD . /app
-RUN pip3 install --no-cache-dir /app gunicorn
+RUN pip3 install --no-cache-dir /app[full] gunicorn
 
 CMD gunicorn --bind 0.0.0.0:8080 -w 4 --preload --access-logfile - morss

README.md

@@ -1,5 +1,7 @@
 # Morss - Get full-text RSS feeds
 
+[![Build Status](https://ci.pictuga.com/api/badges/pictuga/morss/status.svg)](https://ci.pictuga.com/pictuga/morss)
+
 _GNU AGPLv3 code_  
 _Provided logo is CC BY-NC-SA 4.0_
 
@@ -46,28 +48,25 @@ Some features of morss:
 
 ### Python package
 
+Simple install (without optional dependencies)
+
 ```shell
 pip install git+https://git.pictuga.com/pictuga/morss.git
 ```
 
+Full installation (including optional dependencies)
+
+```shell
+pip install git+https://git.pictuga.com/pictuga/morss.git#[full]
+```
+
+The full install includes mysql and redis (possible cache backends). Otherwise,
+only in-memory and sqlite3 caches are available.
+
 The dependency `lxml` is fairly long to install (especially on Raspberry Pi, as
 C code needs to be compiled). If possible on your distribution, try installing
 it with the system package manager.
 
-Dependencies:
-
-- [python](http://www.python.org/) >= 2.6 (python 3 is supported)
-- [lxml](http://lxml.de/) for xml parsing
-- [bs4](https://pypi.org/project/bs4/) for badly-formatted html pages
-- [dateutil](http://labix.org/python-dateutil) to parse feed dates
-- [chardet](https://pypi.python.org/pypi/chardet)
-- [six](https://pypi.python.org/pypi/six), a dependency of chardet
-- pymysql
-
-You may also need:
-- Apache, with python-cgi support, to run on a server
-- a fast internet connection
-
 ### Docker
 
 Build & run
@@ -263,11 +262,12 @@ arguments to morss is explained in Run above.
 The list of arguments can be obtained by running `morss --help`
 
 ```
-usage: morss [-h] [--post STRING] [--format {rss,json,html,csv}]
+usage: morss [-h] [--post STRING] [--xpath XPATH]
-             [--search STRING] [--clip] [--indent] [--cache] [--force]
+             [--format {rss,json,html,csv}] [--search STRING] [--clip]
-             [--proxy] [--newest] [--firstlink] [--resolve] [--items XPATH]
+             [--indent] [--cache] [--force] [--proxy] [--newest] [--firstlink]
-             [--item_link XPATH] [--item_title XPATH] [--item_content XPATH]
+             [--resolve] [--items XPATH] [--item_link XPATH]
-             [--item_time XPATH] [--nolink] [--noref] [--silent]
+             [--item_title XPATH] [--item_content XPATH] [--item_time XPATH]
+             [--nolink] [--noref] [--silent]
              url
 
 Get full-text RSS feeds
@@ -278,6 +278,7 @@ positional arguments:
 optional arguments:
   -h, --help            show this help message and exit
   --post STRING         POST request
+  --xpath XPATH         xpath rule to manually detect the article
 
 output:
   --format {rss,json,html,csv}
@@ -373,6 +374,8 @@ will be cleared every time the program is run). Path can be defined with
 `SQLITE_PATH`.
 - `CACHE=mysql`: MySQL cache. Connection can be defined with the following
 environment variables: `MYSQL_USER`, `MYSQL_PWD`, `MYSQL_DB`, `MYSQL_HOST`
+- `CACHE=redis`: Redis cache. Connection can be defined with the following
+environment variables: `REDIS_HOST`, `REDIS_PORT`, `REDIS_DB`, `REDIS_PWD`
 
 To limit the size of the cache:
 

morss/__init__.py

@@ -16,5 +16,8 @@
 # with this program. If not, see <https://www.gnu.org/licenses/>.
 
 # ran on `import morss`
+
+# pylint: disable=unused-import,unused-variable
+
 from .morss import *
 from .wsgi import application

morss/__main__.py

@@ -20,9 +20,7 @@
 import os
 import sys
 
-from . import wsgi
+from . import cli, wsgi
-from . import cli
-
 from .morss import MorssException
 
 

morss/caching.py

@@ -0,0 +1,188 @@
+# This file is part of morss
+#
+# Copyright (C) 2013-2020 pictuga <contact@pictuga.com>
+#
+# This program is free software: you can redistribute it and/or modify it under
+# the terms of the GNU Affero General Public License as published by the Free
+# Software Foundation, either version 3 of the License, or (at your option) any
+# later version.
