Merge lp:~vila/bzr/940164-native-texinfo into lp:bzr/2.5

[Vincent Ladeuil]

Sphinx now provides a texinfo builder better than ours.

While debugging the doc.canonical.bazaar.com issues I needed to reproduce
some doc generation locally and was faced with a Yak refusing to go away
without a shave...

It turned out disabling our texinfo sphinx builder was a one-liner but
digging only slightly revealed that the texinfo builder provided by sphinx
was far better than ours.

So, bitting the bullet I dug a bit further and this is the result: we can
now process all the documentation we processed for the latex builder
(i.e. the vast majority) and get some .info files that emacs seem quite
happy to process. I didn't check them all but this is a (very) significant
step.

This patch does mainly two things:

- remove our texinfo builder and its tests,

- better wire the Makefiles so one can use ``make texinfo-sphinx`` to
  produce both the ``.texi`` files and the ``.info`` ones.

I'm targetting 2.5 because there is a risk that our texinfo builder break
some builders running the test suite by being unlucky enough to have sphinx
installed.

This doesn't impact bzr/bzrlib themselves so I think it's a safe bet and
will avoid troubles during the whole 2.5 series life cycle.

[Martin Packman (gz)]

Code changes seem reasonable, as does a 2.5 landing. Does anything need to be done when running against a too old sphinx to at least give a reasonable error?

[Vincent Ladeuil (vila)]

Well, against a too old sphinx, the error will be: 'texinfo ? What's that ?' which seems fine given that the files produced were aimed at debugging the proposed builder. The files produced by the sphinx's texinfo are not perfect yet, nor is our use of cross-references, so there is still work to be done but it should happen with the official builder, not ours.

[Martin Packman (gz)]

That's probably a clear enough error.

review: Approve

[Vincent Ladeuil (vila)]

sent to pqm by email

[Vincent Ladeuil (vila)]

sent to pqm by email

=== modified file 'Makefile'
--- Makefile	2011-12-08 19:00:14 +0000
+++ Makefile	2012-03-12 14:56:53 +0000
@@ -192,6 +192,14 @@
 	cd doc/developers && make htmlhelp
 
 
+# Build the texinfo files using Sphinx.
+texinfo-sphinx: $(SPHINX_DEPENDENCIES)
+	cd doc/en && make texinfo
+	cd doc/es && make texinfo
+	cd doc/ru && make texinfo
+	cd doc/ja && make texinfo
+	cd doc/developers && make texinfo
+
 ### Documentation Website ###
 
 # Where to build the website

=== removed directory 'bzrlib/doc_generate/builders'
=== removed file 'bzrlib/doc_generate/builders/__init__.py'
--- bzrlib/doc_generate/builders/__init__.py	2011-12-19 13:23:58 +0000
+++ bzrlib/doc_generate/builders/__init__.py	1970-01-01 00:00:00 +0000
@@ -1,20 +0,0 @@
-# Copyright (C) 2010 Canonical Ltd
-#
-# This program is free software; you can redistribute it and/or modify
-# it under the terms of the GNU General Public License as published by
-# the Free Software Foundation; either version 2 of the License, or
-# (at your option) any later version.
-#
-# This program is distributed in the hope that it will be useful,
-# but WITHOUT ANY WARRANTY; without even the implied warranty of
-# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
-# GNU General Public License for more details.
-#
-# You should have received a copy of the GNU General Public License
-# along with this program; if not, write to the Free Software
-# Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA
-
-"""Placeholder to allow sphinx builders imports."""
-
-from __future__ import absolute_import
-

=== removed file 'bzrlib/doc_generate/builders/texinfo.py'
--- bzrlib/doc_generate/builders/texinfo.py	2012-02-24 13:35:31 +0000
+++ bzrlib/doc_generate/builders/texinfo.py	1970-01-01 00:00:00 +0000
@@ -1,44 +0,0 @@
-# Copyright (C) 2010 Canonical Ltd
-#
-# This program is free software; you can redistribute it and/or modify
-# it under the terms of the GNU General Public License as published by
-# the Free Software Foundation; either version 2 of the License, or
-# (at your option) any later version.
-#
-# This program is distributed in the hope that it will be useful,
-# but WITHOUT ANY WARRANTY; without even the implied warranty of
-# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
-# GNU General Public License for more details.
-#
-# You should have received a copy of the GNU General Public License
-# along with this program; if not, write to the Free Software
-# Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA
-
-"""A sphinx builder producing texinfo output."""
-
-from __future__ import absolute_import
-
-from sphinx import builders
-from sphinx.builders import text as _text_builder
-
-from bzrlib.doc_generate.writers import texinfo as texinfo_writer
-
-class TexinfoBuilder(_text_builder.TextBuilder):
-
-    name = 'texinfo'
-    format = 'texinfo'
-    out_suffix = '.texi'
-
-    def prepare_writing(self, docnames):
-        self.writer = texinfo_writer.TexinfoWriter(self)
-
-    def get_target_uri(self, docname, typ=None):
-        # FIXME: Revisit when info file generation is defined (the suffix is
-        # left here for clarity but the final version may just get rid of
-        # it). And we probalby will join several files into bigger info files
-        # anyway. -- vila 20100506
-        return docname + '.info'
-
-
-def setup(app):
-    app.add_builder(TexinfoBuilder)

=== modified file 'bzrlib/doc_generate/conf.py'
--- bzrlib/doc_generate/conf.py	2011-12-18 15:28:38 +0000
+++ bzrlib/doc_generate/conf.py	2012-03-12 14:56:53 +0000
@@ -21,8 +21,7 @@
 
 # Add any Sphinx extension module names here, as strings. They can be extensions
 # coming with Sphinx (named 'sphinx.ext.*') or your custom ones.
-extensions = ['sphinx.ext.ifconfig',
-              'bzrlib.doc_generate.builders.texinfo']
+extensions = ['sphinx.ext.ifconfig',]
 
 # Add any paths that contain templates here, relative to this directory.
 templates_path = ['_templates']

=== removed directory 'bzrlib/doc_generate/writers'
=== removed file 'bzrlib/doc_generate/writers/__init__.py'
--- bzrlib/doc_generate/writers/__init__.py	2011-12-19 13:23:58 +0000
+++ bzrlib/doc_generate/writers/__init__.py	1970-01-01 00:00:00 +0000
@@ -1,20 +0,0 @@
-# Copyright (C) 2010 Canonical Ltd
-#
-# This program is free software; you can redistribute it and/or modify
-# it under the terms of the GNU General Public License as published by
-# the Free Software Foundation; either version 2 of the License, or
-# (at your option) any later version.
-#
-# This program is distributed in the hope that it will be useful,
-# but WITHOUT ANY WARRANTY; without even the implied warranty of
-# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
-# GNU General Public License for more details.
-#
-# You should have received a copy of the GNU General Public License
-# along with this program; if not, write to the Free Software
-# Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA
-
-"""Placeholder to allow sphinx writer imports."""
-
-from __future__ import absolute_import
-

