Class ParserRegistry
Keeps track of response parsers for each content type. Each parser
should should be a closure that accepts an HttpResponse
instance,
and returns whatever handler is appropriate for reading the response
data for that content-type. For example, a plain-text response should
probably be parsed with a Reader
, while an XML response
might be parsed by an XmlSlurper, which would then be passed to the
response closure.
Note that all methods in this class assume HttpResponse.getEntity()
return a non-null value. It is the job of the HTTPBuilder instance to ensure
a NullPointerException is not thrown by passing a response that contains no
entity.
You can see the list of content-type parsers that are built-in to the
ParserRegistry class in buildDefaultParserMap()
.
- See Also:
-
Field Summary
FieldsModifier and TypeFieldDescriptionprotected static org.apache.xml.resolver.tools.CatalogResolver
This CatalogResolver is static to avoid the overhead of re-parsing the catalog definition file every time.static final String
The default charset to use when no charset is given in the Content-Type header of a response.protected final groovy.lang.Closure
The default parser used for unregistered content-types.private static String
private groovy.lang.Closure
protected static final org.apache.commons.logging.Log
-
Constructor Summary
Constructors -
Method Summary
Modifier and TypeMethodDescriptionstatic void
addCatalog
(URL catalogLocation) Add a new XML catalog definiton to the static XML resolver catalog.Returns a map of default parsers.groovy.lang.Closure
Retrieve a parser for the given response content-type string.static org.apache.xml.resolver.tools.CatalogResolver
Access the default catalog used by all HTTPBuilder instances.static String
getCharset
(org.apache.http.HttpResponse resp) Helper method to get the charset from the response.static String
getContentType
(org.apache.http.HttpResponse resp) Helper method to get the content-type string from the response (no charset).groovy.lang.Closure
Get the default parser used for unregistered content-types.iterator()
Iterate over the entire parser mapparseForm
(org.apache.http.HttpResponse resp) Default parser used to decode a URL-encoded response.groovy.util.slurpersupport.GPathResult
parseHTML
(org.apache.http.HttpResponse resp) Parse an HTML document by passing it through the NekoHTML parser.parseJSON
(org.apache.http.HttpResponse resp) Default parser used to decode a JSON response.parseStream
(org.apache.http.HttpResponse resp) Default parser used for binary data.parseText
(org.apache.http.HttpResponse resp) Default parser used to handle plain text data.groovy.util.slurpersupport.GPathResult
parseXML
(org.apache.http.HttpResponse resp) Default parser used to decode an XML response.groovy.lang.Closure
propertyMissing
(Object key) Alias forgetAt(Object)
to allow property-style access.void
propertyMissing
(Object key, groovy.lang.Closure value) Alias forputAt(Object, Closure)
to allow property-style access.void
Register a new parser for the given content-type.static void
setDefaultCharset
(String charset) Set the charset to use for parsing character streams when no charset is given in the Content-Type header.void
setDefaultParser
(groovy.lang.Closure defaultParser) Set the default parser used for unregistered content-types.
-
Field Details
-
DEFAULT_PARSER
protected final groovy.lang.Closure DEFAULT_PARSERThe default parser used for unregistered content-types. This is a copy ofparseStream(HttpResponse)
, which is like a no-op that just returns the unaltered response stream. -
DEFAULT_CHARSET
The default charset to use when no charset is given in the Content-Type header of a response. This can be modifid viasetDefaultCharset(String)
.- See Also:
-
defaultParser
private groovy.lang.Closure defaultParser -
registeredParsers
-
defaultCharset
-
log
protected static final org.apache.commons.logging.Log log -
catalogResolver
protected static org.apache.xml.resolver.tools.CatalogResolver catalogResolverThis CatalogResolver is static to avoid the overhead of re-parsing the catalog definition file every time. Unfortunately, there's no way to share a single Catalog instance between resolvers. TheCatalog
class is technically not thread-safe, but as long as you do not parse catalog files while using the resolver, it should be fine.
