Creates a package definition from the framework version being built,
and uses composer to install it into an installer project, as well
as the required modules for testing.
Pull requests are always on a branch, and this branch
typically is not present on the installer.
This changes means we need to be careful when merging into 3.1
and master, but that's a necessary evil.
Pull requests are always on a branch, and this branch
typically is not present on the installer.
This changes means we need to be careful when merging into 3.1
and master, but that's a necessary evil.
Pull requests are always on a branch, and this branch
typically is not present on the installer.
This changes means we need to be careful when merging into 3.1
and master, but that's a necessary evil.
RequestHandler#handleAction now exists. It takes the request, and
the action to call on itself. All calls from handleRequest to call an action
will go through this method
Controller#handleAction has had it's signature changed to
match new RequestHandler#handleAction
RequestHandler#findAction has been added, which extracts the
"match URL to rules to find action" portion of RequestHandler#handleRequest
into a separate, overrideable function
GridField#handleAction has beeen renamed to handleAlterAction and
CMSBatchActionHandler#handleAction has been renamed to handleBatchAction to
avoid name clash with new RequestHandler#handleAction
Reason for change: The exact behaviour of request handling depended heavily
on whether you inherited from RequestHandler or Controller, and whether the
rule extracted it's action directly (like "foo/$ID" => 'foo') or dynamically
(like "$Action/$ID" => "handleAction"). This cleans up behaviour so
all calls follow the same path through handleRequest and handleAction, and
the additional behaviour that Controller adds is clear.
allowed_actions is now only allowed to reference public methods defined
on the same Controller as the allowed_actions static, and
the wildcard "*" has been deprecated
This will prevent empty passwords to set the encryption to 'none',
which in turn will store any subsequent password changes in cleartext.
Reproduceable e.g. with ConfirmedPasswordField and setCanBeEmpty(true).
Controller (and subclasses) failed to enforce $allowed_action restrictions
on parent classes if a child class didn't have it explicitly defined.
Controllers which are extended with $allowed_actions (through an Extension)
now deny access to methods defined on the controller, unless this class also has them in its own
$allowed_actions definition.
Shortcodes have traditionally had a problem that they are inside <p> tags,
but generate block level elements. This breaks HTML compliance.
This makes the shortcode parser now mutate the DOM based on the "class" attribute on
the shortcode to insert the generated block level element at the right place in the DOM
- for "left" and "right" elements it puts them just before the block level
element they are inside
- for "leftAlone" and "center" elements it splits the DOM around the shortcode.
The trade off is that shortcodes are no longer "text level" features. They need
knowledge of the HTML they are in to perform this transformation, so they can
only be used in (valid) HTML
Because I removed completely the static setting of SSL_VERIFYPEER I've
added the ability to declare default curl options on the class. This
means that users that really want to one line turn off SSL_VERIFYPEER
can do so without needing to pass a custom option in every request()
call.
Before now, the RestfulService_Response object was never sent the
response headers. For APIs that rely on the response headers to send
back information (signatures, pagination info, etc).
This change makes the curl response have the full HTTP response
(including Headers). We then extract the body and the header information
and assign them to relevant vars and then construct the response as
before (with the addition of the headers array).
This change required two new functions:
extractResponse: This extracts the HTTP Headers and the payload from the
curl response and assigns it to the relevany vars that are passed by
reference
parseRawHeaders: This was designed to mimic http_parse_headers (a
non-standard php class). It converts the headers into an associative
array.
If creating an object using Injector::create() and constructor arguments
are passed through, in some cases where the object being created had a yml
configuration set for it, the passed in constructor arguments weren't being
passed through to the instantiation of the object.
Lazy loading no longer loads fields from the versions table when querying. This could lead to incorrect data being displayed if the data on the object and the version it pointed to did not match.
API methods to allow setting of the context of the query that generated the DataObject on that object (used by the lazy loading mechanism to correctly query the Stage, Live, or Versions tables)
See https://github.com/silverstripe/sapphire/pull/1178 for context.
Lazy loading no longer loads fields from the versions table when querying. This could lead to incorrect data being displayed if the data on the object and the version it pointed to did not match.
API methods to allow setting of the context of the query that generated the DataObject on that object (used by the lazy loading mechanism to correctly query the Stage, Live, or Versions tables)
See https://github.com/silverstripe/sapphire/pull/1178 for context.
urlRewriter will expect a callable as a second parameter,
but will work with the current api and simply raise a deprecation error.
HTTP::absoluteURLs now correctly rewrites urls into absolute urls. Resolves introduced in c56a80d6ce
HTTP::absoluteURLs now handles additional cases where urls were not translated.
Test cases for HTTP::absoluteURLs added for both css and attribute links.
Cleaned up replacement expression and improved documentation.
This reverts commit 356a367eb5.
We can't use headers_sent() to determine an accurate
content length, since PHP defaults to buffering a couple of bytes
even without ob_start() (see "output_buffering" setting).
This makes the patch harmful, since it breaks any responses relying
on more structure data, like removing closing brackets from JSON.
Which in turn breaks the CMS in horrible ways (see #8010).
See #7574 for context.
This is the companion setting to canUpload, letting you control whether existing files from the asset store can be referenced. It's particularly useful when using UploadField on the front-end.
Otherwise conditional logic will only succeed
when run through "sake dev/tests", not when
run through phpunit directly (which is the recommended way now)
This causes a 'Fatal error: Call to a member function hasMethod() on a non-object'.
This can happen when displaying a field in a gridfield on a belongs_to relationship.
+ has a special meaning in the URLs so overall it's a good idea to
strip them out. Otherwise they would need to appear in their ugly url
encoded form "%2B".
Refer: http://open.silverstripe.org/ticket/7929
BUG Issue with deleted records not being queried properly.
API DataQuery::expressionForField no longer requires a second parameter. Rather the query object is inferred from the DataQuery itself. This should improve consistency of use of this function.
Introduced new FormField->castedCopy() method
which tries to replicate the existing form field instance
as closely as possible.
Primarily, the fix was targeted at consistently passing
through FormField->description to all of its variations.
In 3.0 there was some confusion about whether DataLists and ArrayLists
were mutable or not. If DataLists were immutable, they'd return the result, and your code
would look like
$list = $list->filter(....);
If DataLists were mutable, they'd operate on themselves, returning nothing, and your code
would look like
$list->filter(....);
This makes all DataLists and ArrayList immutable for all _searching_ operations.
Operations on DataList that modify the underlying SQL data store remain mutating.
- These functions no longer mutate the existing object, and if you do not capture the value
returned by them will have no effect:
ArrayList#reverse
ArrayList#sort
ArrayList#filter
ArrayList#exclude
DataList#dataQuery (use DataList#alterDataQuery to modify dataQuery in a safe manner)
DataList#where
DataList#limit
DataList#sort
DataList#addFilter
DataList#applyFilterContext
DataList#innerJoin
DataList#leftJoin
DataList#find
DataList#byIDs
DataList#reverse
- DataList#setDataQueryParam has been added as syntactic sugar around the most common
cause of accessing the dataQuery directly - setting query parameters
- RelationList#setForeignID has been removed. Always use RelationList#forForeignID
when querying, and overload RelationList#foreignIDList when subclassing.
- Relatedly,the protected variable RelationList->foreignID has been removed, as the ID is
now stored on a query parameter. Use RelationList#getForeignID to read it.
Session is not initialized by the time we need to use
the setting in DB::connect(). Cookie values get initialized
automatically for each request.
Tightened name format validation to ensure it can only
be used for temporary databases, rather than switching
the browser session to a different production database.
Encrypting token for secure cookie usage.
Added dev/generatesecuretoken to generate this token.
Not storing in YML config directly because of web access issues.
Fixed issue where convertServiceProperty is called when creating objects
with user-supplied constructor arguments, so that it's only called when
creating objects using injector configuration. This reduces the overhead
of unnecessary calls to convertServiceProperty.
Updated test cases to validate behaviour
Enables more generic use of the fixture facilities
without dependency on the YAML format, for example
when creating fixtures from Behat step definitions.
