#include <serialize.hh>
Public Member Functions | |
| template<typename Base , typename T > | |
| void | serializeBase (T &t) |
| Is this archive a loader or a saver. More... | |
| template<typename Base , typename T > | |
| void | serializeInlinedBase (T &t, unsigned version) |
| Serialize the base class of this classtype. More... | |
| bool | needVersion () const |
| Does this archive store version information. More... | |
| bool | isReverseSnapshot () const |
| Is this a reverse-snapshot? More... | |
| bool | translateEnumToString () const |
| Does this archive store enums as strings. More... | |
| template<typename T > | |
| void | attribute (const char *name, T &t) |
| Load/store an attribute from/in the archive. More... | |
| void | attribute (const char *name, const char *value) |
| bool | canHaveOptionalAttributes () const |
| Some archives (like XML archives) can store optional attributes. More... | |
| bool | hasAttribute (const char *) |
| Check the presence of a (optional) attribute. More... | |
| bool | findAttribute (const char *, unsigned &) |
| Optimization: combination of hasAttribute() and getAttribute(). More... | |
| bool | canCountChildren () const |
| Some archives (like XML archives) can count the number of subtags that belong to the current tag. More... | |
| int | countChildren () const |
| Count the number of child tags. More... | |
| void | beginTag (const char *) |
| Indicate begin of a tag. More... | |
| void | endTag (const char *) |
| Indicate begin of a tag. More... | |
Protected Member Functions | |
| Derived & | self () |
| Returns a reference to the most derived class. More... | |
Detailed Description
template<typename Derived>
class openmsx::ArchiveBase< Derived >
Definition at line 65 of file serialize.hh.
Member Function Documentation
◆ attribute() [1/2]
|
inline |
Load/store an attribute from/in the archive.
Depending on the underlying concrete stream, attributes are either stored like XML attributes or as regular values. Because of this (and thus unlike XML attributes) the order of attributes matter. It also matters whether an attribute is present or not.
Definition at line 212 of file serialize.hh.
Referenced by openmsx::ArchiveBase< XmlInputArchive >::attribute(), openmsx::XmlOutputArchive::attribute(), openmsx::XmlInputArchive::attribute(), openmsx::XmlOutputArchive::attributeImpl(), and openmsx::XmlInputArchive::attributeImpl().
◆ attribute() [2/2]
| void openmsx::ArchiveBase< Derived >::attribute | ( | const char * | name, |
| const char * | value | ||
| ) |
Definition at line 23 of file serialize.cc.
◆ beginTag()
|
inline |
Indicate begin of a tag.
Only XML archives use this, other archives ignore it. XML saver uses it as a name for the current tag, it doesn't interpret the name in any way. XML loader uses it only as a check: it checks whether the current tag name matches the given name. So we will NOT search the tag with the given name, the tags have to be in the correct order.
Definition at line 269 of file serialize.hh.
Referenced by openmsx::XmlOutputArchive::canCountChildren(), openmsx::XmlInputArchive::canCountChildren(), openmsx::InputArchiveBase< XmlInputArchive >::doSerialize(), openmsx::OutputArchiveBase< MemOutputArchive >::serialize(), openmsx::InputArchiveBase< XmlInputArchive >::serialize(), openmsx::OutputArchiveBase< MemOutputArchive >::serializeChar(), openmsx::InputArchiveBase< XmlInputArchive >::serializeChar(), openmsx::OutputArchiveBase< MemOutputArchive >::serializePointerID(), openmsx::InputArchiveBase< XmlInputArchive >::serializePointerID(), and openmsx::OutputArchiveBase< MemOutputArchive >::serializeWithID().
◆ canCountChildren()
|
inline |
Some archives (like XML archives) can count the number of subtags that belong to the current tag.
This method indicates whether that's the case for this archive or not. This can for example be used to make the XML files look prettier in case of serialization of collections: in that case we don't need to explictly store the size of the collection, it can be derived from the number of subtags.
Definition at line 250 of file serialize.hh.
◆ canHaveOptionalAttributes()
|
inline |
Some archives (like XML archives) can store optional attributes.
This method indicates whether that's the case or not. This can be used to for example in XML files don't store attributes with default values (thus to make the XML look prettier).
Definition at line 223 of file serialize.hh.
◆ countChildren()
|
inline |
Count the number of child tags.
It's only allowed to call this method on archives that have support for this operation.
Definition at line 256 of file serialize.hh.
Referenced by openmsx::XmlInputArchive::attribute().
◆ endTag()
|
inline |
Indicate begin of a tag.
Only XML archives use this, other archives ignore it. Both XML loader and saver only use the given tag name to do some internal checks (with checks disabled, the tag parameter has no influence at all on loading or saving of the stream).
Definition at line 279 of file serialize.hh.
Referenced by openmsx::XmlOutputArchive::canCountChildren(), openmsx::XmlInputArchive::canCountChildren(), openmsx::InputArchiveBase< XmlInputArchive >::doSerialize(), openmsx::OutputArchiveBase< MemOutputArchive >::serialize(), openmsx::InputArchiveBase< XmlInputArchive >::serialize(), openmsx::OutputArchiveBase< MemOutputArchive >::serializeChar(), openmsx::InputArchiveBase< XmlInputArchive >::serializeChar(), openmsx::OutputArchiveBase< MemOutputArchive >::serializePointerID(), openmsx::InputArchiveBase< XmlInputArchive >::serializePointerID(), and openmsx::OutputArchiveBase< MemOutputArchive >::serializeWithID().
◆ findAttribute()
|
inline |
Optimization: combination of hasAttribute() and getAttribute().
Returns true if hasAttribute() and (if so) also fills in the value.
Definition at line 237 of file serialize.hh.
Referenced by openmsx::XmlInputArchive::attribute().
◆ hasAttribute()
|
inline |
Check the presence of a (optional) attribute.
It's only allowed to call this method on archives that can have optional attributes.
Definition at line 229 of file serialize.hh.
Referenced by openmsx::XmlInputArchive::attribute().
◆ isReverseSnapshot()
|
inline |
Is this a reverse-snapshot?
Definition at line 199 of file serialize.hh.
◆ needVersion()
|
inline |
Does this archive store version information.
Definition at line 196 of file serialize.hh.
◆ self()
|
inlineprotected |
Returns a reference to the most derived class.
Helper function to implement static polymorphism.
Definition at line 322 of file serialize.hh.
◆ serializeBase()
|
inline |
Is this archive a loader or a saver.
bool isLoader() const; Serialize the base class of this classtype. Should preferably be called as the first statement in the implementation of a serialize() method of a class type. See also serializeInlinedBase() below.
Definition at line 77 of file serialize.hh.
◆ serializeInlinedBase()
|
inline |
Serialize the base class of this classtype.
Should preferably be called as the first statement in the implementation of a serialize() method of a class type. See also serializeBase() above.
The differece between serializeBase() and serializeInlinedBase() is only relevant for versioned archives (see needVersion(), e.g. XML archives). In XML archives serializeBase() will put the base class in a new subtag, serializeInlinedBase() puts the members of the base class (inline) in the current tag. The advantage of serializeBase() is that the base class can be versioned seperatly from the subclass. The disadvantage is that it exposes an internal implementation detail in the XML file, and thus makes it harder to for example change the class hierarchy or move members from base to subclass or vice-versa.
Definition at line 100 of file serialize.hh.
◆ translateEnumToString()
|
inline |
Does this archive store enums as strings.
See also struct serialize_as_enum.
Definition at line 204 of file serialize.hh.
The documentation for this class was generated from the following files:
