Also dddxfxixiasdflkjthed export.

Author: joka921

src/engine/sparqlExpressions/LiteralExpression.h

@@ -50,9 +50,8 @@ class LiteralExpression : public SparqlExpression {
       }
       auto id = context->_qec.getIndex().getId(s);
       IdOrString result =
-          id.has_value()
-              ? IdOrString{id.value()}
-              : IdOrString{std::string{s.toInternalRepresentation()}};
+          id.has_value() ? IdOrString{id.value()}
+                         : IdOrString{std::string{s.toStringRepresentation()}};
       auto ptrForCache = std::make_unique<IdOrString>(result);
       ptrForCache.reset(std::atomic_exchange_explicit(
           &cachedResult_, ptrForCache.release(), std::memory_order_relaxed));
@@ -104,9 +103,9 @@ class LiteralExpression : public SparqlExpression {
     } else if constexpr (std::is_same_v<T, ValueId>) {
       return absl::StrCat("#valueId ", _value.getBits(), "#");
     } else if constexpr (std::is_same_v<T, TripleComponent::Literal>) {
-      return absl::StrCat("#literal: ", _value.toInternalRepresentation());
+      return absl::StrCat("#literal: ", _value.toStringRepresentation());
     } else if constexpr (std::is_same_v<T, TripleComponent::Iri>) {
-      return absl::StrCat("#iri: ", _value.toInternalRepresentation());
+      return absl::StrCat("#iri: ", _value.toStringRepresentation());
     } else if constexpr (std::is_same_v<T, VectorWithMemoryLimit<ValueId>>) {
       // We should never cache this, as objects of this type of expression are
       // used exactly *once* in the HashMap optimization of the GROUP BY

src/index/IndexBuilderTypes.h

@@ -240,7 +240,7 @@ auto getIdMapLambdas(
     // This is not necessary for the actual QLever code, but certain unit tests
     // currently fail without it.
     itemArray[j]->getId(TripleComponent{
-        ad_utility::triple_component::Iri::iriref(LANGUAGE_PREDICATE)});
+        ad_utility::triple_component::Iri::fromIriref(LANGUAGE_PREDICATE)});
   }
   using OptionalIds = std::array<std::optional<std::array<Id, 3>>, 3>;
 
@@ -278,8 +278,9 @@ auto getIdMapLambdas(
         // extra triple <object> ql:language-tag <@language>
         res[2].emplace(std::array<Id, 3>{
             spoIds[2],
-            map.getId(TripleComponent{
-                ad_utility::triple_component::Iri::iriref(LANGUAGE_PREDICATE)}),
+            map.getId(
+                TripleComponent{ad_utility::triple_component::Iri::fromIriref(
+                    LANGUAGE_PREDICATE)}),
             langTagId});
       }
       return res;

src/index/IndexImpl.cpp

@@ -1295,7 +1295,7 @@ std::optional<string> IndexImpl::idToOptionalString(WordVocabIndex id) const {
 std::optional<Id> IndexImpl::getIdImpl(const auto& element) const {
   VocabIndex vocabIndex;
   auto success =
-      getVocab().getId(element.toInternalRepresentation(), &vocabIndex);
+      getVocab().getId(element.toStringRepresentation(), &vocabIndex);
   if (!success) {
     return std::nullopt;
   }

src/parser/Iri.cpp

@@ -25,7 +25,7 @@ NormalizedStringView Iri::getContent() const {
 }
 
 // __________________________________________
-Iri Iri::iriref(std::string_view stringWithBrackets) {
+Iri Iri::fromIriref(std::string_view stringWithBrackets) {
   auto first = stringWithBrackets.find('<');
   AD_CORRECTNESS_CHECK(first != std::string_view::npos);
   return Iri{
@@ -35,19 +35,19 @@ Iri Iri::iriref(std::string_view stringWithBrackets) {
 }
 
 // __________________________________________
-Iri Iri::prefixed(const Iri& prefix, std::string_view suffix) {
+Iri Iri::fromPrefixAndSuffix(const Iri& prefix, std::string_view suffix) {
   auto suffixNormalized = RdfEscaping::unescapePrefixedIri(suffix);
   return Iri{prefix, asNormalizedStringViewUnsafe(suffixNormalized)};
 }
 
 // __________________________________________
-Iri Iri::fromInternalRepresentation(std::string_view s) {
+Iri Iri::fromStringRepresentation(std::string_view s) {
   AD_CORRECTNESS_CHECK(s.starts_with("<") || s.starts_with("@"));
   return Iri{NormalizedString{asNormalizedStringViewUnsafe(s)}};
 }
 
 // __________________________________________
-std::string_view Iri::toInternalRepresentation() const {
+std::string_view Iri::toStringRepresentation() const {
   return asStringViewUnsafe(iri_);
 }
 

src/parser/Iri.h

@@ -36,15 +36,15 @@ class Iri {
                       asStringViewUnsafe(NormalizedStringView(iri.iri_)));
   }
   bool operator==(const Iri&) const = default;
-  static Iri fromInternalRepresentation(std::string_view s);
+  static Iri fromStringRepresentation(std::string_view s);
 
