.github/workflows/format.yml

@@ -19,10 +19,12 @@ jobs:
       - name: Install dependencies
         run: |
           python -m pip install --upgrade pip
-          pip install -r requirements/dev.txt
+          pip install flake8 black
 
-      - name: Lint with pylint
+      - name: Lint with flake8
-        run: pylint app --extension-pkg-allow-list=lxml
+        run: |
+          flake8 . --count --select=E9,F63,F7,F82 --show-source --statistics
+          flake8 . --count --exit-zero --max-complexity=10 --max-line-length=127 --statistics
 
       - name: Format with black
         run: black .

.gitignore

@@ -1,4 +1,3 @@
 .venv
 __pycache__/
-.vscode
+.vscode
-.vercel

Dockerfile

@@ -1,11 +1,13 @@
-FROM python:alpine
+FROM python
 
 WORKDIR /srv
 
-COPY ./requirements /srv/requirements
+COPY ./requirements.txt /srv/requirements.txt
 
-RUN pip install -r requirements/prod.txt
+RUN pip install -r requirements.txt
+
+EXPOSE 80
 
 COPY ./app /srv/app
 
-CMD uvicorn app.main:app --host 0.0.0.0 --port ${PORT:-8081}
+CMD ["uvicorn", "app.main:app", "--host", "0.0.0.0", "--port", "80"]

README.md

@@ -10,54 +10,19 @@ Backend for online ebook viewer publite
 
 ## Deploy
 
-Run app locally (development only!)
-
-```bash
-# install requirements
-pip install -r requirements/dev.txt
-
-# run app with uvicorn
-uvicorn app.main:app --reload --port <port>
-```
-
-Run app locally (test prod)
-
-```bash
-# install requirements
-pip install -r requirements/prod.txt
-
-# run app with uvicorn
-uvicorn app.main:app --port <port>
-
-# or
-
-# run with python script
-python run.py
-```
-
 Simple docker deployment
 
 ```bash
-# build docker image
+# build docker image 
 docker build . -t publite_backend
 
 # run it with docker
-docker run -p <port>:8081 publite_backend
+docker run -p <port>:80 publite_backend
 ```
 
 Dokku deployment with image from Docker Hub
 
 ```bash
 dokku apps:create publitebackend
-
-# increase file size limit to be able to upload bigger books
-dokku nginx:set publitebackend client_max_body_size 50m
-
 dokku git:from-image publitebackend publite/backend:latest
-```
+```
-
-# TODO
-
-- Separate epub and fb2 files to python modules
-- Rewrite own `.opf` file parsing to get rid of dependency on EbookLib
-- Add cli interfaces for epub and fb2 libs

api

@@ -1 +0,0 @@
-app

app/epub.py

@@ -2,19 +2,19 @@
 Module for EPUB file conversion to html
 """
 
-import html
-import os
 from base64 import b64encode
 from functools import cache
+import html
+import os
 from tempfile import SpooledTemporaryFile
 
-import aiofiles
+import aiofiles as aiof
-import ebooklib
-from ebooklib import epub
 from fastapi import HTTPException
 from lxml import etree
+import ebooklib
+from ebooklib import epub
 
-from .utils import DocumentTokens, HTMLBook, strip_whitespace
+from .utils import DocumentTokens, strip_whitespace, HTMLBook
 
 parser = etree.XMLParser(recover=True)
 
@@ -61,7 +61,7 @@ async def epub_to_tokens(
 
     tokens = {}
 
-    async with aiofiles.tempfile.NamedTemporaryFile() as tmp:
+    async with aiof.tempfile.NamedTemporaryFile() as tmp:
         await tmp.write(file.read())
 
         # Reading book file
@@ -108,6 +108,7 @@ async def epub_to_tokens(
 
 
 def read_metadata(book: epub.EpubBook) -> dict[str, str]:
+
     """
     Reads metadata from xml to dict
     """
@@ -120,6 +121,7 @@ def read_metadata(book: epub.EpubBook) -> dict[str, str]:
 
 
 def convert_list(titles_list: list[tuple[str, dict[str, str]]]) -> str:
+
     """
     Joins titles list to one string
     """
@@ -132,6 +134,7 @@ def convert_list(titles_list: list[tuple[str, dict[str, str]]]) -> str:
 
 
 def set_cover(tokens: DocumentTokens) -> None:
+
     """
     Converts cover file name to base64 image stored in `tokens`
     """
@@ -142,6 +145,7 @@ def set_cover(tokens: DocumentTokens) -> None:
 
 
 def epub_tokens2html(spine: list[tuple[str, str]], tokens: DocumentTokens) -> bytes:
+
     """
     Joins chapters in `spice` to one html string
     """
@@ -153,10 +157,11 @@ def epub_tokens2html(spine: list[tuple[str, str]], tokens: DocumentTokens) -> by
         if file_path:
             res += process_xhtml(file_path, tokens)
 
-    return html.unescape(res)
+    return html.escape(html.unescape(res))
 
 
 def process_xhtml(path: str, tokens: DocumentTokens) -> bytes:
+
     """
     Processes content of one xml body
     """
@@ -174,6 +179,7 @@ def process_xhtml(path: str, tokens: DocumentTokens) -> bytes:
 
 
 def process_content(node: etree.Element, path: str, tokens: DocumentTokens) -> None:
+
     """
     Recursive function for xml element convertion to valid html
     """
@@ -213,6 +219,7 @@ def process_content(node: etree.Element, path: str, tokens: DocumentTokens) -> N
 
 
 def process_a_element(node: etree.Element, path: str):
+
     r"""
     Converts `filed` links to ids in \<a\> element
     """
@@ -230,6 +237,7 @@ def process_a_element(node: etree.Element, path: str):
 
 
 def process_media_element(node: etree.Element, path: str, tokens: DocumentTokens):
+
     """
     Replaces file paths to base64 encoded media in `src` and `srcset` tags
     """
@@ -248,6 +256,7 @@ def process_media_element(node: etree.Element, path: str, tokens: DocumentTokens
 
 
 def rel_to_abs_path(parent: str, rel: str):
+
     """
     Helper for relative path to media convertion to absolute
     """
@@ -257,6 +266,7 @@ def rel_to_abs_path(parent: str, rel: str):
 
 @cache
 def path_to_name(path: str) -> str:
+
     """
     Helper function for getting file name
     """
@@ -265,6 +275,7 @@ def path_to_name(path: str) -> str:
 
 
 def children_to_html(root: etree.Element) -> bytes:
+
     """
     Converts all xml children of element to string and joins them
     """

app/fb2.py

@@ -2,15 +2,16 @@
 Module for FB2 file conversion to html
 """
 
-import html
-import xml.etree.ElementTree as ET
 from tempfile import SpooledTemporaryFile
-from typing import Optional
+import xml.etree.ElementTree as ET
 from xml.etree.ElementTree import Element
+from typing import Optional
+import html
 
 from fastapi import HTTPException
 
-from .utils import DocumentTokens, HTMLBook, strip_whitespace
+from .utils import DocumentTokens, strip_whitespace, HTMLBook
+
 
 namespaces = {
     "": "http://www.gribuser.ru/xml/fictionbook/2.0",
@@ -32,7 +33,7 @@ async def fb22html(file: SpooledTemporaryFile) -> HTMLBook:
 
         return {
             **(tokens["metadata"]),
-            "content": html.unescape(html_content.decode()),
+            "content": html.escape(html.unescape(html_content.decode())),
         }
 
     except Exception as err:
@@ -42,6 +43,7 @@ async def fb22html(file: SpooledTemporaryFile) -> HTMLBook:
 
 
 def fb22tokens(file: SpooledTemporaryFile) -> DocumentTokens:
+
     r"""
     Parses fb2 file as xml document.
     It puts book metadata, its content and media to `tokens` dictionary and returns it.
@@ -77,8 +79,7 @@ def fb22tokens(file: SpooledTemporaryFile) -> DocumentTokens:
         metadata = {}
         metadata["title"] = book_info.find("./book-title", namespaces).text
         metadata["author"] = get_author(book_info.find("./author", namespaces))
-        metadata["cover"] = get_cover(
+        metadata["cover"] = get_cover(book_info.find("./coverpage", namespaces))
-            book_info.find("./coverpage", namespaces))
         if "cover" not in metadata.keys():
             metadata.pop("cover")
 
@@ -104,6 +105,7 @@ def fb22tokens(file: SpooledTemporaryFile) -> DocumentTokens:
 
 
 def get_author(author: Element) -> str:
+
     """
     Converts author xml structure to string
     """
@@ -115,7 +117,7 @@ def get_author(author: Element) -> str:
         "last-name",
     ):
         tag = author.find("./" + tag_name, namespaces)
-        if tag is not None and tag.text is not None:
+        if tag is not None:
             res.append(tag.text)
     if len(res) == 0:
         res = author.find("./nickname", namespaces).text
@@ -126,6 +128,7 @@ def get_author(author: Element) -> str:
 
 
 def get_cover(coverpage: Optional[Element]) -> Optional[str]:
+
     """
     Extracts cover image id if exists
     """
@@ -146,6 +149,7 @@ def set_cover(tokens: DocumentTokens) -> None:
 
 
 def fb2body2html(tokens: DocumentTokens) -> str:
+
     """
     Convert fb2 xml to html, joins bodies into one string
     """
@@ -160,6 +164,7 @@ def fb2body2html(tokens: DocumentTokens) -> str:
 
 
 def process_section(body: Element, tokens: DocumentTokens) -> str:
+
     """
     Processes individual sections, recursively goes throw sections tree
     """
@@ -187,6 +192,7 @@ def process_section(body: Element, tokens: DocumentTokens) -> str:
 
 
 def children_to_html(root: Element) -> str:
+
     """
     Converts xml tag children to string
     """
@@ -200,6 +206,7 @@ def children_to_html(root: Element) -> str:
 
 
 def process_image(element: Element, tokens: DocumentTokens) -> None:
+
     r"""
     Converts fb2 \<image /\> to html \<img /\>. Replaces xlink:href with src="\<base64_image_data\>"
     """
@@ -230,6 +237,7 @@ tag_with_class = {
 
 
 def process_content(root: Element, tokens: DocumentTokens) -> None:
+
     """
     Converts fb2 xml tag names to html equivalents and my own styled elements.
     Resolves binary data dependencies

app/main.py

@@ -2,18 +2,13 @@
 
 from datetime import datetime
 
-from fastapi import FastAPI, File, HTTPException, UploadFile
+from fastapi import FastAPI, File, UploadFile, HTTPException
-from fastapi.middleware.cors import CORSMiddleware
 from pydantic import BaseModel  # pylint: disable=no-name-in-module
 
 from .epub import epub2html
 from .fb2 import fb22html
 from .utils import HashedHTMLBook, add_hash
 
-origins = (
-    "*"
-)
-
 
 class DebugInfo(BaseModel):  # pylint: disable=too-few-public-methods
     """Main handler return types"""
@@ -23,14 +18,6 @@ class DebugInfo(BaseModel):  # pylint: disable=too-few-public-methods
 
 app = FastAPI()
 
-app.add_middleware(
-    CORSMiddleware,
-    allow_origins=origins,
-    allow_credentials=True,
-    allow_methods=["*"],
-    allow_headers=["*"],
-)
-
 start_time = datetime.now()
 
 
@@ -60,8 +47,7 @@ async def create_upload_file(file: UploadFile = File(...)):
     elif file.filename.endswith(".epub"):
         content = await epub2html(file.file)
     else:
-        raise HTTPException(
+        raise HTTPException(status_code=415, detail="Error! Unsupported file type")
-            status_code=415, detail="Error! Unsupported file type")
 
     h_content = add_hash(content)
 

app/utils.py

@@ -3,9 +3,9 @@ Utils for publite_backend module
 """
 
 
+from typing import Union, Optional
 import re
 from hashlib import sha256
-from typing import Optional, Union
 
 from pydantic import BaseModel  # pylint: disable=no-name-in-module
 
@@ -17,7 +17,7 @@ class HTMLBook(BaseModel):  # pylint: disable=too-few-public-methods
 
     title: str
     author: str
-    cover: Optional[str] = None
+    cover: Optional[str]
     content: str
 
 

requirements.txt

@@ -1 +1,23 @@
--r requirements/prod.txt
+aiofiles==0.7.0
+appdirs==1.4.4
+asgiref==3.4.0
+black==21.6b0
+click==8.0.1
+EbookLib==0.17.1
+fastapi==0.65.2
+flake8==3.9.2
+h11==0.12.0
+lxml==4.6.3
+mccabe==0.6.1
+mypy-extensions==0.4.3
+pathspec==0.8.1
+pycodestyle==2.7.0
+pydantic==1.8.2
+pyflakes==2.3.1
+python-multipart==0.0.5
+regex==2021.7.1
+six==1.16.0
+starlette==0.14.2
+toml==0.10.2
+typing-extensions==3.10.0.0
+uvicorn==0.14.0

requirements/dev.txt

@@ -1,4 +0,0 @@
--r prod.txt
-pylint
-rope
-black

requirements/prod.txt

@@ -1,7 +0,0 @@
-fastapi
-uvicorn
-aiofiles
-ebooklib
-python-multipart
-lxml
-pydantic

run.py

@@ -1,4 +0,0 @@
-import uvicorn
-
-if __name__ == "__main__":
-    uvicorn.run("app.main:app")

vercel.json

@@ -1,5 +0,0 @@
-{
-    "rewrites": [
-        { "source": "/(.*)", "destination": "/api/main"}
-    ]
-}