=== removed file 'bzrlib/doc_generate/writers/texinfo.py'
--- bzrlib/doc_generate/writers/texinfo.py	2011-12-19 13:23:58 +0000
+++ bzrlib/doc_generate/writers/texinfo.py	1970-01-01 00:00:00 +0000
@@ -1,534 +0,0 @@
-# Copyright (C) 2010 Canonical Ltd
-#
-# This program is free software; you can redistribute it and/or modify
-# it under the terms of the GNU General Public License as published by
-# the Free Software Foundation; either version 2 of the License, or
-# (at your option) any later version.
-#
-# This program is distributed in the hope that it will be useful,
-# but WITHOUT ANY WARRANTY; without even the implied warranty of
-# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
-# GNU General Public License for more details.
-#
-# You should have received a copy of the GNU General Public License
-# along with this program; if not, write to the Free Software
-# Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA
-
-"""A sphinx/docutil writer producing texinfo output."""
-
-from __future__ import absolute_import
-
-from docutils import (
-    nodes,
-    writers,
-    )
-
-DEBUG = 0
-
-class TexinfoWriter(writers.Writer):
-
-    supported = ('texinfo',)
-    settings_spec = ('No options here.', '', ())
-    settings_defaults = {}
-
-    output = None
-
-    def __init__(self, builder):
-        writers.Writer.__init__(self)
-        self.builder = builder
-
-    def translate(self):
-        visitor = TexinfoTranslator(self.document, self.builder)
-        self.document.walkabout(visitor)
-        self.output = visitor.body
-
-
-class TexinfoTranslator(nodes.NodeVisitor):
-
-    section_names = ['chapter', 'section', 'subsection', 'subsubsection']
-    """texinfo section names differ from the sphinx ones.
-
-    Since this can be confusing, the correspondences are shown below
-    (shpinx -> texinfo):
-    part       -> chapter
-    chapter    -> section
-    section    -> subsection
-    subsection -> subsubsection
-
-    Additionally, sphinx defines subsubsections and paragraphs which are
-    handled as @heading (unnumbered).
-    """
-
-    def __init__(self, document, builder):
-        nodes.NodeVisitor.__init__(self, document)
-        self.builder = builder
-        # toctree uses some nodes for different purposes (namely:
-        # compact_paragraph, bullet_list) that needs to know when they are
-        # processing a toctree.
-        self.in_toctree = False
-        # sections can be embedded and produce different directives depending
-        # on the depth.
-        self.section_level = -1
-        # By default paragraghs are separated by newlines, but there are some
-        # exceptions that set it to '' for some subtrees instead
-        self.paragraph_sep = '\n'
-
-    # The whole document
-
-    def visit_document(self, node):
-        if DEBUG:
-            import sys
-            sys.stdout.write(node.pformat().encode('utf8'))
-        set_item_list_collector(node, 'text')
-
-    def depart_document(self, node):
-        # FIXME: info requires a Top node for each info file, but unless we
-        # chose a global layout to divide the overall documentation into a set
-        # of info files, there is no criteria to decide for a title.
-        top_cmd = '''\
-This file has been converted using a beta rst->texinfo converter. 
-Most of the info links are currently bogus, don't report bugs about them,
-this is currently worked on.
-@node Top
-@top Placeholder
-'''
-        self.body = top_cmd + ''.join(node['text'])
-
-    # Layout
-
-    def visit_section(self, node):
-        self.section_level += 1
-        set_item_list_collector(node, 'text')
-
-    def depart_section(self, node):
-        title = node['title']
-        ids = node.get('ids', [])
-        try:
-            section_name = self.section_names[self.section_level]
-        except IndexError:
-            # Just use @heading, it's not numbered anyway
-            section_name = 'heading'
-        if ids:
-            # There shouldn't be different ids for a section, so until we
-            # encounter bugs, just take the first one.
-            node_cmd = '@node %s\n' % (ids[0],)
-        else:
-            node_cmd = ''
-        section_cmd = '@%s %s\n' % (section_name, title)
-        text = ''.join(node['text'])
-        node.parent.collect_text(node_cmd + section_cmd + text)
-        self.section_level -= 1
-
-    def visit_topic(self, node):
-        pass
-
-    def depart_topic(self, node):
-        pass
-
-    def visit_compound(self, node):
-        # compound is new in sphinx >= 1.0 and just add a optional layer so we
-        # relay the text to the parent when it occurs. This may requires a
-        # cleaner approach once we settle on which sphinx versions we want to
-        # support.
-        set_item_list_collector(node, 'text')
-
-    def depart_compound(self, node):
-        text = ''.join(node['text'])
-        node.parent.collect_text(text)
-
-    def visit_paragraph(self, node):
-        set_item_list_collector(node, 'text')
-
-    def depart_paragraph(self, node):
-        # End the paragraph with a new line (or '' depending on the parent) and
-        # leave a blank line after it.
-        text = ''.join(node['text']) + self.paragraph_sep * 2
-        node.parent.collect_text(text)
-
-    def visit_compact_paragraph(self, node):
-        set_item_list_collector(node, 'text')
-        if node.has_key('toctree'):
-            self.in_toctree = True
-        elif self.in_toctree:
-            set_item_collector(node, 'reference')
-
-    def depart_compact_paragraph(self, node):
-        # FIXME: Using a different visitor specific to toctree may be a better
-        # design and makes code clearer. -- vila 20100708
-        if node.has_key('toctree'):
-            if node['text']:
-                node.parent.collect_text('@menu\n')
-                node.parent.collect_text(''.join(node['text']))
-                node.parent.collect_text('@end menu\n')
-            self.in_toctree = False
-        elif self.in_toctree:
-            # * FIRST-ENTRY-NAME:(FILENAME)NODENAME.     DESCRIPTION
-            # XXX: the file name should probably be adjusted to the targeted
-            # info file name
-            node_name, file_name, entry_name = node['reference']
-            if not node_name:
-                node_name = entry_name
-            description = '' # We can't specify a description in rest AFAICS
-            # XXX: What if :maxdepth: is not 1 ?
-            text = '* %s: (%s)%s. %s\n' % (entry_name, file_name,
-                                           node_name, description)
-            node.parent.collect_text(text)
-        else:
-            # End the paragraph with a new line (or '' depending on the parent)
-            # and leave a blank line after it.
-            text = ''.join(node['text']) + self.paragraph_sep * 2
-            node.parent.collect_text(text)
-
-    def visit_literal_block(self, node):
-        set_item_collector(node, 'text')
-
-    def depart_literal_block(self, node):
-        text = '@samp{%s}' % ''.join(node['text']) + self.paragraph_sep * 2
-        node.parent.collect_text(text)
-
-    def visit_block_quote(self, node):
-        set_item_list_collector(node, 'text')
-
-    def depart_block_quote(self, node):
-        node.parent.collect_text('@example\n')
-        node.parent.collect_text(''.join(node['text']))
-        node.parent.collect_text('@end example\n\n')
-
-    def depart_warning(self, node):
-        pass
-
-    def visit_warning(self, node):
-        raise nodes.SkipNode # Not implemented yet
-
-    def visit_note(self, node):
-        raise nodes.SkipNode # Not implemented yet
-
-    def depart_note(self, node):
-        pass
-
-    def visit_footnote(self, node):
-        raise nodes.SkipNode # Not implemented yet
-
-    def depart_footnote(self, node):
-        pass
-
-    def visit_comment(self, node):
-        raise nodes.SkipNode # Not implemented yet
-
-    # Attributes
-
-    def visit_title(self, node):
-        set_item_collector(node, 'text')
-
-    def depart_title(self, node):
-        node.parent['title'] = node['text']
-
-    def visit_label(self, node):
-        raise nodes.SkipNode # Not implemented yet
-
-    def visit_substitution_definition(self, node):
-        raise nodes.SkipNode # Not implemented yet
-
-    # Plain text
-
-    def visit_Text(self, node):
-        pass
-
-    def depart_Text(self, node):
-        text = node.astext()
-        if '@' in text:
-            text = text.replace('@', '@@')
-        if '{' in text:
-            text = text.replace('{', '@{')
-        if '}' in text:
-            text = text.replace('}', '@}')
-        node.parent.collect_text(text)
-
-
-    # Styled text
-
