Save auto version number

Fixed #108

.drone.yml

@@ -1,94 +0,0 @@
----
-kind: pipeline
-name: test
-
-steps:
-- name: lint
-  image: alpine:edge
-  commands:
-  - apk add --no-cache python3 py3-lxml py3-setproctitle py3-pip py3-wheel py3-enchant hunspell-en
-  - pip3 install --no-cache-dir .[full] .[dev]
-  - isort --check-only --diff .
-  - pylint morss --rcfile=.pylintrc --disable=C,R,W --fail-under=8
-  - pytest --cov=morss tests
-
----
-kind: pipeline
-name: python
-
-steps:
-- name: publish
-  image: plugins/pypi
-  settings:
-    username:
-      from_secret: pypi_user
-    password:
-      from_secret: pypi_pwd
-  commands:
-    - /bin/drone-pypi
-    - cp dist/morss-*.tar.gz dist/morss.tar.gz
-
-- name: push
-  image: appleboy/drone-scp
-  settings:
-    host:
-      from_secret: ssh_host
-    username:
-      from_secret: ssh_user
-    key:
-      from_secret: ssh_key
-    source:
-      - dist/morss.tar.gz
-    target: /home/ubuntu
-
-- name: deploy
-  image: appleboy/drone-ssh
-  settings:
-    host:
-      from_secret: ssh_host
-    username:
-      from_secret: ssh_user
-    key:
-      from_secret: ssh_key
-    script_stop: true
-    script:
-      - sudo pip install --upgrade dist/morss.tar.gz[full]
-      - sudo rm -r dist
-      - sudo morss-helper reload
-
-trigger:
-  branch:
-  - master
-  event:
-  - push
-
-depends_on:
-- test
-
----
-kind: pipeline
-name: docker
-
-steps:
-- name: publish
-  image: thegeeklab/drone-docker-buildx
-  # NB. this requires qemu installed on host
-  privileged: true
-  settings:
-    username:
-      from_secret: docker_user
-    password:
-      from_secret: docker_pwd
-    repo:
-      from_secret: docker_repo
-    tags: latest
-    platforms: linux/amd64,linux/arm64,linux/arm/v7
-
-trigger:
-  branch:
-  - master
-  event:
-  - push
-
-depends_on:
-- test

.github/workflows/default.yml

@@ -0,0 +1,78 @@
+name: default
+on:
+    push:
+        branches:
+            - master
+
+jobs:
+    test-lint:
+        runs-on: ubuntu-latest
+        steps:
+            - name: Checkout
+              uses: actions/checkout@v3
+              with:
+                  fetch-depth: 0
+
+            - name: Prepare image
+              run: apt-get -y update && apt-get -y install python3-pip libenchant-2-2 aspell-en
+
+            - name: Install dependencies
+              run: pip3 install .[full] .[dev]
+            - run: isort --check-only --diff .
+            - run: pylint morss --rcfile=.pylintrc --disable=C,R,W --fail-under=8
+            - run: pytest --cov=morss tests
+
+    python-publish:
+        runs-on: ubuntu-latest
+        steps:
+            - name: Checkout
+              uses: actions/checkout@v3
+              with:
+                  fetch-depth: 0
+
+            - name: Prepare image
+              run: apt-get -y update && apt-get -y install python3-pip python3-build
+
+            - name: Build package
+              run: python3 -m build
+
+            - name: Publish package
+              uses: https://github.com/pypa/gh-action-pypi-publish@release/v1
+              with:
+                  password: ${{ secrets.pypi_api_token }}
+
+    docker-publish-deploy:
+        runs-on: ubuntu-latest
+        container:
+            image: catthehacker/ubuntu:act-latest
+        steps:
+            - name: Checkout
+              uses: actions/checkout@v3
+
+            - name: Set up QEMU
+              uses: https://github.com/docker/setup-qemu-action@v2
+
+            - name: Set up Docker Buildx
+              uses: https://github.com/docker/setup-buildx-action@v2
+
+            - name: Login to Docker Hub
+              uses: https://github.com/docker/login-action@v2
+              with:
+                  username: ${{ secrets.docker_user }}
+                  password: ${{ secrets.docker_pwd }}
+
+            - name: Build and push
+              uses: https://github.com/docker/build-push-action@v4
+              with:
+                  context: .
+                  platforms: linux/amd64,linux/arm64,linux/arm/v7
+                  push: true
+                  tags: ${{ secrets.docker_repo }}
+
+            - name: Deploy on server
+              uses: https://github.com/appleboy/ssh-action@v0.1.10
+              with:
+                  host: ${{ secrets.ssh_host }}
+                  username: ${{ secrets.ssh_user }}
+                  key: ${{ secrets.ssh_key }}
+                  script: morss-update

