12345678910111213141516171819202122232425262728293031323334353637383940414243444546474849505152535455565758596061626364656667686970717273747576777879808182838485868788899091929394959697989910010110210310410510610710810911011111211311411511611711811912012112212312412512612712812913013113213313413513613713813914014114214314414514614714814915015115215315415515615715815916016116216316416516616716816917017117217317417517617717817918018118218318418518618718818919019119219319419519619719819920020120220320420520620720820921021121221321421521621721821922022122222322422522622722822923023123223323423523623723823924024124224324424524624724824925025125225325425525625725825926026126226326426526626726826927027127227327427527627727827928028128228328428528628728828929029129229329429529629729829930030130230330430530630730830931031131231331431531631731831932032132232332432532632732832933033133233333433533633733833934034134234334434534634734834935035135235335435535635735835936036136236336436536636736836937037137237337437537637737837938038138238338438538638738838939039139239339439539639739839940040140240340440540640740840941041141241341441541641741841942042142242342442542642742842943043143243343443543643743843944044144244344444544644744844945045145245345445545645745845946046146246346446546646746846947047147247347447547647747847948048148248348448548648748848949049149249349449549649749849950050150250350450550650750850951051151251351451551651751851952052152252352452552652752852953053153253353453553653753853954054154254354454554654754854955055155255355455555655755855956056156256356456556656756856957057157257357457557657757857958058158258358458558658758858959059159259359459559659759859960060160260360460560660760860961061161261361461561661761861962062162262362462562662762862963063163263363463563663763863964064164264364464564664764864965065165265365465565665765865966066166266366466566666766866967067167267367467567667767867968068168268368468568668768868969069169269369469569669769869970070170270370470570670770870971071171271371471571671771871972072172272372472572672772872973073173273373473573673773873974074174274374474574674774874975075175275375475575675775875976076176276376476576676776876977077177277377477577677777877978078178278378478578678778878979079179279379479579679779879980080180280380480580680780880981081181281381481581681781881982082182282382482582682782882983083183283383483583683783883984084184284384484584684784884985085185285385485585685785885986086186286386486586686786886987087187287387487587687787887988088188288388488588688788888989089189289389489589689789889990090190290390490590690790890991091191291391491591691791891992092192292392492592692792892993093193293393493593693793893994094194294394494594694794894995095195295395495595695795895996096196296396496596696796896997097197297397497597697797897998098198298398498598698798898999099199299399499599699799899910001001100210031004100510061007100810091010101110121013101410151016101710181019102010211022102310241025102610271028102910301031103210331034103510361037103810391040104110421043104410451046104710481049105010511052105310541055105610571058105910601061106210631064106510661067106810691070107110721073107410751076107710781079108010811082108310841085108610871088108910901091109210931094109510961097109810991100110111021103110411051106110711081109111011111112111311141115111611171118111911201121112211231124112511261127112811291130113111321133113411351136113711381139114011411142114311441145114611471148114911501151115211531154115511561157115811591160116111621163116411651166116711681169117011711172117311741175117611771178117911801181118211831184118511861187118811891190119111921193119411951196119711981199120012011202120312041205120612071208120912101211121212131214121512161217121812191220122112221223122412251226122712281229123012311232123312341235123612371238123912401241 |
- /*
- * Summary: the core parser module
- * Description: Interfaces, constants and types related to the XML parser
- *
- * Copy: See Copyright for the status of this software.
- *
- * Author: Daniel Veillard
- */
- #ifndef __XML_PARSER_H__
- #define __XML_PARSER_H__
- #include <stdarg.h>
- #include <libxml/xmlversion.h>
- #include <libxml/tree.h>
- #include <libxml/dict.h>
- #include <libxml/hash.h>
- #include <libxml/valid.h>
- #include <libxml/entities.h>
- #include <libxml/xmlerror.h>
- #include <libxml/xmlstring.h>
- #ifdef __cplusplus
- extern "C" {
- #endif
- /**
- * XML_DEFAULT_VERSION:
- *
- * The default version of XML used: 1.0
- */
- #define XML_DEFAULT_VERSION "1.0"
- /**
- * xmlParserInput:
- *
- * An xmlParserInput is an input flow for the XML processor.
- * Each entity parsed is associated an xmlParserInput (except the
- * few predefined ones). This is the case both for internal entities
- * - in which case the flow is already completely in memory - or
- * external entities - in which case we use the buf structure for
- * progressive reading and I18N conversions to the internal UTF-8 format.
- */
- /**
- * xmlParserInputDeallocate:
- * @str: the string to deallocate
- *
- * Callback for freeing some parser input allocations.
- */
- typedef void (* xmlParserInputDeallocate)(xmlChar *str);
- struct _xmlParserInput {
- /* Input buffer */
- xmlParserInputBufferPtr buf; /* UTF-8 encoded buffer */
- const char *filename; /* The file analyzed, if any */
- const char *directory; /* the directory/base of the file */
- const xmlChar *base; /* Base of the array to parse */
- const xmlChar *cur; /* Current char being parsed */
- const xmlChar *end; /* end of the array to parse */
- int length; /* length if known */
- int line; /* Current line */
- int col; /* Current column */
- /*
- * NOTE: consumed is only tested for equality in the parser code,
- * so even if there is an overflow this should not give troubles
- * for parsing very large instances.
- */
- unsigned long consumed; /* How many xmlChars already consumed */
- xmlParserInputDeallocate free; /* function to deallocate the base */
- const xmlChar *encoding; /* the encoding string for entity */
- const xmlChar *version; /* the version string for entity */
- int standalone; /* Was that entity marked standalone */
- int id; /* an unique identifier for the entity */
- };
- /**
- * xmlParserNodeInfo:
- *
- * The parser can be asked to collect Node informations, i.e. at what
- * place in the file they were detected.
- * NOTE: This is off by default and not very well tested.