-    def visit_emphasis(self, node):
-        set_item_collector(node, 'text')
-
-    def depart_emphasis(self, node):
-        text = '@emph{%s}' % node['text']
-        node.parent.collect_text(text)
-
-    def visit_strong(self, node):
-        set_item_collector(node, 'text')
-
-    def depart_strong(self, node):
-        text = '@strong{%s}' % node['text']
-        node.parent.collect_text(text)
-
-    def visit_literal(self, node):
-        set_item_collector(node, 'text')
-
-    def depart_literal(self, node):
-        text = '@code{%s}' % node['text']
-        node.parent.collect_text(text)
-
-    # Lists
-
-    def _decorate_list(self, item_list, collect, item_fmt='%s',
-                       head=None, foot=None):
-        if head is not None:
-            collect(head)
-        for item in item_list:
-            collect(item_fmt % item)
-        if foot is not None:
-            collect(foot)
-
-    def visit_bullet_list(self, node):
-        set_item_list_collector(node, 'list_item')
-
-    def depart_bullet_list(self, node):
-        l = node['list_item']
-        if self.in_toctree:
-            self._decorate_list(node['list_item'], node.parent.collect_text)
-        else:
-            self._decorate_list(node['list_item'], node.parent.collect_text,
-                                '@item\n%s',
-                                # FIXME: Should respect the 'bullet' attribute
-                                '@itemize @bullet\n', '@end itemize\n')
-
-    def visit_enumerated_list(self, node):
-        set_item_list_collector(node, 'list_item')
-
-    def depart_enumerated_list(self, node):
-        self._decorate_list(node['list_item'], node.parent.collect_text,
-                            '@item\n%s',
-                            '@enumerate\n', '@end enumerate\n')
-
-    def visit_definition_list(self, node):
-        raise nodes.SkipNode # Not implemented yet
-
-    def depart_definition_list(self, node):
-        raise nodes.SkipNode # Not implemented yet
-
-    def visit_definition_list_item(self, node):
-        raise nodes.SkipNode # Not implemented yet
-
-    def depart_definition_list_item(self, node):
-        pass
-
-    def visit_term(self, node):
-        raise nodes.SkipNode # Not implemented yet
-
-    def depart_term(self, node):
-        pass
-
-    def visit_definition(self, node):
-        raise nodes.SkipNode # Not implemented yet
-
-    def depart_definition(self, node):
-        pass
-
-    def visit_field_list(self, node):
-        raise nodes.SkipNode # Not implemented yet
-
-    def depart_field_list(self, node):
-        pass
-
-    def visit_field(self, node):
-        raise nodes.SkipNode # Not implemented yet
-
-    def depart_field(self, node):
-        pass
-
-    def visit_field_name(self, node):
-        raise nodes.SkipNode # Not implemented yet
-
-    def depart_field_name(self, node):
-        pass
-
-    def visit_field_body(self, node):
-        raise nodes.SkipNode # Not implemented yet
-
-    def depart_field_body(self, node):
-        pass
-
-    def visit_list_item(self, node):
-        set_item_list_collector(node, 'text')
-
-    def depart_list_item(self, node):
-        text = ''.join(node['text'])
-        node.parent.collect_list_item(text)
-
-    def visit_option_list(self, node):
-        raise nodes.SkipNode # Not implemented yet
-
-    def depart_option_list(self, node):
-        pass
-
-    def visit_option_list_item(self, node):
-        pass
-
-    def depart_option_list_item(self, node):
-        pass
-
-    def visit_option_group(self, node):
-        pass
-
-    def depart_option_group(self, node):
-        pass
-
-    def visit_option(self, node):
-        pass
-
-    def depart_option(self, node):
-        pass
-
-    def visit_option_string(self, node):
-        pass
-    def depart_option_string(self, node):
-        pass
-
-    def visit_option_argument(self, node):
-        pass
-
-    def depart_option_argument(self, node):
-        pass
-
-    def visit_description(self, node):
-        pass
-    def depart_description(self, node):
-        pass
-
-    # Tables
-    def visit_table(self, node):
-        set_item_collector(node, 'table')
-
-    def depart_table(self, node):
-        node.parent.collect_text(node['table'])
-
-    def visit_tgroup(self, node):
-        set_item_list_collector(node, 'colspec')
-        set_item_collector(node, 'head_entries')
-        set_item_collector(node, 'body_rows')
-
-    def depart_tgroup(self, node):
-        header = []
-        # The '@multitable {xxx}{xxx}' line
-        self._decorate_list(node['colspec'], header.append,
-                            '{%s}', '@multitable ', '\n')
-        # The '@headitem xxx @tab yyy...' line
-        head_entries = node['head_entries']
-        if head_entries is not None:
-            # Not all tables define titles for the columns... rest parser bug ?
-            # FIXME: need a test
-            self._decorate_list(head_entries[1:], header.append,
-                                ' @tab %s',
-                                '@headitem %s' % head_entries[0], '\n')
-        header = ''.join(header)
-        # The '@item xxx\n @tab yyy\n ...' lines
-        body_rows = node['body_rows']
-        rows = []
-        for r in body_rows:
-            self._decorate_list(r[1:], rows.append,
-                                '@tab %s\n', '@item %s\n' % r[0])
-        footer = '@end multitable\n'
-        node.parent.collect_table(header + ''.join(rows) + footer)
-
-    def visit_colspec(self, node):
-        pass
-
-    def depart_colspec(self, node):
-        node.parent.collect_colspec('x' * node['colwidth'])
-
-    def visit_thead(self, node):
-        set_item_collector(node, 'row')
-
-    def depart_thead(self, node):
-        node.parent.collect_head_entries(node['row'])
-
-    def visit_tbody(self, node):
-        set_item_list_collector(node, 'row')
-
-    def depart_tbody(self, node):
-        node.parent.collect_body_rows(node['row'])
-
-    def visit_row(self, node):
-        set_item_list_collector(node, 'entry')
-
-    def depart_row(self, node):
-        node.parent.collect_row(node['entry'])
-
-    def visit_entry(self, node):
-        set_item_list_collector(node, 'text')
-        node['par_sep_orig'] = self.paragraph_sep
-        self.paragraph_sep = ''
-
-    def depart_entry(self, node):
-        node.parent.collect_entry(''.join(node['text']))
-        self.paragraph_sep = node['par_sep_orig']
-
-    # References
-
-    def visit_reference(self, node):
-        for c in node.children:
-            if getattr(c, 'parent', None) is None:
-                # Bug sphinx
-                node.setup_child(c)
-        set_item_collector(node, 'text')
-
-    def depart_reference(self, node):
-        anchorname = node.get('anchorname', None)
-        refuri = node.get('refuri', None)
-        refid = node.get('refid', None)
-        text = ''.join(node['text'])
-        collect = getattr(node.parent, 'collect_reference', None)
-        if collect is not None:
-            if not self.in_toctree:
-                raise AssertionError('collect_reference is specific to toctree')
-            if anchorname is None:
-                anchorname = ''
-            if refuri is None:
-                refuri = ''
-            collect((anchorname, refuri, text))
-        elif refuri is not None:
-            node.parent.collect_text('@uref{%s,%s}' % (refuri, text))
-        elif refid is not None:
-            # Info format requires that a reference is followed by some
-            # punctuation char ('.', ','. ')', etc). Rest is more liberal. To
-            # accommodate, we use pxref inside parenthesis.
-            node.parent.collect_text('%s (@pxref{%s})' % (text, refid))
-
-    def visit_footnote_reference(self, node):
-        raise nodes.SkipNode # Not implemented yet
-
-    def visit_citation_reference(self, node):
-        raise nodes.SkipNode # Not implemented yet
-
-    def visit_title_reference(self, node):
-        raise nodes.SkipNode # Not implemented yet
-
-    def depart_title_reference(self, node):
-        pass
-
-    def visit_target(self, node):
-        raise nodes.SkipNode # Not implemented yet
-
-    def depart_target(self, node):
-        pass
-
-    def visit_image(self, node):
-        raise nodes.SkipNode # Not implemented yet
-
-# Helpers to collect data in parent node
-
-def set_item_collector(node, name):
-    node[name] = None
-    def set_item(item):
-        node[name] = item
-    setattr(node, 'collect_' + name, set_item)
-
-
-def set_item_list_collector(node, name, sep=''):
-    node[name] = []
-    node[name + '_sep'] = sep
-    def append_item(item):
-        node[name].append(item)
-    setattr(node, 'collect_' + name, append_item)
-
-

