Prior to this commit, StandardServletEnvironment's servlet context
PropertySource remained stubbed out until it the ServletContext became
available and could be replaced during the refresh() of its enclosing
WebApplicationContext. This behavior is acceptable in most cases.
However, if the user has declared an ApplicationContextInitializer that
attempts to access servlet context-params via the Environment API, this
result in a kind of 'false negative', i.e. the context-param key and
value are actually present in the ServletContext, but the PropertySource
representing servlet context params is still a stub at this point,
meaning that it returns an empty result in all cases.
With this change, WebApplicationContextUtils#initServletPropertySources
is invoked eagerly by the ContextLoader if any ACI classes have been
declared. This swaps out the servlet context property source stub for
the real thing just in time for ACIs to use it if necessary.
Extra guard logic has been added to #initServletPropertySources to
ensure idempotency -- once the stub has been replaced, the method
never attempts the replacement again, e.g. during the normal context
refresh() when this method will be called again.
Issue: SPR-8991
- Perform early check whether any ACI classes have been declared and
exit immediately if not, avoiding any other processing
- Polish method names in ContextLoaderTests
Issue: SPR-8991
A list of "known" session attributes (listed in @SessionAttributes)
was gradually built as attributes get added to the model. In a
failover scenario that knowledge is lost causing session attributes
to be potentially re-initialized via @ModelAttribute methods.
With this change @SessionAttributes listed by name are immediately
added to he list of "known" session attributes thus this knowledge
is not lost after a failover. Attributes listed by type however
still must be discovered as they get added to the model.
- Eliminate trailing whitespace
- Update long method signatures to follow framework whitespace
conventions
Based on the following search,
$ git grep -A3 '^.public .* .*([^\{;]*$' */src/main
the strong convention throughout the framework when dealing with
methods having long signatures (i.e. many parameters) is to break
immediately after the opening paren, indent two tabs deeper and break
lines around 90 characters as necessary. Such signatures should also
be followed by a newline after the opening curly brace to break
things up visually.
The files edited in this commit had a particularly different style of
intenting arguments to align with each other vertically, but the
alignment only worked if one's tabstop is set at four spaces.
When viewed at a different tabstop value, the effect is is jarring,
both in that it is misaligned and significantly different from most
of the framework. The convention described above reads well at any
tabstop value.
Prior to this change, roughly 5% (~300 out of 6000+) of files under the
source tree had CRLF line endings as opposed to the majority which have
LF endings.
This change normalizes these files to LF for consistency going forward.
Command used:
$ git ls-files | xargs file | grep CRLF | cut -d":" -f1 | xargs dos2unix
Issue: SPR-5608
Prior to this fix, ContextLoader(Listener)'s would overwrite any
value set directly against a WebApplicationContext's #setConfigLocation
method. This is a likely scenario when using Spring 3.1's new
WebApplicationInitializer support.
Now a check is performed to ensure that the ContextLoader init-param
value is non-null before doing the overwriting.
Added tests to ensure that all expected precedence, overwriting and
defaulting of context config locations works as expected.
Issue: SPR-8510
This method allows a view to access the combined context path and
servlet mapping path for prefixing URLs without having to specify
the literal part of a servlet mapping such as "/main/*")
explicitly everywhere. For example:
${requestContext.pathToServlet}/css/main.css
The AbstractHttpMessageConverter was using the requested Content-Type
rather than the actual response Content-Type to determine the length
of the content. This can lead to a problem when a controller returns
a ResponseEntity with a Content-Type header that ignores (overrides)
the requested Content-Type. The fix ensures that actual response
Content-Type is the one used both to write to the response and to
determine the length of the content.
The UriComponentsBuilder instance passed into the method is initialized
with current request information including host, scheme, port, context
path, and the servlet mapping's literal part.
Also added shortcut methods to buildAndExpand in UriComponentsBuilder.
The initial solution kept these three in full sync at all times:
contentType field, characterEncoding field, 'Content-Type' header.
That is correct behavior, however it breaks existing tests that rely
on contentType and characterEncoding being equal to exactly what
they were set to.
For example, consider:
response.setContentType("text/plain");
response.setCharacterEncoding("UTF-8");
Ideally both contentType and the 'Content-Type' header would now be
"text/plain;charset=UTF-8". However, existing tests would expect
that contentType is equal to "text/plain".
To avoid breaking existing tests, contentType and characterEncoding
will continue to be equal to exactly what they were set to while
the 'Content-Type' header will always include both the content
type and the charset.
The only exception to this rule is when a 'Content-Type' header
is set explicitly, the contentType and characterEncoding fields will
be updated accordingly, possibly overriding the existing values.
The Content-Type header and the contentType field in HttpServletRequest/Response
are now always in sync. When a header is added the contentType field is updated
as well and vice versa.
Similarly when the Content-Type header or the contentType field includes a charset
field, the character encoding is updated and vice versa.
Juergen Hoeller
Arjen Poutsma
Chris Beams
Rossen Stoyanchev