CFXMLParser.h

   1 /*
   2  * Copyright (c) 2008 Apple Inc. All rights reserved.
   3  *
   4  * @APPLE_LICENSE_HEADER_START@
   5  *
   6  * This file contains Original Code and/or Modifications of Original Code
   7  * as defined in and that are subject to the Apple Public Source License
   8  * Version 2.0 (the 'License'). You may not use this file except in
   9  * compliance with the License. Please obtain a copy of the License at
  10  * http://www.opensource.apple.com/apsl/ and read it before using this
  11  * file.
  12  *
  13  * The Original Code and all software distributed under the License are
  14  * distributed on an 'AS IS' basis, WITHOUT WARRANTY OF ANY KIND, EITHER
  15  * EXPRESS OR IMPLIED, AND APPLE HEREBY DISCLAIMS ALL SUCH WARRANTIES,
  16  * INCLUDING WITHOUT LIMITATION, ANY WARRANTIES OF MERCHANTABILITY,
  17  * FITNESS FOR A PARTICULAR PURPOSE, QUIET ENJOYMENT OR NON-INFRINGEMENT.
  18  * Please see the License for the specific language governing rights and
  19  * limitations under the License.
  20  *
  21  * @APPLE_LICENSE_HEADER_END@
  22  */
  23 /*      CFXMLParser.h
  24         Copyright (c) 1998-2007, Apple Inc. All rights reserved.
  25 */
  26
  27 #if !defined(__COREFOUNDATION_CFXMLPARSER__)
  28 #define __COREFOUNDATION_CFXMLPARSER__ 1
  29
  30 #include <CoreFoundation/CFBase.h>
  31 #include <CoreFoundation/CFArray.h>
  32 #include <CoreFoundation/CFData.h>
  33 #include <CoreFoundation/CFDictionary.h>
  34 #include <CoreFoundation/CFTree.h>
  35 #include <CoreFoundation/CFURL.h>
  36 #include <CoreFoundation/CFXMLNode.h>
  37
  38 CF_EXTERN_C_BEGIN
  39
  40 typedef struct __CFXMLParser * CFXMLParserRef;
  41
  42 /* These are the various options you can configure the parser with.  These are
  43    chosen such that an option flag of 0 (kCFXMLParserNoOptions) leaves the XML
  44    as "intact" as possible (reports all structures; performs no replacements).
  45    Hence, to make the parser do the most work, returning only the pure element
  46    tree, set the option flag to kCFXMLParserAllOptions.
  47
  48 kCFXMLParserValidateDocument -
  49    validate the document against its grammar from the DTD, reporting any errors.
  50    Currently not supported.
  51
  52 kCFXMLParserSkipMetaData -
  53    silently skip over metadata constructs (the DTD and comments)
  54
  55 kCFXMLParserReplacePhysicalEntities -
  56    replace declared entities like &lt;.  Note that other than the 5 predefined
  57    entities (lt, gt, quot, amp, apos), these must be defined in the DTD.
  58    Currently not supported.
  59
  60 kCFXMLParserSkipWhitespace -
  61    skip over all whitespace that does not abut non-whitespace character data.
  62    In other words, given <foo>  <bar> blah </bar></foo>, the whitespace between
  63    foo's open tag and bar's open tag would be suppressed, but the whitespace
  64    around blah would be preserved.
  65
  66 kCFXMLParserAddImpliedAttributes -
  67    where the DTD specifies implied attribute-value pairs for a particular element,
  68    add those pairs to any occurances of the element in the element tree.