+#
+# This program is distributed in the hope that it will be useful, but WITHOUT
+# ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or FITNESS
+# FOR A PARTICULAR PURPOSE. See the GNU Affero General Public License for more
+# details.
+#
+# You should have received a copy of the GNU Affero General Public License along
+# with this program. If not, see <https://www.gnu.org/licenses/>.
+
+import os
+import pickle
+import threading
+import time
+from collections import OrderedDict
+
+CACHE_SIZE = int(os.getenv('CACHE_SIZE', 1000)) # max number of items in cache (default: 1k items)
+CACHE_LIFESPAN = int(os.getenv('CACHE_LIFESPAN', 60)) # how often to auto-clear the cache (default: 1min)
+
+
+class BaseCache:
+    """ Subclasses must behave like a dict """
+
+    def trim(self):
+        pass
+
+    def autotrim(self, delay=CACHE_LIFESPAN):
+        # trim the cache every so often
+
+        self.trim()
+
+        t = threading.Timer(delay, self.autotrim)
+        t.daemon = True
+        t.start()
+
+    def __contains__(self, url):
+        try:
+            self[url]
+
+        except KeyError:
+            return False
+
+        else:
+            return True
+
+
+try:
+    import sqlite3 # isort:skip
+except ImportError:
+    pass
+
+
+class SQLiteCache(BaseCache):
+    def __init__(self, filename=':memory:'):
+        self.con = sqlite3.connect(filename, detect_types=sqlite3.PARSE_DECLTYPES, check_same_thread=False)
+
+        with self.con:
+            self.con.execute('CREATE TABLE IF NOT EXISTS data (ky UNICODE PRIMARY KEY, data BLOB, timestamp INT)')
+            self.con.execute('pragma journal_mode=WAL')
+
+        self.trim()
+
+    def __del__(self):
+        self.con.close()
+
+    def trim(self):
+        with self.con:
+            self.con.execute('DELETE FROM data WHERE timestamp <= ( SELECT timestamp FROM ( SELECT timestamp FROM data ORDER BY timestamp DESC LIMIT 1 OFFSET ? ) foo )', (CACHE_SIZE,))
+
+    def __getitem__(self, key):
+        row = self.con.execute('SELECT * FROM data WHERE ky=?', (key,)).fetchone()
+
+        if not row:
+            raise KeyError
+
+        return row[1]
+
+    def __setitem__(self, key, data):
+        with self.con:
+            self.con.execute('INSERT INTO data VALUES (?,?,?) ON CONFLICT(ky) DO UPDATE SET data=?, timestamp=?', (key, data, time.time(), data, time.time()))
+
+
+try:
+    import pymysql.cursors # isort:skip
+except ImportError:
+    pass
+
+
+class MySQLCacheHandler(BaseCache):
+    def __init__(self, user, password, database, host='localhost'):
+        self.user = user
+        self.password = password
+        self.database = database
+        self.host = host
+
+        with self.cursor() as cursor:
+            cursor.execute('CREATE TABLE IF NOT EXISTS data (ky VARCHAR(255) NOT NULL PRIMARY KEY, data MEDIUMBLOB, timestamp INT)')
+
+        self.trim()
+
+    def cursor(self):
+        return pymysql.connect(host=self.host, user=self.user, password=self.password, database=self.database, charset='utf8', autocommit=True).cursor()
+
+    def trim(self):
+        with self.cursor() as cursor:
+            cursor.execute('DELETE FROM data WHERE timestamp <= ( SELECT timestamp FROM ( SELECT timestamp FROM data ORDER BY timestamp DESC LIMIT 1 OFFSET %s ) foo )', (CACHE_SIZE,))
+
+    def __getitem__(self, key):
+        cursor = self.cursor()
+        cursor.execute('SELECT * FROM data WHERE ky=%s', (key,))
+        row = cursor.fetchone()
+
+        if not row:
+            raise KeyError
+
+        return row[1]
+
+    def __setitem__(self, key, data):
+        with self.cursor() as cursor:
+            cursor.execute('INSERT INTO data VALUES (%s,%s,%s) ON DUPLICATE KEY UPDATE data=%s, timestamp=%s',
+                (key, data, time.time(), data, time.time()))
+
+
+class CappedDict(OrderedDict, BaseCache):
+    def trim(self):
+        if CACHE_SIZE >= 0:
+            for i in range( max( len(self) - CACHE_SIZE , 0 )):
+                self.popitem(False)
+
+    def __setitem__(self, key, data):