- */
- typedef struct _xmlParserNodeInfo xmlParserNodeInfo;
- typedef xmlParserNodeInfo *xmlParserNodeInfoPtr;
- struct _xmlParserNodeInfo {
- const struct _xmlNode* node;
- /* Position & line # that text that created the node begins & ends on */
- unsigned long begin_pos;
- unsigned long begin_line;
- unsigned long end_pos;
- unsigned long end_line;
- };
- typedef struct _xmlParserNodeInfoSeq xmlParserNodeInfoSeq;
- typedef xmlParserNodeInfoSeq *xmlParserNodeInfoSeqPtr;
- struct _xmlParserNodeInfoSeq {
- unsigned long maximum;
- unsigned long length;
- xmlParserNodeInfo* buffer;
- };
- /**
- * xmlParserInputState:
- *
- * The parser is now working also as a state based parser.
- * The recursive one use the state info for entities processing.
- */
- typedef enum {
- XML_PARSER_EOF = -1, /* nothing is to be parsed */
- XML_PARSER_START = 0, /* nothing has been parsed */
- XML_PARSER_MISC, /* Misc* before int subset */
- XML_PARSER_PI, /* Within a processing instruction */
- XML_PARSER_DTD, /* within some DTD content */
- XML_PARSER_PROLOG, /* Misc* after internal subset */
- XML_PARSER_COMMENT, /* within a comment */
- XML_PARSER_START_TAG, /* within a start tag */
- XML_PARSER_CONTENT, /* within the content */
- XML_PARSER_CDATA_SECTION, /* within a CDATA section */
- XML_PARSER_END_TAG, /* within a closing tag */
- XML_PARSER_ENTITY_DECL, /* within an entity declaration */
- XML_PARSER_ENTITY_VALUE, /* within an entity value in a decl */
- XML_PARSER_ATTRIBUTE_VALUE, /* within an attribute value */
- XML_PARSER_SYSTEM_LITERAL, /* within a SYSTEM value */
- XML_PARSER_EPILOG, /* the Misc* after the last end tag */
- XML_PARSER_IGNORE, /* within an IGNORED section */
- XML_PARSER_PUBLIC_LITERAL /* within a PUBLIC value */
- } xmlParserInputState;
- /**
- * XML_DETECT_IDS:
- *
- * Bit in the loadsubset context field to tell to do ID/REFs lookups.
- * Use it to initialize xmlLoadExtDtdDefaultValue.
- */
- #define XML_DETECT_IDS 2
- /**
- * XML_COMPLETE_ATTRS:
- *
- * Bit in the loadsubset context field to tell to do complete the
- * elements attributes lists with the ones defaulted from the DTDs.
- * Use it to initialize xmlLoadExtDtdDefaultValue.
- */
- #define XML_COMPLETE_ATTRS 4
- /**
- * XML_SKIP_IDS:
- *
- * Bit in the loadsubset context field to tell to not do ID/REFs registration.
- * Used to initialize xmlLoadExtDtdDefaultValue in some special cases.
- */
- #define XML_SKIP_IDS 8
- /**
- * xmlParserMode:
- *
- * A parser can operate in various modes
- */
- typedef enum {
- XML_PARSE_UNKNOWN = 0,
- XML_PARSE_DOM = 1,
- XML_PARSE_SAX = 2,
- XML_PARSE_PUSH_DOM = 3,
- XML_PARSE_PUSH_SAX = 4,
- XML_PARSE_READER = 5
- } xmlParserMode;
- /**
- * xmlParserCtxt:
- *
- * The parser context.
- * NOTE This doesn't completely define the parser state, the (current ?)
- * design of the parser uses recursive function calls since this allow
- * and easy mapping from the production rules of the specification
- * to the actual code. The drawback is that the actual function call
- * also reflect the parser state. However most of the parsing routines
- * takes as the only argument the parser context pointer, so migrating
- * to a state based parser for progressive parsing shouldn't be too hard.
- */
- struct _xmlParserCtxt {
- struct _xmlSAXHandler *sax; /* The SAX handler */
- void *userData; /* For SAX interface only, used by DOM build */
- xmlDocPtr myDoc; /* the document being built */
- int wellFormed; /* is the document well formed */
- int replaceEntities; /* shall we replace entities ? */
- const xmlChar *version; /* the XML version string */
- const xmlChar *encoding; /* the declared encoding, if any */
- int standalone; /* standalone document */
- int html; /* an HTML(1)/Docbook(2) document
- * 3 is HTML after <head>
- * 10 is HTML after <body>
- */
- /* Input stream stack */
- xmlParserInputPtr input; /* Current input stream */
- int inputNr; /* Number of current input streams */
- int inputMax; /* Max number of input streams */
- xmlParserInputPtr *inputTab; /* stack of inputs */
- /* Node analysis stack only used for DOM building */
- xmlNodePtr node; /* Current parsed Node */
- int nodeNr; /* Depth of the parsing stack */
- int nodeMax; /* Max depth of the parsing stack */
- xmlNodePtr *nodeTab; /* array of nodes */
- int record_info; /* Whether node info should be kept */
- xmlParserNodeInfoSeq node_seq; /* info about each node parsed */
- int errNo; /* error code */
- int hasExternalSubset; /* reference and external subset */
- int hasPErefs; /* the internal subset has PE refs */
- int external; /* are we parsing an external entity */
- int valid; /* is the document valid */
- int validate; /* shall we try to validate ? */
- xmlValidCtxt vctxt; /* The validity context */
- xmlParserInputState instate; /* current type of input */
- int token; /* next char look-ahead */
- char *directory; /* the data directory */
- /* Node name stack */
- const xmlChar *name; /* Current parsed Node */
- int nameNr; /* Depth of the parsing stack */
- int nameMax; /* Max depth of the parsing stack */
- const xmlChar **nameTab; /* array of nodes */
- long nbChars; /* number of xmlChar processed */
- long checkIndex; /* used by progressive parsing lookup */
- int keepBlanks; /* ugly but ... */
- int disableSAX; /* SAX callbacks are disabled */
- int inSubset; /* Parsing is in int 1/ext 2 subset */
- const xmlChar * intSubName; /* name of subset */
- xmlChar * extSubURI; /* URI of external subset */
- xmlChar * extSubSystem; /* SYSTEM ID of external subset */
- /* xml:space values */
- int * space; /* Should the parser preserve spaces */
- int spaceNr; /* Depth of the parsing stack */
- int spaceMax; /* Max depth of the parsing stack */
- int * spaceTab; /* array of space infos */
