X-Git-Url: https://git.mdrn.pl/wolnelektury.git/blobdiff_plain/6e75a30246598a40cabb8ca1498c3a4f6f146c91..e2df026c95a4f2751d7355d3e43676dcc0ad7512:/apps/catalogue/models.py diff --git a/apps/catalogue/models.py b/apps/catalogue/models.py index 565b759ce..f8d5f1b08 100644 --- a/apps/catalogue/models.py +++ b/apps/catalogue/models.py @@ -8,10 +8,11 @@ from django.db import models from django.db.models import permalink, Q import django.dispatch from django.core.cache import cache +from django.core.files.storage import DefaultStorage from django.utils.translation import ugettext_lazy as _ from django.contrib.auth.models import User -from django.core.files import File from django.template.loader import render_to_string +from django.utils.datastructures import SortedDict from django.utils.safestring import mark_safe from django.utils.translation import get_language from django.core.urlresolvers import reverse @@ -22,14 +23,12 @@ from django.conf import settings from newtagging.models import TagBase, tags_updated from newtagging import managers from catalogue.fields import JSONField, OverwritingFileField -from catalogue.utils import ExistingFile, ORMDocProvider, create_zip_task, remove_zip +from catalogue.utils import create_zip +from shutil import copy +from glob import glob +import re +from os import path -from librarian import dcparser, html, epub, NoDublinCore -import mutagen -from mutagen import id3 -from slughifi import slughifi -from sortify import sortify -from os import unlink TAG_CATEGORIES = ( ('author', _('author')), @@ -176,21 +175,46 @@ class Tag(TagBase): return '/'.join((Tag.categories_dict[self.category], self.slug)) +def get_dynamic_path(media, filename, ext=None, maxlen=100): + from slughifi import slughifi + + # how to put related book's slug here? + if not ext: + if media.type == 'daisy': + ext = 'daisy.zip' + else: + ext = media.type + if media is None or not media.name: + name = slughifi(filename.split(".")[0]) + else: + name = slughifi(media.name) + return 'book/%s/%s.%s' % (ext, name[:maxlen-len('book/%s/.%s' % (ext, ext))-4], ext) + + # TODO: why is this hard-coded ? def book_upload_path(ext=None, maxlen=100): - def get_dynamic_path(media, filename, ext=ext): - # how to put related book's slug here? - if not ext: - if media.type == 'daisy': - ext = 'daisy.zip' - else: - ext = media.type - if not media.name: - name = slughifi(filename.split(".")[0]) - else: - name = slughifi(media.name) - return 'book/%s/%s.%s' % (ext, name[:maxlen-len('book/%s/.%s' % (ext, ext))-4], ext) - return get_dynamic_path + return lambda *args: get_dynamic_path(*args, ext=ext, maxlen=maxlen) + + +def get_customized_pdf_path(book, customizations): + """ + Returns a MEDIA_ROOT relative path for a customized pdf. The name will contain a hash of customization options. + """ + customizations.sort() + h = hash(tuple(customizations)) + pdf_name = '%s-custom-%s' % (book.slug, h) + pdf_file = get_dynamic_path(None, pdf_name, ext='pdf') + return pdf_file + + +def get_existing_customized_pdf(book): + """ + Returns a list of paths to generated customized pdf of a book + """ + pdf_glob = '%s-custom-' % (book.slug,) + pdf_glob = get_dynamic_path(None, pdf_glob, ext='pdf') + pdf_glob = re.sub(r"[.]([a-z0-9]+)$", "*.\\1", pdf_glob) + return glob(path.join(settings.MEDIA_ROOT, pdf_glob)) class BookMedia(models.Model): @@ -211,6 +235,9 @@ class BookMedia(models.Model): verbose_name_plural = _('book media') def save(self, *args, **kwargs): + from slughifi import slughifi + from catalogue.utils import ExistingFile, remove_zip + try: old = BookMedia.objects.get(pk=self.pk) except BookMedia.DoesNotExist, e: @@ -220,10 +247,11 @@ class BookMedia(models.Model): if slughifi(self.name) != slughifi(old.name): self.file.save(None, ExistingFile(self.file.path), save=False, leave=True) + super(BookMedia, self).save(*args, **kwargs) + # remove the zip package for book with modified media remove_zip(self.book.slug) - super(BookMedia, self).save(*args, **kwargs) extra_info = self.get_extra_info_value() extra_info.update(self.read_meta()) self.set_extra_info_value(extra_info) @@ -234,6 +262,8 @@ class BookMedia(models.Model): """ Reads some metadata from the audiobook. """ + import mutagen + from mutagen import id3 artist_name = director_name = project = funded_by = '' if self.type == 'mp3': @@ -266,6 +296,8 @@ class BookMedia(models.Model): """ Reads source file SHA1 from audiobok metadata. """ + import mutagen + from mutagen import id3 if filetype == 'mp3': try: @@ -305,6 +337,7 @@ class Book(models.Model): tags = managers.TagDescriptor(Tag) html_built = django.dispatch.Signal() + published = django.dispatch.Signal() class AlreadyExists(Exception): pass @@ -318,6 +351,8 @@ class Book(models.Model): return self.title def save(self, force_insert=False, force_update=False, reset_short_html=True, **kwargs): + from sortify import sortify + self.sort_key = sortify(self.title) ret = super(Book, self).save(force_insert, force_update) @@ -458,38 +493,56 @@ class Book(models.Model): has_daisy_file.short_description = 'DAISY' has_daisy_file.boolean = True - def build_pdf(self): + def build_pdf(self, customizations=None, file_name=None): """ (Re)builds the pdf file. - + customizations - customizations which are passed to LaTeX class file. + file_name - save the pdf file under a different name and DO NOT save it in db. """ - from librarian import pdf from tempfile import NamedTemporaryFile - import os - - # remove zip with all pdf files - remove_zip(settings.ALL_PDF_ZIP) + from os import unlink + from django.core.files import File + from librarian import pdf + from catalogue.utils import ORMDocProvider, remove_zip try: pdf_file = NamedTemporaryFile(delete=False) pdf.transform(ORMDocProvider(self), file_path=str(self.xml_file.path), output_file=pdf_file, + customizations=customizations ) - self.pdf_file.save('%s.pdf' % self.slug, File(open(pdf_file.name))) + if file_name is None: + # we'd like to be sure not to overwrite changes happening while + # (timely) pdf generation is taking place (async celery scenario) + current_self = Book.objects.get(id=self.id) + current_self.pdf_file.save('%s.pdf' % self.slug, File(open(pdf_file.name))) + self.pdf_file = current_self.pdf_file + + # remove cached downloadables + remove_zip(settings.ALL_PDF_ZIP) + + for customized_pdf in get_existing_customized_pdf(self): + unlink(customized_pdf) + else: + print "save %s to: %s" % (file_name, DefaultStorage().path(file_name)) + + pdf_django_file = File(open(pdf_file.name)) + DefaultStorage().save(file_name, pdf_django_file) + pdf_django_file.close() finally: + pass unlink(pdf_file.name) def build_mobi(self): """ (Re)builds the MOBI file. """ - from librarian import mobi from tempfile import NamedTemporaryFile - import os - - # remove zip with all pdf files - remove_zip(settings.ALL_MOBI_ZIP) + from os import unlink + from django.core.files import File + from librarian import mobi + from catalogue.utils import ORMDocProvider, remove_zip try: mobi_file = NamedTemporaryFile(suffix='.mobi', delete=False) @@ -502,6 +555,9 @@ class Book(models.Model): finally: unlink(mobi_file.name) + # remove zip with all mobi files + remove_zip(settings.ALL_MOBI_ZIP) + def build_epub(self, remove_descendants=True): """ (Re)builds the epub file. If book has a parent, does nothing. @@ -510,14 +566,13 @@ class Book(models.Model): from StringIO import StringIO from hashlib import sha1 from django.core.files.base import ContentFile + from librarian import epub, NoDublinCore + from catalogue.utils import ORMDocProvider, remove_zip if self.parent: # don't need an epub return - # remove zip package with all epub files - remove_zip(settings.ALL_EPUB_ZIP) - epub_file = StringIO() try: epub.transform(ORMDocProvider(self), self.slug, output_file=epub_file) @@ -535,6 +590,9 @@ class Book(models.Model): child_book.save() book_descendants += list(child_book.children.all()) + # remove zip package with all epub files + remove_zip(settings.ALL_EPUB_ZIP) + def build_txt(self): from StringIO import StringIO from django.core.files.base import ContentFile @@ -548,6 +606,9 @@ class Book(models.Model): def build_html(self): from tempfile import NamedTemporaryFile from markupstring import MarkupString + from django.core.files import File + from slughifi import slughifi + from librarian import html meta_tags = list(self.tags.filter( category__in=('author', 'epoch', 'genre', 'kind'))) @@ -601,52 +662,32 @@ class Book(models.Model): return False @staticmethod - def zip_epub(): - books = Book.objects.all() - - paths = filter(lambda x: x is not None, - map(lambda b: b.epub_file and b.epub_file.path or None, books)) - if settings.USE_CELERY: - result = create_zip_task.delay(paths, settings.ALL_EPUB_ZIP) - return result.wait() - else: - result = create_zip_task(paths, settings.ALL_EPUB_ZIP) - return result - - @staticmethod - def zip_pdf(): - books = Book.objects.all() - - paths = filter(lambda x: x is not None, - map(lambda b: b.pdf_file and b.pdf_file.path or None, books)) - if settings.USE_CELERY: - result = create_zip_task.delay(paths, settings.ALL_PDF_ZIP) - return result.wait() - else: - result = create_zip_task(paths, settings.ALL_PDF_ZIP) - return result - - @staticmethod - def zip_mobi(): - books = Book.objects.all() - - paths = filter(lambda x: x is not None, - map(lambda b: b.mobi_file and b.mobi_file.path or None, books)) - result = create_zip_task.delay(paths, settings.ALL_MOBI_ZIP) - return settings.MEDIA_URL + result.wait() + def zip_format(format_): + def pretty_file_name(book): + return "%s/%s.%s" % ( + b.get_extra_info_value()['author'], + b.slug, + format_) + + field_name = "%s_file" % format_ + books = Book.objects.filter(parent=None).exclude(**{field_name: ""}) + paths = [(pretty_file_name(b), getattr(b, field_name).path) + for b in books] + result = create_zip.delay(paths, + getattr(settings, "ALL_%s_ZIP" % format_.upper())) + return result.wait() def zip_audiobooks(self): - bm = BookMedia.objects.filter(book=self) - paths = map(lambda bm: bm.file.path, bm) - if settings.USE_CELERY: - result = create_zip_task.delay(paths, self.slug) - return result.wait() - else: - result = create_zip_task(paths, self.slug) - return result + bm = BookMedia.objects.filter(book=self, type='mp3') + paths = map(lambda bm: (None, bm.file.path), bm) + result = create_zip.delay(paths, self.slug) + return result.wait() @classmethod def from_xml_file(cls, xml_file, **kwargs): + from django.core.files import File + from librarian import dcparser + # use librarian to parse meta-data book_info = dcparser.parse(xml_file) @@ -662,6 +703,8 @@ class Book(models.Model): def from_text_and_meta(cls, raw_file, book_info, overwrite=False, build_epub=True, build_txt=True, build_pdf=True, build_mobi=True): import re + from slughifi import slughifi + from sortify import sortify # check for parts before we do anything children = [] @@ -756,6 +799,7 @@ class Book(models.Model): book.reset_tag_counter() book.reset_theme_counter() + cls.published.send(sender=book) return book def reset_tag_counter(self): @@ -849,6 +893,56 @@ class Book(models.Model): return objects + @classmethod + def book_list(cls, filter=None): + """Generates a hierarchical listing of all books. + + Books are optionally filtered with a test function. + + """ + + books_by_parent = {} + books = cls.objects.all().order_by('parent_number', 'sort_key').only('title', 'parent', 'slug') + if filter: + books = books.filter(filter).distinct() + book_ids = set((book.pk for book in books)) + for book in books: + parent = book.parent_id + if parent not in book_ids: + parent = None + books_by_parent.setdefault(parent, []).append(book) + else: + for book in books: + books_by_parent.setdefault(book.parent_id, []).append(book) + + orphans = [] + books_by_author = SortedDict() + for tag in Tag.objects.filter(category='author'): + books_by_author[tag] = [] + + for book in books_by_parent.get(None,()): + authors = list(book.tags.filter(category='author')) + if authors: + for author in authors: + books_by_author[author].append(book) + else: + orphans.append(book) + + return books_by_author, orphans, books_by_parent + + _audiences_pl = { + "SP1": (1, u"szkoła podstawowa"), + "SP2": (1, u"szkoła podstawowa"), + "P": (1, u"szkoła podstawowa"), + "G": (2, u"gimnazjum"), + "L": (3, u"liceum"), + "LP": (3, u"liceum"), + } + def audiences_pl(self): + audiences = self.get_extra_info_value().get('audiences', []) + audiences = sorted(set([self._audiences_pl[a] for a in audiences])) + return [a[1] for a in audiences] + def _has_factory(ftype): has = lambda self: bool(getattr(self, "%s_file" % ftype))