Calibre-Web-Automated/cps/uploader.py

# -*- coding: utf-8 -*-
# Calibre-Web Automated – fork of Calibre-Web
# Copyright (C) 2018-2025 Calibre-Web contributors
# Copyright (C) 2024-2025 Calibre-Web Automated contributors
# SPDX-License-Identifier: GPL-3.0-or-later
# See CONTRIBUTORS for full list of authors.

import os
import hashlib
from flask_babel import gettext as _

from . import logger, comic, isoLanguages
from .constants import BookMeta
from .helper import split_authors
from .file_helper import get_temp_dir
from .string_helper import strip_whitespaces

log = logger.create()

try:
    from wand.image import Image, Color
    from wand import version as ImageVersion
    from wand.exceptions import PolicyError
    use_generic_pdf_cover = False
except (ImportError, RuntimeError) as e:
    log.debug('Cannot import Image, generating pdf covers for pdf uploads will not work: %s', e)
    use_generic_pdf_cover = True

try:
    from pypdf import PdfReader
    from pypdf.generic import NullObject
    use_pdf_meta = True
except ImportError as ex:
    log.debug('PyPDF is recommended for best performance in metadata extracting from pdf files: %s', ex)
    try:
        from PyPDF2 import PdfReader
        from pypdf.generic import NullObject
        use_pdf_meta = True
    except ImportError as ex:
        log.debug('PyPDF is recommended for best performance in metadata extracting from pdf files: %s', ex)
        log.debug('PyPdf2 is also possible for metadata extracting from pdf files, but not recommended anymore')
        try:
            from PyPDF3 import PdfFileReader as PdfReader
            from pypdf.generic import NullObject
            use_pdf_meta = True
        except ImportError as e:
            log.debug('Cannot import PyPDF3/PyPDF2, extracting pdf metadata will not work: %s / %s', e)
            use_pdf_meta = False

try:
    from . import epub
    use_epub_meta = True
except ImportError as e:
    log.debug('Cannot import epub, extracting epub metadata will not work: %s', e)
    use_epub_meta = False

try:
    from . import fb2
    use_fb2_meta = True
except ImportError as e:
    log.debug('Cannot import fb2, extracting fb2 metadata will not work: %s', e)
    use_fb2_meta = False

try:
    from . import audio
    use_audio_meta = True
except ImportError as e:
    log.debug('Cannot import mutagen, extracting audio metadata will not work: %s', e)
    use_audio_meta = False


def process(tmp_file_path, original_file_name, original_file_extension, rar_executable, no_cover=False):
    meta = default_meta(tmp_file_path, original_file_name, original_file_extension)
    extension_upper = original_file_extension.upper()
    try:
        if ".PDF" == extension_upper:
            meta = pdf_meta(tmp_file_path, original_file_name, original_file_extension, no_cover)
        elif extension_upper in [".KEPUB", ".EPUB"] and use_epub_meta is True:
            meta = epub.get_epub_info(tmp_file_path, original_file_name, original_file_extension, no_cover)
        elif ".FB2" == extension_upper and use_fb2_meta is True:
            meta = fb2.get_fb2_info(tmp_file_path, original_file_extension)
        elif extension_upper in ['.CBZ', '.CBT', '.CBR', ".CB7"]:
            meta = comic.get_comic_info(tmp_file_path,
                                        original_file_name,
                                        original_file_extension,
                                        rar_executable,
                                        no_cover)
        elif extension_upper in [".MP3", ".OGG", ".FLAC", ".WAV", ".AAC", ".AIFF", ".ASF", ".MP4",
                                 ".M4A", ".M4B", ".OGV", ".OPUS"] and use_audio_meta:
            meta = audio.get_audio_file_info(tmp_file_path, original_file_extension, original_file_name, no_cover)
    except Exception as ex:
        log.warning('cannot parse metadata, using default: %s', ex)

    if not strip_whitespaces(meta.title):
        meta = meta._replace(title=original_file_name)
    if not strip_whitespaces(meta.author) or meta.author.lower() == 'unknown':
        meta = meta._replace(author=_('Unknown'))
    return meta


def default_meta(tmp_file_path, original_file_name, original_file_extension):
    return BookMeta(
        file_path=tmp_file_path,
        extension=original_file_extension,
        title=original_file_name,
        author=_('Unknown'),
        cover=None,
        description="",
        tags="",
        series="",
        series_id="",
        languages="",
        publisher="",
        pubdate="",
        identifiers=[]
        )