- int depth; /* to prevent entity substitution loops */
- xmlParserInputPtr entity; /* used to check entities boundaries */
- int charset; /* encoding of the in-memory content
- actually an xmlCharEncoding */
- int nodelen; /* Those two fields are there to */
- int nodemem; /* Speed up large node parsing */
- int pedantic; /* signal pedantic warnings */
- void *_private; /* For user data, libxml won't touch it */
- int loadsubset; /* should the external subset be loaded */
- int linenumbers; /* set line number in element content */
- void *catalogs; /* document's own catalog */
- int recovery; /* run in recovery mode */
- int progressive; /* is this a progressive parsing */
- xmlDictPtr dict; /* dictionnary for the parser */
- const xmlChar **atts; /* array for the attributes callbacks */
- int maxatts; /* the size of the array */
- int docdict; /* use strings from dict to build tree */
- /*
- * pre-interned strings
- */
- const xmlChar *str_xml;
- const xmlChar *str_xmlns;
- const xmlChar *str_xml_ns;
- /*
- * Everything below is used only by the new SAX mode
- */
- int sax2; /* operating in the new SAX mode */
- int nsNr; /* the number of inherited namespaces */
- int nsMax; /* the size of the arrays */
- const xmlChar **nsTab; /* the array of prefix/namespace name */
- int *attallocs; /* which attribute were allocated */
- void **pushTab; /* array of data for push */
- xmlHashTablePtr attsDefault; /* defaulted attributes if any */
- xmlHashTablePtr attsSpecial; /* non-CDATA attributes if any */
- int nsWellFormed; /* is the document XML Nanespace okay */
- int options; /* Extra options */
- /*
- * Those fields are needed only for treaming parsing so far
- */
- int dictNames; /* Use dictionary names for the tree */
- int freeElemsNr; /* number of freed element nodes */
- xmlNodePtr freeElems; /* List of freed element nodes */
- int freeAttrsNr; /* number of freed attributes nodes */
- xmlAttrPtr freeAttrs; /* List of freed attributes nodes */
- /*
- * the complete error informations for the last error.
- */
- xmlError lastError;
- xmlParserMode parseMode; /* the parser mode */
- unsigned long nbentities; /* number of entities references */
- unsigned long sizeentities; /* size of parsed entities */
- /* for use by HTML non-recursive parser */
- xmlParserNodeInfo *nodeInfo; /* Current NodeInfo */
- int nodeInfoNr; /* Depth of the parsing stack */
- int nodeInfoMax; /* Max depth of the parsing stack */
- xmlParserNodeInfo *nodeInfoTab; /* array of nodeInfos */
- int input_id; /* we need to label inputs */
- unsigned long sizeentcopy; /* volume of entity copy */
- };
- /**
- * xmlSAXLocator:
- *
- * A SAX Locator.
- */
- struct _xmlSAXLocator {
- const xmlChar *(*getPublicId)(void *ctx);
- const xmlChar *(*getSystemId)(void *ctx);
- int (*getLineNumber)(void *ctx);
- int (*getColumnNumber)(void *ctx);
- };
- /**
- * xmlSAXHandler:
- *
- * A SAX handler is bunch of callbacks called by the parser when processing
- * of the input generate data or structure informations.
- */
- /**
- * resolveEntitySAXFunc:
- * @ctx: the user data (XML parser context)
- * @publicId: The public ID of the entity
- * @systemId: The system ID of the entity
- *
- * Callback:
- * The entity loader, to control the loading of external entities,
- * the application can either:
- * - override this resolveEntity() callback in the SAX block
- * - or better use the xmlSetExternalEntityLoader() function to
- * set up it's own entity resolution routine
- *
- * Returns the xmlParserInputPtr if inlined or NULL for DOM behaviour.
- */
- typedef xmlParserInputPtr (*resolveEntitySAXFunc) (void *ctx,
- const xmlChar *publicId,
- const xmlChar *systemId);
- /**
- * internalSubsetSAXFunc:
- * @ctx: the user data (XML parser context)
- * @name: the root element name
- * @ExternalID: the external ID
- * @SystemID: the SYSTEM ID (e.g. filename or URL)
- *
- * Callback on internal subset declaration.
- */
- typedef void (*internalSubsetSAXFunc) (void *ctx,
- const xmlChar *name,
- const xmlChar *ExternalID,
- const xmlChar *SystemID);
- /**
- * externalSubsetSAXFunc:
- * @ctx: the user data (XML parser context)
- * @name: the root element name
- * @ExternalID: the external ID
- * @SystemID: the SYSTEM ID (e.g. filename or URL)
- *
- * Callback on external subset declaration.
- */
- typedef void (*externalSubsetSAXFunc) (void *ctx,
- const xmlChar *name,
- const xmlChar *ExternalID,
- const xmlChar *SystemID);
- /**
- * getEntitySAXFunc:
- * @ctx: the user data (XML parser context)
- * @name: The entity name
- *
- * Get an entity by name.
- *
- * Returns the xmlEntityPtr if found.
- */
- typedef xmlEntityPtr (*getEntitySAXFunc) (void *ctx,
- const xmlChar *name);
- /**
- * getParameterEntitySAXFunc:
- * @ctx: the user data (XML parser context)
- * @name: The entity name
- *
- * Get a parameter entity by name.
- *
- * Returns the xmlEntityPtr if found.
- */
- typedef xmlEntityPtr (*getParameterEntitySAXFunc) (void *ctx,
- const xmlChar *name);
- /**
- * entityDeclSAXFunc:
- * @ctx: the user data (XML parser context)
- * @name: the entity name
- * @type: the entity type
- * @publicId: The public ID of the entity
- * @systemId: The system ID of the entity
- * @content: the entity value (without processing).
- *
- * An entity definition has been parsed.
- */
- typedef void (*entityDeclSAXFunc) (void *ctx,
- const xmlChar *name,
- int type,
- const xmlChar *publicId,
- const xmlChar *systemId,
- xmlChar *content);
- /**
- * notationDeclSAXFunc:
- * @ctx: the user data (XML parser context)
- * @name: The name of the notation
- * @publicId: The public ID of the entity
- * @systemId: The system ID of the entity
- *
- * What to do when a notation declaration has been parsed.