=== modified file 'bzrlib/tests/__init__.py'
--- bzrlib/tests/__init__.py	2012-02-15 22:43:50 +0000
+++ bzrlib/tests/__init__.py	2012-03-12 14:56:53 +0000
@@ -3909,7 +3909,6 @@
         'bzrlib.doc',
         'bzrlib.tests.blackbox',
         'bzrlib.tests.commands',
-        'bzrlib.tests.doc_generate',
         'bzrlib.tests.per_branch',
         'bzrlib.tests.per_bzrdir',
         'bzrlib.tests.per_controldir',

=== removed directory 'bzrlib/tests/doc_generate'
=== removed file 'bzrlib/tests/doc_generate/__init__.py'
--- bzrlib/tests/doc_generate/__init__.py	2011-01-10 22:20:12 +0000
+++ bzrlib/tests/doc_generate/__init__.py	1970-01-01 00:00:00 +0000
@@ -1,113 +0,0 @@
-# Copyright (C) 2010, 2011 Canonical Ltd
-#
-# This program is free software; you can redistribute it and/or modify
-# it under the terms of the GNU General Public License as published by
-# the Free Software Foundation; either version 2 of the License, or
-# (at your option) any later version.
-#
-# This program is distributed in the hope that it will be useful,
-# but WITHOUT ANY WARRANTY; without even the implied warranty of
-# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
-# GNU General Public License for more details.
-#
-# You should have received a copy of the GNU General Public License
-# along with this program; if not, write to the Free Software
-# Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA
-
-"""Documentation generation tests."""
-
-import os
-from bzrlib import tests
-from bzrlib.doc_generate import (
-    # FIXME: doc/en/conf.py should be used here, or rather merged into
-    # bzrlib/doc_generate/conf.py -- vila 20100429
-    conf,
-    )
-from bzrlib.tests import features
-
-
-def load_tests(basic_tests, module, loader):
-    suite = loader.suiteClass()
-    # add the tests for this module
-    suite.addTests(basic_tests)
-
-    testmod_names = [
-        'builders',
-        'writers',
-        ]
-    # add the tests for the sub modules
-    suite.addTests(loader.loadTestsFromModuleNames(
-            ['bzrlib.tests.doc_generate.' + name
-             for name in testmod_names]))
-
-    return suite
-
-
-class TestSphinx(tests.TestCaseInTempDir):
-    """Base class for sphinx tests.
-
-    This is used for both the builder and the writer until a better solution is
-    found to test at a lower level.
-    """
-
-    _test_needs_features = [features.sphinx]
-
-    def sphinx_version(self):
-        # Convert to a tuple to avoid traps in string comparison
-        # ( '1.12' < '1.6' but (1, 12) > (1, 6) )
-        return tuple(map(int, features.sphinx.module.__version__.split('.')))
-
-    def make_sphinx(self):
-        out = tests.StringIOWrapper()
-        err = tests.StringIOWrapper()
-        from sphinx import application
-        app = application.Sphinx(
-            '.', confdir=os.path.dirname(conf.__file__), outdir='.',
-            doctreedir='.',
-            buildername='texinfo',
-            confoverrides={},
-            status=out, warning=err,
-            freshenv=True)
-        return app, out, err
-
-    def build(self, app, all_files=True, file_names=None):
-        if file_names is None:
-            file_names = []
-        app.build(all_files, file_names)
-
-    # FIXME: something smells wrong here as we can't process a single file
-    # alone. On top of that, it seems the doc tree must contain an index.txt
-    # file. We may need a texinfo builder ? -- vila 20100505
-
-    def create_content(self, content):
-        """Put content into a single file.
-
-        This is appropriate for simple tests about the content of a single file.
-        """
-        app, out, err = self.make_sphinx()
-        self.build_tree_contents([('index.txt', content),])
-        self.build(app)
-
-    def assertContent(self, expected, header=None, end=None):
-        """Check the content of the file created with creste_content().
-
-        Most texinfo constructs can be tested this way without caring for any
-        boilerplate that texinfo may require at the beginning or the end of the
-        file.
-        """
-        if header is None:
-            # default boilerplate
-            header = '''\
-This file has been converted using a beta rst->texinfo converter. 
-Most of the info links are currently bogus, don't report bugs about them,
-this is currently worked on.
-@node Top
-@top Placeholder
-'''
-        if end is None:
-            # By default we test constructs that are embedded into a paragraph
-            # which always end with two \n (even if the input has none)
-            end = '\n\n'
-        self.assertFileEqual(header + expected + end, 'index.texi')
-
-

=== removed directory 'bzrlib/tests/doc_generate/builders'
=== removed file 'bzrlib/tests/doc_generate/builders/__init__.py'
--- bzrlib/tests/doc_generate/builders/__init__.py	2010-04-29 16:10:59 +0000
+++ bzrlib/tests/doc_generate/builders/__init__.py	1970-01-01 00:00:00 +0000
@@ -1,36 +0,0 @@
-# Copyright (C) 2010 Canonical Ltd
-#
-# This program is free software; you can redistribute it and/or modify
-# it under the terms of the GNU General Public License as published by
-# the Free Software Foundation; either version 2 of the License, or
-# (at your option) any later version.
-#
-# This program is distributed in the hope that it will be useful,
-# but WITHOUT ANY WARRANTY; without even the implied warranty of
-# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
-# GNU General Public License for more details.
-#
-# You should have received a copy of the GNU General Public License
-# along with this program; if not, write to the Free Software
-# Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA
-
-
-"""Sphinx builders tests."""
-
-from bzrlib.tests import features
-
-def load_tests(basic_tests, module, loader):
-    suite = loader.suiteClass()
-    # add the tests for this module
-    suite.addTests(basic_tests)
-
-    if features.sphinx.available():
-        testmod_names = [
-            'texinfo',
-            ]
-        # add the tests for the sub modules
-        suite.addTests(loader.loadTestsFromModuleNames(
-                ['bzrlib.tests.doc_generate.builders.test_' + name
-                 for name in testmod_names]))
-
-    return suite