-  std::string_view toInternalRepresentation() const;
+  std::string_view toStringRepresentation() const;
 
   // Create a new iri given an iri with brackets
-  static Iri iriref(std::string_view stringWithBrackets);
+  static Iri fromIriref(std::string_view stringWithBrackets);
 
   // Create a new iri given a prefix iri and its suffix
-  static Iri prefixed(const Iri& prefix, std::string_view suffix);
+  static Iri fromPrefixAndSuffix(const Iri& prefix, std::string_view suffix);
 
   // Return the string value of the iri object without any leading or trailing
   // angled brackets.

src/parser/Literal.cpp

@@ -9,13 +9,14 @@
 
 #include "parser/LiteralOrIri.h"
 
+static constexpr NormalizedChar quote{'"'};
+static constexpr NormalizedChar at{'@'};
+static constexpr NormalizedChar hat{'^'};
+
 namespace ad_utility::triple_component {
 // __________________________________________
 Literal::Literal(NormalizedString content, size_t beginOfSuffix)
     : content_{std::move(content)}, beginOfSuffix_{beginOfSuffix} {
-  NormalizedChar quote{'"'};
-  NormalizedChar at{'@'};
-  NormalizedChar hat{'^'};
   AD_CORRECTNESS_CHECK(content_.starts_with(quote));
   AD_CORRECTNESS_CHECK(beginOfSuffix_ >= 2);
   AD_CORRECTNESS_CHECK(content_[beginOfSuffix_ - 1] == quote);
@@ -25,14 +26,10 @@ Literal::Literal(NormalizedString content, size_t beginOfSuffix)
 }
 
 // __________________________________________
-bool Literal::hasLanguageTag() const {
-  return getSuffix().starts_with(NormalizedChar{'@'});
-}
+bool Literal::hasLanguageTag() const { return getSuffix().starts_with(at); }
 
 // __________________________________________
-bool Literal::hasDatatype() const {
-  return getSuffix().starts_with(NormalizedChar{'^'});
-}
+bool Literal::hasDatatype() const { return getSuffix().starts_with(hat); }
 
 // __________________________________________
 NormalizedStringView Literal::getContent() const {
@@ -60,9 +57,9 @@ NormalizedStringView Literal::getLanguageTag() const {
 }
 
 // __________________________________________
-Literal Literal::literalWithQuotes(
+Literal Literal::fromEscapedRdfLiteral(
     std::string_view rdfContentWithQuotes,
-    std::optional<std::variant<Iri, string>> descriptor) {
+    std::optional<std::variant<Iri, std::string>> descriptor) {
   NormalizedString content =
       RdfEscaping::normalizeLiteralWithQuotes(rdfContentWithQuotes);
 
@@ -107,30 +104,30 @@ Literal Literal::literalWithNormalizedContent(
 
 // __________________________________________
 void Literal::addLanguageTag(std::string_view languageTag) {
-  content_.push_back(NormalizedChar{'@'});
+  content_.push_back(at);
   content_.append(RdfEscaping::normalizeLanguageTag(languageTag));
 }
 
 // __________________________________________
 void Literal::addDatatype(const Iri& datatype) {
   content_.append(asNormalizedStringViewUnsafe("^^"));
   content_.append(
-      asNormalizedStringViewUnsafe(datatype.toInternalRepresentation()));
+      asNormalizedStringViewUnsafe(datatype.toStringRepresentation()));
 }
 
 // __________________________________________
-std::string_view Literal::toInternalRepresentation() const {
+std::string_view Literal::toStringRepresentation() const {
   return asStringViewUnsafe(content_);
 }
 
 // __________________________________________
-Literal Literal::fromInternalRepresentation(std::string_view input) {
+Literal Literal::fromStringRepresentation(std::string_view internal) {
   // TODO<joka921> This is a little dangerous as there might be quotes in the
   // IRI which might lead to unexpected results here.
-  AD_CORRECTNESS_CHECK(input.starts_with('"'));
-  auto endIdx = input.rfind('"');
+  AD_CORRECTNESS_CHECK(internal.starts_with('"'));
+  auto endIdx = internal.rfind('"');
   AD_CORRECTNESS_CHECK(endIdx > 0);
-  return Literal{NormalizedString{asNormalizedStringViewUnsafe(input)},
+  return Literal{NormalizedString{asNormalizedStringViewUnsafe(internal)},
                  endIdx + 1};
 }
 

src/parser/Literal.h

@@ -11,21 +11,28 @@ namespace ad_utility::triple_component {
 // A class to hold literal values.
 class Literal {
  private:
-  // Store the string value of the literal without the surrounding quotation
-  // marks or trailing descriptor.
-  //  "Hello World"@en -> Hello World
+  // Store the normalized version of the literal, including possible datatypes
+  // and descriptors.
+  //  For example `"Hello World"@en`  or `"With"Quote"^^<someDatatype>` (note
+  //  that the quote in the middle is unescaped because this is the normalized
+  //  form that QLever stores.
   NormalizedString content_;
+  // The position after the closing `"`, so either the size of the string, or
+  // the position of the `@` or `^^` for literals with language tags or
+  // datatypes.
   std::size_t beginOfSuffix_;
 
