summaryrefslogtreecommitdiff
diff options
context:
space:
mode:
Diffstat
-rw-r--r--src/core/model/Ontology.cpp45
-rw-r--r--src/core/model/Ontology.hpp87
-rw-r--r--src/core/model/Syntax.cpp22
-rw-r--r--src/core/model/Syntax.hpp35
-rw-r--r--test/core/model/OntologyTest.cpp44
5 files changed, 120 insertions, 113 deletions
diff --git a/src/core/model/Ontology.cpp b/src/core/model/Ontology.cpp
index e56d628..bc7b1a7 100644
--- a/src/core/model/Ontology.cpp
+++ b/src/core/model/Ontology.cpp
@@ -315,24 +315,22 @@ bool FieldDescriptor::doValidate(Logger &logger) const
} else {
valid = valid & validateName(logger);
}
- // check start and end token.
- if (!startToken.special && !startToken.token.empty() &&
- !Utils::isUserDefinedToken(startToken.token)) {
+ // check open and close token.
+ if (!openToken.isValid()) {
// TODO: Correct error message.
logger.error(std::string("Field \"") + getNameOrDefaultName() +
"\" of descriptor \"" + parentName +
- "\" has an invalid custom start token: " +
- startToken.token,
+ "\" has an invalid custom open token: " +
+ openToken.token,
*this);
valid = false;
}
- if (!endToken.special && !endToken.token.empty() &&
- !Utils::isUserDefinedToken(endToken.token)) {
+ if (!closeToken.isValid()) {
// TODO: Correct error message.
logger.error(std::string("Field \"") + getNameOrDefaultName() +
"\" of descriptor \"" + parentName +
- "\" has an invalid custom end token: " +
- endToken.token,
+ "\" has an invalid custom close token: " +
+ closeToken.token,
*this);
valid = false;
}
@@ -524,19 +522,17 @@ bool Descriptor::doValidate(Logger &logger) const
}
// check start and end token.
- if (!startToken.special && !startToken.token.empty() &&
- !Utils::isUserDefinedToken(startToken.token)) {
+ if (!openToken.isValid()) {
logger.error(std::string("Descriptor \"") + getName() +
"\" has an invalid custom start token: " +
- startToken.token,
+ openToken.token,
*this);
valid = false;
}
- if (!endToken.special && !endToken.token.empty() &&
- !Utils::isUserDefinedToken(endToken.token)) {
+ if (!closeToken.isValid()) {
logger.error(std::string("Descriptor \"") + getName() +
"\" has an invalid custom end token: " +
- endToken.token,
+ closeToken.token,
*this);
valid = false;
}
@@ -818,8 +814,7 @@ bool StructuredClass::doValidate(Logger &logger) const
}
// check short token.
- if (!shortToken.special && !shortToken.token.empty() &&
- !Utils::isUserDefinedToken(shortToken.token)) {
+ if (!shortToken.isValid()) {
logger.error(std::string("Descriptor \"") + getName() +
"\" has an invalid custom short form token: " +
shortToken.token,
@@ -1092,22 +1087,22 @@ static void gatherTokenDescriptors(
std::unordered_set<FieldDescriptor *> &visited)
{
// add the TokenDescriptors for the Descriptor itself.
- if (!desc->getStartToken().isEmpty()) {
- res.push_back(desc->getStartTokenPointer());
+ if (!desc->getOpenToken().isEmpty()) {
+ res.push_back(desc->getOpenTokenPointer());
}
- if (!desc->getEndToken().isEmpty()) {
- res.push_back(desc->getEndTokenPointer());
+ if (!desc->getCloseToken().isEmpty()) {
+ res.push_back(desc->getCloseTokenPointer());
}
// add the TokenDescriptors for its FieldDescriptors.
