This is a pre-production deployment of Warehouse. Changes made here affect the production instance of PyPI (pypi.python.org).
Help us improve Python packaging - Donate today!
Project Description

This package implements the ‘blocks’ rendering model, by providing several transform stages that hook into plone.transformchain.

The rendering stages are:

plone.app.blocks.parsexml (order 8000)
Turns the response in a repoze.xmliter XMLSerializer object. This is then used by the subsequent stages. If the input is not HTML, the transformation is aborted.
plone.app.blocks.mergepanels (order 8100)
Looks up the site layout and executes the panel merge algorithm. Sets a request variable (‘plone.app.blocks.merged’) to indicate that it has done its job.
plone.app.blocks.tiles (order 8500)
Resolve tiles and place them directly into the merged layout. This is the fallback for views that do not opt into ITilePageRendered.
plone.app.blocks.esirender (order 9900)
Only executed if the request key plone.app.blocks.esi is set and its value is true, as would be the case if any ESI-rendered tiles are included and ESI rendering is enabled globally. This step will serialise the response down to a string and perform some substitution to make ESI rendering work.

Site layouts

The package also registers the sitelayout plone.resource resource type, allowing site layouts to be created easily as static HTML files served from resource directories. The URL to a site layout is typically something like:

/++sitelayout++my.layout/site.html

See plone.resource for more information about how to register resource directories. For site layouts, the type of the resource directory is sitelayout.

It is possible to provide a manifest file that gives a title, description and alternative default file for a site layout HTML file in a resource directory. To create such a manifest, put a manifest.cfg file in the layout directory with the following structure:

[sitelayout]
title = My layout title
description = Some description
file = some-html-file.html
  • All keys are optional.
  • The file defaults to site.html.
  • Single manifest may contain multiple [sitelayout] sections.

A vocabulary factory called plone.availableSiteLayouts is registered to allow lookup of all registered site layouts. The terms in this vocabulary use the URL as a value, the resource directory name as a token, and the title from the manifest (falling back on a sanitised version of the resource directory name) as the title.

The current default site layout can be identified by the plone.registry key plone.defaultSiteLayout, which is set to None by default. To always use the current site default, use:

<html data-layout="./@@default-site-layout">

The @@default-site-layout view will render the current default site layout.

Content layouts

The package also registers the contentlayout plone.resource resource type, allowing shared content area layouts to be created easily as static HTML files served from resource directories. The URL to a content layout is typically something like:

/++contentlayout++my.layout/content.html

See plone.resource for more information about how to register resource directories. For site layouts, the type of the resource directory is contentlayout.

It is possible to provide a manifest file that gives a title, description and alternative default file for a site layout HTML file in a resource directory. To create such a manifest, put a manifest.cfg file in the layout directory with the following structure:

[contentlayout]
title = My layout title
description = Some description
file = some-html-file.html
screenshot = mylayout.png
for = Document,Folder
  • All keys are optional.
  • The file defaults to content.html.
  • Single manifest may contain multiple [contentlayout] sections.

A vocabulary factory called plone.availableContentLayouts is registered to allow lookup of all registered content layouts. The terms in this vocabulary use the URL as a value, the resource directory name as a token, and the title from the manifest (falling back on a sanitised version of the resource directory name) as the title.

The default content layout can be identified by the plone.registry key plone.app.blocks.default_layout, and the default content layout for some specific content type with key plone.app.blocks.default_layout.my_type. The default content layout is supported by the built-in layout_view browser view for content with ILayoutAware behavior.

ILayoutAware behavior

It is possible for the default site layout to be overridden per section, by having parent objects provide or be adaptable to plone.app.blocks.layoutbehavior.ILayoutAware. As the module name implies, this interface can be used as a plone.behavior behavior, but it can also be implemented directly or used as a standard adapter.

The ILayoutAware interface defines three properties:

content
which contains the body of the page to be rendered.
contentLayout
which contains the path to the selected static content layout, which is used instead of content when set.
pageSiteLayout
which contains the path to the site layout to be used for the given page. It can be None if the default is to be used.
sectionSiteLayout
which contains the path to the site layout to be used for pages underneath the given page (but not for the page itself). Again, it can be None if the default is to be used.

To make use of the page site layout, use the following:

<html data-layout="./@@default-site-layout">

See rendering.rst for detailed examples of how the processing is applied, and esi.rst for details about how Edge Side Includes can be supported.

Blocks rendering in detail

This doctest illustrates the blocks rendering process. At a high level, it consists of the following steps:

  1. Obtain the content page, an HTML document.

  2. Look for a site layout link in the content page.

    This takes the form of an attribute on the html tag like <html data-layout="..." />.

    Usually, the site layout URL will refer to a resource in a resource directory of type sitelayout, e.g. /++sitelayout++foo/site.html, although the layout can be any URL. An absolute path like this will be adjusted so that it is always relative to the Plone site root.

  3. Resolve and obtain the site layout.

    This is another HTML document.

  4. Extract panels from the site layout.

    A panel is an element (usually a <div />) in the layout page with a data-panel attribute, for example: <div data-panel="panel1" />. The attribute specifies an id which may be used in the content page.

  5. Merge panels.

    This is the process which applies the layout to the unstyled page. All panels in the layout page that have a matching element in the content page are replaced by the content page element. The rest of the content page is discarded.

  6. Resolve and obtain tiles.

    A tile is a placeholder element in the page which will be replaced by the contents of a document referenced by a URL.

    A tile is identified by a placeholder element with a data-tile attribute containing the tile URL.

    Note that at this point, panel merging has taken place, so if a panel in the content page contains tiles, they will be carried over into the merge page. Also note that it is possible to have tiles outside of panels - the two concepts are not directly related.

    The plone.tiles package provides a framework for writing tiles, although in reality a tile can be any HTML page.