  69    Currently not supported.
  70 */
  71
  72 enum {
  73     kCFXMLParserValidateDocument = (1 << 0),
  74     kCFXMLParserSkipMetaData = (1 << 1),
  75     kCFXMLParserReplacePhysicalEntities = (1 << 2),
  76     kCFXMLParserSkipWhitespace = (1 << 3),
  77     kCFXMLParserResolveExternalEntities = (1 << 4),
  78     kCFXMLParserAddImpliedAttributes = (1 << 5),
  79     kCFXMLParserAllOptions = 0x00FFFFFF,
  80     kCFXMLParserNoOptions = 0
  81 };
  82 typedef CFOptionFlags CFXMLParserOptions;
  83
  84 /* This list is expected to grow */
  85 enum {
  86     kCFXMLStatusParseNotBegun = -2,
  87     kCFXMLStatusParseInProgress = -1,
  88     kCFXMLStatusParseSuccessful = 0,
  89     kCFXMLErrorUnexpectedEOF = 1,
  90     kCFXMLErrorUnknownEncoding,
  91     kCFXMLErrorEncodingConversionFailure,
  92     kCFXMLErrorMalformedProcessingInstruction,
  93     kCFXMLErrorMalformedDTD,
  94     kCFXMLErrorMalformedName,
  95     kCFXMLErrorMalformedCDSect,
  96     kCFXMLErrorMalformedCloseTag,
  97     kCFXMLErrorMalformedStartTag,
  98     kCFXMLErrorMalformedDocument,
  99     kCFXMLErrorElementlessDocument,
 100     kCFXMLErrorMalformedComment,
 101     kCFXMLErrorMalformedCharacterReference,
 102     kCFXMLErrorMalformedParsedCharacterData,
 103     kCFXMLErrorNoData
 104 };
 105 typedef CFIndex CFXMLParserStatusCode;
 106
 107
 108 /*  These functions are called as a parse progresses.
 109
 110 createXMLStructure -
 111   called as new XML structures are encountered by the parser.  May return NULL to indicate
 112   that the given structure should be skipped; if NULL is returned for a given structure,
 113   only minimal parsing is done for that structure (enough to correctly determine its end,
 114   and to extract any data necessary for the remainder of the parse, such as Entity definitions).
 115   createXMLStructure (or indeed, any of the tree-creation callbacks) will not be called for any
 116   children of the skipped structure.  The only exception is that the top-most element will always
 117   be reported even if NULL was returned for the document as a whole.  NOTE: for performance reasons,
 118   the node passed to createXMLStructure cannot be safely retained by the client; the node as
 119   a whole must be copied (via CFXMLNodeCreateCopy), or its contents must be extracted and copied.
 120
 121 addChild -
 122   called as children are parsed and are ready to be added to the tree.  If createXMLStructure
 123   returns NULL for a given structure, that structure is omitted entirely, and addChild will
 124   NOT be called for either a NULL child or parent.
 125
 126 endXMLStructure -
 127   called once a structure (and all its children) are completely parsed.  As elements are encountered,
 128   createXMLStructure is called for them first, then addChild to add the new structure to its parent,
 129   then addChild (potentially several times) to add the new structure's children to it, then finally
 130   endXMLStructure to show that the structure has been fully parsed.
 131
 132 createXMLStructure, addChild, and endXMLStructure are all REQUIRED TO BE NON-NULL.
 133
 134 resolveExternalEntity -
 135   called when external entities are referenced (NOT when they are simply defined).  If the function
 136   pointer is NULL, the parser uses its internal routines to try and resolve the entity.  If the
 137   function pointer is set, and the function returns NULL, a place holder for the external entity
 138   is inserted into the tree.  In this manner, the parser's client can prevent any external network
 139   or file accesses.
 140
 141 handleError - called as errors/warnings are encountered in the data stream.  At some point, we will
 142   have an enum of the expected errors, some of which will be fatal, others of which will not.  If
 143   the function pointer is NULL, the parser will silently attempt to recover.  The
 144   handleError function may always return false to force the parser to stop; if handleError returns
 145   true, the parser will attempt to recover (fatal errors will still cause the parse to abort
 146   immediately).
 147 */
 148
 149 typedef void *          (*CFXMLParserCreateXMLStructureCallBack)(CFXMLParserRef parser, CFXMLNodeRef nodeDesc, void *info);
 150 typedef void            (*CFXMLParserAddChildCallBack)(CFXMLParserRef parser, void *parent, void *child, void *info);
 151 typedef void            (*CFXMLParserEndXMLStructureCallBack)(CFXMLParserRef parser, void *xmlType, void *info);
 152 typedef CFDataRef       (*CFXMLParserResolveExternalEntityCallBack)(CFXMLParserRef parser, CFXMLExternalID *extID, void *info);
 153 typedef Boolean         (*CFXMLParserHandleErrorCallBack)(CFXMLParserRef parser, CFXMLParserStatusCode error, void *info);
 154 typedef struct {
 155     CFIndex                                  version;
 156     CFXMLParserCreateXMLStructureCallBack    createXMLStructure;
 157     CFXMLParserAddChildCallBack              addChild;
 158     CFXMLParserEndXMLStructureCallBack       endXMLStructure;
 159     CFXMLParserResolveExternalEntityCallBack resolveExternalEntity;
 160     CFXMLParserHandleErrorCallBack           handleError;
 161 } CFXMLParserCallBacks;
 162
 163 typedef const void *    (*CFXMLParserRetainCallBack)(const void *info);
 164 typedef void            (*CFXMLParserReleaseCallBack)(const void *info);
 165 typedef CFStringRef     (*CFXMLParserCopyDescriptionCallBack)(const void *info);
 166 typedef struct {
 167     CFIndex                             version;
 168     void *                              info;
 169     CFXMLParserRetainCallBack           retain;
 170     CFXMLParserReleaseCallBack          release;
 171     CFXMLParserCopyDescriptionCallBack  copyDescription;
 172 } CFXMLParserContext;
 173
 174 CF_EXPORT
 175 CFTypeID CFXMLParserGetTypeID(void);
 176
 177 /* Creates a parser which will parse the given data with the given options.  xmlData may not be NULL.