Note: The YamlFixture class needs to be created via
Injector::inst()->create('YamlFixture') now,
direct instantiation is no longer supported.
This bug will surface when using the ORM and adding an join to DataList
where a DataObject inherits another DataObject.
If you for example want to restrict the number of pages that only have a
related Staff object:
$list = DataList::create('Page')
->InnerJoin('Staff', '"Staff"."ID" = "Page"."StaffID");
This will create a SQL query where the INNER JOIN is before the
LEFT JOIN of Page and SiteTree in the resulting SQL string. In MySQL
and PostgreSQL this will create an invalid query.
This patch solves the problem by sorting the joins.
Resolves an issue where if not using the themes directory (i.e just a single app folder) you cannot override module templates.
Changes the SS_TemplateManifest constructor with a new $project argument.
Avoid PHPUnit throwing "test didn't run any assertions"
notices in PHP. If nothing else, it keeps test output
looking less broken by default, making it more likely
that actual errors do get noticed.
Quoted table / column names to make test cases work in postgres
BUG Fixed issue with SQLQuery::lastRow crashing on empty set. Added test cases for lastRow and firstRow.
Quoted table / column names to make test cases work in postgres
Merge branch '3.0-sqlquery-lastrow-fix' of github.com:tractorcow/sapphire into 3.0-sqlquery-lastrow-fix
Refactor the code to make it clear the distinction is made between a
plaintext token and a hashed version. Rename fields so it is more
obvious what is being written and what sent out to the user.
This reuses the salt and algorithm from the Member, which are kept
constant throughout the Member lifetime in a normal scenario. If they do
change, users will need to re-request so the hashes can be regenerated.
The existence of .ss-tabset triggers JS which applies $.tabs(),
and in turn interprets the first available link as the tab navigation.
jQuery UI subsequently tries to ajax-load this link, which is not
desired. Instead, $.tabs() should *only* be applied to a container
DOM element with .cms-tabset applied.
It was setting a NULL empty string when constructing the field,
which shouldn't call setEmptyField() in the first place.
This logical error somehow just surfaced when the HTML output
wasn't run through tidy.
See https://github.com/silverstripe/sapphire/pull/886
When a user renames a file to "__test.txt" (two underscores or more),
then FileNameFilter will only remove the very first underscore from the
filename. This is not sufficient, as any number of underscores in the
filename will be problematic when Filesystem::sync() is called, it will
remove that File record thinking it's an internal file. This fixes it
so any number of underscores are stripped out at the start of the filename.
API Added Convert::nl2os function to normalise end of line characters across systems with tests
BUG Fixed i18n unit tests in non-unix systems constantly failing
BUG Fixed problems with HTMLCleaner tests failing in non-unix systems
The specific situation where this is useful is where populateDefaults on
DataObjects needs to query the database. This will break the dev/build
when it tries to create the object via singleton - the query will not be
able to be executed if the table is not there or its schema has changed.
For an example of such use case see Translatable::populateDefaults.
If formatOutput is set to TRUE, then the regexes in getContent()
will not match the newlines, and the output will include html, body
and meta tags. Introduce a few new tests to ensure the output is
correct, and fix the regex.
Using late static binding makes it possible to override SS_Log to create
logs which are separate to the main Silverstripe log but still use the
built in functionality.
Add test for SS_Log subclassing.
At this stage, the test just checks line-length and line-endings, but previous commits have ensured that framework actually passes those tests. We can add more tests as we actually correct the code to pass those tests, and grow the test suite, as we had for unit tests.
The entire framework repo (with the exception of system-generated files) has been amended to respect the 120c line-length limit. This is in preparation for the enforcement of this rule with PHP_CodeSniffer.
These APIs are primarily intended to let developers write custom 404 handlers. They can define an onBeforeHTTPError404() method on an Extension that gets added to Controller or RequestHandler.
The SS_HTTPResponse_Exception object has also been tidied up to override the status info of any SS_HTTPResponse object that might get passed. This is mainly to make it easier for callers (such as ContentController and ModelAsController) to use RequestHandler::httpError() more consistently.
