first commit

.github/workflows/python-ci.yml

@@ -0,0 +1,56 @@
+name: CI
+
+on:
+  push:
+    branches: [ main, master ]
+    tags:
+      - 'v*.*.*'
+  pull_request:
+    branches: [ main, master ]
+
+jobs:
+  test:
+    runs-on: ubuntu-latest
+    strategy:
+      matrix:
+        python-version: ['3.8', '3.11']
+    steps:
+      - uses: actions/checkout@v2
+      - name: Set up Python ${{ matrix.python-version }}
+        uses: actions/setup-python@v2
+        with:
+          python-version: ${{ matrix.python-version }}
+      - name: Install dependencies
+        run: |
+          python -m pip install --upgrade pip
+          python -m pip install '.[dev]'
+      - name: Check code with black
+        run: |
+          black --check .
+      - name: Lint code
+        run: |
+          pylama
+      - name: Test with pytest
+        run: |
+          #pytest
+
+  deploy:
+    needs: test
+    runs-on: ubuntu-latest
+    steps:
+      - uses: actions/checkout@v2
+      - uses: actions/setup-python@v2
+        with:
+          python-version: '3.x'
+      - id: check-tag
+        run: |
+          if [[ "${{ github.event.ref }}" =~ ^refs/tags/v[0-9]+\.[0-9]+\.[0-9]+$ ]]; then
+              echo ::set-output name=match::true
+          fi
+      - name: Create PyPI release
+        uses: casperdcl/deploy-pypi@v2
+        with:
+          password: ${{ secrets.PYPI_TOKEN }}
+          build: true
+          # only upload if a tag is pushed (otherwise just build & check)
+          upload: ${{ github.event_name == 'push' && steps.check-tag.outputs.match == 'true' }}

.gitignore

@@ -0,0 +1,13 @@
+__pycache__
+.cache
+.mypy_cache
+.pytest_cache
+build
+dist
+.eggs
+*.egg-info
+*/_version.py
+
+*~
+.#*
+*#

README.md

@@ -0,0 +1,52 @@
+# Web to PDF
+
+[![Latest Release](https://img.shields.io/pypi/v/web2pdf.svg)](https://pypi.org/project/web2pdf)
+[![CI](https://github.com/deltachat-bot/web2pdf/actions/workflows/python-ci.yml/badge.svg)](https://github.com/deltachat-bot/web2pdf/actions/workflows/python-ci.yml)
+[![Code style: black](https://img.shields.io/badge/code%20style-black-000000.svg)](https://github.com/psf/black)
+
+Delta Chat bot that allows to fetch websites as PDF. Just send any URL to the bot
+in private or add it to get a PDF in any message containing an URL.
+
+## Install
+
+```sh
+pip install web2pdf
+```
+
+You also need to install wkhtmltopdf package. Debian/Ubuntu example:
+
+```sh
+sudo apt-get install wkhtmltopdf
+```
+
+**Warning!** Version in debian/ubuntu repos have reduced functionality (because it
+is compiled without the wkhtmltopdf QT patches), such as adding outlines, headers,
+footers, TOC etc. To use this options you should install static binary from
+wkhtmltopdf site: https://wkhtmltopdf.org/
+
+## Usage
+
+Configure the bot's Delta Chat account:
+
+```sh
+web2pdf init [email protected] PASSWORD
+```
+
+You can run `web2pdf init` several times to add multiple different accounts to the
+bot so it can be reached in more than one email address.
+
+The bot's display name, avatar and status/signature can also be tweaked:
+
+```
+web2pdf config selfavatar "/path/to/avatar.png"
+web2pdf config displayname "My Bot"
+web2pdf config selfstatus "Hi, I am a Delta Chat bot"
+```
+
+To run the bot so it starts processing messages:
+
+```sh
+web2pdf serve
+```
+
+To see all available options run `web2pdf --help`

pylama.ini

@@ -0,0 +1,3 @@
+[pylama]
+linters=mccabe,pyflakes,pylint,isort,mypy
+skip=.*,build/*,tests/*,*/_version.py,*/flycheck_*

pyproject.toml

@@ -0,0 +1,50 @@
+[build-system]
+requires = ["setuptools>=64", "setuptools_scm>=8"]
+build-backend = "setuptools.build_meta"
+
+[project]
+name = "web2pdf"
+description = "Download websites as PDF in Delta Chat"
+dynamic = ["version"]
+readme = "README.md"
+requires-python = ">=3.8"
+keywords = ["deltachat", "bot", "web", "pdf"]
+authors = [
+  {name = "adbenitez", email = "[email protected]"},
+]
+classifiers = [
+  "Development Status :: 4 - Beta",
+  "Programming Language :: Python",
+  "License :: OSI Approved :: GNU General Public License v3 or later (GPLv3+)",
+]
+dependencies = [
+    "deltabot-cli>=6.2.0,<7.0",
+    "pdfkit",
+]
+
+[project.urls]
+Homepage = "https://github.com/deltachat-bot/web2pdf"
+
+[project.optional-dependencies]
+dev = [
+  "black",
+  "mypy",
+  "isort",
+  "pylint",
+  "pylama",
+  "pytest",
+  "types-requests",
+]
+
+[project.scripts]
+web2pdf = "web2pdf:main"
+
+[tool.setuptools_scm]
+# can be empty if no extra settings are needed, presence enables setuptools_scm
+version_file = "web2pdf/_version.py"
+
+[tool.isort]
+profile = "black"
+
+[tool.mypy]
+ignore_missing_imports = "True"