for (auto fd : desc->getFieldDescriptors()) {
if (!visited.insert(fd.get()).second) {
continue;
}
- if (!fd->getStartToken().isEmpty()) {
- res.push_back(fd->getStartTokenPointer());
+ if (!fd->getOpenToken().isEmpty()) {
+ res.push_back(fd->getOpenTokenPointer());
}
- if (!fd->getEndToken().isEmpty()) {
- res.push_back(fd->getEndTokenPointer());
+ if (!fd->getCloseToken().isEmpty()) {
+ res.push_back(fd->getCloseTokenPointer());
}
}
}
diff --git a/src/core/model/Ontology.hpp b/src/core/model/Ontology.hpp
index c90093c..079640c 100644
--- a/src/core/model/Ontology.hpp
+++ b/src/core/model/Ontology.hpp
@@ -228,8 +228,8 @@ private:
Owned<Type> primitiveType;
bool optional;
bool primitive;
- TokenDescriptor startToken;
- TokenDescriptor endToken;
+ TokenDescriptor openToken;
+ TokenDescriptor closeToken;
WhitespaceMode whitespaceMode;
protected:
@@ -469,7 +469,7 @@ public:
}
/**
- * Returns a pointer to the start TokenDescriptor. This Token is used as a
+ * Returns a pointer to the open TokenDescriptor. This Token is used as a
* signifier during parsing that an instance of this FieldDescriptor starts.
*
* Note that this does not invalidate the FieldDescriptor. So use with
@@ -477,54 +477,54 @@ public:
*
* @return a pointer to the start TokenDescriptor.
*/
- TokenDescriptor *getStartTokenPointer() { return &startToken; }
+ TokenDescriptor *getOpenTokenPointer() { return &openToken; }
/**
- * Returns a copy of the start TokenDescriptor. This Token is used as a
+ * Returns a copy of the open TokenDescriptor. This Token is used as a
* signifier during parsing that an instance of this FieldDescriptor starts.
*
* @return a copy of the start TokenDescriptor.
*/
- TokenDescriptor getStartToken() const { return startToken; }
+ TokenDescriptor getOpenToken() const { return openToken; }
/**
- * Sets the start TokenDescriptor. This Token is used as a
- * signifier during parsing that an instance of this FieldDescriptor starts.
+ * Sets the open TokenDescriptor. This Token is used as a signifier during
+ * parsing that an instance of this FieldDescriptor starts.
*
- * @param st the new start TokenDescriptor.
+ * @param t the new open TokenDescriptor.
*/
- void setStartToken(TokenDescriptor st)
+ void setOpenToken(TokenDescriptor t)
{
invalidate();
- startToken = st;
+ openToken = t;
}
/**
- * Returns a pointer to the end TokenDescriptor. This Token is used as a
+ * Returns a pointer to the close TokenDescriptor. This Token is used as a
* signifier during parsing that an instance of this FieldDescriptor ends.
*
- * @return a pointer to the end TokenDescriptor.
+ * @return a pointer to the close TokenDescriptor.
*/
- TokenDescriptor *getEndTokenPointer() { return &endToken; }
+ TokenDescriptor *getCloseTokenPointer() { return &closeToken; }
/**
- * Returns a copy of the end TokenDescriptor. This Token is used as a
+ * Returns a copy of the close TokenDescriptor. This Token is used as a
* signifier during parsing that an instance of this FieldDescriptor ends.
*
- * @return a copy of the end TokenDescriptor.
+ * @return a copy of the close TokenDescriptor.
*/
- TokenDescriptor getEndToken() const { return endToken; }
+ TokenDescriptor getCloseToken() const { return closeToken; }
/**
- * Sets the end TokenDescriptor. This Token is used as a
+ * Sets the close TokenDescriptor. This Token is used as a
* signifier during parsing that an instance of this FieldDescriptor ends.
*
- * @param e the new end TokenDescriptor.
+ * @param t the new close TokenDescriptor.