README.md

@@ -41,7 +41,7 @@ Some features of morss:
 - Follow 301/meta redirects
 - Recover xml feeds with corrupt encoding
 - Supports gzip-compressed http content
-- HTTP caching with different backends (in-memory/sqlite/mysql/redis/diskcache)
+- HTTP caching with different backends (in-memory/redis/diskcache)
 - Works as server/cli tool
 - Deobfuscate various tracking links
 
@@ -81,9 +81,9 @@ From git
 pip install git+https://git.pictuga.com/pictuga/morss.git#egg=morss[full]
 ```
 
-The full install includes all the cache backends. Otherwise, only in-memory and
-sqlite3 caches are available. The full install also includes gunicorn (for more
-efficient HTTP handling).
+The full install includes all the cache backends. Otherwise, only in-memory
+cache is available. The full install also includes gunicorn (for more efficient
+HTTP handling).
 
 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
@@ -103,7 +103,7 @@ With cli
 docker pull pictuga/morss
 ```
 
-With docker-compose
+With docker-compose **(recommended)**
 
 ```yml
 services:
@@ -215,7 +215,7 @@ From source
 docker run -p 8000:8000 morss
 ```
 
-With docker-compose
+With docker-compose **(recommended)**
 
 ```shell
 docker-compose up
@@ -353,7 +353,7 @@ Using cache and passing arguments:
 ```python
 >>> import morss
 >>> url = 'http://feeds.bbci.co.uk/news/rss.xml'
->>> cache = '/tmp/morss-cache.db' # sqlite cache location
+>>> cache = '/tmp/morss-cache' # diskcache cache location
 >>> options = {'csv':True}
 >>> xml_string = morss.process(url, cache, options)
 >>> xml_string[:50]
@@ -367,11 +367,10 @@ under the hood.
 Doing it step-by-step:
 
 ```python
-import morss, morss.crawler
+import morss
 
 url = 'http://newspaper.example/feed.xml'
 options = morss.Options(csv=True) # arguments
-morss.crawler.sqlite_default = '/tmp/morss-cache.db' # sqlite cache location
 
 url, rss = morss.FeedFetch(url, options) # this only grabs the RSS feed
 rss = morss.FeedGather(rss, url, options) # this fills the feed and cleans it up
@@ -395,8 +394,8 @@ usage: morss [-h] [--post STRING] [--xpath XPATH]
              [--indent] [--cache] [--force] [--proxy]
              [--order {first,last,newest,oldest}] [--firstlink] [--resolve]
              [--items XPATH] [--item_link XPATH] [--item_title XPATH]
-             [--item_content XPATH] [--item_time XPATH] [--nolink] [--noref]
-             [--silent]
+             [--item_content XPATH] [--item_time XPATH]
+             [--mode {xml,html,json}] [--nolink] [--noref] [--silent]
              url
 
 Get full-text RSS feeds
@@ -440,6 +439,8 @@ custom feeds:
   --item_content XPATH  entry's content
   --item_time XPATH     entry's date & time (accepts a wide range of time
                         formats)
+  --mode {xml,html,json}
+                        parser to use for the custom feeds
 
 misc:
   --nolink              drop links, but keeps links' inner text
@@ -501,11 +502,6 @@ be dropped from the feed, even if they're cached. `-1` for unlimited.
 morss uses caching to make loading faster. There are 3 possible cache backends:
 
 - `(nothing/default)`: a simple python in-memory dict-like object.
-- `CACHE=sqlite`: sqlite3 cache. Default file location is in-memory (i.e. it
-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`
 - `CACHE=diskcache`: disk-based cache. Target directory canbe defined with

morss.service

@@ -0,0 +1,13 @@
+[Unit]
+Description=morss server (gunicorn)
+After=network.target
+
+[Service]
+ExecStart=/usr/local/bin/morss-helper run
+ExecReload=/usr/local/bin/morss-helper reload
+KillMode=process
+Restart=always
+User=http
+
+[Install]
+WantedBy=multi-user.target

morss/__init__.py

@@ -19,5 +19,7 @@
 
 # pylint: disable=unused-import,unused-variable
 
+__version__ = ""
+
 from .morss import *
 from .wsgi import application

morss/caching.py

@@ -16,7 +16,6 @@
 # with this program. If not, see <https://www.gnu.org/licenses/>.
 
 import os
-import pickle
 import threading
 import time
 from collections import OrderedDict
@@ -51,83 +50,6 @@ class BaseCache:
             return True
 
 
-try:
-    import sqlite3 # isort:skip
-except ImportError:
-    pass
-
-
-class SQLiteCache(BaseCache):
-    def __init__(self, path=':memory:'):