  7. Place tiles into the page.

    The tile should resolve to a full HTML document. Any content found in the <head /> of the tile content will be merged into the <head /> of the rendered content. The contents of the <body /> of the tile content are put into the rendered document at the tile placeholder.

Rendering step-by-step

Let us now illustrate the rendering process. We’ll need a few variables defined first:

>>> from plone.testing.z2 import Browser
>>> import transaction

>>> app = layer['app']
>>> portal = layer['portal']

>>> browser = Browser(app)
>>> browser.handleErrors = False

Creating a site layout

The most common approach for managing site layouts is to use a resource registered using a plone.resource directory of type sitelayout, and then use the @@default-site-layout view to reference the content. We will illustrate this below, but it is important to realise that plone.app.blocks works by post-processing responses rendered by Zope. The content and layout pages could just as easily be created by views of content objects, or even resources external to Zope/Plone.

First, we will create a resource representing the site layout and its panels. This includes some resources and other elements in the <head />, <link /> tags which identify tile placeholders and panels, as well as content inside and outside panels. The tiles in this case are managed by plone.tiles, and are both of the same type.

>>> layoutHTML = """\
... <!DOCTYPE html PUBLIC "-//W3C//DTD XHTML 1.0 Transitional//EN" "http://www.w3.org/TR/xhtml1/DTD/xhtml1-transitional.dtd">
... <html>
...     <head>
...         <title>Layout title</title>
...         <link rel="stylesheet" href="/layout/style.css" />
...         <script type="text/javascript">alert('layout');</script>
...
...         <style type="text/css">
...         div {
...             margin: 5px;
...             border: dotted black 1px;
...             padding: 5px;
...         }
...         </style>
...
...         <link rel="stylesheet" data-tile="./@@test.tile_nobody/tile_css" />
...     </head>
...     <body>
...         <h1>Welcome!</h1>
...         <div data-panel="panel1">Layout panel 1</div>
...         <div data-panel="panel2">
...             Layout panel 2
...             <div id="layout-tile1" data-tile="./@@test.tile1/tile1">Layout tile 1 placeholder</div>
...         </div>
...         <div data-panel="panel3">
...             Layout panel 3
...             <div id="layout-tile2" data-tile="./@@test.tile1/tile2">Layout tile 2 placeholder</div>
...         </div>
...     </body>
... </html>
... """

We can create an in-ZODB resource directory of type sitelayout that contains this layout. Another way would be to register a resource directory in a package using ZCML, or use a global resource directory. See plone.resource for more details.

>>> from Products.CMFCore.utils import getToolByName
>>> from Products.BTreeFolder2.BTreeFolder2 import BTreeFolder2
>>> from StringIO import StringIO
>>> from OFS.Image import File

>>> resources = getToolByName(portal, 'portal_resources')
>>> resources._setOb('sitelayout', BTreeFolder2('sitelayout'))
>>> resources['sitelayout']._setOb('mylayout', BTreeFolder2('mylayout'))
>>> resources['sitelayout']['mylayout']._setOb('site.html', File('site.html', 'site.html', StringIO(layoutHTML)))

>>> transaction.commit()

This resource can now be accessed using the path /++sitelayout++mylayout/site.html. Let’s render it on its own to verify that.

>>> browser.open(portal.absolute_url() + '/++sitelayout++mylayout/site.html')
>>> print browser.contents
<!DOCTYPE html PUBLIC "-//W3C//DTD XHTML 1.0 Transitional//EN" "http://www.w3.org/TR/xhtml1/DTD/xhtml1-transitional.dtd">
<html>
    <head>
        <title>Layout title</title>
        <link rel="stylesheet" href="/layout/style.css" />
        <script type="text/javascript">alert('layout');</script>
<BLANKLINE>
        <style type="text/css">
        div {
            margin: 5px;
            border: dotted black 1px;
            padding: 5px;
        }
        </style>
<BLANKLINE>
        <link rel="stylesheet" data-tile="./@@test.tile_nobody/tile_css" />
    </head>
    <body>
        <h1>Welcome!</h1>
        <div data-panel="panel1">Layout panel 1</div>
        <div data-panel="panel2">
            Layout panel 2
            <div id="layout-tile1" data-tile="./@@test.tile1/tile1">Layout tile 1 placeholder</div>
        </div>
        <div data-panel="panel3">
            Layout panel 3
            <div id="layout-tile2" data-tile="./@@test.tile1/tile2">Layout tile 2 placeholder</div>
        </div>
    </body>
</html>

We can now set this as the site-wide default layout by setting the registry key plone.defaultSiteLayout. There are two indirection views, @@default-site-layout and @@page-site-layout, that respect this registry setting. By using one of these views to reference the layout of a given page, we can manage the default site layout centrally.

>>> from zope.component import getUtility
>>> from plone.registry.interfaces import IRegistry
>>> registry = getUtility(IRegistry)
>>> registry['plone.defaultSiteLayout'] = '/++sitelayout++mylayout/site.html'
>>> transaction.commit()

Creating a page layout and tiles

Next, we will define the markup of a content page that uses this layout via the @@default-site-layout indirection view:

>>> pageHTML = """\
... <!DOCTYPE html PUBLIC "-//W3C//DTD XHTML 1.0 Transitional//EN" "http://www.w3.org/TR/xhtml1/DTD/xhtml1-transitional.dtd">
... <html data-layout="./@@default-site-layout">
...     <body>
...         <h1>Welcome!</h1>
...         <div data-panel="panel1">
...             Page panel 1
...             <div id="page-tile2" data-tile="./@@test.tile1/tile2?magicNumber:int=2">Page tile 2 placeholder</div>
...         </div>
...         <div data-panel="panel2">
...             Page panel 2
...             <div id="page-tile3" data-tile="./@@test.tile1/tile3">Page tile 3 placeholder</div>
...         </div>
...         <div data-panel="panel4">
...             Page panel 4 (ignored)
...             <div id="page-tile4" data-tile="./@@test.tile1/tile4">Page tile 4 placeholder</div>
...         </div>
...     </body>
... </html>
... """

We then register a view that simply return this HTML, and a tile type which we can use to test tile rendering.