=== removed file 'bzrlib/tests/doc_generate/builders/test_texinfo.py'
--- bzrlib/tests/doc_generate/builders/test_texinfo.py	2011-05-13 12:51:05 +0000
+++ bzrlib/tests/doc_generate/builders/test_texinfo.py	1970-01-01 00:00:00 +0000
@@ -1,71 +0,0 @@
-# Copyright (C) 2010, 2011 Canonical Ltd
-#
-# This program is free software; you can redistribute it and/or modify
-# it under the terms of the GNU General Public License as published by
-# the Free Software Foundation; either version 2 of the License, or
-# (at your option) any later version.
-#
-# This program is distributed in the hope that it will be useful,
-# but WITHOUT ANY WARRANTY; without even the implied warranty of
-# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
-# GNU General Public License for more details.
-#
-# You should have received a copy of the GNU General Public License
-# along with this program; if not, write to the Free Software
-# Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA
-
-"""sphinx texinfo builder tests."""
-
-from bzrlib import tests
-from bzrlib.doc_generate import (
-    # FIXME: doc/en/conf.py should be used here, or rather merged into
-    # bzrlib/doc_generate/conf.py -- vila 20100429
-    conf,
-    )
-from bzrlib.tests import (
-    doc_generate as test_dg, # Avoid clash with from bzrlib import doc_generate
-    )
-
-
-class TestBuilderDefined(tests.TestCase):
-
-    def test_builder_defined(self):
-        self.assertTrue('bzrlib.doc_generate.builders.texinfo'
-                        in conf.extensions)
-
-class TestBuilderLoaded(test_dg.TestSphinx):
-
-    def test_builder_loaded(self):
-        app, out, err = self.make_sphinx()
-        self.assertTrue('texinfo' in app.builderclasses)
-
-
-class TestFileProduction(test_dg.TestSphinx):
-
-    def test_files_generated(self):
-        self.build_tree_contents(
-            [('index.txt', """
-Table of Contents
-=================
-
-.. toctree::
-   :maxdepth: 1
-
-   content
-"""),
-             ('content.txt', """
-
-bzr 0.0.8
-*********
-
-Improvements
-============
-
-* Adding a file whose parent directory is not versioned will
-  implicitly add the parent, and so on up to the root.
-"""),
-             ])
-        app, out, err = self.make_sphinx()
-        self.build(app)
-        self.assertPathExists('index.texi')
-        self.assertPathExists('content.texi')

=== removed directory 'bzrlib/tests/doc_generate/writers'
=== removed file 'bzrlib/tests/doc_generate/writers/__init__.py'
--- bzrlib/tests/doc_generate/writers/__init__.py	2010-07-06 13:22:32 +0000
+++ bzrlib/tests/doc_generate/writers/__init__.py	1970-01-01 00:00:00 +0000
@@ -1,36 +0,0 @@
-# Copyright (C) 2010 Canonical Ltd
-#
-# This program is free software; you can redistribute it and/or modify
-# it under the terms of the GNU General Public License as published by
-# the Free Software Foundation; either version 2 of the License, or
-# (at your option) any later version.
-#
-# This program is distributed in the hope that it will be useful,
-# but WITHOUT ANY WARRANTY; without even the implied warranty of
-# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
-# GNU General Public License for more details.
-#
-# You should have received a copy of the GNU General Public License
-# along with this program; if not, write to the Free Software
-# Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA
-
-
-"""Sphinx writers tests."""
-
-from bzrlib.tests import features
-
-def load_tests(basic_tests, module, loader):
-    suite = loader.suiteClass()
-    # add the tests for this module
-    suite.addTests(basic_tests)
-
-    if features.sphinx.available():
-        testmod_names = [
-            'texinfo',
-            ]
-        # add the tests for the sub modules
-        suite.addTests(loader.loadTestsFromModuleNames(
-                ['bzrlib.tests.doc_generate.writers.test_' + name
-                 for name in testmod_names]))
-
-    return suite

