Search in stagin phase - waiting for final guidelines on new wl

[wolnelektury.git] / apps / catalogue / models.py

diff --git a/apps/catalogue/models.py b/apps/catalogue/models.py
index bb0067f..1a2e8f8 100644 (file)
--- a/apps/catalogue/models.py
+++ b/apps/catalogue/models.py
@@ -31,6 +31,8 @@ from slughifi import slughifi
 from sortify import sortify
 from os import unlink
 
+import search
+
 TAG_CATEGORIES = (
     ('author', _('author')),
     ('epoch', _('epoch')),
@@ -220,10 +222,11 @@ 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)
@@ -466,9 +469,6 @@ class Book(models.Model):
         from tempfile import NamedTemporaryFile
         import os
 
-        # remove zip with all pdf files
-        remove_zip(settings.ALL_PDF_ZIP)
-
         try:
             pdf_file = NamedTemporaryFile(delete=False)
             pdf.transform(ORMDocProvider(self),
@@ -480,6 +480,9 @@ class Book(models.Model):
         finally:
             unlink(pdf_file.name)
 
+        # remove zip with all pdf files
+        remove_zip(settings.ALL_PDF_ZIP)
+
     def build_mobi(self):
         """ (Re)builds the MOBI file.
 
@@ -488,9 +491,6 @@ class Book(models.Model):
         from tempfile import NamedTemporaryFile
         import os
 
-        # remove zip with all pdf files
-        remove_zip(settings.ALL_MOBI_ZIP)
-
         try:
             mobi_file = NamedTemporaryFile(suffix='.mobi', delete=False)
             mobi.transform(ORMDocProvider(self), verbose=1,
@@ -502,6 +502,9 @@ 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.
@@ -515,9 +518,6 @@ class Book(models.Model):
             # 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 +535,9 @@ 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
@@ -601,38 +604,42 @@ 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))
-        result = create_zip_task.delay(paths, settings.ALL_EPUB_ZIP)
+    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()
 
-    @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))
-        result = create_zip_task.delay(paths, settings.ALL_PDF_ZIP)
-        return result.wait()
-
-    @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_audiobooks(self):
-        bm = BookMedia.objects.filter(book=self)
-        paths = map(lambda bm: bm.file.path, bm)
-        result = create_zip_task.delay(paths, self.slug)
+        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()
 
+    def search_index(self):
+        if settings.SEARCH_INDEX_PARALLEL:
+            if instance(settings.SEARCH_INDEX_PARALLEL, int):
+                idx = search.ReusableIndex(threads=4)
+            else:
+                idx = search.ReusableIndex()
+        else:
+            idx = search.Index()
+            
+        idx.open()
+        try:
+            idx.index_book(self)
+        finally:
+            idx.close()
+
     @classmethod
     def from_xml_file(cls, xml_file, **kwargs):
         # use librarian to parse meta-data
@@ -648,7 +655,8 @@ class Book(models.Model):
 
     @classmethod
     def from_text_and_meta(cls, raw_file, book_info, overwrite=False,
-            build_epub=True, build_txt=True, build_pdf=True, build_mobi=True):
+            build_epub=True, build_txt=True, build_pdf=True, build_mobi=True,
+            search_index=True):
         import re
 
         # check for parts before we do anything
@@ -727,6 +735,9 @@ class Book(models.Model):
         if not settings.NO_BUILD_MOBI and build_mobi:
             book.build_mobi()
 
+        if not settings.NO_SEARCH_INDEX and search_index:
+            book.search_index()
+
         book_descendants = list(book.children.all())
         # add l-tag to descendants and their fragments
         # delete unnecessary EPUB files