+        # https://docs.python.org/2/library/collections.html#ordereddict-examples-and-recipes
+        if key in self:
+            del self[key]
+        OrderedDict.__setitem__(self, key, data)
+
+
+try:
+    import redis # isort:skip
+except ImportError:
+    pass
+
+
+class RedisCacheHandler(BaseCache):
+    def __init__(self, host='localhost', port=6379, db=0, password=None):
+        self.r = redis.Redis(host=host, port=port, db=db, password=password)
+
+    def __getitem__(self, key):
+        return self.r.get(key)
+
+    def __setitem__(self, key, data):
+        self.r.set(key, data)
+
+
+if 'CACHE' in os.environ:
+    if os.environ['CACHE'] == 'mysql':
+        default_cache = MySQLCacheHandler(
+            user = os.getenv('MYSQL_USER'),
+            password = os.getenv('MYSQL_PWD'),
+            database = os.getenv('MYSQL_DB'),
+            host = os.getenv('MYSQL_HOST', 'localhost')
+        )
+
+    elif os.environ['CACHE'] == 'sqlite':
+        if 'SQLITE_PATH' in os.environ:
+            path = os.getenv('SQLITE_PATH')
+
+        else:
+            path = ':memory:'
+
+        default_cache = SQLiteCache(path)
+
+    elif os.environ['CACHE'] == 'redis':
+        default_cache = RedisCacheHandler(
+            host = os.getenv('REDIS_HOST', 'localhost'),
+            port = int(os.getenv('REDIS_PORT', 6379)),
+            db = int(os.getenv('REDIS_DB', 0)),
+            password = os.getenv('REDIS_PWD', None)
+        )
+
+else:
+        default_cache = CappedDict()

morss/cli.py

@@ -15,12 +15,11 @@
 # You should have received a copy of the GNU Affero General Public License along
 # with this program. If not, see <https://www.gnu.org/licenses/>.
 
-import sys
-import os.path
 import argparse
+import os.path
+import sys
 
-from .morss import FeedFetch, FeedGather, FeedFormat
+from .morss import FeedFetch, FeedFormat, FeedGather, Options
-from .morss import Options
 
 
 def cli_app():
@@ -33,6 +32,7 @@ def cli_app():
     parser.add_argument('url', help='feed url')
 
     parser.add_argument('--post', action='store', type=str, metavar='STRING', help='POST request')
+    parser.add_argument('--xpath', action='store', type=str, metavar='XPATH', help='xpath rule to manually detect the article')
 
     group = parser.add_argument_group('output')
     group.add_argument('--format', default='rss', choices=('rss', 'json', 'html', 'csv'), help='output format')

morss/crawler.py

@@ -16,30 +16,36 @@
 # with this program. If not, see <https://www.gnu.org/licenses/>.
 
 import os
-import sys
+import pickle
-
-import zlib
-from io import BytesIO, StringIO
-import re
-import chardet
-from cgi import parse_header
-import time
-import threading
 import random
+import re
+import sys
+import time
+import zlib
+from cgi import parse_header
 from collections import OrderedDict
+from io import BytesIO, StringIO
+
+import chardet
+
+from .caching import default_cache
 
 try:
     # python 2
-    from urllib2 import BaseHandler, HTTPCookieProcessor, Request, addinfourl, parse_keqv_list, parse_http_list, build_opener
     from urllib import quote
-    from urlparse import urlparse, urlunparse
+
     import mimetools
+    from urllib2 import (BaseHandler, HTTPCookieProcessor, HTTPRedirectHandler,
+                         Request, addinfourl, build_opener, parse_http_list,
+                         parse_keqv_list)
+    from urlparse import urlparse, urlunparse
 except ImportError:
     # python 3
-    from urllib.request import BaseHandler, HTTPCookieProcessor, Request, addinfourl, parse_keqv_list, parse_http_list, build_opener
-    from urllib.parse import quote
-    from urllib.parse import urlparse, urlunparse
     import email
+    from urllib.parse import quote, urlparse, urlunparse
+    from urllib.request import (BaseHandler, HTTPCookieProcessor,
+                                HTTPRedirectHandler, Request, addinfourl,
+                                build_opener, parse_http_list, parse_keqv_list)
 
 try:
     # python 2
@@ -49,10 +55,6 @@ except NameError:
     basestring = unicode = str
 
 
-CACHE_SIZE = int(os.getenv('CACHE_SIZE', 1000)) # max number of items in cache (default: 1k items)
-CACHE_LIFESPAN = int(os.getenv('CACHE_LIFESPAN', 60)) # how often to auto-clear the cache (default: 1min)
-
-
 MIMETYPE = {
     'xml': ['text/xml', 'application/xml', 'application/rss+xml', 'application/rdf+xml', 'application/atom+xml', 'application/xhtml+xml'],
     'rss': ['application/rss+xml', 'application/rdf+xml', 'application/atom+xml'],
@@ -131,6 +133,7 @@ def custom_opener(follow=None, delay=None):
     handlers.append(SizeLimitHandler(500*1024)) # 500KiB
     handlers.append(HTTPCookieProcessor())
     handlers.append(GZIPHandler())
+    handlers.append(HTTPAllRedirectHandler())
     handlers.append(HTTPEquivHandler())
     handlers.append(HTTPRefreshHandler())
     handlers.append(UAHandler(random.choice(DEFAULT_UAS)))
@@ -397,6 +400,11 @@ class HTTPEquivHandler(RespStrHandler):
                     resp.headers[meta.get('http-equiv').lower()] = meta.get('content')
 
 
+class HTTPAllRedirectHandler(HTTPRedirectHandler):
+    def http_error_308(self, req, fp, code, msg, headers):
+        return self.http_error_301(req, fp, 301, msg, headers)
+
+
 class HTTPRefreshHandler(BaseHandler):
     handler_order = 700 # HTTPErrorProcessor has a handler_order of 1000
 
@@ -447,37 +455,46 @@ class CacheHandler(BaseHandler):
 
     def load(self, url):
         try:
-            out = list(self.cache[url])
+            data = pickle.loads(self.cache[url])
+
         except KeyError:
-            out = [None, None, unicode(), bytes(), 0]
+            data = None
 
-        if sys.version_info[0] >= 3:
-            out[2] = email.message_from_string(out[2] or unicode()) # headers
         else:
-            out[2] = mimetools.Message(StringIO(out[2] or unicode()))
+            if sys.version_info[0] >= 3:
+                data['headers'] = email.message_from_string(data['headers'] or unicode()) # headers
+            else:
+                data['headers'] = mimetools.Message(StringIO(data['headers'] or unicode()))
 
-        return out
+        return data
 
-    def save(self, url, code, msg, headers, data, timestamp):
+    def save(self, key, data):
-        self.cache[url] = (code, msg, unicode(headers), data, timestamp)
+        data['headers'] = unicode(data['headers'])
+        self.cache[key] = pickle.dumps(data, 0)
 
-    def is_cached(self, url):
+    def is_cached(self, key):
-        return self.load(url)[0] is not None
+        return self.load(key) is not None
 
     def cached_response(self, req):
         # this does NOT check whether it's already cached, use with care
-        (code, msg, headers, data, timestamp) = self.load(req.get_full_url())
+        data = self.load(req.get_full_url())
 
         # return the cache as a response
-        resp = addinfourl(BytesIO(data), headers, req.get_full_url(), code)
+        resp = addinfourl(BytesIO(data['data']), data['headers'], req.get_full_url(), data['code'])
-        resp.msg = msg
+        resp.msg = data['msg']
 
         return resp
 
     def save_response(self, req, resp):
         data = resp.read()
 
-        self.save(req.get_full_url(), resp.code, resp.msg, resp.headers, data, time.time())
+        self.save(req.get_full_url(), {
+            'code': resp.code,
+            'msg': resp.msg,
+            'headers': resp.headers,
+            'data': data,
+            'timestamp': time.time()
+            })
 
         fp = BytesIO(data)
         old_resp = resp
@@ -487,13 +504,14 @@ class CacheHandler(BaseHandler):
         return resp
 
     def http_request(self, req):
-        (code, msg, headers, data, timestamp) = self.load(req.get_full_url())
+        data = self.load(req.get_full_url())
 
-        if 'etag' in headers:
+        if data is not None:
-            req.add_unredirected_header('If-None-Match', headers['etag'])
+            if 'etag' in data['headers']:
+                req.add_unredirected_header('If-None-Match', data['headers']['etag'])
 
-        if 'last-modified' in headers:
+            if 'last-modified' in data['headers']:
-            req.add_unredirected_header('If-Modified-Since', headers.get('last-modified'))
+                req.add_unredirected_header('If-Modified-Since', data['headers']['last-modified'])
 
         return req
 
@@ -502,33 +520,33 @@ class CacheHandler(BaseHandler):
         # If 'None' is returned, try your chance with the next-available handler
         # If a 'resp' is returned, stop there, and proceed with 'http_response'
 
-        (code, msg, headers, data, timestamp) = self.load(req.get_full_url())
+        data = self.load(req.get_full_url())
+
+        if data is None:
+            # cache empty, refresh
+            return None
 
         # some info needed to process everything
-        cache_control = parse_http_list(headers.get('cache-control', ()))
+        cache_control = parse_http_list(data['headers'].get('cache-control', ()))
-        cache_control += parse_http_list(headers.get('pragma', ()))
+        cache_control += parse_http_list(data['headers'].get('pragma', ()))
 
         cc_list = [x for x in cache_control if '=' not in x]
         cc_values = parse_keqv_list([x for x in cache_control if '=' in x])
 
-        cache_age = time.time() - timestamp
+        cache_age = time.time() - data['timestamp']
 
         # list in a simple way what to do when
         if self.force_min == -2:
-            if code is not None:
+            if data['code'] is not None:
                 # already in cache, perfect, use cache
                 return self.cached_response(req)
 
             else:
                 # raise an error, via urllib handlers
-                resp = addinfourl(BytesIO(), headers, req.get_full_url(), 409)
+                resp = addinfourl(BytesIO(), data['headers'], req.get_full_url(), 409)
                 resp.msg = 'Conflict'
                 return resp
 
-        elif code is None:
-            # cache empty, refresh
-            return None
-
         elif self.force_min == -1:
             # force use cache
             return self.cached_response(req)
@@ -537,7 +555,7 @@ class CacheHandler(BaseHandler):
             # force refresh
             return None
 
-        elif code == 301 and cache_age < 7*24*3600:
+        elif data['code'] == 301 and cache_age < 7*24*3600:
             # "301 Moved Permanently" has to be cached...as long as we want
             # (awesome HTTP specs), let's say a week (why not?). Use force_min=0
             # if you want to bypass this (needed for a proper refresh)
@@ -594,142 +612,6 @@ class CacheHandler(BaseHandler):
     https_response = http_response
 
 
-class BaseCache:
-    """ Subclasses must behave like a dict """
-
-    def trim(self):
-        pass
-
-    def autotrim(self, delay=CACHE_LIFESPAN):
-        # trim the cache every so often
-
-        self.trim()
-
-        t = threading.Timer(delay, self.autotrim)
-        t.daemon = True
-        t.start()
-
-    def __contains__(self, url):
-        try:
-            self[url]
-
-        except KeyError:
-            return False
-
-        else:
-            return True
-
-
-import sqlite3
-
-
-class SQLiteCache(BaseCache):
-    def __init__(self, filename=':memory:'):
-        self.con = sqlite3.connect(filename, detect_types=sqlite3.PARSE_DECLTYPES, check_same_thread=False)
-
-        with self.con:
-            self.con.execute('CREATE TABLE IF NOT EXISTS data (url UNICODE PRIMARY KEY, code INT, msg UNICODE, headers UNICODE, data BLOB, timestamp INT)')
-            self.con.execute('pragma journal_mode=WAL')
-
-        self.trim()
-
-    def __del__(self):
-        self.con.close()
-
-    def trim(self):
-        with self.con:
-            self.con.execute('DELETE FROM data WHERE timestamp <= ( SELECT timestamp FROM ( SELECT timestamp FROM data ORDER BY timestamp DESC LIMIT 1 OFFSET ? ) foo )', (CACHE_SIZE,))
-
-    def __getitem__(self, url):
-        row = self.con.execute('SELECT * FROM data WHERE url=?', (url,)).fetchone()
-
-        if not row:
-            raise KeyError