=== removed file 'bzrlib/tests/doc_generate/writers/test_texinfo.py'
--- bzrlib/tests/doc_generate/writers/test_texinfo.py	2011-03-11 11:33:33 +0000
+++ bzrlib/tests/doc_generate/writers/test_texinfo.py	1970-01-01 00:00:00 +0000
@@ -1,359 +0,0 @@
-# Copyright (C) 2010, 2011 Canonical Ltd
-#
-# This program is free software; you can redistribute it and/or modify
-# it under the terms of the GNU General Public License as published by
-# the Free Software Foundation; either version 2 of the License, or
-# (at your option) any later version.
-#
-# This program is distributed in the hope that it will be useful,
-# but WITHOUT ANY WARRANTY; without even the implied warranty of
-# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
-# GNU General Public License for more details.
-#
-# You should have received a copy of the GNU General Public License
-# along with this program; if not, write to the Free Software
-# Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA
-
-"""sphinx texinfo writer tests."""
-
-from bzrlib import tests
-from bzrlib.tests import (
-    doc_generate as test_dg, # Avoid clash with from bzrlib import doc_generate
-    )
-
-
-class TestTextGeneration(test_dg.TestSphinx):
-
-    def test_special_chars(self):
-        self.create_content("A '@' a '{' and a '}'")
-        self.assertContent("A '@@' a '@{' and a '@}'")
-
-    def test_emphasis(self):
-        self.create_content('*important*')
-        self.assertContent('@emph{important}')
-
-    def test_strong(self):
-        self.create_content('**very important**')
-        self.assertContent('@strong{very important}')
-
-    def test_literal(self):
-        self.create_content('the command is ``foo``')
-        self.assertContent('the command is @code{foo}')
-
-    def test_paragraphs(self):
-        self.create_content('''\
-This is a paragraph.
-
-This is another one.
-''')
-        self.assertContent('''\
-This is a paragraph.
-
-This is another one.''')
-
-    def test_literal_block(self):
-        self.create_content('''\
-Do this::
-
-   bzr xxx
-   bzr yyy
-''')
-        self.assertContent('''\
-Do this:
-
-@samp{bzr xxx
-bzr yyy}
-
-''',
-                           end='')
-
-    def test_block_quote(self):
-        self.create_content('''\
-This is an ordinary paragraph, introducing a block quote.
-
-    "It is my business to know things.  That is my trade."
-
-This is another ordinary paragraph.
-''')
-        self.assertContent('''\
-This is an ordinary paragraph, introducing a block quote.
-
-@example
-"It is my business to know things.  That is my trade."
-
-@end example
-
-This is another ordinary paragraph.
-
-''',
-                           # examples are not followed by an empty line
-                           end='')
-
-
-class TestDocumentAttributesGeneration(test_dg.TestSphinx):
-
-    def test_title(self):
-        self.create_content('''\
-####################
-Bazaar Release Notes
-####################
-''')
-        self.assertContent('''\
-@node bazaar-release-notes
-@chapter Bazaar Release Notes
-''',
-                           end='')
-
-
-class TestListGeneration(test_dg.TestSphinx):
-
-    def test_bullet_list(self):
-        self.create_content('''\
-* This is a bulleted list.
-* It has two items, the second
-  item uses two lines.
-''')
-        self.assertContent('''\
-@itemize @bullet
-@item
-This is a bulleted list.
-
-@item
-It has two items, the second
-item uses two lines.
-
-@end itemize
-''',
-                           end='')
-
-    def test_enumerated_list(self):
-        self.create_content('''\
-#. This is a numbered list.
-#. It has two items, the second
-   item uses two lines.
-''')
-        self.assertContent('''\
-@enumerate
-@item
-This is a numbered list.
-
-@item
-It has two items, the second
-item uses two lines.
-
-@end enumerate
-''',
-                           end='')
-
-
-class TestTableGeneration(test_dg.TestSphinx):
-
-    def test_table(self):
-        self.create_content('''\
-  ===========         ================
-  Prefix              Description
-  ===========         ================
-  first               The first
-  second              The second
-  last                The last
-  ===========         ================
-''')
-        # FIXME: Sphinx bug ? Why are tables enclosed in a block_quote
-        # (translated as an @example).
-        self.assertContent('''\
-@example
-@multitable {xxxxxxxxxxx}{xxxxxxxxxxxxxxxx}
-@headitem Prefix @tab Description
-@item first
-@tab The first
-@item second
-@tab The second
-@item last
-@tab The last
-@end multitable
-@end example''')
-
-
-class TestTocTreeGeneration(test_dg.TestSphinx):
-
-    def test_toctree(self):
-        self.build_tree_contents(
-            [('index.txt', """
-Table of Contents
-=================
-
-.. toctree::
-   :maxdepth: 1
-
-   bzr 0.0.8 <bzr-0.0.8>
-"""),
-             ('bzr-0.0.8.txt', """
-
-bzr 0.0.8
-*********
-
-Improvements
-============
-
-* Adding a file whose parent directory is not versioned will
-  implicitly add the parent, and so on up to the root.
-"""),
-             ])
-        app, out, err = self.make_sphinx()
-        self.build(app)
-        self.assertFileEqual("""\
-This file has been converted using a beta rst->texinfo converter. 
-Most of the info links are currently bogus, don't report bugs about them,
-this is currently worked on.
-@node Top
-@top Placeholder
-@node table-of-contents
-@chapter Table of Contents
-@menu
-* bzr 0.0.8: (bzr-0.0.8.info)bzr 0.0.8. 
-@end menu
-""",
-                             'index.texi')
-        self.assertFileEqual("""\
-This file has been converted using a beta rst->texinfo converter. 
-Most of the info links are currently bogus, don't report bugs about them,
-this is currently worked on.
-@node Top
-@top Placeholder
-@node bzr-0-0-8
-@chapter bzr 0.0.8
-@node improvements
-@section Improvements
-@itemize @bullet
-@item
-Adding a file whose parent directory is not versioned will
-implicitly add the parent, and so on up to the root.
-
-@end itemize
-""",
-                             'bzr-0.0.8.texi')
-
-    def test_toctree_empty(self):
-        self.build_tree_contents(
-            [('index.txt', """
-Table of Contents
-=================
-
-.. toctree::
-   :maxdepth: 1
-""")])
-        app, out, err = self.make_sphinx()
-        self.build(app)
-        self.assertFileEqual("""\
-This file has been converted using a beta rst->texinfo converter. 
-Most of the info links are currently bogus, don't report bugs about them,
-this is currently worked on.
-@node Top
-@top Placeholder
-@node table-of-contents
-@chapter Table of Contents
-""",
-                             'index.texi')
-
-
-class TestSections(test_dg.TestSphinx):
-
-    def test_sections(self):
-        self.create_content('''\
-###########
-Chapter one
-###########
-
-Chapter introduction.
-
-***********
-section one
-***********
-
-The first section.
-
-
-subsection one
-==============
-
-The first subsection.
-
-subsection two
-==============
-
-The second subsection.
-
-subsubsection one
------------------
-
-Here is sus sub section one.
-
-blob one
-^^^^^^^^
-
-Far tooo deep to get a name
-
-thing one
-"""""""""
-
-No idea how to call that, but sphinx says it's a paragraph.
-''')
-        self.assertContent('''\
-@node chapter-one
-@chapter Chapter one
-Chapter introduction.
-
-@node section-one
-@section section one
-The first section.
-
-@node subsection-one
-@subsection subsection one
-The first subsection.
-
-@node subsection-two
-@subsection subsection two
-The second subsection.
-
-@node subsubsection-one
-@subsubsection subsubsection one
-Here is sus sub section one.
-
-@node blob-one
-@heading blob one
-Far tooo deep to get a name
-
-@node thing-one
-@heading thing one
-No idea how to call that, but sphinx says it's a paragraph.''')
-
-
-class TestReferences(test_dg.TestSphinx):
-
-    def test_external_reference(self):
-        self.create_content('''\
-The `example web site`_ is nice.
-
-.. _example web site: http://www.example.com/
-''')
-        self.assertContent('''\
-The @uref{http://www.example.com/,example web site} is nice.''')
-
-
-    def test_internal_reference(self):
-        self.create_content('''\
-The `example web site`_ contains more examples.
-
-Example web site
-----------------
-
-Here we have a lot of nice examples.
-
-''')
-        self.assertContent('''\
-The example web site (@pxref{example-web-site}) contains more examples.
-
-@node example-web-site
-@chapter Example web site
-Here we have a lot of nice examples.''')
-

=== modified file 'bzrlib/tests/features.py'
--- bzrlib/tests/features.py	2011-12-05 14:21:55 +0000
+++ bzrlib/tests/features.py	2012-03-12 14:56:53 +0000
@@ -352,7 +352,6 @@
 paramiko = ModuleAvailableFeature('paramiko')
 pycurl = ModuleAvailableFeature('pycurl')
 pywintypes = ModuleAvailableFeature('pywintypes')
-sphinx = ModuleAvailableFeature('sphinx')
 subunit = ModuleAvailableFeature('subunit')
 testtools = ModuleAvailableFeature('testtools')
 

=== modified file 'doc/developers/conf.py'
--- doc/developers/conf.py	2010-07-07 15:03:14 +0000
+++ doc/developers/conf.py	2012-03-12 14:56:53 +0000
@@ -33,16 +33,26 @@
 
 # Grouping the document tree into LaTeX files. List of tuples
 # (source start file, target name, title, author, documentclass [howto/manual]).
+bzr_documents = [
+  ('HACKING', 'bzr-en-developer-guide', u'Bazaar Developer Guide',
+   u'Bazaar Developers', 'manual'),
+  ('testing', 'bzr-en-testing-guide', u'Bazaar Testing Guide',
+   u'Bazaar Developers', 'manual'),
+  ('overview', 'bzr-en-architecture-overview', u'Bazaar Architecture Overview',
+   u'Bazaar Developers', 'howto'),
+  ('integration', 'bzr-en-integration-guide', u'Bazaar Integration Guide',
+   u'Bazaar Developers', 'howto'),
+]
+
 latex_documents = [
-  ('HACKING', 'bzr-en-developer-guide.tex', u'Bazaar Developer Guide',
-   u'Bazaar Developers', 'manual'),
-  ('testing', 'bzr-en-testing-guide.tex', u'Bazaar Testing Guide',
-   u'Bazaar Developers', 'manual'),
-  ('overview', 'bzr-en-architecture-overview.tex', u'Bazaar Architecture Overview',
-   u'Bazaar Developers', 'howto'),
-  ('integration', 'bzr-en-integration-guide.tex', u'Bazaar Integration Guide',
-   u'Bazaar Developers', 'howto'),
-]
+    (start, target+'.tex', title, author, doc_class)
+    for start, target, title, author, doc_class in bzr_documents
+    ]
+
+texinfo_documents = [
+    (start, target, title, author, doc_class)
+    for start, target, title, author, doc_class in bzr_documents
+    ]
 
 # List of documents that shouldn't be included in the build.
 # Note: Maybe some of them *ought* to be linked in somewhere?