We do this in code for the purposes of the test, and we have to apply security because we will shortly render those pages using the test publisher. In real life, these could be registered using the standard <browser:page /> and <plone:tile /> directives.

>>> from zope.publisher.browser import BrowserView
>>> from zope.interface import Interface, implements
>>> from zope import schema
>>> from plone.tiles import Tile
>>> from plone.app.blocks.interfaces import IBlocksTransformEnabled

>>> class Page(BrowserView):
...     implements(IBlocksTransformEnabled)
...     __name__ = 'test-page'
...     def __call__(self):
...         return pageHTML

>>> class ITestTile(Interface):
...     magicNumber = schema.Int(title=u"Magic number", required=False)

>>> class TestTile(Tile):
...     __name__ = 'test.tile1' # normally set by ZCML handler
...
...     def __call__(self):
...         # fake a page template to keep things simple in the test
...         return """\
... <html>
...     <head>
...         <meta name="tile-name" content="%(name)s" />
...     </head>
...     <body>
...         <p>
...             This is a demo tile with id %(name)s
...         </p>
...         <p>
...             Magic number: %(number)d; Form: %(form)s; Query string: %(queryString)s; URL: %(url)s
...         </p>
...     </body>
... </html>""" % dict(name=self.id, number=self.data['magicNumber'] or -1,
...                   form=sorted(self.request.form.items()), queryString=self.request['QUERY_STRING'], url=self.request.getURL())

Let’s add another tile, this time only a head part. This could for example be a tile that only needs to insert some CSS.

>>> class TestTileNoBody(Tile):
...     __name__ = 'test.tile_nobody'
...
...     def __call__(self):
...         return """\
... <html>
...     <head>
...         <link rel="stylesheet" type="text/css" href="tiled.css" />
...     </head>
... </html>"""

We register these views and tiles in the same way the ZCML handlers for <browser:page /> and <plone:tile /> would:

>>> from plone.tiles.type import TileType
>>> from Products.Five.security import protectClass
>>> from App.class_init import InitializeClass
>>> from zope.component import provideAdapter, provideUtility
>>> from zope.interface import Interface

>>> testTileType = TileType(
...     name=u'test.tile1',
...     title=u"Test tile",
...     description=u"A tile used for testing",
...     add_permission="cmf.ManagePortal",
...     view_permission="zope2.View",
...     schema=ITestTile)

>>> testTileTypeNoBody = TileType(
...     name=u'test.tile_nobody',
...     title=u"Test tile using only a header",
...     description=u"Another tile used for testing",
...     add_permission="cmf.ManagePortal",
...     view_permission="zope2.View")

>>> protectClass(Page, 'zope2.View')
>>> protectClass(TestTile, 'zope2.View')

>>> InitializeClass(Page)
>>> InitializeClass(TestTile)

>>> provideAdapter(Page, (Interface, Interface,), Interface, u'test-page')
>>> provideAdapter(TestTile, (Interface, Interface,), Interface, u'test.tile1',)
>>> provideAdapter(TestTileNoBody, (Interface, Interface,), Interface, u'test.tile_nobody',)
>>> provideUtility(testTileType, name=u'test.tile1')
>>> provideUtility(testTileTypeNoBody, name=u'test.tile_nobody')

Rendering the page

We can now render the page. Provided plone.app.blocks is installed and working, it should perform its magic. We make sure that Zope is in “development mode” to get pretty-printed output.

>>> browser.open(portal.absolute_url() + '/@@test-page')
>>> print browser.contents.replace('<head><meta', '<head>\n\t<meta')
<!DOCTYPE html PUBLIC "-//W3C//DTD XHTML 1.0 Transitional//EN" "http://www.w3.org/TR/xhtml1/DTD/xhtml1-transitional.dtd">
<html xmlns="http://www.w3.org/1999/xhtml">
  <head>
    <meta http-equiv="Content-Type" content="text/html; charset=ASCII" />
    <title>Layout title</title>
    <link rel="stylesheet" href="/layout/style.css" />
    <script type="text/javascript">alert('layout');</script>
    <style type="text/css">
        div {
            margin: 5px;
            border: dotted black 1px;
            padding: 5px;
        }
        </style>
    <link rel="stylesheet" type="text/css" href="tiled.css" />
    <meta name="tile-name" content="tile2" />
    <meta name="tile-name" content="tile3" />
    <meta name="tile-name" content="tile2" />
  </head>
  <body>
        <h1>Welcome!</h1>
        <div data-panel="panel1">
            Page panel 1
        <p>
            This is a demo tile with id tile2
        </p>
        <p>
            Magic number: 2; Form: [('magicNumber', 2)]; Query string: magicNumber:int=2; URL: http://nohost/plone/@@test.tile1/tile2
        </p>
        </div>
        <div data-panel="panel2">
            Page panel 2
        <p>
            This is a demo tile with id tile3
        </p>
        <p>
            Magic number: -1; Form: []; Query string: ; URL: http://nohost/plone/@@test.tile1/tile3
        </p>
        </div>
        <div data-panel="panel3">
            Layout panel 3
        <p>
            This is a demo tile with id tile2
        </p>
        <p>
            Magic number: -1; Form: []; Query string: ; URL: http://nohost/plone/@@test.tile1/tile2
        </p>
        </div>
    </body>
</html>
<BLANKLINE>

Notice how:

  • Panels from the page have been merged into the layout, replacing the corresponding panels there.
  • The <head /> sections of the two documents have been merged
  • The rest of the layout page is intact
  • The rest of the content page is discarded
  • The tiles have been rendered, replacing the relevant placeholders
  • The <head /> section from the rendered tiles has been merged into the <head /> of the output page.