Adds three extra methods to Data/SQLQuery query that allow for starting
a disjunctive subgroup, a conjunctive subgroup and for ending a subgroup.
Database::sqlWhereToString() now builds up the WHERE clause one by one
instead of with a straight implode. Uses a stack to know which conenctive
to use.
This is a fix for ticket #7670. Some hosting situations don't
allow write access to the system temp path. tiny_mce_gzip.php is currently
using sys_get_temp_dir() by default, and not using a local silverstripe-cache
folder that may exist in the SilverStripe project.
This change moves the getTempFolder() function into a common file, and
includes that in core/Core.php, as well as thirdparty/tinymce/tiny_mce_gzip.php
so both locations share the same code to work out the temp path.
The issue was raised in #7628, where an anchor tag was being changed from
<a name="anchor"></a> to <a name="anchor"/> by SS_HTMLValue, when
HtmlEditorField::saveInto() parses the HTML fragments.
This is because SS_HTMLValue uses DOMDocument::saveXML(), which is fine
for saving an XML document, but not suitable for HTML. This fix changes
that to use DOMDocument::saveHTML() instead.
Note that we can't use the parameter to saveHTML() for selecting a single
node only, as that's only supported in PHP 5.3.6+, SilverStripe 3.0 supports
PHP 5.3.2 as a minimum. The workaround for this shortcoming is to replace
unncessary output by DOMDocument with a regular expression.
ADDED: Test cases correctly checking for changes (and no changes) to the data model for both fields and indexes.
FIXED: References to indexes throughough the code that probably should have quoted field names. This prevents a lot of 'spam' during dev build. This includes an updated FulltextSearchable test case.
CMSProfileController currently checks canView() which ensures that a logged in CMS
Member can access the profile controller, but when saving the record on Member_ProfileForm
there is no check for canEdit(), so extended permissions don't get respected.
This adds a check for canEdit() in Member_ProfileForm, and adds some functional tests
to check permissions.
The table name in the join was being escaped, though table
names aren't escaped anywhere else. This breaks
namespaced classes, which rely on unescaped backslashes.
The Travis config will now run tests on the following instances
* 5.3 + SQLite
* 5.3 + MySQL
* 5.3 + PostgreSQL
* 5.4 + MySQL
In other words, with the exception of Windows tech (MSSQL + Win server) this is a wide-coverage build config.
ADDED: HTTP_Request::params() to retrieve all (shifted) params used in the request
FIXED: Issue where route-table level arguments would not be accessible without using non-deprecated API.
ADDED: Test case to test the above items
UPDATED: Extended Director::test to allow for the retrieval of the request object
UPDATED: Deprecated notice on Director::urlParam and Director::urlParams
REMOVED: Unused variable
FIXED: Coding convention conformity
The values in $values aren't returned in any order, so this test can randomly fail. This
changes the check from expected = $values to $values \cap expected = $value. PHP's array_intersect
maintains the keys of the first array, so order is preserved. The intersect also guarentees that the
only accepted values are the expected ones.
This allows arguments to be passed along in an $allowed_actions deceleration of
the form 'action' => '->method' in the same way that arguments can be passed to
extension constructors when adding them using $extensions or
Object::add_extension.
I.e. 'action' => '->checkerMethod(false, 7, 2, "yesterday") would call the
checkerMethod method with the boolean false the numbers 7 and 2 and the string
"yesterday" as its arguments.
The Requirements class currently treats only absolute URLs as URLs, and
tries to interpret anything else as a filesystem path. This prevents
using scheme-relative URLs for requirements.
Example:
<% require javascript(//ajax.googleapis.com/ajax/libs/jquery/1.7.2/jquery.min.js) %>
This forces the unfortunate choice of not using a CDN for common
scripts, always using an https absolute URL, or accepting that some
browsers will throw a security warning when viewing the site in https.
This change allows scheme-relative URLs & updates RequirementsTest.
This changes the behaviour of output to browser to use the standard SilverStripe rendering process rather than an echo statement to enable easier testability.