- */
- typedef void (*notationDeclSAXFunc)(void *ctx,
- const xmlChar *name,
- const xmlChar *publicId,
- const xmlChar *systemId);
- /**
- * attributeDeclSAXFunc:
- * @ctx: the user data (XML parser context)
- * @elem: the name of the element
- * @fullname: the attribute name
- * @type: the attribute type
- * @def: the type of default value
- * @defaultValue: the attribute default value
- * @tree: the tree of enumerated value set
- *
- * An attribute definition has been parsed.
- */
- typedef void (*attributeDeclSAXFunc)(void *ctx,
- const xmlChar *elem,
- const xmlChar *fullname,
- int type,
- int def,
- const xmlChar *defaultValue,
- xmlEnumerationPtr tree);
- /**
- * elementDeclSAXFunc:
- * @ctx: the user data (XML parser context)
- * @name: the element name
- * @type: the element type
- * @content: the element value tree
- *
- * An element definition has been parsed.
- */
- typedef void (*elementDeclSAXFunc)(void *ctx,
- const xmlChar *name,
- int type,
- xmlElementContentPtr content);
- /**
- * unparsedEntityDeclSAXFunc:
- * @ctx: the user data (XML parser context)
- * @name: The name of the entity
- * @publicId: The public ID of the entity
- * @systemId: The system ID of the entity
- * @notationName: the name of the notation
- *
- * What to do when an unparsed entity declaration is parsed.
- */
- typedef void (*unparsedEntityDeclSAXFunc)(void *ctx,
- const xmlChar *name,
- const xmlChar *publicId,
- const xmlChar *systemId,
- const xmlChar *notationName);
- /**
- * setDocumentLocatorSAXFunc:
- * @ctx: the user data (XML parser context)
- * @loc: A SAX Locator
- *
- * Receive the document locator at startup, actually xmlDefaultSAXLocator.
- * Everything is available on the context, so this is useless in our case.
- */
- typedef void (*setDocumentLocatorSAXFunc) (void *ctx,
- xmlSAXLocatorPtr loc);
- /**
- * startDocumentSAXFunc:
- * @ctx: the user data (XML parser context)
- *
- * Called when the document start being processed.
- */
- typedef void (*startDocumentSAXFunc) (void *ctx);
- /**
- * endDocumentSAXFunc:
- * @ctx: the user data (XML parser context)
- *
- * Called when the document end has been detected.
- */
- typedef void (*endDocumentSAXFunc) (void *ctx);
- /**
- * startElementSAXFunc:
- * @ctx: the user data (XML parser context)
- * @name: The element name, including namespace prefix
- * @atts: An array of name/value attributes pairs, NULL terminated
- *
- * Called when an opening tag has been processed.
- */
- typedef void (*startElementSAXFunc) (void *ctx,
- const xmlChar *name,
- const xmlChar **atts);
- /**
- * endElementSAXFunc:
- * @ctx: the user data (XML parser context)
- * @name: The element name
- *
- * Called when the end of an element has been detected.
- */
- typedef void (*endElementSAXFunc) (void *ctx,
- const xmlChar *name);
- /**
- * attributeSAXFunc:
- * @ctx: the user data (XML parser context)
- * @name: The attribute name, including namespace prefix
- * @value: The attribute value
- *
- * Handle an attribute that has been read by the parser.
- * The default handling is to convert the attribute into an
- * DOM subtree and past it in a new xmlAttr element added to
- * the element.
- */
- typedef void (*attributeSAXFunc) (void *ctx,
- const xmlChar *name,
- const xmlChar *value);
- /**
- * referenceSAXFunc:
- * @ctx: the user data (XML parser context)
- * @name: The entity name
- *
- * Called when an entity reference is detected.
- */
- typedef void (*referenceSAXFunc) (void *ctx,
- const xmlChar *name);
- /**
- * charactersSAXFunc:
- * @ctx: the user data (XML parser context)
- * @ch: a xmlChar string
- * @len: the number of xmlChar
- *
- * Receiving some chars from the parser.
- */
- typedef void (*charactersSAXFunc) (void *ctx,
- const xmlChar *ch,
- int len);
- /**
- * ignorableWhitespaceSAXFunc:
- * @ctx: the user data (XML parser context)
- * @ch: a xmlChar string
- * @len: the number of xmlChar
- *
- * Receiving some ignorable whitespaces from the parser.
- * UNUSED: by default the DOM building will use characters.
- */
- typedef void (*ignorableWhitespaceSAXFunc) (void *ctx,
- const xmlChar *ch,
- int len);
- /**
- * processingInstructionSAXFunc:
- * @ctx: the user data (XML parser context)
- * @target: the target name
- * @data: the PI data's
- *
- * A processing instruction has been parsed.
- */
- typedef void (*processingInstructionSAXFunc) (void *ctx,
- const xmlChar *target,
- const xmlChar *data);
- /**
- * commentSAXFunc:
- * @ctx: the user data (XML parser context)
- * @value: the comment content
- *
- * A comment has been parsed.
- */
- typedef void (*commentSAXFunc) (void *ctx,
- const xmlChar *value);
- /**
- * cdataBlockSAXFunc:
- * @ctx: the user data (XML parser context)
- * @value: The pcdata content
- * @len: the block length
- *
- * Called when a pcdata block has been parsed.
- */
- typedef void (*cdataBlockSAXFunc) (
- void *ctx,
- const xmlChar *value,
- int len);
- /**
- * warningSAXFunc:
- * @ctx: an XML parser context
- * @msg: the message to display/transmit
- * @...: extra parameters for the message display
- *
- * Display and format a warning messages, callback.
- */
- typedef void (XMLCDECL *warningSAXFunc) (void *ctx,
- const char *msg, ...) LIBXML_ATTR_FORMAT(2,3);
- /**
- * errorSAXFunc:
- * @ctx: an XML parser context
- * @msg: the message to display/transmit
- * @...: extra parameters for the message display
- *
- * Display and format an error messages, callback.