Using VHM

Make sure to have a clean browser:

>>> browser = Browser(app)
>>> browser.handleErrors = False

Using Virtual Host Monster we rewrite the url to consider all content being under /:

>>> vhm_url = 'http://nohost/VirtualHostBase/http/nohost:80/plone/VirtualHostRoot/'
>>> browser.open(vhm_url + '/@@test-page')

Tiles should return an url according to this:

>>> 'Magic number: -1; Form: []; Query string: ; URL: http://nohost/@@test.tile1/tile2' in browser.contents
True

Now we deal with _vh_* arguments. We expect our site to be under a subdir with id subplone:

>>> vhm_url = 'http://nohost/VirtualHostBase/http/nohost:80/plone/VirtualHostRoot/_vh_subplone'
>>> browser.open(vhm_url + '/@@test-page')

Tiles should return an url according to this:

>>> 'Magic number: -1; Form: []; Query string: ; URL: http://nohost/subplone/@@test.tile1/tile2' in browser.contents
True

ESI rendering

Blocks supports rendering of tiles for Edge Side Includes (ESI). A tile will be rendered to ESI provided that:

  • The tile itself is marked with the IESIRendered marker interface. See plone.tiles for more details.
  • The plone.app.blocks.interfaces.IBlocksSettings.esi record in the registry is set to True. It is False by default. To switch this through-the-web, you can visit the configuration registry control panel in Plone.

Note that if a tile is rendered using ESI, it’s <head /> contents are ignored, instead of being merged into the final page. That is, only the @@esi-body view form plone.tiles is used by default.

An ESI link looks like this:

<esi:include src="http://example.com/plone/@@some.tile/tile-1/@@esi-body?param1=value1" />

A fronting server such as Varnish will be able to load this on demand and compose the page from fragments that may be cached individually.

Test setup

Let’s first register a two very simple tiles. One uses ESI, one does not.

>>> from plone.tiles.esi import ESITile
>>> from plone.tiles import Tile
>>> from plone.tiles.type import TileType

>>> class NonESITile(Tile):
...     __name__ = 'test.tile2' # normally set by ZCML handler
...
...     def __call__(self):
...         return """\
... <html>
...     <head>
...         <meta name="tile-name" content="%(name)s" />
...     </head>
...     <body>
...         <p>
...             Non-ESI tile with query string %(queryString)s
...         </p>
...     </body>
... </html>""" % dict(name=self.id, queryString=self.request['QUERY_STRING'])

>>> testTile2Type = TileType(
...     name=u'test.tile2',
...     title=u"Test tile 2",
...     description=u"A tile used for testing",
...     add_permission="cmf.ManagePortal",
...     view_permission="zope2.View")

>>> class SimpleESITile(ESITile):
...     __name__ = 'test.tile3' # normally set by ZCML handler
...
...     def render(self):
...         return """\
... <html>
...     <head>
...         <meta name="tile-name" content="%(name)s" />
...     </head>
...     <body>
...         <p>
...             ESI tile with query string %(queryString)s
...         </p>
...     </body>
... </html>""" % dict(name=self.id, queryString=self.request['QUERY_STRING'])

>>> testTile3Type = TileType(
...     name=u'test.tile3',
...     title=u"Test tile 3",
...     description=u"A tile used for testing",
...     add_permission="cmf.ManagePortal",
...     view_permission="zope2.View")

Register these in the same way that the ZCML handlers would, more or less.

>>> from Products.Five.security import protectClass
>>> protectClass(NonESITile, 'zope2.View')
>>> protectClass(SimpleESITile, 'zope2.View')

>>> from App.class_init import InitializeClass
>>> InitializeClass(NonESITile)
>>> InitializeClass(SimpleESITile)

>>> from zope.component import provideAdapter, provideUtility
>>> from zope.interface import Interface
>>> provideAdapter(NonESITile, (Interface, Interface,), Interface, u'test.tile2',)
>>> provideUtility(testTile2Type, name=u'test.tile2')
>>> provideAdapter(SimpleESITile, (Interface, Interface,), Interface, u'test.tile3',)
>>> provideUtility(testTile3Type, name=u'test.tile3')

We will also register a simple layout and a simple page using these tiles.

>>> layoutHTML = u"""\
... <!DOCTYPE html PUBLIC "-//W3C//DTD XHTML 1.0 Transitional//EN" "http://www.w3.org/TR/xhtml1/DTD/xhtml1-transitional.dtd">
... <html>
...     <head>
...         <title>Layout title</title>
...     </head>
...     <body>
...         <h1>Welcome!</h1>
...         <div data-panel="panel1">Content goes here</div>
...         <div id="layout-non-esi-tile" data-tile="./@@test.tile2/tile1">Layout tile 1 placeholder</div>
...         <div id="layout-esi-tile" data-tile="./@@test.tile3/tile2">Layout tile 2 placeholder</div>
...     </body>
... </html>
... """

To keep things simple, we’ll skip the resource directory and layout indirection view, instead just referencing a view containing the layout directly.

>>> from zope.publisher.browser import BrowserView
>>> class Layout(BrowserView):
...     __name__ = 'test-layout'
...     def __call__(self):
...         return layoutHTML

>>> protectClass(Layout, 'zope2.View')
>>> InitializeClass(Layout)
>>> provideAdapter(Layout, (Interface, Interface,), Interface, u'test-layout',)

>>> pageHTML = u"""\
... <!DOCTYPE html PUBLIC "-//W3C//DTD XHTML 1.0 Transitional//EN" "http://www.w3.org/TR/xhtml1/DTD/xhtml1-transitional.dtd">
... <html data-layout="./@@test-layout">
...     <body>
...         <div data-panel="panel1">
...             <div id="page-non-esi-tile" data-tile="./@@test.tile2/tile3?foo=bar">Page tile 3 placeholder</div>
...             <div id="page-esi-tile" data-tile="./@@test.tile3/tile4?foo=bar">Page tile 4 placeholder</div>
...         </div>
...     </body>
... </html>
... """

>>> from zope.interface import implements
>>> from plone.app.blocks.interfaces import IBlocksTransformEnabled
>>> class Page(BrowserView):
...     implements(IBlocksTransformEnabled)
...     __name__ = 'test-page'
...     def __call__(self):
...         return pageHTML

>>> protectClass(Page, 'zope2.View')
>>> InitializeClass(Page)
>>> provideAdapter(Page, (Interface, Interface,), Interface, u'test-page',)

ESI disabled

We first render the page without enabling ESI. The ESI-capable tiles should be rendered as normal.