 178    dataSource should be the URL from which the data came, and may be NULL; it is used to resolve any
 179    relative references found in xmlData. versionOfNodes determines which version CFXMLNodes are produced
 180    by the parser; see CFXMLNode.h for more details.  callBacks are the callbacks called by the parser as
 181    the parse progresses; callBacks, callBacks->createXMLStructure, callBacks->addChild, and
 182    callBacks->endXMLStructure must all be non-NULL.  context determines what if any info pointer is
 183    passed to the callbacks as the parse progresses; context may be NULL.  */
 184 CF_EXPORT
 185 CFXMLParserRef CFXMLParserCreate(CFAllocatorRef allocator, CFDataRef xmlData, CFURLRef dataSource, CFOptionFlags parseOptions, CFIndex versionOfNodes, CFXMLParserCallBacks *callBacks, CFXMLParserContext *context);
 186
 187 /* Arguments as above, except that the data to be parsed is loaded directly
 188    from dataSource.  dataSource may not be NULL.  */
 189 CF_EXPORT
 190 CFXMLParserRef CFXMLParserCreateWithDataFromURL(CFAllocatorRef allocator, CFURLRef dataSource, CFOptionFlags parseOptions, CFIndex versionOfNodes, CFXMLParserCallBacks *callBacks, CFXMLParserContext *context);
 191
 192 CF_EXPORT
 193 void CFXMLParserGetContext(CFXMLParserRef parser, CFXMLParserContext *context);
 194
 195 CF_EXPORT
 196 void CFXMLParserGetCallBacks(CFXMLParserRef parser, CFXMLParserCallBacks *callBacks);
 197
 198 CF_EXPORT
 199 CFURLRef CFXMLParserGetSourceURL(CFXMLParserRef parser);
 200
 201 /* Returns the character index of the current parse location */
 202 CF_EXPORT
 203 CFIndex CFXMLParserGetLocation(CFXMLParserRef parser);
 204
 205 /* Returns the line number of the current parse location */
 206 CF_EXPORT
 207 CFIndex CFXMLParserGetLineNumber(CFXMLParserRef parser);
 208
 209 /* Returns the top-most object returned by the createXMLStructure callback */
 210 CF_EXPORT
 211 void *CFXMLParserGetDocument(CFXMLParserRef parser);
 212
 213 /* Get the status code or a user-readable description of the last error that occurred in a parse.
 214    If no error has occurred, a null description string is returned.  See the enum above for
 215    possible status returns */
 216 CF_EXPORT
 217 CFXMLParserStatusCode CFXMLParserGetStatusCode(CFXMLParserRef parser);
 218
 219 CF_EXPORT
 220 CFStringRef CFXMLParserCopyErrorDescription(CFXMLParserRef parser);
 221
 222 /* Cause any in-progress parse to abort with the given error code and description.  errorCode
 223    must be positive, and errorDescription may not be NULL.  Cannot be called asynchronously
 224    (i.e. must be called from within a parser callback) */
 225 CF_EXPORT
 226 void CFXMLParserAbort(CFXMLParserRef parser, CFXMLParserStatusCode errorCode, CFStringRef errorDescription);
 227
 228 /* Starts a parse of the data the parser was created with; returns success or failure.