-
-
Constructor Details
-
ParserRegistry
public ParserRegistry()
-
-
Method Details
-
setDefaultCharset
Set the charset to use for parsing character streams when no charset is given in the Content-Type header.- Parameters:
charset
- the charset to use, ornull
to useDEFAULT_CHARSET
-
getCharset
Helper method to get the charset from the response. This should be done when manually parsing any text response to ensure it is decoded using the correct charset. For instance:Reader reader = new InputStreamReader( resp.getEntity().getContent(), ParserRegistry.getCharset( resp ) );
- Parameters:
resp
-
-
getContentType
Helper method to get the content-type string from the response (no charset).- Parameters:
resp
-
-
parseStream
Default parser used for binary data. This simply returns the underlying response InputStream.- Parameters:
resp
-- Returns:
- an InputStream the binary response stream
- Throws:
IllegalStateException
IOException
- See Also:
-
parseText
Default parser used to handle plain text data. The response text is decoded using the charset passed in the response content-type header.- Parameters:
resp
-- Returns:
- Throws:
UnsupportedEncodingException
IllegalStateException
IOException
- See Also:
-
parseForm
Default parser used to decode a URL-encoded response.- Parameters:
resp
-- Returns:
- Throws:
IOException
- See Also:
-
parseHTML
public groovy.util.slurpersupport.GPathResult parseHTML(org.apache.http.HttpResponse resp) throws IOException, SAXException Parse an HTML document by passing it through the NekoHTML parser.- Parameters:
resp
- HTTP response from which to parse content- Returns:
- the
GPathResult
from callingXmlSlurper.parse(Reader)
- Throws:
IOException
SAXException
- See Also:
-
parseXML
public groovy.util.slurpersupport.GPathResult parseXML(org.apache.http.HttpResponse resp) throws IOException, SAXException, ParserConfigurationException Default parser used to decode an XML response.- Parameters:
resp
- HTTP response from which to parse content- Returns:
- the
GPathResult
from callingXmlSlurper.parse(Reader)
- Throws:
IOException
SAXException
ParserConfigurationException
- See Also:
-
parseJSON
Default parser used to decode a JSON response.- Parameters:
resp
-- Returns:
- Throws:
IOException
- See Also:
-
buildDefaultParserMap
Returns a map of default parsers. Override this method to change what parsers are registered by default. A 'parser' is really just a closure that acceipts an
HttpResponse
instance and returns some parsed data. You can of course callsuper.buildDefaultParserMap()
and then add or remove from that result as well.Default registered parsers are:
-
addCatalog
Add a new XML catalog definiton to the static XML resolver catalog. See the HTTPBuilder source catalog for an example.- Parameters:
catalogLocation
- URL of a catalog definition file- Throws:
IOException
- if the given URL cannot be parsed or accessed for whatever reason.
-
getCatalogResolver
public static org.apache.xml.resolver.tools.CatalogResolver getCatalogResolver()Access the default catalog used by all HTTPBuilder instances.- Returns:
- the static
CatalogResolver
instance
-
getDefaultParser
public groovy.lang.Closure getDefaultParser()Get the default parser used for unregistered content-types.- Returns:
-
setDefaultParser
public void setDefaultParser(groovy.lang.Closure defaultParser) Set the default parser used for unregistered content-types.- Parameters:
defaultParser
- if
-
getAt
Retrieve a parser for the given response content-type string. This is called by HTTPBuildre to retrieve the correct parser for a given content-type. The parser is then used to decode the response data prior to passing it to a response handler.- Parameters:
contentType
-- Returns:
- parser that can interpret the given response content type, or the default parser if no parser is registered for the given content-type.
-
putAt
Register a new parser for the given content-type. The parser closure should accept anHttpResponse
argument and return a type suitable to be passed as the 'parsed data' argument of aresponse handler
closure.- Parameters:
contentType
-content-type
stringvalue
- code that will parse the HttpResponse and return parsed data to the response handler.
-
propertyMissing
Alias forgetAt(Object)
to allow property-style access.- Parameters:
key
- content-type string- Returns:
-
propertyMissing
Alias forputAt(Object, Closure)
to allow property-style access.- Parameters:
key
- content-type stringvalue
- parser closure
-
iterator
Iterate over the entire parser map- Returns:
-