=== modified file 'doc/en/Makefile'
--- doc/en/Makefile	2010-07-05 12:40:28 +0000
+++ doc/en/Makefile	2012-03-12 14:56:53 +0000
@@ -114,5 +114,8 @@
 	      "results in _build/doctest/output.txt."
 
 texinfo:
+	$(make_output_dirs)
 	$(SPHINXBUILD) -b texinfo $(ALLSPHINXOPTS) _build/texinfo
+        # Now build the info files using the Makefile provided by Sphinx
+	$(MAKE) -C _build/texinfo
 

=== modified file 'doc/en/conf.py'
--- doc/en/conf.py	2010-10-13 06:14:37 +0000
+++ doc/en/conf.py	2012-03-12 14:56:53 +0000
@@ -47,35 +47,45 @@
 # Output file base name for HTML help builder.
 htmlhelp_basename = 'bzr-%s' % (bzr_locale,)
 
-# Grouping the document tree into LaTeX files. List of tuples
+# Grouping the document tree into files. List of tuples
 # (source start file, target name, title, author, documentclass [howto/manual]).
-latex_documents = [
+bzr_documents = [
   # Manuals
-  ('user-guide/index', 'bzr-%s-user-guide.tex' % (bzr_locale,),
+  ('user-guide/index', 'bzr-%s-user-guide' % (bzr_locale,),
     bzr_title(u'Bazaar User Guide'), bzr_team, 'manual'),
-  ('user-reference/index', 'bzr-%s-user-reference.tex' % (bzr_locale,),
+  ('user-reference/index', 'bzr-%s-user-reference' % (bzr_locale,),
     bzr_title(u'Bazaar User Reference'), bzr_team, 'manual'),
-  ('release-notes/index', 'bzr-%s-release-notes.tex' % (bzr_locale,),
+  ('release-notes/index', 'bzr-%s-release-notes' % (bzr_locale,),
     bzr_title(u'Bazaar Release Notes'), bzr_team, 'manual'),
-  ('upgrade-guide/index', 'bzr-%s-upgrade-guide.tex' % (bzr_locale,),
+  ('upgrade-guide/index', 'bzr-%s-upgrade-guide' % (bzr_locale,),
     bzr_title(u'Bazaar Upgrade Guide'), bzr_team, 'manual'),
-  ('admin-guide/index', 'bzr-%s-admin-guide.tex' % (bzr_locale,),
+  ('admin-guide/index', 'bzr-%s-admin-guide' % (bzr_locale,),
     bzr_title(u"Bazaar System Administrator's Guide"), bzr_team, 'manual'),
   # Tutorials
-  ('mini-tutorial/index', 'bzr-%s-tutorial-mini.tex' % (bzr_locale,),
+  ('mini-tutorial/index', 'bzr-%s-tutorial-mini' % (bzr_locale,),
     bzr_title(u'Bazaar in five minutes'), bzr_team, 'howto'),
-  ('tutorials/tutorial', 'bzr-%s-tutorial.tex' % (bzr_locale,),
+  ('tutorials/tutorial', 'bzr-%s-tutorial' % (bzr_locale,),
     bzr_title(u'Bazaar Tutorial'), bzr_team, 'howto'),
   ('tutorials/using_bazaar_with_launchpad',
-    'bzr-%s-tutorial-with-launchpad.tex' % (bzr_locale,),
+    'bzr-%s-tutorial-with-launchpad' % (bzr_locale,),
     bzr_title(u'Using Bazaar With Launchpad'), bzr_team, 'howto'),
   ('tutorials/centralized_workflow',
-    'bzr-%s-tutorial-centralized.tex' % (bzr_locale,),
+    'bzr-%s-tutorial-centralized' % (bzr_locale,),
     bzr_title(u'Centralized Workflow Tutorial'), bzr_team, 'howto'),
-  ('whats-new/whats-new-in-2.1', 'bzr-%s-whats-new.tex' % (bzr_locale,),
+  ('whats-new/whats-new-in-2.1', 'bzr-%s-whats-new' % (bzr_locale,),
     bzr_title(u"What's New in Bazaar 2.1?"), bzr_team, 'howto'),
 ]
 
+latex_documents = [
+    (start, target+'.tex', title, author, doc_class)
+    for start, target, title, author, doc_class in bzr_documents
+    ]
+
+texinfo_documents = [
+    (start, target, title, author, doc_class)
+    for start, target, title, author, doc_class in bzr_documents
+    ]
+
 # List of documents that shouldn't be included in the build.
 unused_docs = [
     # Subtopics that get included

=== modified file 'doc/en/release-notes/bzr-2.5.txt'
--- doc/en/release-notes/bzr-2.5.txt	2012-02-28 05:02:15 +0000
+++ doc/en/release-notes/bzr-2.5.txt	2012-03-12 14:56:53 +0000
@@ -38,7 +38,11 @@
 Documentation
 *************
 
-.. Improved or updated documentation.
+* The alpha-quality texinfo sphinx builder has been deprecated. Sphinx >=
+  1.1.2 now provides a better one. Most of the documentation can now be
+  generated to the texinfo format with ``make texinfo-sphinx``. This will
+  generate both the ``.texi`` files and the ``.info`` ones.
+  (Vincent Ladeuil, #940164)
 
 API Changes
 ***********

=== modified file 'doc/en/whats-new/whats-new-in-2.5.txt'
--- doc/en/whats-new/whats-new-in-2.5.txt	2012-03-09 11:22:08 +0000
+++ doc/en/whats-new/whats-new-in-2.5.txt	2012-03-12 14:56:53 +0000
@@ -134,6 +134,13 @@
 ``bzr help po_merge`` for documentation on how to
 enable it and what it can do.
 
+Documentation in texinfo format
+*******************************
+
+Sphinx (>= 1.1.2) provides a better texinfo builder than the alpha-quality
+one provided with bzr sources. Most of the bzr documentation can be
+processed and produce valid ``.info`` files.
+
 Further information
 *******************
 

=== modified file 'doc/es/conf.py'
--- doc/es/conf.py	2010-07-07 15:03:14 +0000
+++ doc/es/conf.py	2012-03-12 14:56:53 +0000
@@ -52,29 +52,39 @@
 
 # Grouping the document tree into LaTeX files. List of tuples
 # (source start file, target name, title, author, documentclass [howto/manual]).
-latex_documents = [
+bzr_documents = [
   # Manuals
-  ('user-guide/index', 'bzr-%s-user-guide.tex' % (bzr_locale,),
+  ('user-guide/index', 'bzr-%s-user-guide' % (bzr_locale,),
     bzr_title(u'Bazaar User Guide'), bzr_team, 'manual'),
-  #('user-reference/bzr_man', 'bzr-%s-user-reference.tex' % (bzr_locale,),
+  #('user-reference/bzr_man', 'bzr-%s-user-reference' % (bzr_locale,),
   #  bzr_title(u'Bazaar User Reference'), bzr_team, 'manual'),
-  #('release-notes/NEWS', 'bzr-%s-release-notes.tex' % (bzr_locale,),
+  #('release-notes/NEWS', 'bzr-%s-release-notes' % (bzr_locale,),
   #  bzr_title(u'Bazaar Release Notes'), bzr_team, 'manual'),
-  #('upgrade-guide/index', 'bzr-%s-upgrade-guide.tex' % (bzr_locale,),
+  #('upgrade-guide/index', 'bzr-%s-upgrade-guide' % (bzr_locale,),
   #  bzr_title(u'Bazaar Upgrade Guide'), bzr_team, 'manual'),
   # Tutorials
-  ('mini-tutorial/index', 'bzr-%s-tutorial-mini.tex' % (bzr_locale,),
+  ('mini-tutorial/index', 'bzr-%s-tutorial-mini' % (bzr_locale,),
     bzr_title(u'Bazaar in five minutes'), bzr_team, 'howto'),
-  #('tutorials/tutorial', 'bzr-%s-tutorial.tex' % (bzr_locale,),
+  #('tutorials/tutorial', 'bzr-%s-tutorial' % (bzr_locale,),
   #  bzr_title(u'Bazaar Tutorial'), bzr_team, 'howto'),
   #('tutorials/using_bazaar_with_launchpad',
-  #  'bzr-%s-tutorial-with-launchpad.tex' % (bzr_locale,),
+  #  'bzr-%s-tutorial-with-launchpad' % (bzr_locale,),
   #  bzr_title(u'Using Bazaar With Launchpad'), bzr_team, 'howto'),
   #('tutorials/centralized_workflow',
-  #  'bzr-%s-tutorial-centralized.tex' % (bzr_locale,),
+  #  'bzr-%s-tutorial-centralized' % (bzr_locale,),
   #  bzr_title(u'Centralized Workflow Tutorial'), bzr_team, 'howto'),
 ]
 