- */
- typedef void (XMLCDECL *errorSAXFunc) (void *ctx,
- const char *msg, ...) LIBXML_ATTR_FORMAT(2,3);
- /**
- * fatalErrorSAXFunc:
- * @ctx: an XML parser context
- * @msg: the message to display/transmit
- * @...: extra parameters for the message display
- *
- * Display and format fatal error messages, callback.
- * Note: so far fatalError() SAX callbacks are not used, error()
- * get all the callbacks for errors.
- */
- typedef void (XMLCDECL *fatalErrorSAXFunc) (void *ctx,
- const char *msg, ...) LIBXML_ATTR_FORMAT(2,3);
- /**
- * isStandaloneSAXFunc:
- * @ctx: the user data (XML parser context)
- *
- * Is this document tagged standalone?
- *
- * Returns 1 if true
- */
- typedef int (*isStandaloneSAXFunc) (void *ctx);
- /**
- * hasInternalSubsetSAXFunc:
- * @ctx: the user data (XML parser context)
- *
- * Does this document has an internal subset.
- *
- * Returns 1 if true
- */
- typedef int (*hasInternalSubsetSAXFunc) (void *ctx);
- /**
- * hasExternalSubsetSAXFunc:
- * @ctx: the user data (XML parser context)
- *
- * Does this document has an external subset?
- *
- * Returns 1 if true
- */
- typedef int (*hasExternalSubsetSAXFunc) (void *ctx);
- /************************************************************************
- * *
- * The SAX version 2 API extensions *
- * *
- ************************************************************************/
- /**
- * XML_SAX2_MAGIC:
- *
- * Special constant found in SAX2 blocks initialized fields
- */
- #define XML_SAX2_MAGIC 0xDEEDBEAF
- /**
- * startElementNsSAX2Func:
- * @ctx: the user data (XML parser context)
- * @localname: the local name of the element
- * @prefix: the element namespace prefix if available
- * @URI: the element namespace name if available
- * @nb_namespaces: number of namespace definitions on that node
- * @namespaces: pointer to the array of prefix/URI pairs namespace definitions
- * @nb_attributes: the number of attributes on that node
- * @nb_defaulted: the number of defaulted attributes. The defaulted
- * ones are at the end of the array
- * @attributes: pointer to the array of (localname/prefix/URI/value/end)
- * attribute values.
- *
- * SAX2 callback when an element start has been detected by the parser.
- * It provides the namespace informations for the element, as well as
- * the new namespace declarations on the element.
- */
- typedef void (*startElementNsSAX2Func) (void *ctx,
- const xmlChar *localname,
- const xmlChar *prefix,
- const xmlChar *URI,
- int nb_namespaces,
- const xmlChar **namespaces,
- int nb_attributes,
- int nb_defaulted,
- const xmlChar **attributes);
- /**
- * endElementNsSAX2Func:
- * @ctx: the user data (XML parser context)
- * @localname: the local name of the element
- * @prefix: the element namespace prefix if available
- * @URI: the element namespace name if available
- *
- * SAX2 callback when an element end has been detected by the parser.
- * It provides the namespace informations for the element.
- */
- typedef void (*endElementNsSAX2Func) (void *ctx,
- const xmlChar *localname,
- const xmlChar *prefix,
- const xmlChar *URI);
- struct _xmlSAXHandler {
- internalSubsetSAXFunc internalSubset;
- isStandaloneSAXFunc isStandalone;
- hasInternalSubsetSAXFunc hasInternalSubset;
- hasExternalSubsetSAXFunc hasExternalSubset;
- resolveEntitySAXFunc resolveEntity;
- getEntitySAXFunc getEntity;
- entityDeclSAXFunc entityDecl;
- notationDeclSAXFunc notationDecl;
- attributeDeclSAXFunc attributeDecl;
- elementDeclSAXFunc elementDecl;
- unparsedEntityDeclSAXFunc unparsedEntityDecl;
- setDocumentLocatorSAXFunc setDocumentLocator;
- startDocumentSAXFunc startDocument;
- endDocumentSAXFunc endDocument;
- startElementSAXFunc startElement;
- endElementSAXFunc endElement;
- referenceSAXFunc reference;
- charactersSAXFunc characters;
- ignorableWhitespaceSAXFunc ignorableWhitespace;
- processingInstructionSAXFunc processingInstruction;
- commentSAXFunc comment;
- warningSAXFunc warning;
- errorSAXFunc error;
- fatalErrorSAXFunc fatalError; /* unused error() get all the errors */
- getParameterEntitySAXFunc getParameterEntity;
- cdataBlockSAXFunc cdataBlock;
- externalSubsetSAXFunc externalSubset;
- unsigned int initialized;
- /* The following fields are extensions available only on version 2 */
- void *_private;
- startElementNsSAX2Func startElementNs;
- endElementNsSAX2Func endElementNs;
- xmlStructuredErrorFunc serror;
- };
- /*
- * SAX Version 1
- */
- typedef struct _xmlSAXHandlerV1 xmlSAXHandlerV1;
- typedef xmlSAXHandlerV1 *xmlSAXHandlerV1Ptr;
- struct _xmlSAXHandlerV1 {
- internalSubsetSAXFunc internalSubset;
- isStandaloneSAXFunc isStandalone;
- hasInternalSubsetSAXFunc hasInternalSubset;
- hasExternalSubsetSAXFunc hasExternalSubset;
- resolveEntitySAXFunc resolveEntity;
- getEntitySAXFunc getEntity;
- entityDeclSAXFunc entityDecl;
- notationDeclSAXFunc notationDecl;
- attributeDeclSAXFunc attributeDecl;
- elementDeclSAXFunc elementDecl;
- unparsedEntityDeclSAXFunc unparsedEntityDecl;
- setDocumentLocatorSAXFunc setDocumentLocator;
- startDocumentSAXFunc startDocument;
- endDocumentSAXFunc endDocument;
- startElementSAXFunc startElement;
- endElementSAXFunc endElement;
- referenceSAXFunc reference;
- charactersSAXFunc characters;
- ignorableWhitespaceSAXFunc ignorableWhitespace;
- processingInstructionSAXFunc processingInstruction;
- commentSAXFunc comment;
- warningSAXFunc warning;
- errorSAXFunc error;
- fatalErrorSAXFunc fatalError; /* unused error() get all the errors */
- getParameterEntitySAXFunc getParameterEntity;
- cdataBlockSAXFunc cdataBlock;
- externalSubsetSAXFunc externalSubset;
- unsigned int initialized;
- };
- /**
- * xmlExternalEntityLoader:
- * @URL: The System ID of the resource requested
- * @ID: The Public ID of the resource requested
- * @context: the XML parser context
- *
- * External entity loaders types.