def parse_xmp(pdf_file):
    """
    Parse XMP Metadata and prepare for BookMeta object
    """
    try:
        xmp_info = pdf_file.xmp_metadata
    except Exception as ex:
        log.debug('Can not read PDF XMP metadata {}'.format(ex))
        return None

    if xmp_info:
        try:
            xmp_author = xmp_info.dc_creator # list
        except AttributeError:
            xmp_author = ['Unknown']

        if xmp_info.dc_title:
            xmp_title = xmp_info.dc_title['x-default']
        else:
            xmp_title = ''

        if xmp_info.dc_description:
            xmp_description = xmp_info.dc_description['x-default']
        else:
            xmp_description = ''

        languages = []
        try:
            for i in xmp_info.dc_language:
                languages.append(isoLanguages.get_lang3(i))
        except AttributeError:
            languages.append('')

        xmp_tags = ', '.join(xmp_info.dc_subject)
        xmp_publisher = ', '.join(xmp_info.dc_publisher)

        return {'author': xmp_author,
                'title': xmp_title,
                'subject': xmp_description,
                'tags': xmp_tags,
                'languages': languages,
                'publisher': xmp_publisher
                }


def pdf_meta(tmp_file_path, original_file_name, original_file_extension, no_cover_processing):
    doc_info = None
    xmp_info = None
    create_date = ""

    if use_pdf_meta:
        with open(tmp_file_path, 'rb') as f:
            pdf_file = PdfReader(f)
            try:
                doc_info = pdf_file.metadata
            except Exception as exc:
                log.debug('Can not read PDF DocumentInfo {}'.format(exc))
            xmp_info = parse_xmp(pdf_file)

    if xmp_info:
        author = ' & '.join(split_authors(xmp_info['author']))
        title = xmp_info['title']
        subject = xmp_info['subject']
        tags = xmp_info['tags']
        languages = xmp_info['languages']
        publisher = xmp_info['publisher']
    else:
        author = 'Unknown'
        title = ''
        languages = [""]
        publisher = ""
        subject = ""
        tags = ""

    if doc_info:
        if author == '':
            author = ' & '.join(split_authors([doc_info.author])) if doc_info.author else 'Unknown'
        if title == '':
            title = doc_info.title if doc_info.title else original_file_name
        if subject == '':
            subject = doc_info.subject or ""
        if tags == '' and '/Keywords' in doc_info:
            keywords = doc_info['/Keywords']
            if not isinstance(keywords, NullObject):
                if isinstance(keywords, bytes):
                    tags = keywords.decode('utf-8')
                else:
                    tags = keywords
        if create_date == '':
            create_date = doc_info.creation_date

    else:
        title = original_file_name

    return BookMeta(
        file_path=tmp_file_path,
        extension=original_file_extension,
        title=title,
        author=author,
        cover=pdf_preview(tmp_file_path, original_file_name) if not no_cover_processing else None,
        description=subject,
        tags=tags,
        series="",
        series_id="",
        languages=','.join(languages),
        publisher=publisher,
        pubdate=str(create_date),
        identifiers=[])


def pdf_preview(tmp_file_path, tmp_dir):
    if use_generic_pdf_cover:
        return None
    try:
        cover_file_name = os.path.join(os.path.dirname(tmp_file_path), "cover.jpg")
        with Image() as img:
            img.options["pdf:use-cropbox"] = "true"
            img.read(filename=tmp_file_path + '[0]', resolution=150)
            img.compression_quality = 88
            if img.alpha_channel:
                img.alpha_channel = 'remove'
                img.background_color = Color('white')
            img.save(filename=os.path.join(tmp_dir, cover_file_name))
        return cover_file_name
    except PolicyError as ex:
        log.warning('Pdf extraction forbidden by Imagemagick policy: %s', ex)
        return None
    except Exception as ex:
        log.warning('Cannot extract cover image, using default: %s', ex)
        log.warning('On Windows this error could be caused by missing ghostscript')
        return None


def get_magick_version():
    ret = dict()
    if not use_generic_pdf_cover:
        ret['Image Magick'] = ImageVersion.MAGICK_VERSION
    else:
        ret['Image Magick'] = 'not installed'
    return ret


def upload(uploadfile, rar_excecutable):
    tmp_dir = get_temp_dir()

    filename = uploadfile.filename
    filename_root, file_extension = os.path.splitext(filename)
    md5 = hashlib.md5(filename.encode('utf-8')).hexdigest()  # nosec
    tmp_file_path = os.path.join(tmp_dir, md5)
    log.debug("Temporary file: %s", tmp_file_path)
    uploadfile.save(tmp_file_path)
    return process(tmp_file_path, filename_root, file_extension, rar_excecutable)