X-Git-Url: https://git.mdrn.pl/wolnelektury.git/blobdiff_plain/1405761d13e0ba62db4b1f9a9d5fa1472142b6d0..0e79f07515678e3e678c3637124a27cb81ee8a94:/apps/catalogue/utils.py diff --git a/apps/catalogue/utils.py b/apps/catalogue/utils.py index 1ac0ee8d8..bcc5a0b33 100644 --- a/apps/catalogue/utils.py +++ b/apps/catalogue/utils.py @@ -2,8 +2,8 @@ # This file is part of Wolnelektury, licensed under GNU Affero GPLv3 or later. # Copyright © Fundacja Nowoczesna Polska. See NOTICE for more information. # -from __future__ import with_statement - +from collections import defaultdict +import hashlib import random import re import time @@ -13,7 +13,6 @@ from django.http import HttpResponse from django.core.files.uploadedfile import UploadedFile from django.core.files.storage import DefaultStorage from django.utils.encoding import force_unicode -from django.utils.hashcompat import sha_constructor from django.conf import settings from os import mkdir, path, unlink from errno import EEXIST, ENOENT @@ -31,36 +30,28 @@ MAX_SESSION_KEY = 18446744073709551616L # 2 << 63 def get_random_hash(seed): - sha_digest = sha_constructor('%s%s%s%s' % + sha_digest = hashlib.sha1('%s%s%s%s' % (randrange(0, MAX_SESSION_KEY), time.time(), unicode(seed).encode('utf-8', 'replace'), settings.SECRET_KEY)).digest() return urlsafe_b64encode(sha_digest).replace('=', '').replace('_', '-').lower() -def split_tags(tags): - result = {} - for tag in tags: - result.setdefault(tag.category, []).append(tag) - return result - - -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: - # BookMedia case - ext = media.formats[media.type].ext - if media is None or not media.name: - name = slughifi(filename.split(".")[0]) +def split_tags(*tag_lists): + if len(tag_lists) == 1: + result = defaultdict(list) + for tag in tag_lists[0]: + result[tag.category].append(tag) 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): - return lambda *args: get_dynamic_path(*args, ext=ext, maxlen=maxlen) + result = defaultdict(dict) + for tag_list in tag_lists: + for tag in tag_list: + try: + result[tag.category][tag.pk].count += tag.count + except KeyError: + result[tag.category][tag.pk] = tag + for k, v in result.items(): + result[k] = sorted(v.values(), key=lambda tag: tag.sort_key) + return result class ExistingFile(UploadedFile): @@ -144,7 +135,7 @@ def remove_zip(zip_slug): class AttachmentHttpResponse(HttpResponse): """Response serving a file to be downloaded. """ - def __init__ (self, file_path, file_name, mimetype): + def __init__(self, file_path, file_name, mimetype): super(AttachmentHttpResponse, self).__init__(mimetype=mimetype) self['Content-Disposition'] = 'attachment; filename=%s' % file_name self.file_path = file_path @@ -158,18 +149,18 @@ class MultiQuerySet(object): def __init__(self, *args, **kwargs): self.querysets = args self._count = None - + def count(self): if not self._count: self._count = sum(len(qs) for qs in self.querysets) return self._count - + def __len__(self): return self.count() - + def __getitem__(self, item): try: - indices = (offset, stop, step) = item.indices(self.count()) + (offset, stop, step) = item.indices(self.count()) except AttributeError: # it's not a slice - make it one return self[item : item + 1][0] @@ -187,6 +178,55 @@ class MultiQuerySet(object): stop = total_len - len(items) continue +class SortedMultiQuerySet(MultiQuerySet): + def __init__(self, *args, **kwargs): + self.order_by = kwargs.pop('order_by', None) + self.sortfn = kwargs.pop('sortfn', None) + if self.order_by is not None: + self.sortfn = lambda a, b: cmp((getattr(a, f) for f in self.order_by), + (getattr(b, f) for f in self.order_by)) + super(SortedMultiQuerySet, self).__init__(*args, **kwargs) + + def __getitem__(self, item): + sort_heads = [0] * len(self.querysets) + try: + (offset, stop, step) = item.indices(self.count()) + except AttributeError: + # it's not a slice - make it one + return self[item : item + 1][0] + items = [] + total_len = stop - offset + skipped = 0 + i_s = range(len(sort_heads)) + + while len(items) < total_len: + candidate = None + candidate_i = None + for i in i_s: + def get_next(): + return self.querysets[i][sort_heads[i]] + try: + if candidate is None: + candidate = get_next() + candidate_i = i + else: + competitor = get_next() + if self.sortfn(candidate, competitor) > 0: + candidate = competitor + candidate_i = i + except IndexError: + continue # continue next sort_head + # we have no more elements: + if candidate is None: + break + sort_heads[candidate_i] += 1 + if skipped < offset: + skipped += 1 + continue # continue next item + items.append(candidate) + + return items + def truncate_html_words(s, num, end_text='...'): """Truncates HTML to a certain number of words (not counting tags and @@ -300,3 +340,19 @@ class AppSettings(object): if hasattr(self, more): value = getattr(self, more)(value) return value + + +def trim_query_log(trim_to=25): + """ +connection.queries includes all SQL statements -- INSERTs, UPDATES, SELECTs, etc. Each time your app hits the database, the query will be recorded. +This can sometimes occupy lots of memory, so trim it here a bit. + """ + if settings.DEBUG: + from django.db import connection + connection.queries = trim_to > 0 \ + and connection.queries[-trim_to:] \ + or [] + + +def delete_from_cache_by_language(cache, key_template): + cache.delete_many([key_template % lc for lc, ln in settings.LANGUAGES])