- *
- * Returns the entity input parser.
- */
- typedef xmlParserInputPtr (*xmlExternalEntityLoader) (const char *URL,
- const char *ID,
- xmlParserCtxtPtr context);
- #ifdef __cplusplus
- }
- #endif
- #include <libxml/encoding.h>
- #include <libxml/xmlIO.h>
- #include <libxml/globals.h>
- #ifdef __cplusplus
- extern "C" {
- #endif
- /*
- * Init/Cleanup
- */
- XMLPUBFUN void XMLCALL
- xmlInitParser (void);
- XMLPUBFUN void XMLCALL
- xmlCleanupParser (void);
- /*
- * Input functions
- */
- XMLPUBFUN int XMLCALL
- xmlParserInputRead (xmlParserInputPtr in,
- int len);
- XMLPUBFUN int XMLCALL
- xmlParserInputGrow (xmlParserInputPtr in,
- int len);
- /*
- * Basic parsing Interfaces
- */
- #ifdef LIBXML_SAX1_ENABLED
- XMLPUBFUN xmlDocPtr XMLCALL
- xmlParseDoc (const xmlChar *cur);
- XMLPUBFUN xmlDocPtr XMLCALL
- xmlParseFile (const char *filename);
- XMLPUBFUN xmlDocPtr XMLCALL
- xmlParseMemory (const char *buffer,
- int size);
- #endif /* LIBXML_SAX1_ENABLED */
- XMLPUBFUN int XMLCALL
- xmlSubstituteEntitiesDefault(int val);
- XMLPUBFUN int XMLCALL
- xmlKeepBlanksDefault (int val);
- XMLPUBFUN void XMLCALL
- xmlStopParser (xmlParserCtxtPtr ctxt);
- XMLPUBFUN int XMLCALL
- xmlPedanticParserDefault(int val);
- XMLPUBFUN int XMLCALL
- xmlLineNumbersDefault (int val);
- #ifdef LIBXML_SAX1_ENABLED
- /*
- * Recovery mode
- */
- XMLPUBFUN xmlDocPtr XMLCALL
- xmlRecoverDoc (const xmlChar *cur);
- XMLPUBFUN xmlDocPtr XMLCALL
- xmlRecoverMemory (const char *buffer,
- int size);
- XMLPUBFUN xmlDocPtr XMLCALL
- xmlRecoverFile (const char *filename);
- #endif /* LIBXML_SAX1_ENABLED */
- /*
- * Less common routines and SAX interfaces
- */
- XMLPUBFUN int XMLCALL
- xmlParseDocument (xmlParserCtxtPtr ctxt);
- XMLPUBFUN int XMLCALL
- xmlParseExtParsedEnt (xmlParserCtxtPtr ctxt);
- #ifdef LIBXML_SAX1_ENABLED
- XMLPUBFUN int XMLCALL
- xmlSAXUserParseFile (xmlSAXHandlerPtr sax,
- void *user_data,
- const char *filename);
- XMLPUBFUN int XMLCALL
- xmlSAXUserParseMemory (xmlSAXHandlerPtr sax,
- void *user_data,
- const char *buffer,
- int size);
- XMLPUBFUN xmlDocPtr XMLCALL
- xmlSAXParseDoc (xmlSAXHandlerPtr sax,
- const xmlChar *cur,
- int recovery);
- XMLPUBFUN xmlDocPtr XMLCALL
- xmlSAXParseMemory (xmlSAXHandlerPtr sax,
- const char *buffer,
- int size,
- int recovery);
- XMLPUBFUN xmlDocPtr XMLCALL
- xmlSAXParseMemoryWithData (xmlSAXHandlerPtr sax,
- const char *buffer,
- int size,
- int recovery,
- void *data);
- XMLPUBFUN xmlDocPtr XMLCALL
- xmlSAXParseFile (xmlSAXHandlerPtr sax,
- const char *filename,
- int recovery);
- XMLPUBFUN xmlDocPtr XMLCALL
- xmlSAXParseFileWithData (xmlSAXHandlerPtr sax,
- const char *filename,
- int recovery,
- void *data);
- XMLPUBFUN xmlDocPtr XMLCALL
- xmlSAXParseEntity (xmlSAXHandlerPtr sax,
- const char *filename);
- XMLPUBFUN xmlDocPtr XMLCALL
- xmlParseEntity (const char *filename);
- #endif /* LIBXML_SAX1_ENABLED */
- #ifdef LIBXML_VALID_ENABLED
- XMLPUBFUN xmlDtdPtr XMLCALL
- xmlSAXParseDTD (xmlSAXHandlerPtr sax,
- const xmlChar *ExternalID,
- const xmlChar *SystemID);
- XMLPUBFUN xmlDtdPtr XMLCALL
- xmlParseDTD (const xmlChar *ExternalID,
- const xmlChar *SystemID);
- XMLPUBFUN xmlDtdPtr XMLCALL
- xmlIOParseDTD (xmlSAXHandlerPtr sax,
- xmlParserInputBufferPtr input,
- xmlCharEncoding enc);
- #endif /* LIBXML_VALID_ENABLE */
- #ifdef LIBXML_SAX1_ENABLED
- XMLPUBFUN int XMLCALL
- xmlParseBalancedChunkMemory(xmlDocPtr doc,
- xmlSAXHandlerPtr sax,
- void *user_data,
- int depth,
- const xmlChar *string,
- xmlNodePtr *lst);
- #endif /* LIBXML_SAX1_ENABLED */
- XMLPUBFUN xmlParserErrors XMLCALL
- xmlParseInNodeContext (xmlNodePtr node,
- const char *data,
- int datalen,
- int options,
- xmlNodePtr *lst);
- #ifdef LIBXML_SAX1_ENABLED
- XMLPUBFUN int XMLCALL
- xmlParseBalancedChunkMemoryRecover(xmlDocPtr doc,
- xmlSAXHandlerPtr sax,
- void *user_data,
- int depth,
- const xmlChar *string,
- xmlNodePtr *lst,
- int recover);
- XMLPUBFUN int XMLCALL
- xmlParseExternalEntity (xmlDocPtr doc,
- xmlSAXHandlerPtr sax,
- void *user_data,
- int depth,
- const xmlChar *URL,
- const xmlChar *ID,
- xmlNodePtr *lst);
- #endif /* LIBXML_SAX1_ENABLED */
- XMLPUBFUN int XMLCALL
- xmlParseCtxtExternalEntity(xmlParserCtxtPtr ctx,
- const xmlChar *URL,
- const xmlChar *ID,
- xmlNodePtr *lst);
- /*
- * Parser contexts handling.