*/
- void setEndToken(TokenDescriptor e)
+ void setCloseToken(TokenDescriptor t)
{
invalidate();
- endToken = e;
+ closeToken = t;
}
/**
@@ -555,7 +555,7 @@ public:
*/
SyntaxDescriptor getSyntaxDescriptor(ssize_t depth = -1)
{
- SyntaxDescriptor stx{startToken.id, endToken.id, Tokens::Empty,
+ SyntaxDescriptor stx{openToken.id, closeToken.id, Tokens::Empty,
const_cast<FieldDescriptor *>(this), depth};
return stx;
}
@@ -608,8 +608,8 @@ class Descriptor : public Node {
private:
Owned<StructType> attributesDescriptor;
NodeVector<FieldDescriptor> fieldDescriptors;
- TokenDescriptor startToken;
- TokenDescriptor endToken;
+ TokenDescriptor openToken;
+ TokenDescriptor closeToken;
bool addAndSortFieldDescriptor(Handle<FieldDescriptor> fd, Logger &logger);
@@ -860,31 +860,31 @@ public:
NodeVector<StructuredClass> getPermittedChildren() const;
/**
- * Returns a pointer to the start TokenDescriptor. This Token is used as a
+ * Returns a pointer to the open TokenDescriptor. This Token is used as a
* signifier during parsing that an instance of this FieldDescriptor starts.
*
- * @return a pointer to the start TokenDescriptor.
+ * @return a pointer to the open TokenDescriptor.
*/
- TokenDescriptor *getStartTokenPointer() { return &startToken; }
+ TokenDescriptor *getOpenTokenPointer() { return &openToken; }
/**
- * Returns a copy of the start TokenDescriptor. This Token is used as a
+ * Returns a copy of the open TokenDescriptor. This Token is used as a
* signifier during parsing that an instance of this FieldDescriptor starts.
*
- * @return a copy of the start TokenDescriptor.
+ * @return a copy of the open TokenDescriptor.
*/
- TokenDescriptor getStartToken() const { return startToken; }
+ TokenDescriptor getOpenToken() const { return openToken; }
/**
- * Sets the start TokenDescriptor. This Token is used as a
+ * Sets the open TokenDescriptor. This Token is used as a
* signifier during parsing that an instance of this FieldDescriptor starts.
*
- * @param st the new start TokenDescriptor.
+ * @param t the new start TokenDescriptor.
*/
- void setStartToken(TokenDescriptor st)
+ void setOpenToken(TokenDescriptor t)
{
invalidate();
- startToken = st;
+ openToken = t;
}
/**
@@ -893,7 +893,7 @@ public:
*
* @return a pointer to the end TokenDescriptor.
*/
- TokenDescriptor *getEndTokenPointer() { return &endToken; }
+ TokenDescriptor *getCloseTokenPointer() { return &closeToken; }
/**
* Returns a copy of the end TokenDescriptor. This Token is used as a
@@ -901,18 +901,18 @@ public:
*
* @return a copy of the end TokenDescriptor.
*/
- TokenDescriptor getEndToken() const { return endToken; }
+ TokenDescriptor getCloseToken() const { return closeToken; }
/**
* Sets the end TokenDescriptor. This Token is used as a
* signifier during parsing that an instance of this FieldDescriptor ends.
*
- * @param e the new end TokenDescriptor.
+ * @param t the new end TokenDescriptor.
*/
- void setEndToken(TokenDescriptor e)
+ void setCloseToken(TokenDescriptor t)
{
invalidate();
- endToken = e;
+ closeToken = t;
}
/**
@@ -922,7 +922,7 @@ public:
*/
virtual SyntaxDescriptor getSyntaxDescriptor(ssize_t depth = -1)
{
- SyntaxDescriptor stx{startToken.id, endToken.id, Tokens::Empty,
+ SyntaxDescriptor stx{openToken.id, closeToken.id, Tokens::Empty,
const_cast<Descriptor *>(this), depth};
return stx;
}
@@ -938,11 +938,6 @@ public:
*/
std::vector<SyntaxDescriptor> getPermittedTokens() const;
};
-/*
- * TODO: We should discuss Cardinalities one more time. Is it smart to define
- * cardinalities independent of context? Should we not have at least have the
- * possibility to define it context-dependently?