-        self.con = sqlite3.connect(path, 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:
@@ -182,20 +104,7 @@ class DiskCacheHandler(BaseCache):
 
 
 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':
-        default_cache = SQLiteCache(
-            os.getenv('SQLITE_PATH', ':memory:')
-        )
-
-    elif os.environ['CACHE'] == 'redis':
+    if os.environ['CACHE'] == 'redis':
         default_cache = RedisCacheHandler(
             host = os.getenv('REDIS_HOST', 'localhost'),
             port = int(os.getenv('REDIS_PORT', 6379)),

morss/cli.py

@@ -54,6 +54,7 @@ def cli_app():
     group.add_argument('--item_title', action='store', type=str, metavar='XPATH', help='entry\'s title')
     group.add_argument('--item_content', action='store', type=str, metavar='XPATH', help='entry\'s content')
     group.add_argument('--item_time', action='store', type=str, metavar='XPATH', help='entry\'s date & time (accepts a wide range of time formats)')
+    group.add_argument('--mode', default=None, choices=('xml', 'html', 'json'), help='parser to use for the custom feeds')
 
     group = parser.add_argument_group('misc')
     group.add_argument('--nolink', action='store_true', help='drop links, but keeps links\' inner text')

morss/crawler.py

@@ -368,7 +368,7 @@ class BrowserlyHeaderHandler(BaseHandler):
 def iter_html_tag(html_str, tag_name):
     " To avoid parsing whole pages when looking for a simple tag "
 
-    re_tag = r'<%s(\s*[^>])*>' % tag_name
+    re_tag = r'<%s\s+[^>]+>' % tag_name
     re_attr = r'(?P<key>[^=\s]+)=[\'"](?P<value>[^\'"]+)[\'"]'
 
     for tag_match in re.finditer(re_tag, html_str):
@@ -425,7 +425,7 @@ class HTTPRefreshHandler(BaseHandler):
     def http_response(self, req, resp):
         if 200 <= resp.code < 300:
             if resp.headers.get('refresh'):
-                regex = r'(?i)^(?P<delay>[0-9]+)\s*;\s*url=(["\']?)(?P<url>.+)\2$'
+                regex = r'(?i)^(?P<delay>[0-9]+)\s*;\s*url\s*=\s*(["\']?)(?P<url>.+)\2$'
                 match = re.search(regex, resp.headers.get('refresh'))
 
                 if match:

morss/morss.py

@@ -287,6 +287,9 @@ def FeedFetch(url, options):
 
         ruleset['items'] = options.items
 
+        if options.mode:
+            ruleset['mode'] = options.mode
+
         ruleset['title'] = options.get('title', '//head/title')
         ruleset['desc'] = options.get('desc', '//head/meta[@name="description"]/@content')
 
@@ -425,7 +428,7 @@ def process(url, cache=None, options=None):
     options = Options(options)
 
     if cache:
-        caching.default_cache = caching.SQLiteCache(cache)
+        caching.default_cache = caching.DiskCacheHandler(cache)
 
     url, rss = FeedFetch(url, options)
     rss = FeedGather(rss, url, options)

setup.py

@@ -3,11 +3,33 @@ from glob import glob
 
 from setuptools import setup
 
+
+def get_version():
+    with open('morss/__init__.py', 'r+') as file:
+        lines = file.readlines()
+
+        # look for hard coded version number
+        for i in range(len(lines)):
+            if lines[i].startswith('__version__'):
+                version = lines[i].split('"')[1]
+                break
+
+        # create (& save) one if none found
+        if version == '':
+            version = datetime.now().strftime('%Y%m%d.%H%M')
+            lines[i] = '__version__ = "' + version + '"\n'
+
+            file.seek(0)
+            file.writelines(lines)
+
+        # return version number
+        return version
+
 package_name = 'morss'
 
 setup(
     name = package_name,
-    version = datetime.now().strftime('%Y%m%d.%H%M'),
+    version = get_version(),
     description = 'Get full-text RSS feeds',
     long_description = open('README.md').read(),
     long_description_content_type = 'text/markdown',
@@ -22,8 +44,8 @@ setup(
     packages = [package_name],
     install_requires = ['lxml', 'bs4', 'python-dateutil', 'chardet'],
     extras_require = {
-        'full': ['pymysql', 'redis', 'diskcache', 'gunicorn', 'setproctitle'],
-        'dev': ['pylint', 'pytest', 'pytest-cov'],
+        'full': ['redis', 'diskcache', 'gunicorn', 'setproctitle'],
+        'dev': ['pylint', 'pyenchant', 'pytest', 'pytest-cov'],
     },
     python_requires = '>=2.7',
     package_data = {package_name: ['feedify.ini']},