web2pdf/__init__.py

@@ -0,0 +1,11 @@
+"""Web2PDF bot."""
+
+from .hooks import cli
+
+
+def main() -> None:
+    """Run the application."""
+    try:
+        cli.start()
+    except KeyboardInterrupt:
+        pass

web2pdf/__main__.py

@@ -0,0 +1,5 @@
+"""Support for package execution."""
+
+from . import main
+
+main()

web2pdf/hooks.py

@@ -0,0 +1,101 @@
+"""Event Hooks"""
+
+import re
+from argparse import Namespace
+from tempfile import NamedTemporaryFile
+
+import pdfkit
+from deltabot_cli import BotCli
+from deltachat2 import Bot, ChatType, CoreEvent, EventType, MsgData, NewMsgEvent, events
+from rich.logging import RichHandler
+
+from ._version import __version__
+
+cli = BotCli("web2pdf")
+cli.add_generic_option("-v", "--version", action="version", version=__version__)
+cli.add_generic_option(
+    "--no-time",
+    help="do not display date timestamp in log messages",
+    action="store_false",
+)
+
+
+@cli.on_init
+def _on_init(bot: Bot, args: Namespace) -> None:
+    bot.logger.handlers = [
+        RichHandler(show_path=False, omit_repeated_times=False, show_time=args.no_time)
+    ]
+    for accid in bot.rpc.get_all_account_ids():
+        if not bot.rpc.get_config(accid, "displayname"):
+            bot.rpc.set_config(accid, "displayname", "Web to PDF")
+            status = "I am a Delta Chat bot, send me /help for more info"
+            bot.rpc.set_config(accid, "selfstatus", status)
+            bot.rpc.set_config(accid, "delete_device_after", str(60 * 60 * 24))
+
+
+@cli.on(events.RawEvent)
+def _log_event(bot: Bot, accid: int, event: CoreEvent) -> None:
+    if event.kind == EventType.INFO:
+        bot.logger.debug(event.msg)
+    elif event.kind == EventType.WARNING:
+        bot.logger.warning(event.msg)
+    elif event.kind == EventType.ERROR:
+        bot.logger.error(event.msg)
+    elif event.kind == EventType.MSG_DELIVERED:
+        bot.rpc.delete_messages(accid, [event.msg_id])
+    elif event.kind == EventType.SECUREJOIN_INVITER_PROGRESS:
+        if event.progress == 1000:
+            if not bot.rpc.get_contact(accid, event.contact_id).is_bot:
+                bot.logger.debug("QR scanned by contact id=%s", event.contact_id)
+                chatid = bot.rpc.create_chat_by_contact_id(accid, event.contact_id)
+                _send_help(bot, accid, chatid)
+
+
+@cli.after(events.NewMessage)
+def delete_msgs(bot: Bot, accid: int, event: NewMsgEvent) -> None:
+    """Delete messages after processing"""
+    msg = event.msg
+    bot.rpc.delete_messages(accid, [msg.id])
+    bot.logger.debug(f"[chat={msg.chat_id}] Deleted message={msg.id}")
+
+
+@cli.on(events.NewMessage(is_info=False))
+def _web2pdf_filter(bot: Bot, accid: int, event: NewMsgEvent) -> None:
+    if bot.has_command(event.command):
+        return
+    msg = event.msg
+    chat = bot.rpc.get_basic_chat_info(accid, msg.chat_id)
+    match = re.search(
+        r"http[s]?://(?:[a-zA-Z]|[0-9]|[$-_@.&+]|[!*(),]|"
+        r"(?:%[0-9a-fA-F][0-9a-fA-F]))+",
+        msg.text,
+    )
+    if match:
+        url = match.group()
+    elif msg.text and chat.chat_type == ChatType.SINGLE:
+        url = msg.text
+    else:
+        return
+    with NamedTemporaryFile(suffix=".pdf") as file:
+        try:
+            pdfkit.from_url(url, file.name)
+            reply = MsgData(file=file.name, quoted_message_id=msg.id)
+            bot.rpc.send_msg(accid, msg.chat_id, reply)
+        except Exception as ex:  # noqa: W0718
+            bot.logger.exception(ex)
+            text = "Failed to retrieve web site, is the URL correct?"
+            reply = MsgData(text=text, quoted_message_id=msg.id)
+            bot.rpc.send_msg(accid, msg.chat_id, reply)
+
+
+@cli.on(events.NewMessage(command="/help"))
+def _help(bot: Bot, accid: int, event: NewMsgEvent) -> None:
+    _send_help(bot, accid, event.msg.chat_id)
+
+
+def _send_help(bot: Bot, accid: int, chatid: int) -> None:
+    text = (
+        "I'm a bot, I allow to retrieve HTML pages as PDF."
+        " Just send me any link to a website you would like to save as PDF."
+    )
+    bot.rpc.send_msg(accid, chatid, MsgData(text=text))