Add InventoryFile.dump()

sphinx/builders/html.py

@@ -31,6 +31,7 @@ from docutils.readers.doctree import Reader as DoctreeReader
 from sphinx import package_dir, __display_version__
 from sphinx.util import jsonimpl, logging, status_iterator
 from sphinx.util.i18n import format_date
+from sphinx.util.inventory import InventoryFile
 from sphinx.util.osutil import SEP, os_path, relative_uri, ensuredir, \
     movefile, copyfile
 from sphinx.util.nodes import inline_all_toctrees
@@ -896,33 +897,8 @@ class StandaloneHTMLBuilder(Builder):
 
     def dump_inventory(self):
         # type: () -> None
-        def safe_name(string):
-            return re.sub("\s+", " ", string)
-
         logger.info(bold('dumping object inventory... '), nonl=True)
-        with open(path.join(self.outdir, INVENTORY_FILENAME), 'wb') as f:
-            f.write((u'# Sphinx inventory version 2\n'
-                     u'# Project: %s\n'
-                     u'# Version: %s\n'
-                     u'# The remainder of this file is compressed using zlib.\n'
-                     % (safe_name(self.config.project),
-                        safe_name(self.config.version))).encode('utf-8'))
-            compressor = zlib.compressobj(9)
-            for domainname, domain in sorted(self.env.domains.items()):
-                for name, dispname, type, docname, anchor, prio in \
-                        sorted(domain.get_objects()):
-                    if anchor.endswith(name):
-                        # this can shorten the inventory by as much as 25%
-                        anchor = anchor[:-len(name)] + '$'
-                    uri = self.get_target_uri(docname)
-                    if anchor:
-                        uri += '#' + anchor
-                    if dispname == name:
-                        dispname = u'-'
-                    f.write(compressor.compress(
-                        (u'%s %s:%s %s %s %s\n' % (name, domainname, type,
-                                                   prio, uri, dispname)).encode('utf-8')))
-            f.write(compressor.flush())
+        InventoryFile.dump(path.join(self.outdir, INVENTORY_FILENAME), self.env, self)
         logger.info('done')
 
     def dump_search_index(self):

sphinx/util/inventory.py

@@ -9,11 +9,14 @@
     :license: BSD, see LICENSE for details.
 """
 import re
+import os
 import zlib
 import codecs
 
 from six import PY3
 
+from sphinx.util import logging
+
 if False:
     # For type annotation
     from typing import Callable, Dict, IO, Iterator, Tuple  # NOQA
@@ -27,6 +30,8 @@ if False:
 BUFSIZE = 16 * 1024
 UTF8StreamReader = codecs.lookup('utf-8')[2]
 
+logger = logging.getLogger(__name__)
+
 
 class ZlibReader(object):
     """Compressed file reader."""
@@ -121,3 +126,35 @@ class InventoryFile(object):
             invdata.setdefault(type, {})[name] = (projname, version,
                                                   location, dispname)
         return invdata
+
+    @classmethod
+    def dump(cls, filename, env, builder):
+        # type: (unicode, BuildEnvironment, Builder) -> None
+        def escape(string):
+            # type: (unicode) -> unicode
+            return re.sub("\s+", " ", string).encode('utf-8')
+
+        with open(os.path.join(filename), 'wb') as f:
+            # header
+            f.write('# Sphinx inventory version 2\n')
+            f.write('# Project: %s\n' % escape(env.config.project))
+            f.write('# Version: %s\n' % escape(env.config.version))
+            f.write('# The remainder of this file is compressed using zlib.\n')
+
+            # body
+            compressor = zlib.compressobj(9)
+            for domainname, domain in sorted(env.domains.items()):
+                for name, dispname, typ, docname, anchor, prio in \
+                        sorted(domain.get_objects()):
+                    if anchor.endswith(name):
+                        # this can shorten the inventory by as much as 25%
+                        anchor = anchor[:-len(name)] + '$'
+                    uri = builder.get_target_uri(docname)
+                    if anchor:
+                        uri += '#' + anchor
+                    if dispname == name:
+                        dispname = u'-'
+                    entry = (u'%s %s:%s %s %s %s\n' %
+                             (name, domainname, typ, prio, uri, dispname))
+                    f.write(compressor.compress(entry.encode('utf-8')))
+            f.write(compressor.flush())

tests/test_build_html.py

@@ -16,6 +16,8 @@ from itertools import cycle, chain
 from six import PY3
 
 from sphinx import __display_version__
+from sphinx.util.inventory import InventoryFile
+
 from util import remove_unicode_literals, strip_escseq
 import xml.etree.cElementTree as ElementTree
 from html5lib import getTreeBuilder, HTMLParser
@@ -1149,3 +1151,29 @@ def test_html_entity(app):
     content = (app.outdir / 'index.html').text()
     for entity in re.findall(r'&([a-z]+);', content, re.M):
         assert entity not in valid_entities
+
+
+@pytest.mark.sphinx('html', testroot='basic')
+def test_html_inventory(app):
+    app.builder.build_all()
+    with open(app.outdir / 'objects.inv') as f:
+        invdata = InventoryFile.load(f, 'http://example.com', os.path.join)
+    assert invdata.keys() == ['std:label', 'std:doc']
+    assert invdata['std:label'].keys() == ['modindex', 'genindex', 'search']
+    assert invdata['std:label']['modindex'] == ('Python',
+                                                '',
+                                                'http://example.com/py-modindex.html',
+                                                'Module Index')
+    assert invdata['std:label']['genindex'] == ('Python',
+                                                '',
+                                                'http://example.com/genindex.html',
+                                                'Index')
+    assert invdata['std:label']['search'] == ('Python',
+                                              '',
+                                              'http://example.com/search.html',
+                                              'Search Page')
+    assert invdata['std:doc'].keys() == ['index']
+    assert invdata['std:doc']['index'] == ('Python',
+                                           '',
+                                           'http://example.com/index.html',
+                                           'The basic Sphinx documentation for testing')