- */
/**
* A StructuredClass specifies nodes in the StructureTree of a document that
@@ -1220,7 +1215,7 @@ public:
*/
SyntaxDescriptor getSyntaxDescriptor(ssize_t depth = -1) override
{
- SyntaxDescriptor stx{getStartToken().id, getEndToken().id,
+ SyntaxDescriptor stx{getOpenToken().id, getCloseToken().id,
shortToken.id, const_cast<StructuredClass *>(this),
depth};
return stx;
diff --git a/src/core/model/Syntax.cpp b/src/core/model/Syntax.cpp
index bd17bff..a97acf7 100644
--- a/src/core/model/Syntax.cpp
+++ b/src/core/model/Syntax.cpp
@@ -16,13 +16,21 @@
along with this program. If not, see <http://www.gnu.org/licenses/>.
*/
-#include "Syntax.hpp"
+#include <core/common/Utils.hpp>
#include "Ontology.hpp"
+#include "Syntax.hpp"
namespace ousia {
-/* Class TokenSyntaxDescriptor */
+/* Class TokenDescriptor */
+
+bool TokenDescriptor::isValid() const
+{
+ return special || isEmpty() || Utils::isUserDefinedToken(token);
+}
+
+/* Class SyntaxDescriptor */
bool SyntaxDescriptor::isAnnotation() const
{
@@ -39,11 +47,11 @@ bool SyntaxDescriptor::isStruct() const
void SyntaxDescriptor::insertIntoTokenSet(TokenSet &set) const
{
- if (start != Tokens::Empty) {
- set.insert(start);
+ if (open != Tokens::Empty) {
+ set.insert(open);
}
- if (end != Tokens::Empty) {
- set.insert(end);
+ if (close != Tokens::Empty) {
+ set.insert(close);
}
if (shortForm != Tokens::Empty) {
set.insert(shortForm);
@@ -52,7 +60,7 @@ void SyntaxDescriptor::insertIntoTokenSet(TokenSet &set) const
bool SyntaxDescriptor::isEmpty() const
{
- return start == Tokens::Empty && end == Tokens::Empty &&
+ return open == Tokens::Empty && close == Tokens::Empty &&
shortForm == Tokens::Empty;
}
}
diff --git a/src/core/model/Syntax.hpp b/src/core/model/Syntax.hpp
index 4da3408..4adb329 100644
--- a/src/core/model/Syntax.hpp
+++ b/src/core/model/Syntax.hpp
@@ -74,6 +74,15 @@ struct TokenDescriptor {
* @return true if and only if neither a string nor an ID is given.
*/
bool isEmpty() const { return token.empty() && id == Tokens::Empty; }
+
+ /**
+ * Returns true if the token is valid, which is the case if this class is
+ * either marked as special token or is empty or does have a valid token
+ * string set.
+ *
+ * @return true if the token descriptor is valid, false otherwise.
+ */
+ bool isValid() const;
};
/**
@@ -86,17 +95,17 @@ struct TokenDescriptor {
*/
struct SyntaxDescriptor {
/**
- * Possible start token or Tokens::Empty if no token is set.
+ * Possible open token or Tokens::Empty if no token is set.
*/
- TokenId start;
+ TokenId open;
/**
- * Possible end token or Tokens::Empty if no token is set.
+ * Possible close token or Tokens::Empty if no token is set.
*/
- TokenId end;
+ TokenId close;
/**
- * Possible representation token or Tokens::Empty if no token is set.
+ * Possible short form token or Tokens::Empty if no token is set.
*/
TokenId shortForm;
@@ -118,8 +127,8 @@ struct SyntaxDescriptor {
* descriptor handle to nullptr.
*/
SyntaxDescriptor()
- : start(Tokens::Empty),
- end(Tokens::Empty),
+ : open(Tokens::Empty),
+ close(Tokens::Empty),
shortForm(Tokens::Empty),
descriptor(nullptr),
depth(-1)
@@ -129,18 +138,18 @@ struct SyntaxDescriptor {
/**
* Member initializer constructor.
*
- * @param start is a possible start token.