- */
- XMLPUBFUN xmlParserCtxtPtr XMLCALL
- xmlNewParserCtxt (void);
- XMLPUBFUN int XMLCALL
- xmlInitParserCtxt (xmlParserCtxtPtr ctxt);
- XMLPUBFUN void XMLCALL
- xmlClearParserCtxt (xmlParserCtxtPtr ctxt);
- XMLPUBFUN void XMLCALL
- xmlFreeParserCtxt (xmlParserCtxtPtr ctxt);
- #ifdef LIBXML_SAX1_ENABLED
- XMLPUBFUN void XMLCALL
- xmlSetupParserForBuffer (xmlParserCtxtPtr ctxt,
- const xmlChar* buffer,
- const char *filename);
- #endif /* LIBXML_SAX1_ENABLED */
- XMLPUBFUN xmlParserCtxtPtr XMLCALL
- xmlCreateDocParserCtxt (const xmlChar *cur);
- #ifdef LIBXML_LEGACY_ENABLED
- /*
- * Reading/setting optional parsing features.
- */
- XMLPUBFUN int XMLCALL
- xmlGetFeaturesList (int *len,
- const char **result);
- XMLPUBFUN int XMLCALL
- xmlGetFeature (xmlParserCtxtPtr ctxt,
- const char *name,
- void *result);
- XMLPUBFUN int XMLCALL
- xmlSetFeature (xmlParserCtxtPtr ctxt,
- const char *name,
- void *value);
- #endif /* LIBXML_LEGACY_ENABLED */
- #ifdef LIBXML_PUSH_ENABLED
- /*
- * Interfaces for the Push mode.
- */
- XMLPUBFUN xmlParserCtxtPtr XMLCALL
- xmlCreatePushParserCtxt(xmlSAXHandlerPtr sax,
- void *user_data,
- const char *chunk,
- int size,
- const char *filename);
- XMLPUBFUN int XMLCALL
- xmlParseChunk (xmlParserCtxtPtr ctxt,
- const char *chunk,
- int size,
- int terminate);
- #endif /* LIBXML_PUSH_ENABLED */
- /*
- * Special I/O mode.
- */
- XMLPUBFUN xmlParserCtxtPtr XMLCALL
- xmlCreateIOParserCtxt (xmlSAXHandlerPtr sax,
- void *user_data,
- xmlInputReadCallback ioread,
- xmlInputCloseCallback ioclose,
- void *ioctx,
- xmlCharEncoding enc);
- XMLPUBFUN xmlParserInputPtr XMLCALL
- xmlNewIOInputStream (xmlParserCtxtPtr ctxt,
- xmlParserInputBufferPtr input,
- xmlCharEncoding enc);
- /*
- * Node infos.
- */
- XMLPUBFUN const xmlParserNodeInfo* XMLCALL
- xmlParserFindNodeInfo (const xmlParserCtxtPtr ctxt,
- const xmlNodePtr node);
- XMLPUBFUN void XMLCALL
- xmlInitNodeInfoSeq (xmlParserNodeInfoSeqPtr seq);
- XMLPUBFUN void XMLCALL
- xmlClearNodeInfoSeq (xmlParserNodeInfoSeqPtr seq);
- XMLPUBFUN unsigned long XMLCALL
- xmlParserFindNodeInfoIndex(const xmlParserNodeInfoSeqPtr seq,
- const xmlNodePtr node);
- XMLPUBFUN void XMLCALL
- xmlParserAddNodeInfo (xmlParserCtxtPtr ctxt,
- const xmlParserNodeInfoPtr info);
- /*
- * External entities handling actually implemented in xmlIO.
- */
- XMLPUBFUN void XMLCALL
- xmlSetExternalEntityLoader(xmlExternalEntityLoader f);
- XMLPUBFUN xmlExternalEntityLoader XMLCALL
- xmlGetExternalEntityLoader(void);
- XMLPUBFUN xmlParserInputPtr XMLCALL
- xmlLoadExternalEntity (const char *URL,
- const char *ID,
- xmlParserCtxtPtr ctxt);
- /*
- * Index lookup, actually implemented in the encoding module
- */
- XMLPUBFUN long XMLCALL
- xmlByteConsumed (xmlParserCtxtPtr ctxt);
- /*
- * New set of simpler/more flexible APIs
- */
- /**
- * xmlParserOption:
- *
- * This is the set of XML parser options that can be passed down
- * to the xmlReadDoc() and similar calls.