 229    Upon success, use CFXMLParserGetDocument() to get the product of the parse.  Upon
 230    failure, use CFXMLParserGetErrorCode() or CFXMLParserCopyErrorDescription() to get
 231    information about the error.  It is an error to call CFXMLParserParse() while a
 232    parse is already underway. */
 233 CF_EXPORT
 234 Boolean CFXMLParserParse(CFXMLParserRef parser);
 235
 236 /* These functions provide a higher-level interface.  The XML data is parsed to a
 237    special CFTree (an CFXMLTree) with known contexts and callbacks.  See CFXMLNode.h
 238    for full details on using an CFXMLTree and the CFXMLNodes contained therein.
 239 */
 240 /* Parse to an CFXMLTreeRef.  parseOptions are as above. versionOfNodes determines
 241    what version CFXMLNodes are used to populate the tree.  */
 242 CF_EXPORT
 243 CFXMLTreeRef CFXMLTreeCreateFromData(CFAllocatorRef allocator, CFDataRef xmlData, CFURLRef dataSource, CFOptionFlags parseOptions, CFIndex versionOfNodes);
 244
 245 /* As above, with the additional by-reference pass of a CFDictionaryRef containing
 246    various error information (see below). The caller is responsible for releasing the
 247    returned dictionary. If the error dictionary is not desired, pass NULL. */
 248 CF_EXPORT
 249 CFXMLTreeRef CFXMLTreeCreateFromDataWithError(CFAllocatorRef allocator, CFDataRef xmlData, CFURLRef dataSource, CFOptionFlags parseOptions, CFIndex versionOfNodes, CFDictionaryRef *errorDict) AVAILABLE_MAC_OS_X_VERSION_10_3_AND_LATER;
 250
 251 /* Loads the data to be parsed directly from dataSource.  Arguments as above. */
 252 CF_EXPORT
 253 CFXMLTreeRef CFXMLTreeCreateWithDataFromURL(CFAllocatorRef allocator, CFURLRef dataSource, CFOptionFlags parseOptions, CFIndex versionOfNodes);
 254
 255 /* Generate the XMLData (ready to be written to whatever permanent storage is to be
 256    used) from an CFXMLTree.  Will NOT regenerate entity references (except those
 257    required for syntactic correctness) if they were replaced at the parse time;
 258    clients that wish this should walk the tree and re-insert any entity references
 259    that should appear in the final output file. */
 260 CF_EXPORT
 261 CFDataRef CFXMLTreeCreateXMLData(CFAllocatorRef allocator, CFXMLTreeRef xmlTree);
 262
 263 /* Escaping and unescaping XML entities in CFStrings. The standard XML entities
 264    are always replaced.  */
 265 /* Creates a CFString by replacing entities that appear in the entities dictionary.
 266    Dictionary keys are the entities themselves, and the values should be CFStrings
 267    containing the expansion. Pass NULL for entitiesDictionary to indicate no entities
 268    other than the standard five. */
 269 CF_EXPORT
 270 CFStringRef CFXMLCreateStringByEscapingEntities(CFAllocatorRef allocator, CFStringRef string, CFDictionaryRef entitiesDictionary) AVAILABLE_MAC_OS_X_VERSION_10_3_AND_LATER;
 271
 272 CF_EXPORT
 273 CFStringRef CFXMLCreateStringByUnescapingEntities(CFAllocatorRef allocator, CFStringRef string, CFDictionaryRef entitiesDictionary) AVAILABLE_MAC_OS_X_VERSION_10_3_AND_LATER;
 274
 275 /* CFXMLTreeCreateFromDataWithError error dictionary key constants. */
 276 CF_EXPORT const CFStringRef kCFXMLTreeErrorDescription          AVAILABLE_MAC_OS_X_VERSION_10_3_AND_LATER;
 277     /* value is a CFString containing the readable error string. */
 278
 279 CF_EXPORT const CFStringRef kCFXMLTreeErrorLineNumber           AVAILABLE_MAC_OS_X_VERSION_10_3_AND_LATER;
 280     /* value is a CFNumber containing the line on which the error appears. */
 281
 282 CF_EXPORT const CFStringRef kCFXMLTreeErrorLocation             AVAILABLE_MAC_OS_X_VERSION_10_3_AND_LATER;
 283     /* value is a CFNumber containing the byte location at which the error occurred. */
 284
 285 CF_EXPORT const CFStringRef kCFXMLTreeErrorStatusCode           AVAILABLE_MAC_OS_X_VERSION_10_3_AND_LATER;
 286     /* value is a CFNumber containing the error status code. */
 287
 288 CF_EXTERN_C_END
 289
 290 #endif /* ! __COREFOUNDATION_CFXMLPARSER__ */
 291