>>> from plone.testing.z2 import Browser
>>> app = layer['app']
>>> browser = Browser(app)
>>> browser.handleErrors = False

>>> portal = layer['portal']
>>> browser.open(portal.absolute_url() + '/@@test-page')

Some cleanup is needed to cover lxml platform discrepancies…

>>> print browser.contents.replace('<head><meta', '<head>\n\t<meta')
<!DOCTYPE html PUBLIC "-//W3C//DTD XHTML 1.0 Transitional//EN" "http://www.w3.org/TR/xhtml1/DTD/xhtml1-transitional.dtd">
<html xmlns="http://www.w3.org/1999/xhtml">
    <head>
    <meta http-equiv="Content-Type" content="text/html; charset=ASCII" />
    <title>Layout title</title>
    <meta name="tile-name" content="tile3" />
    <meta name="tile-name" content="tile4" />
    <meta name="tile-name" content="tile1" />
    <meta name="tile-name" content="tile2" />
    </head>
    <body>
        <h1>Welcome!</h1>
        <div data-panel="panel1">
        <p>
            Non-ESI tile with query string foo=bar
        </p>
        <p>
            ESI tile with query string foo=bar
        </p>
        </div>
        <p>
            Non-ESI tile with query string
        </p>
        <p>
            ESI tile with query string
        </p>
    </body>
</html>
<BLANKLINE>

ESI enabled

We can now enable ESI. This could be done using GenericSetup (with the registry.xml import step), or through the configuration registry control panel. In code, it is done like so:

>>> from zope.component import getUtility
>>> from plone.registry.interfaces import IRegistry
>>> from plone.app.blocks.interfaces import IBlocksSettings
>>> registry = getUtility(IRegistry)
>>> registry.forInterface(IBlocksSettings).esi = True
>>> import transaction
>>> transaction.commit()

We can now perform the same rendering again. This time, the ESI-capable tiles should be rendered as ESI links. See plone.tiles for more details.

>>> browser.open(portal.absolute_url() + '/@@test-page')
>>> print browser.contents.replace('<head><meta', '<head>\n\t<meta')
<!DOCTYPE html PUBLIC "-//W3C//DTD XHTML 1.0 Transitional//EN" "http://www.w3.org/TR/xhtml1/DTD/xhtml1-transitional.dtd">
<html xmlns:esi="http://www.edge-delivery.org/esi/1.0" xmlns="http://www.w3.org/1999/xhtml">
    <head>
    <meta http-equiv="Content-Type" content="text/html; charset=ASCII" />
    <title>Layout title</title>
    <meta name="tile-name" content="tile3" />
    <meta name="tile-name" content="tile1" />
    </head>
    <body>
        <h1>Welcome!</h1>
        <div data-panel="panel1">
        <p>
            Non-ESI tile with query string foo=bar
        </p>
        <esi:include src="http://nohost/plone/@@test.tile3/tile4/@@esi-body?foo=bar" />
        </div>
        <p>
            Non-ESI tile with query string
        </p>
        <esi:include src="http://nohost/plone/@@test.tile3/tile2/@@esi-body?" />
    </body>
</html>
<BLANKLINE>

When ESI rendering takes place, the following URLs will be called:

>>> browser.open("http://nohost/plone/@@test.tile3/tile4/@@esi-body?foo=bar")
>>> print browser.contents
<p>
    ESI tile with query string foo=bar
</p>

>>> browser.open("http://nohost/plone/@@test.tile3/tile2/@@esi-body?")
>>> print browser.contents
<p>
    ESI tile with query string
</p>

Changelog

4.0.6 (2017-02-09)

Fixes:

  • Fix issue where layout related fields could have been acquired (only sectionSiteLayout can be allowed to be acquired) [datakurre]

4.0.5 (2017-02-08)

Fixes:

  • Fix issue where page site layout could have been accidentally acquired (page site layout should never be acquired) [datakurre]
  • Fix transforms to comply with plone.transformchain.interfaces.ITransform [datakurre]

4.0.4 (2017-01-30)

Fixes:

  • Fix issue where ESIRender has been broken since plone.protect’s ProtectTransform was introduced, because of protect transform breaking ESI-tags; Change ESIRender transform order from 8900 to 9900 [datakurre]

4.0.3 (2017-01-15)

Fixes:

  • Fix issue where default layouts paths were not found if they were stored unicode (TextLine) instead of str (ASCIILine or BytesLine) [datakurre]
  • Fix issue where tiles merge failed for addresses with space, because subrequest was called with quoted (‘%20’) paths [datakurre]

4.0.2 (2017-01-03)

Fixes:

  • Fix issue where error in diazo transform for a single tile aborted tile merge as whole [datakurre]

4.0.1 (2016-12-28)

Fixes:

  • Fix issue where tile data storage decoded HTML primary fields using ASCII instead of utf-8 causing broken broken latin characters in attribute values [datakurre]

4.0.0 (2016-12-13)

Incompatibilities:

  • Remove grid transform, because it did not serve its purpose as as well expected and required HTML-syntax not editable by humans; Instead using grid framework agnostic CSS class names and building CSS grid against those class names is recommended [agitator]
  • Remove IOmittedField marker from layout behavior fields not meant to be displayed on legacy Deco UIs [jensens]
  • Rename ILayoutAware.content to ILayoutAware.customContentLayout [datakurre]
  • Move functions getDefaultAjaxLayout, getDefaultSiteLayout, getLayout and getLayoutAwareSiteLayout to .layoutbehavior in order to avoid circular imports (all deprecated now, see section New). [jensens]
  • Move views from .layoutbehavior to new module .layoutviews in order to avoid circular imports. Deprecated deferred imports are in place. [jensens]

New:

  • Add ILayoutAware.content as layout independent “layout like” tile configuration and data storage for all serializable tile configurations [datakurre]
  • Add @@layout_preview view for previewing currently drafted layout aware content [datakurre]
  • ILayoutAware is now also responsible to lookup the behaviors. [jensens]
  • Get layouts always by adapting with ILayoutAware. This introduces a generic adapter and a behavior adapter. Deprecated the formerly used functions getLayout getDefaultSiteLayout just calls ILayoutAware().site_layout and is deprected. getLayout just calls ILayoutAware().content_layout and is deprecated. [jensens]
  • Behavior shortname plone.layoutaware added. [jensens]

Fixes:

  • Handle missing content layouts so they do not cause an error [vangheem]
  • A tile raising an 401 Unauthorized on traversal, results in a status rewriting to a 302 which results in 200 login form. The whole login form page then is rendered as the tile contents. This patch catches the 401 by providing a custom exception handler. The 401 is catched and ignored. This is not pefect yet and need some work, but it at least does not break design and intended behavior of tiles. [jensens]

Refactoring:

  • Housekeeping: ZCA decorators, sorted imports, line-lengths and related. [jensens]
  • Reformat documentation. [gforcada]
  • Update travis configuration. [gforcada]

3.1.0 (2016-03-28)

New:

  • Don’t make a tile exception break other tiles (closes #27). [rodfersou, datakurre]
  • Provide new getLayoutsFromDirectory utility to get layouts from any plone.resource directory, not just the base resource directory [vangheem]
  • Index layout data; When collective.dexteritytextindexer is present, its Dynamic SearchableText indexer behavior must be enabled for content type [vangheem, datakurre]
  • Cleanup tile data on save/edit [vangheem]

3.0.1 (2015-09-23)

  • Remove the default ‘Custom layout’ display menu registration for ‘layout_view’, because it was not possible to customize it with more exact registration [datakurre]
  • Fix the default view to report template name as ‘template-layout’ [datakurre]

3.0.0 (2015-09-16)

  • Change layout behavior default view name from view to layout_view [datakurre]
  • Add to be able to set default grid system in registry settings [vangheem]
  • Add support for provide more than one layout with a layout directory and manifest (replaces removed layout variants) [vangheem]
  • Add contentlayout resource type with plone.availableContentLayouts vocabulary and ++contentlayout++ traverser [vangheem]
  • Add contentLayout field to layoutbehavior to select the rendered layout from centrally managed content layouts [vangheem]
  • Add content type specific registry configuration with key plone.app.blocks.default_layout.portal_type for used default content layout when custom layout is not defined [vangheem]
  • Add to check plone.app.blocks.default_layout registry key for a default content layout path when content type specific default content layout path is not set [datakurre]
  • Fixed layout behavior to apply Plone outputfilters for rendered content [datakurre]
  • Add default grid system registry setting [vangheem]
  • Restore support for Plone 4.2.x [datakurre]
  • Remove layout variants introduced in 2.0.0, in favor of ability to provide more than one layout with a layout directory and manifest by using multiple [...layout] directive in the same manifest [vangheem]

2.1.2 (2015-06-10)

  • Fix issue where grid transform did replaced class names instead of appending to them [datakurre]

2.1.1 (2015-06-10)

  • Fix BS3 grid transform to only introduce offset when the tile position is greater than the current position in the current row [datakurre]
  • Fix issue where tiles with empty response or syntax error broke tiles transform (add to log syntax errors instead) [datakurre]

2.1.0 (2015-05-25)

  • Add support for indexing layout field into SearchableText index when collective.dexteritytextindexer is installed and its Dynamic SearchableText indexer behavior is enabled for the indexed content type with Layout support behavior [datakurre]

2.0.0 (2015-04-21)

  • Fix package dependencies; remove dependency on unittest2. [hvelarde]

  • Change blocks transforms to be opt-in for only published objects e.g. views or requests with IBlocksTransformEnabled (marker) interface [fixes #11] [datakurre]

  • Change tags with data-tiles-attrs to be completely replaced (by replace_with_children instad of replace_content) to restore original design and support for site layout tiles in HTML document head tag [datakurre]

  • Change default site layout to be optional by adding an implicit main_template-based site layout when the default site layout is not set [datakurre]

  • Change to retry resolveResources with 301 or 302 response when redirect location is for the same site [datakurre]

  • Add support for AJAX site layout for requests with ajax_load parameter either by getting a layout from a reqistry key plone.defaultAjaxLayout or by using an implicit main_template-based AJAX layout [simahawk, datakurre]

  • Add extensible CSS grid transform with built-in transforms for Deco and Bootstrap 3 grid systems [bloodbare, ACatila]

    <utility
        provides=".gridsystem.IGridSystem"
        component=".gridsystem.DecoGridSystem"
        name="deco"
        />
    
    <html data-gridsystem="deco">
      ...
      <div data-grid='{"type": "row"}'>
        <div data-grid='{"type": "cell",
                         "info": {"xs": "false",
                                  "sm": "False",
                                  "lg": "True",
                         "pos": {"x":1,
                                 "width": 12}}}'>
         </div>
      </div>
    </html>
    
    <div class="row">
       <div class="cell position-1 width-12">
       </div>
    </div>
    
  • Add default view for ILayoutAware content and register a localizable display menu item called Custom layout for it when plone.app.contentmenu is present [datakurre]

  • Add Layout-fieldset for ILayoutAware behavior [datakurre]

  • Add support to use the whole tile as its body when both head and body tags are missing (add support for using Dexterithy display widgets as tiles) [datakurre]

  • Add support for layout variants (for supporting multiple layouts in a single resource folder) [datakurre]

    [sitelayout]
    ...
    