-
-        return row[1:]
-
-    def __setitem__(self, url, value): # value = (code, msg, headers, data, timestamp)
-        value = list(value)
-        value[3] = sqlite3.Binary(value[3]) # data
-        value = tuple(value)
-
-        with self.con:
-            self.con.execute('INSERT INTO data VALUES (?,?,?,?,?,?) ON CONFLICT(url) DO UPDATE SET code=?, msg=?, headers=?, data=?, timestamp=?', (url,) + value + value)
-
-
-import pymysql.cursors
-
-
-class MySQLCacheHandler(BaseCache):
-    def __init__(self, user, password, database, host='localhost'):
-        self.user = user
-        self.password = password
-        self.database = database
-        self.host = host
-
-        with self.cursor() as cursor:
-            cursor.execute('CREATE TABLE IF NOT EXISTS data (url VARCHAR(255) NOT NULL PRIMARY KEY, code INT, msg TEXT, headers TEXT, data BLOB, timestamp INT)')
-
-        self.trim()
-
-    def cursor(self):
-        return pymysql.connect(host=self.host, user=self.user, password=self.password, database=self.database, charset='utf8', autocommit=True).cursor()
-
-    def trim(self):
-        with self.cursor() as cursor:
-            cursor.execute('DELETE FROM data WHERE timestamp <= ( SELECT timestamp FROM ( SELECT timestamp FROM data ORDER BY timestamp DESC LIMIT 1 OFFSET %s ) foo )', (CACHE_SIZE,))
-
-    def __getitem__(self, url):
-        cursor = self.cursor()
-        cursor.execute('SELECT * FROM data WHERE url=%s', (url,))
-        row = cursor.fetchone()
-
-        if not row:
-            raise KeyError
-
-        return row[1:]
-
-    def __setitem__(self, url, value): # (code, msg, headers, data, timestamp)
-        with self.cursor() as cursor:
-            cursor.execute('INSERT INTO data VALUES (%s,%s,%s,%s,%s,%s) ON DUPLICATE KEY UPDATE code=%s, msg=%s, headers=%s, data=%s, timestamp=%s',
-                (url,) + value + value)
-
-
-class CappedDict(OrderedDict, BaseCache):
-    def trim(self):
-        if CACHE_SIZE >= 0:
-            for i in range( max( len(self) - CACHE_SIZE , 0 )):
-                self.popitem(False)
-
-    def __setitem__(self, key, value):
-        # https://docs.python.org/2/library/collections.html#ordereddict-examples-and-recipes
-        if key in self:
-            del self[key]
-        OrderedDict.__setitem__(self, key, value)
-
-
-if 'CACHE' in os.environ:
-    if os.environ['CACHE'] == 'mysql':
-        default_cache = MySQLCacheHandler(
-            user = os.getenv('MYSQL_USER'),
-            password = os.getenv('MYSQL_PWD'),
-            database = os.getenv('MYSQL_DB'),
-            host = os.getenv('MYSQL_HOST', 'localhost')
-        )
-
-    elif os.environ['CACHE'] == 'sqlite':
-        if 'SQLITE_PATH' in os.environ:
-            path = os.getenv('SQLITE_PATH')
-
-        else:
-            path = ':memory:'
-
-        default_cache = SQLiteCache(path)
-
-else:
-        default_cache = CappedDict()
-
-
 if 'IGNORE_SSL' in os.environ:
     import ssl
     ssl._create_default_https_context = ssl._create_unverified_context

morss/feeds.py

@@ -15,35 +15,33 @@
 # You should have received a copy of the GNU Affero General Public License along
 # with this program. If not, see <https://www.gnu.org/licenses/>.
 
-import sys
 import os.path
+import sys
 
-from datetime import datetime
-
-import re
-import json
 import csv
-
+import json
+import re
+from copy import deepcopy
+from datetime import datetime
 from fnmatch import fnmatch
 
-from lxml import etree
-from dateutil import tz
 import dateutil.parser
-from copy import deepcopy
-
 import lxml.html
+from dateutil import tz
+from lxml import etree
+
 from .readabilite import parse as html_parse
 
 json.encoder.c_make_encoder = None
 
 try:
     # python 2
-    from StringIO import StringIO
     from ConfigParser import RawConfigParser
+    from StringIO import StringIO
 except ImportError:
     # python 3
-    from io import StringIO
     from configparser import RawConfigParser
+    from io import StringIO
 
 try:
     # python 2

morss/morss.py