+latex_documents = [
+    (start, target+'.tex', title, author, doc_class)
+    for start, target, title, author, doc_class in bzr_documents
+    ]
+
+texinfo_documents = [
+    (start, target, title, author, doc_class)
+    for start, target, title, author, doc_class in bzr_documents
+    ]
+
 # List of documents that shouldn't be included in the build.
 unused_docs = [
     # Subtopics that get included

=== modified file 'doc/ja/conf.py'
--- doc/ja/conf.py	2010-07-07 15:03:14 +0000
+++ doc/ja/conf.py	2012-03-12 14:56:53 +0000
@@ -52,29 +52,39 @@
 
 # Grouping the document tree into LaTeX files. List of tuples
 # (source start file, target name, title, author, documentclass [howto/manual]).
-latex_documents = [
+bzr_documents = [
   # Manuals
-  ('user-guide/index', 'bzr-%s-user-guide.tex' % (bzr_locale,),
+  ('user-guide/index', 'bzr-%s-user-guide' % (bzr_locale,),
     bzr_title(u'Bazaar User Guide'), bzr_team, 'manual'),
-  ('user-reference/bzr_man', 'bzr-%s-user-reference.tex' % (bzr_locale,),
+  ('user-reference/bzr_man', 'bzr-%s-user-reference' % (bzr_locale,),
     bzr_title(u'Bazaar User Reference'), bzr_team, 'manual'),
-  #('release-notes/NEWS', 'bzr-%s-release-notes.tex' % (bzr_locale,),
+  #('release-notes/NEWS', 'bzr-%s-release-notes' % (bzr_locale,),
   #  bzr_title(u'Bazaar Release Notes'), bzr_team, 'manual'),
-  #('upgrade-guide/index', 'bzr-%s-upgrade-guide.tex' % (bzr_locale,),
+  #('upgrade-guide/index', 'bzr-%s-upgrade-guide' % (bzr_locale,),
   #  bzr_title(u'Bazaar Upgrade Guide'), bzr_team, 'manual'),
   # Tutorials
-  ('mini-tutorial/index', 'bzr-%s-tutorial-mini.tex' % (bzr_locale,),
+  ('mini-tutorial/index', 'bzr-%s-tutorial-mini' % (bzr_locale,),
     bzr_title(u'Bazaar in five minutes'), bzr_team, 'howto'),
-  #('tutorials/tutorial', 'bzr-%s-tutorial.tex' % (bzr_locale,),
+  #('tutorials/tutorial', 'bzr-%s-tutorial' % (bzr_locale,),
   #  bzr_title(u'Bazaar Tutorial'), bzr_team, 'howto'),
   #('tutorials/using_bazaar_with_launchpad',
-  #  'bzr-%s-tutorial-with-launchpad.tex' % (bzr_locale,),
+  #  'bzr-%s-tutorial-with-launchpad' % (bzr_locale,),
   #  bzr_title(u'Using Bazaar With Launchpad'), bzr_team, 'howto'),
   #('tutorials/centralized_workflow',
-  #  'bzr-%s-tutorial-centralized.tex' % (bzr_locale,),
+  #  'bzr-%s-tutorial-centralized' % (bzr_locale,),
   #  bzr_title(u'Centralized Workflow Tutorial'), bzr_team, 'howto'),
 ]
 
+latex_documents = [
+    (start, target+'.tex', title, author, doc_class)
+    for start, target, title, author, doc_class in bzr_documents
+    ]
+
+texinfo_documents = [
+    (start, target, title, author, doc_class)
+    for start, target, title, author, doc_class in bzr_documents
+    ]
+
 # List of documents that shouldn't be included in the build.
 unused_docs = [
     # Subtopics that get included

=== modified file 'doc/ru/conf.py'
--- doc/ru/conf.py	2010-07-07 15:03:14 +0000
+++ doc/ru/conf.py	2012-03-12 14:56:53 +0000
@@ -52,29 +52,40 @@
 
 # Grouping the document tree into LaTeX files. List of tuples
 # (source start file, target name, title, author, documentclass [howto/manual]).
-latex_documents = [
+bzr_documents = [
   # Manuals
-  #('user-guide/index', 'bzr-%s-user-guide.tex' % (bzr_locale,),
+  #('user-guide/index', 'bzr-%s-user-guide' % (bzr_locale,),
   #  bzr_title(u'Bazaar User Guide'), bzr_team, 'manual'),
-  #('user-reference/bzr_man', 'bzr-%s-user-reference.tex' % (bzr_locale,),
+  #('user-reference/bzr_man', 'bzr-%s-user-reference' % (bzr_locale,),
   #  bzr_title(u'Bazaar User Reference'), bzr_team, 'manual'),
-  #('release-notes/NEWS', 'bzr-%s-release-notes.tex' % (bzr_locale,),
+  #('release-notes/NEWS', 'bzr-%s-release-notes' % (bzr_locale,),
   #  bzr_title(u'Bazaar Release Notes'), bzr_team, 'manual'),
-  #('upgrade-guide/index', 'bzr-%s-upgrade-guide.tex' % (bzr_locale,),
+  #('upgrade-guide/index', 'bzr-%s-upgrade-guide' % (bzr_locale,),
   #  bzr_title(u'Bazaar Upgrade Guide'), bzr_team, 'manual'),
   # Tutorials
-  ('mini-tutorial/index', 'bzr-%s-tutorial-mini.tex' % (bzr_locale,),
+  ('mini-tutorial/index', 'bzr-%s-tutorial-mini' % (bzr_locale,),
     bzr_title(u'Bazaar in five minutes'), bzr_team, 'howto'),
-  ('tutorials/tutorial', 'bzr-%s-tutorial.tex' % (bzr_locale,),
+  ('tutorials/tutorial', 'bzr-%s-tutorial' % (bzr_locale,),
     bzr_title(u'Bazaar Tutorial'), bzr_team, 'howto'),
   #('tutorials/using_bazaar_with_launchpad',
-  #  'bzr-%s-tutorial-with-launchpad.tex' % (bzr_locale,),
+  #  'bzr-%s-tutorial-with-launchpad' % (bzr_locale,),
   #  bzr_title(u'Using Bazaar With Launchpad'), bzr_team, 'howto'),
   #('tutorials/centralized_workflow',
-  #  'bzr-%s-tutorial-centralized.tex' % (bzr_locale,),
+  #  'bzr-%s-tutorial-centralized' % (bzr_locale,),
   #  bzr_title(u'Centralized Workflow Tutorial'), bzr_team, 'howto'),
 ]
 
+latex_documents = [
+    (start, target+'.tex', title, author, doc_class)
+    for start, target, title, author, doc_class in bzr_documents
+    ]
+
+texinfo_documents = [
+    (start, target, title, author, doc_class)
+    for start, target, title, author, doc_class in bzr_documents
+    ]
+
+
 # List of documents that shouldn't be included in the build.
 unused_docs = [
     # Subtopics that get included