    [sitelayout:variants]
    document_layout = document.html
    
  • Add experimental support for tile-specific Diazo-rules with data-attribute data-rules="/++sitelayout++name/rules.xml". [datakurre]

  • Fix issue with tile without body-tag breaking the tile composition (fixes issues with some p.a.standardtiles returning only <html/> in some conditions) [datakurre]

  • Fix issue where <![CDATA[…]]> block was quoted (and therefore broken) by lxml serializer [datakurre]

  • Fix issue where XML parser dropped head for layout with CRLF-endings [datakurre]

  • Fix plone.app.blocks re-install to not reset existing plone.defaultSiteLayout and plone.defaultAjaxLayout settings (by setting the values in a custom setuphandler) [datakurre]

  • Fix and update tests, PEP8 [gyst, datakurre, gforcada]

  • Fix to set the merging request flag before testing the merge results to allow staticly placed tiles in content templates to be rendered properly. [cewing]

  • Solve issue with VHM and tile rendering. Fixes https://dev.plone.org/ticket/13581 [ericof]

  • Add z3c.autoinclude support [cdw9, calvinhp]

1.1 (2012-12-17)

  • make sure to use correct url of tile [vangheem]
  • handle not found errors while rendering tiles so layout isn’t borked [vangheem]

1.0 (2012-06-23)

  • initial release. [garbas]
Release History

Release History

4.0.6

This version

History Node

TODO: Figure out how to actually get changelog content.

Changelog content for this version goes here.

Donec et mollis dolor. Praesent et diam eget libero egestas mattis sit amet vitae augue. Nam tincidunt congue enim, ut porta lorem lacinia consectetur. Donec ut libero sed arcu vehicula ultricies a non tortor. Lorem ipsum dolor sit amet, consectetur adipiscing elit.

Show More

4.0.5

History Node

TODO: Figure out how to actually get changelog content.

Changelog content for this version goes here.

Donec et mollis dolor. Praesent et diam eget libero egestas mattis sit amet vitae augue. Nam tincidunt congue enim, ut porta lorem lacinia consectetur. Donec ut libero sed arcu vehicula ultricies a non tortor. Lorem ipsum dolor sit amet, consectetur adipiscing elit.

Show More

4.0.4

History Node

TODO: Figure out how to actually get changelog content.

Changelog content for this version goes here.

Donec et mollis dolor. Praesent et diam eget libero egestas mattis sit amet vitae augue. Nam tincidunt congue enim, ut porta lorem lacinia consectetur. Donec ut libero sed arcu vehicula ultricies a non tortor. Lorem ipsum dolor sit amet, consectetur adipiscing elit.

Show More

4.0.3

History Node

TODO: Figure out how to actually get changelog content.

Changelog content for this version goes here.

Donec et mollis dolor. Praesent et diam eget libero egestas mattis sit amet vitae augue. Nam tincidunt congue enim, ut porta lorem lacinia consectetur. Donec ut libero sed arcu vehicula ultricies a non tortor. Lorem ipsum dolor sit amet, consectetur adipiscing elit.

Show More

4.0.2

History Node

TODO: Figure out how to actually get changelog content.

Changelog content for this version goes here.

Donec et mollis dolor. Praesent et diam eget libero egestas mattis sit amet vitae augue. Nam tincidunt congue enim, ut porta lorem lacinia consectetur. Donec ut libero sed arcu vehicula ultricies a non tortor. Lorem ipsum dolor sit amet, consectetur adipiscing elit.

Show More

4.0.1

History Node

TODO: Figure out how to actually get changelog content.

Changelog content for this version goes here.

Donec et mollis dolor. Praesent et diam eget libero egestas mattis sit amet vitae augue. Nam tincidunt congue enim, ut porta lorem lacinia consectetur. Donec ut libero sed arcu vehicula ultricies a non tortor. Lorem ipsum dolor sit amet, consectetur adipiscing elit.

Show More

4.0.0

History Node

TODO: Figure out how to actually get changelog content.

Changelog content for this version goes here.

Donec et mollis dolor. Praesent et diam eget libero egestas mattis sit amet vitae augue. Nam tincidunt congue enim, ut porta lorem lacinia consectetur. Donec ut libero sed arcu vehicula ultricies a non tortor. Lorem ipsum dolor sit amet, consectetur adipiscing elit.

Show More

4.0.0rc3

History Node

TODO: Figure out how to actually get changelog content.

Changelog content for this version goes here.

Donec et mollis dolor. Praesent et diam eget libero egestas mattis sit amet vitae augue. Nam tincidunt congue enim, ut porta lorem lacinia consectetur. Donec ut libero sed arcu vehicula ultricies a non tortor. Lorem ipsum dolor sit amet, consectetur adipiscing elit.

Show More

4.0.0rc2

History Node

TODO: Figure out how to actually get changelog content.

Changelog content for this version goes here.

Donec et mollis dolor. Praesent et diam eget libero egestas mattis sit amet vitae augue. Nam tincidunt congue enim, ut porta lorem lacinia consectetur. Donec ut libero sed arcu vehicula ultricies a non tortor. Lorem ipsum dolor sit amet, consectetur adipiscing elit.

Show More

4.0.0rc1

History Node

TODO: Figure out how to actually get changelog content.

Changelog content for this version goes here.

Donec et mollis dolor. Praesent et diam eget libero egestas mattis sit amet vitae augue. Nam tincidunt congue enim, ut porta lorem lacinia consectetur. Donec ut libero sed arcu vehicula ultricies a non tortor. Lorem ipsum dolor sit amet, consectetur adipiscing elit.

Show More

3.1.0

History Node

TODO: Figure out how to actually get changelog content.

Changelog content for this version goes here.

Donec et mollis dolor. Praesent et diam eget libero egestas mattis sit amet vitae augue. Nam tincidunt congue enim, ut porta lorem lacinia consectetur. Donec ut libero sed arcu vehicula ultricies a non tortor. Lorem ipsum dolor sit amet, consectetur adipiscing elit.

Show More

3.0.1

History Node

TODO: Figure out how to actually get changelog content.

Changelog content for this version goes here.

Donec et mollis dolor. Praesent et diam eget libero egestas mattis sit amet vitae augue. Nam tincidunt congue enim, ut porta lorem lacinia consectetur. Donec ut libero sed arcu vehicula ultricies a non tortor. Lorem ipsum dolor sit amet, consectetur adipiscing elit.

Show More

3.0.0

History Node

TODO: Figure out how to actually get changelog content.

Changelog content for this version goes here.

Donec et mollis dolor. Praesent et diam eget libero egestas mattis sit amet vitae augue. Nam tincidunt congue enim, ut porta lorem lacinia consectetur. Donec ut libero sed arcu vehicula ultricies a non tortor. Lorem ipsum dolor sit amet, consectetur adipiscing elit.

Show More

2.2.1

History Node

TODO: Figure out how to actually get changelog content.

Changelog content for this version goes here.

Donec et mollis dolor. Praesent et diam eget libero egestas mattis sit amet vitae augue. Nam tincidunt congue enim, ut porta lorem lacinia consectetur. Donec ut libero sed arcu vehicula ultricies a non tortor. Lorem ipsum dolor sit amet, consectetur adipiscing elit.

Show More

2.2.0

History Node

TODO: Figure out how to actually get changelog content.

Changelog content for this version goes here.

Donec et mollis dolor. Praesent et diam eget libero egestas mattis sit amet vitae augue. Nam tincidunt congue enim, ut porta lorem lacinia consectetur. Donec ut libero sed arcu vehicula ultricies a non tortor. Lorem ipsum dolor sit amet, consectetur adipiscing elit.

Show More

2.1.2

History Node

TODO: Figure out how to actually get changelog content.

Changelog content for this version goes here.

Donec et mollis dolor. Praesent et diam eget libero egestas mattis sit amet vitae augue. Nam tincidunt congue enim, ut porta lorem lacinia consectetur. Donec ut libero sed arcu vehicula ultricies a non tortor. Lorem ipsum dolor sit amet, consectetur adipiscing elit.

Show More

2.1.1

History Node

TODO: Figure out how to actually get changelog content.

Changelog content for this version goes here.

Donec et mollis dolor. Praesent et diam eget libero egestas mattis sit amet vitae augue. Nam tincidunt congue enim, ut porta lorem lacinia consectetur. Donec ut libero sed arcu vehicula ultricies a non tortor. Lorem ipsum dolor sit amet, consectetur adipiscing elit.

Show More

2.1.0

History Node

TODO: Figure out how to actually get changelog content.

Changelog content for this version goes here.

Donec et mollis dolor. Praesent et diam eget libero egestas mattis sit amet vitae augue. Nam tincidunt congue enim, ut porta lorem lacinia consectetur. Donec ut libero sed arcu vehicula ultricies a non tortor. Lorem ipsum dolor sit amet, consectetur adipiscing elit.

Show More

2.0.0

History Node

TODO: Figure out how to actually get changelog content.

Changelog content for this version goes here.

Donec et mollis dolor. Praesent et diam eget libero egestas mattis sit amet vitae augue. Nam tincidunt congue enim, ut porta lorem lacinia consectetur. Donec ut libero sed arcu vehicula ultricies a non tortor. Lorem ipsum dolor sit amet, consectetur adipiscing elit.

Show More

1.1.1

History Node

TODO: Figure out how to actually get changelog content.

Changelog content for this version goes here.

Donec et mollis dolor. Praesent et diam eget libero egestas mattis sit amet vitae augue. Nam tincidunt congue enim, ut porta lorem lacinia consectetur. Donec ut libero sed arcu vehicula ultricies a non tortor. Lorem ipsum dolor sit amet, consectetur adipiscing elit.

Show More

1.1

History Node

TODO: Figure out how to actually get changelog content.

Changelog content for this version goes here.

Donec et mollis dolor. Praesent et diam eget libero egestas mattis sit amet vitae augue. Nam tincidunt congue enim, ut porta lorem lacinia consectetur. Donec ut libero sed arcu vehicula ultricies a non tortor. Lorem ipsum dolor sit amet, consectetur adipiscing elit.

Show More

1.0

History Node

TODO: Figure out how to actually get changelog content.

Changelog content for this version goes here.

Donec et mollis dolor. Praesent et diam eget libero egestas mattis sit amet vitae augue. Nam tincidunt congue enim, ut porta lorem lacinia consectetur. Donec ut libero sed arcu vehicula ultricies a non tortor. Lorem ipsum dolor sit amet, consectetur adipiscing elit.

Show More

Download Files

Download Files

Download the file for your platform. If you're not sure which to choose, learn more about installing packages.

File Name & Checksum SHA256 Checksum Help Version File Type Upload Date
plone.app.blocks-4.0.6.tar.gz (66.2 kB) Copy SHA256 Checksum SHA256 Source Feb 9, 2017

Supported By

WebFaction WebFaction Technical Writing Elastic Elastic Search Pingdom Pingdom Monitoring Dyn Dyn DNS Sentry Sentry Error Logging CloudAMQP CloudAMQP RabbitMQ Heroku Heroku PaaS Kabu Creative Kabu Creative UX & Design Fastly Fastly CDN DigiCert DigiCert EV Certificate Rackspace Rackspace Cloud Servers DreamHost DreamHost Log Hosting