@@ -16,30 +16,25 @@
 # with this program. If not, see <https://www.gnu.org/licenses/>.
 
 import os
-
+import re
 import time
 from datetime import datetime
-from dateutil import tz
-
 from fnmatch import fnmatch
-import re
 
 import lxml.etree
 import lxml.html
+from dateutil import tz
 
-from . import feeds
+from . import caching, crawler, feeds, readabilite
-from . import crawler
-from . import readabilite
-
 
 try:
     # python 2
     from httplib import HTTPException
-    from urlparse import urlparse, urljoin, parse_qs
+    from urlparse import parse_qs, urljoin, urlparse
 except ImportError:
     # python 3
     from http.client import HTTPException
-    from urllib.parse import urlparse, urljoin, parse_qs
+    from urllib.parse import parse_qs, urljoin, urlparse
 
 
 MAX_ITEM = int(os.getenv('MAX_ITEM', 5)) # cache-only beyond
@@ -91,12 +86,12 @@ class Options:
         else:
             self.options = options or {}
 
-    def __getattr__(self, key):
+    def __getattr__(self, key, default=None):
         if key in self.options:
             return self.options[key]
 
         else:
-            return None
+            return default
 
     def __setitem__(self, key, value):
         self.options[key] = value
@@ -104,12 +99,7 @@ class Options:
     def __contains__(self, key):
         return key in self.options
 
-    def get(self, key, default=None):
+    get = __getitem__ = __getattr__
-        if key in self.options:
-            return self.options[key]
-
-        else:
-            return default
 
 
 def ItemFix(item, options, feedurl='/'):
@@ -228,7 +218,11 @@ def ItemFill(item, options, feedurl='/', fast=False):
         log('non-text page')
         return True
 
-    out = readabilite.get_article(req['data'], url=req['url'], encoding_in=req['encoding'], encoding_out='unicode')
+    if not req['data']:
+        log('empty page')
+        return True
+
+    out = readabilite.get_article(req['data'], url=req['url'], encoding_in=req['encoding'], encoding_out='unicode', xpath=options.xpath)
 
     if out is not None:
         item.content = out
@@ -417,7 +411,7 @@ def process(url, cache=None, options=None):
     options = Options(options)
 
     if cache:
-        crawler.default_cache = crawler.SQLiteCache(cache)
+        caching.default_cache = caching.SQLiteCache(cache)
 
     url, rss = FeedFetch(url, options)
     rss = FeedGather(rss, url, options)

morss/readabilite.py

@@ -15,10 +15,11 @@
 # You should have received a copy of the GNU Affero General Public License along
 # with this program. If not, see <https://www.gnu.org/licenses/>.
 
+import re
+
 import lxml.etree
 import lxml.html
 from bs4 import BeautifulSoup
-import re
 
 
 def parse(data, encoding=None):
@@ -210,7 +211,7 @@ def clean_node(node, keep_threshold=None):
         return
 
     # high score, so keep
-    if keep_threshold is not None and get_score(node) >= keep_threshold:
+    if keep_threshold is not None and keep_threshold > 0 and get_score(node) >= keep_threshold:
         return
 
     gdparent = parent.getparent()
@@ -293,28 +294,26 @@ def clean_node(node, keep_threshold=None):
             gdparent.insert(gdparent.index(parent)+1, new_node)
 
 
-def lowest_common_ancestor(nodeA, nodeB, max_depth=None):
+def lowest_common_ancestor(node_a, node_b, max_depth=None):
-    ancestorsA = list(nodeA.iterancestors())
+    ancestors_a = list(node_a.iterancestors())
-    ancestorsB = list(nodeB.iterancestors())
+    ancestors_b = list(node_b.iterancestors())
 
     if max_depth is not None:
-        ancestorsA = ancestorsA[:max_depth]
+        ancestors_a = ancestors_a[:max_depth]
-        ancestorsB = ancestorsB[:max_depth]
+        ancestors_b = ancestors_b[:max_depth]
 
-    ancestorsA.insert(0, nodeA)
+    ancestors_a.insert(0, node_a)
-    ancestorsB.insert(0, nodeB)
+    ancestors_b.insert(0, node_b)
 
-    for ancestorA in ancestorsA:
+    for ancestor_a in ancestors_a:
-        if ancestorA in ancestorsB:
+        if ancestor_a in ancestors_b:
-            return ancestorA
+            return ancestor_a
 
-    return nodeA # should always find one tho, at least <html/>, but needed for max_depth
+    return node_a # should always find one tho, at least <html/>, but needed for max_depth
 
 
-def get_article(data, url=None, encoding_in=None, encoding_out='unicode', debug=False, threshold=5):
+def get_best_node(html, threshold=5):
-    " Input a raw html string, returns a raw html string of the article "
+    # score all nodes
-
-    html = parse(data, encoding_in)
     score_all(html)
 
     # rank all nodes (largest to smallest)
@@ -331,9 +330,29 @@ def get_article(data, url=None, encoding_in=None, encoding_out='unicode', debug=
     else:
         best = ranked_nodes[0]
 
+    return best
+
+
+def get_article(data, url=None, encoding_in=None, encoding_out='unicode', debug=False, threshold=5, xpath=None):
+    " Input a raw html string, returns a raw html string of the article "
+
+    html = parse(data, encoding_in)
+
+    if xpath is not None:
+        xpath_match = html.xpath(xpath)
+
+        if len(xpath_match):
+            best = xpath_match[0]
+
+        else:
+            best = get_best_node(html, threshold)
+
+    else:
+        best = get_best_node(html, threshold)
+
     # clean up
     if not debug:
-        keep_threshold = get_score(ranked_nodes[0]) * 3/4
+        keep_threshold = get_score(best) * 3/4
         clean_root(best, keep_threshold)
 
     # check for spammy content (links only)
@@ -352,6 +371,7 @@ def get_article(data, url=None, encoding_in=None, encoding_out='unicode', debug=
 
 if __name__ == '__main__':
     import sys
+
     from . import crawler
 
     req = crawler.adv_get(sys.argv[1] if len(sys.argv) > 1 else 'https://morss.it')

morss/wsgi.py

