From: Radek Czajka Date: Tue, 13 Sep 2011 15:11:27 +0000 (+0200) Subject: packaging zip deflated, X-Git-Url: https://git.mdrn.pl/wolnelektury.git/commitdiff_plain/e47b75737d714a11e22cf3f5bceadddbf340e3b3?ds=sidebyside;hp=e4e3b8bd323e57df11b10e06e2a1fbf1b134ca54 packaging zip deflated, lesmianator update --- diff --git a/apps/catalogue/management/commands/pack.py b/apps/catalogue/management/commands/pack.py index 80f612ad3..a300aff0f 100755 --- a/apps/catalogue/management/commands/pack.py +++ b/apps/catalogue/management/commands/pack.py @@ -53,7 +53,7 @@ class Command(BaseCommand): if exclude: books = [book for book in books if book.slug not in exclude.split(',')] - archive = zipfile.ZipFile(path, 'w') + archive = zipfile.ZipFile(path, 'w', zipfile.ZIP_DEFLATED) processed = skipped = 0 for book in books: diff --git a/apps/lesmianator/management/commands/lesmianator.py b/apps/lesmianator/management/commands/lesmianator.py index c0219214d..5412bee81 100644 --- a/apps/lesmianator/management/commands/lesmianator.py +++ b/apps/lesmianator/management/commands/lesmianator.py @@ -14,7 +14,7 @@ from django.conf import settings from catalogue.models import Book, Tag # extract text from text file -re_text = re_text = re.compile(r'\n{3,}(.*?)\n*-----\n', re.S).search +re_text = re.compile(r'\n{3,}(.*?)\n*-----\n', re.S).search class Command(BaseCommand): @@ -66,21 +66,22 @@ class Command(BaseCommand): print self.style.NOTICE('%s has no TXT file' % book.slug) skipped += 1 continue - with open(book.txt_file.path) as f: - m = re_text(f.read()) - if not m: - print self.style.ERROR("Unknown text format: %s" % book.slug) - skipped += 1 - continue - - processed += 1 - last_word = '' - text = unicode(m.group(1), 'utf-8').lower() - for letter in text: - mydict = lesmianator.setdefault(last_word, {}) - myval = mydict.setdefault(letter, 0) - mydict[letter] += 1 - last_word = last_word[-2:] + letter + f = open(book.txt_file.path) + m = re_text(f.read()) + if not m: + print self.style.ERROR("Unknown text format: %s" % book.slug) + skipped += 1 + continue + + processed += 1 + last_word = '' + text = unicode(m.group(1), 'utf-8').lower() + for letter in text: + mydict = lesmianator.setdefault(last_word, {}) + myval = mydict.setdefault(letter, 0) + mydict[letter] += 1 + last_word = last_word[-2:] + letter + f.close() if not processed: if skipped: diff --git a/apps/lesmianator/models.py b/apps/lesmianator/models.py index 0a59ddee4..5103ebbf1 100644 --- a/apps/lesmianator/models.py +++ b/apps/lesmianator/models.py @@ -47,46 +47,53 @@ class Poem(models.Model): def __unicode__(self): return "%s (%s...)" % (self.slug, self.text[:20]) - @classmethod - def write(cls, continuations=None, length=3, maxlen=1000): - def choose_word(word, continuations): - try: - choices = sum((continuations[word][post] for post in continuations[word])) - r = randint(0, choices - 1) + @staticmethod + def choose_letter(word, continuations): + if word not in continuations: + return u'\n' - for post in continuations[word]: - r -= continuations[word][post] - if r < 0: - return post - except KeyError: - return '' + choices = sum((continuations[word][letter] + for letter in continuations[word])) + r = randint(0, choices - 1) + for letter in continuations[word]: + r -= continuations[word][letter] + if r < 0: + return letter + @classmethod + def write(cls, continuations=None, length=3, min_lines=2, maxlen=1000): if continuations is None: continuations = cls.global_dictionary + if not continuations: + return '' letters = [] word = u'' - empty = -10 - lines = 0 - if not continuations: - maxlen = 0 - # want at least two lines, but let Lesmianator end his stanzas - while (empty < 2 or lines < 2) and maxlen: - letter = choose_word(word, continuations) + + finished_stanza_verses = 0 + current_stanza_verses = 0 + verse_start = True + + char_count = 0 + + # do `min_lines' non-empty verses and then stop, + # but let Lesmianator finish his last stanza. + while finished_stanza_verses < min_lines and char_count < maxlen: + letter = cls.choose_letter(word, continuations) letters.append(letter) - word = word[-length+1:] + letter + word = word[-length + 1:] + letter + char_count += 1 + if letter == u'\n': - # count non-empty lines - if empty == 0: - lines += 1 - # - if lines >= 2: - empty += 1 - lines += 1 + if verse_start: + finished_stanza_verses += current_stanza_verses + current_stanza_verses = 0 + else: + current_stanza_verses += 1 + verse_start = True else: - empty = 0 - maxlen -= 1 + verse_start = False return ''.join(letters).strip()