- */
- typedef enum {
- XML_PARSE_RECOVER = 1<<0, /* recover on errors */
- XML_PARSE_NOENT = 1<<1, /* substitute entities */
- XML_PARSE_DTDLOAD = 1<<2, /* load the external subset */
- XML_PARSE_DTDATTR = 1<<3, /* default DTD attributes */
- XML_PARSE_DTDVALID = 1<<4, /* validate with the DTD */
- XML_PARSE_NOERROR = 1<<5, /* suppress error reports */
- XML_PARSE_NOWARNING = 1<<6, /* suppress warning reports */
- XML_PARSE_PEDANTIC = 1<<7, /* pedantic error reporting */
- XML_PARSE_NOBLANKS = 1<<8, /* remove blank nodes */
- XML_PARSE_SAX1 = 1<<9, /* use the SAX1 interface internally */
- XML_PARSE_XINCLUDE = 1<<10,/* Implement XInclude substitition */
- XML_PARSE_NONET = 1<<11,/* Forbid network access */
- XML_PARSE_NODICT = 1<<12,/* Do not reuse the context dictionnary */
- XML_PARSE_NSCLEAN = 1<<13,/* remove redundant namespaces declarations */
- XML_PARSE_NOCDATA = 1<<14,/* merge CDATA as text nodes */
- XML_PARSE_NOXINCNODE= 1<<15,/* do not generate XINCLUDE START/END nodes */
- XML_PARSE_COMPACT = 1<<16,/* compact small text nodes; no modification of
- the tree allowed afterwards (will possibly
- crash if you try to modify the tree) */
- XML_PARSE_OLD10 = 1<<17,/* parse using XML-1.0 before update 5 */
- XML_PARSE_NOBASEFIX = 1<<18,/* do not fixup XINCLUDE xml:base uris */
- XML_PARSE_HUGE = 1<<19,/* relax any hardcoded limit from the parser */
- XML_PARSE_OLDSAX = 1<<20,/* parse using SAX2 interface before 2.7.0 */
- XML_PARSE_IGNORE_ENC= 1<<21,/* ignore internal document encoding hint */
- XML_PARSE_BIG_LINES = 1<<22 /* Store big lines numbers in text PSVI field */
- } xmlParserOption;
- XMLPUBFUN void XMLCALL
- xmlCtxtReset (xmlParserCtxtPtr ctxt);
- XMLPUBFUN int XMLCALL
- xmlCtxtResetPush (xmlParserCtxtPtr ctxt,
- const char *chunk,
- int size,
- const char *filename,
- const char *encoding);
- XMLPUBFUN int XMLCALL
- xmlCtxtUseOptions (xmlParserCtxtPtr ctxt,
- int options);
- XMLPUBFUN xmlDocPtr XMLCALL
- xmlReadDoc (const xmlChar *cur,
- const char *URL,
- const char *encoding,
- int options);
- XMLPUBFUN xmlDocPtr XMLCALL
- xmlReadFile (const char *URL,
- const char *encoding,
- int options);
- XMLPUBFUN xmlDocPtr XMLCALL
- xmlReadMemory (const char *buffer,
- int size,
- const char *URL,
- const char *encoding,
- int options);
- XMLPUBFUN xmlDocPtr XMLCALL
- xmlReadFd (int fd,
- const char *URL,
- const char *encoding,
- int options);
- XMLPUBFUN xmlDocPtr XMLCALL
- xmlReadIO (xmlInputReadCallback ioread,
- xmlInputCloseCallback ioclose,
- void *ioctx,
- const char *URL,
- const char *encoding,
- int options);
- XMLPUBFUN xmlDocPtr XMLCALL
- xmlCtxtReadDoc (xmlParserCtxtPtr ctxt,
- const xmlChar *cur,
- const char *URL,
- const char *encoding,
- int options);
- XMLPUBFUN xmlDocPtr XMLCALL
- xmlCtxtReadFile (xmlParserCtxtPtr ctxt,
- const char *filename,
- const char *encoding,
- int options);
- XMLPUBFUN xmlDocPtr XMLCALL
- xmlCtxtReadMemory (xmlParserCtxtPtr ctxt,
- const char *buffer,
- int size,
- const char *URL,
- const char *encoding,
- int options);
- XMLPUBFUN xmlDocPtr XMLCALL
- xmlCtxtReadFd (xmlParserCtxtPtr ctxt,
- int fd,
- const char *URL,
- const char *encoding,
- int options);
- XMLPUBFUN xmlDocPtr XMLCALL
- xmlCtxtReadIO (xmlParserCtxtPtr ctxt,
- xmlInputReadCallback ioread,
- xmlInputCloseCallback ioclose,
- void *ioctx,
- const char *URL,
- const char *encoding,
- int options);
- /*
- * Library wide options
- */
- /**
- * xmlFeature:
- *
- * Used to examine the existance of features that can be enabled
- * or disabled at compile-time.
- * They used to be called XML_FEATURE_xxx but this clashed with Expat
- */
- typedef enum {
- XML_WITH_THREAD = 1,
- XML_WITH_TREE = 2,
- XML_WITH_OUTPUT = 3,
- XML_WITH_PUSH = 4,
- XML_WITH_READER = 5,
- XML_WITH_PATTERN = 6,
- XML_WITH_WRITER = 7,
- XML_WITH_SAX1 = 8,
- XML_WITH_FTP = 9,
- XML_WITH_HTTP = 10,
- XML_WITH_VALID = 11,
- XML_WITH_HTML = 12,
- XML_WITH_LEGACY = 13,
- XML_WITH_C14N = 14,
- XML_WITH_CATALOG = 15,
- XML_WITH_XPATH = 16,
- XML_WITH_XPTR = 17,
- XML_WITH_XINCLUDE = 18,
- XML_WITH_ICONV = 19,
- XML_WITH_ISO8859X = 20,
- XML_WITH_UNICODE = 21,
- XML_WITH_REGEXP = 22,
- XML_WITH_AUTOMATA = 23,
- XML_WITH_EXPR = 24,
- XML_WITH_SCHEMAS = 25,
- XML_WITH_SCHEMATRON = 26,
- XML_WITH_MODULES = 27,
- XML_WITH_DEBUG = 28,
- XML_WITH_DEBUG_MEM = 29,
- XML_WITH_DEBUG_RUN = 30,
- XML_WITH_ZLIB = 31,
- XML_WITH_ICU = 32,
- XML_WITH_LZMA = 33,
- XML_WITH_NONE = 99999 /* just to be sure of allocation size */
- } xmlFeature;
- XMLPUBFUN int XMLCALL
- xmlHasFeature (xmlFeature feature);
- #ifdef __cplusplus
- }
- #endif
- #endif /* __XML_PARSER_H__ */
|