@@ -15,16 +15,16 @@
 # You should have received a copy of the GNU Affero General Public License along
 # with this program. If not, see <https://www.gnu.org/licenses/>.
 
-import sys
+import cgitb
+import mimetypes
 import os.path
 import re
-import lxml.etree
+import sys
-
-import cgitb
-import wsgiref.util
-import wsgiref.simple_server
 import wsgiref.handlers
-import mimetypes
+import wsgiref.simple_server
+import wsgiref.util
+
+import lxml.etree
 
 try:
     # python 2
@@ -33,11 +33,9 @@ except ImportError:
     # python 3
     from urllib.parse import unquote
 
-from . import crawler
+from . import caching, crawler, readabilite
-from . import readabilite
+from .morss import (DELAY, TIMEOUT, FeedFetch, FeedFormat, FeedGather,
-from .morss import FeedFetch, FeedGather, FeedFormat
+                    MorssException, Options, log)
-from .morss import Options, log, TIMEOUT, DELAY, MorssException
-
 
 PORT = int(os.getenv('PORT', 8080))
 
@@ -201,7 +199,7 @@ def cgi_get(environ, start_response):
     req = crawler.adv_get(url=url, timeout=TIMEOUT)
 
     if req['contenttype'] in ['text/html', 'application/xhtml+xml', 'application/xml']:
-        if options.get == 'page':
+        if options['get'] == 'page':
             html = readabilite.parse(req['data'], encoding=req['encoding'])
             html.make_links_absolute(req['url'])
 
@@ -213,7 +211,7 @@ def cgi_get(environ, start_response):
 
             output = lxml.etree.tostring(html.getroottree(), encoding='utf-8', method='html')
 
-        elif options.get == 'article':
+        elif options['get'] == 'article':
             output = readabilite.get_article(req['data'], url=req['url'], encoding_in=req['encoding'], encoding_out='utf-8', debug=options.debug)
 
         else:
@@ -289,7 +287,7 @@ class WSGIRequestHandlerRequestUri(wsgiref.simple_server.WSGIRequestHandler):
 
 
 def cgi_start_server():
-    crawler.default_cache.autotrim()
+    caching.default_cache.autotrim()
 
     print('Serving http://localhost:%s/' % PORT)
     httpd = wsgiref.simple_server.make_server('', PORT, application, handler_class=WSGIRequestHandlerRequestUri)
@@ -297,4 +295,4 @@ def cgi_start_server():
 
 
 if 'gunicorn' in os.getenv('SERVER_SOFTWARE', ''):
-    crawler.default_cache.autotrim()
+    caching.default_cache.autotrim()

setup.py

@@ -1,6 +1,7 @@
-from setuptools import setup
 from glob import glob
 
+from setuptools import setup
+
 package_name = 'morss'
 
 setup(
@@ -12,7 +13,8 @@ setup(
     download_url = 'https://git.pictuga.com/pictuga/morss',
     license = 'AGPL v3',
     packages = [package_name],
-    install_requires = ['lxml', 'bs4', 'python-dateutil', 'chardet', 'pymysql'],
+    install_requires = ['lxml', 'bs4', 'python-dateutil', 'chardet'],
+    extras_require = {'full': ['pymysql', 'redis']},
     package_data = {package_name: ['feedify.ini']},
     data_files = [
         ('share/' + package_name, ['README.md', 'LICENSE']),