packaging zip deflated,
authorRadek Czajka <radoslaw.czajka@nowoczesnapolska.org.pl>
Tue, 13 Sep 2011 15:11:27 +0000 (17:11 +0200)
committerRadek Czajka <radoslaw.czajka@nowoczesnapolska.org.pl>
Tue, 13 Sep 2011 15:11:27 +0000 (17:11 +0200)
lesmianator update

apps/catalogue/management/commands/pack.py
apps/lesmianator/management/commands/lesmianator.py
apps/lesmianator/models.py

index 80f612a..a300aff 100755 (executable)
@@ -53,7 +53,7 @@ class Command(BaseCommand):
         if exclude:
             books = [book for book in books if book.slug not in exclude.split(',')]
 
         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:
 
         processed = skipped = 0
         for book in books:
index c021921..5412bee 100644 (file)
@@ -14,7 +14,7 @@ from django.conf import settings
 from catalogue.models import Book, Tag
 
 # extract text from text file
 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):
 
 
 class Command(BaseCommand):
@@ -66,21 +66,22 @@ class Command(BaseCommand):
                     print self.style.NOTICE('%s has no TXT file' % book.slug)
                 skipped += 1
                 continue
                     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:
 
         if not processed:
             if skipped:
index 0a59dde..5103ebb 100644 (file)
@@ -47,46 +47,53 @@ class Poem(models.Model):
     def __unicode__(self):
         return "%s (%s...)" % (self.slug, self.text[:20])
 
     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 continuations is None:
             continuations = cls.global_dictionary
+        if not continuations:
+            return ''
 
         letters = []
         word = u''
 
         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)
             letters.append(letter)
-            word = word[-length+1:] + letter
+            word = word[-length + 1:] + letter
+            char_count += 1
+
             if letter == u'\n':
             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:
             else:
-                empty = 0
-            maxlen -= 1
+                verse_start = False
 
         return ''.join(letters).strip()
 
 
         return ''.join(letters).strip()