- * @param end is a possible end token.
+ * @param open is a possible open token.
+ * @param close is a possible close token.
* @param shortForm is a possible short form token.
* @param descriptor The Descriptor this SyntaxDescriptor belongs to.
* @param depth Given the current leaf in the parsed document the depth of a
* SyntaxDescriptor is defined as the number of transparent elements that
* would be needed to construct an instance of the referenced descriptor.
*/
- SyntaxDescriptor(TokenId start, TokenId end, TokenId shortForm,
+ SyntaxDescriptor(TokenId open, TokenId close, TokenId shortForm,
Handle<Node> descriptor, ssize_t depth)
- : start(start),
- end(end),
+ : open(open),
+ close(close),
shortForm(shortForm),
descriptor(descriptor),
depth(depth)
@@ -193,4 +202,4 @@ struct SyntaxDescriptor {
bool isEmpty() const;
};
}
-#endif \ No newline at end of file
+#endif
diff --git a/test/core/model/OntologyTest.cpp b/test/core/model/OntologyTest.cpp
index c6e0596..21893a1 100644
--- a/test/core/model/OntologyTest.cpp
+++ b/test/core/model/OntologyTest.cpp
@@ -530,8 +530,8 @@ TEST(Descriptor, getSyntaxDescriptor)
Rooted<Ontology> ontology{new Ontology(mgr, sys, "ontology")};
Rooted<StructuredClass> A{new StructuredClass(
mgr, "A", ontology, Cardinality::any(), {nullptr}, false, false)};
- A->setStartToken(TokenDescriptor(Tokens::Indent));
- A->setEndToken(TokenDescriptor(Tokens::Dedent));
+ A->setOpenToken(TokenDescriptor(Tokens::Indent));
+ A->setCloseToken(TokenDescriptor(Tokens::Dedent));
{
TokenDescriptor sh{"<+>"};
sh.id = 1;
@@ -539,8 +539,8 @@ TEST(Descriptor, getSyntaxDescriptor)
}
// check the SyntaxDescriptor
SyntaxDescriptor stx = A->getSyntaxDescriptor();
- ASSERT_EQ(Tokens::Indent, stx.start);
- ASSERT_EQ(Tokens::Dedent, stx.end);
+ ASSERT_EQ(Tokens::Indent, stx.open);
+ ASSERT_EQ(Tokens::Dedent, stx.close);
ASSERT_EQ(1, stx.shortForm);
ASSERT_EQ(A, stx.descriptor);
ASSERT_TRUE(stx.isStruct());
@@ -559,8 +559,8 @@ TEST(Descriptor, getPermittedTokens)
// add one StructuredClass with all tokens set.
Rooted<StructuredClass> A{new StructuredClass(
mgr, "A", ontology, Cardinality::any(), {nullptr}, false, false)};
- A->setStartToken(TokenDescriptor(Tokens::Indent));
- A->setEndToken(TokenDescriptor(Tokens::Dedent));
+ A->setOpenToken(TokenDescriptor(Tokens::Indent));
+ A->setCloseToken(TokenDescriptor(Tokens::Dedent));
{
TokenDescriptor sh{"<+>"};
sh.id = 1;
@@ -568,19 +568,19 @@ TEST(Descriptor, getPermittedTokens)
}
// add a field with one token set.
Rooted<FieldDescriptor> A_field = A->createFieldDescriptor(logger).first;
- A_field->setEndToken(TokenDescriptor(Tokens::Newline));
+ A_field->setCloseToken(TokenDescriptor(Tokens::Newline));
A_field->addChild(A);
// add an annotation with start and end set.
Rooted<AnnotationClass> A_anno = ontology->createAnnotationClass("A");
{
TokenDescriptor start{"<"};
start.id = 7;
- A_anno->setStartToken(start);
+ A_anno->setOpenToken(start);
}
{
TokenDescriptor end{">"};
end.id = 8;
- A_anno->setEndToken(end);
+ A_anno->setCloseToken(end);
}
// add a trivial annotation, which should not be returned.
Rooted<AnnotationClass> B_anno = ontology->createAnnotationClass("B");
@@ -592,16 +592,16 @@ TEST(Descriptor, getPermittedTokens)
// the field should be first, because A itself should not be collected
// directly.
ASSERT_EQ(A_field, stxs[0].descriptor);
- ASSERT_EQ(Tokens::Empty, stxs[0].start);
- ASSERT_EQ(Tokens::Newline, stxs[0].end);
+ ASSERT_EQ(Tokens::Empty, stxs[0].open);
+ ASSERT_EQ(Tokens::Newline, stxs[0].close);
ASSERT_EQ(Tokens::Empty, stxs[0].shortForm);
ASSERT_EQ(A, stxs[1].descriptor);
- ASSERT_EQ(Tokens::Indent, stxs[1].start);
- ASSERT_EQ(Tokens::Dedent, stxs[1].end);
+ ASSERT_EQ(Tokens::Indent, stxs[1].open);
+ ASSERT_EQ(Tokens::Dedent, stxs[1].close);
ASSERT_EQ(1, stxs[1].shortForm);
ASSERT_EQ(A_anno, stxs[2].descriptor);
- ASSERT_EQ(7, stxs[2].start);
- ASSERT_EQ(8, stxs[2].end);
+ ASSERT_EQ(7, stxs[2].open);
+ ASSERT_EQ(8, stxs[2].close);
ASSERT_EQ(Tokens::Empty, stxs[2].shortForm);
}
@@ -720,11 +720,11 @@ TEST(Ontology, validate)
ASSERT_EQ(ValidationState::UNKNOWN, ontology->getValidationState());
ASSERT_TRUE(ontology->validate(logger));
// add an invalid start token.
- base_field->setStartToken(TokenDescriptor("< + >"));
+ base_field->setOpenToken(TokenDescriptor("< + >"));
ASSERT_EQ(ValidationState::UNKNOWN, ontology->getValidationState());
ASSERT_FALSE(ontology->validate(logger));
// make it valid.
- base_field->setStartToken(TokenDescriptor("<"));
+ base_field->setOpenToken(TokenDescriptor("<"));
ASSERT_EQ(ValidationState::UNKNOWN, ontology->getValidationState());
ASSERT_TRUE(ontology->validate(logger));
// add a subclass for our base class.
@@ -796,8 +796,8 @@ TEST(Ontology, getAllTokenDescriptors)
// add one StructuredClass with all tokens set.
Rooted<StructuredClass> A{new StructuredClass(
mgr, "A", ontology, Cardinality::any(), {nullptr}, false, false)};
- A->setStartToken(TokenDescriptor(Tokens::Indent));
- A->setEndToken(TokenDescriptor(Tokens::Dedent));
+ A->setOpenToken(TokenDescriptor(Tokens::Indent));
+ A->setCloseToken(TokenDescriptor(Tokens::Dedent));
{
TokenDescriptor sh{"<+>"};
sh.id = 1;
@@ -805,19 +805,19 @@ TEST(Ontology, getAllTokenDescriptors)
}
// add a field with one token set.
Rooted<FieldDescriptor> A_field = A->createFieldDescriptor(logger).first;
- A_field->setEndToken(TokenDescriptor(Tokens::Newline));
+ A_field->setCloseToken(TokenDescriptor(Tokens::Newline));
A_field->addChild(A);
// add an annotation with start and end set.
Rooted<AnnotationClass> A_anno = ontology->createAnnotationClass("A");
{
TokenDescriptor start{"<"};
start.id = 7;
- A_anno->setStartToken(start);
+ A_anno->setOpenToken(start);
}
{
TokenDescriptor end{">"};
end.id = 8;
- A_anno->setEndToken(end);
+ A_anno->setCloseToken(end);
}
// add a trivial annotation, which should not be returned.
Rooted<AnnotationClass> B_anno = ontology->createAnnotationClass("B");
generated by cgit v1.2.3 (git 2.39.1) at 2025-04-30 12:48:50 +0000