   // Create a new literal without any descriptor
   explicit Literal(NormalizedString content, size_t beginOfSuffix_);
 
-  // Similar to `literalWithQuotes`, except the rdfContent is expected to
+  // Similar to `fromEscapedRdfLiteral`, except the rdfContent is expected to
   // already be normalized
   static Literal literalWithNormalizedContent(
       NormalizedString normalizedRdfContent,
       std::optional<std::variant<Iri, std::string>> descriptor = std::nullopt);
 
+  // Internal helper function. Return either the empty string (for a plain
+  // literal), `@langtag` or `^^<datatypeIri>`.
   NormalizedStringView getSuffix() const {
     NormalizedStringView result = content_;
     result.remove_prefix(beginOfSuffix_);
@@ -39,9 +46,9 @@ class Literal {
   }
   bool operator==(const Literal&) const = default;
 
-  std::string_view toInternalRepresentation() const;
+  std::string_view toStringRepresentation() const;
 
-  static Literal fromInternalRepresentation(std::string_view internal);
+  static Literal fromStringRepresentation(std::string_view internal);
 
   // Return true if the literal has an assigned language tag
   bool hasLanguageTag() const;
@@ -62,8 +69,8 @@ class Literal {
   NormalizedStringView getDatatype() const;
 
   // For documentation, see documentation of function
-  // LiteralORIri::literalWithQuotes
-  static Literal literalWithQuotes(
+  // LiteralORIri::fromEscapedRdfLiteral
+  static Literal fromEscapedRdfLiteral(
       std::string_view rdfContentWithQuotes,
       std::optional<std::variant<Iri, std::string>> descriptor = std::nullopt);
 

src/parser/LiteralOrIri.cpp

@@ -82,21 +82,21 @@ NormalizedStringView LiteralOrIri::getContent() const {
 
 // __________________________________________
 LiteralOrIri LiteralOrIri::iriref(const std::string& stringWithBrackets) {
-  return LiteralOrIri{Iri::iriref(stringWithBrackets)};
+  return LiteralOrIri{Iri::fromIriref(stringWithBrackets)};
 }
 
 // __________________________________________
 LiteralOrIri LiteralOrIri::prefixedIri(const Iri& prefix,
                                        std::string_view suffix) {
-  return LiteralOrIri{Iri::prefixed(prefix, suffix)};
+  return LiteralOrIri{Iri::fromPrefixAndSuffix(prefix, suffix)};
 }
 
 // __________________________________________
 LiteralOrIri LiteralOrIri::literalWithQuotes(
     std::string_view rdfContentWithQuotes,
     std::optional<std::variant<Iri, string>> descriptor) {
-  return LiteralOrIri(
-      Literal::literalWithQuotes(rdfContentWithQuotes, std::move(descriptor)));
+  return LiteralOrIri(Literal::fromEscapedRdfLiteral(rdfContentWithQuotes,
+                                                     std::move(descriptor)));
 }
 
 // __________________________________________

src/parser/LiteralOrIri.h

@@ -39,18 +39,18 @@ class LiteralOrIri {
 
   std::string toInternalRepresentation() const {
     auto impl = [](const auto& val) {
-      return absl::StrCat(val.toInternalRepresentation());
+      return absl::StrCat(val.toStringRepresentation());
     };
     return std::visit(impl, data_);
   }
 
   static LiteralOrIri fromInternalRepresentation(std::string_view internal) {
     char tag = internal.front();
     if (tag == iriPrefixChar) {
-      return LiteralOrIri{Iri::fromInternalRepresentation(internal)};
+      return LiteralOrIri{Iri::fromStringRepresentation(internal)};
     } else {
       AD_CORRECTNESS_CHECK(tag == literalPrefixChar);
-      return LiteralOrIri{Literal::fromInternalRepresentation(internal)};
+      return LiteralOrIri{Literal::fromStringRepresentation(internal)};
     }
   }
   template <typename H>
@@ -109,8 +109,8 @@ class LiteralOrIri {
       std::string_view rdfContentWithQuotes,
       std::optional<std::variant<Iri, std::string>> descriptor = std::nullopt);
 
-  // Similar to `literalWithQuotes`, except the rdfContent is expected to NOT BE
-  // surrounded by quotation marks.
+  // Similar to `fromEscapedRdfLiteral`, except the rdfContent is expected to
+  // NOT BE surrounded by quotation marks.
   static LiteralOrIri literalWithoutQuotes(
       std::string_view rdfContentWithoutQuotes,
       std::optional<std::variant<Iri, std::string>> descriptor = std::nullopt);

src/parser/ParsedQuery.h

@@ -113,7 +113,7 @@ class SparqlTriple : public SparqlTripleBase<PropertyPath> {
     TripleComponent p =
         isVariable(p_._iri)
             ? TripleComponent{Variable{p_._iri}}
-            : TripleComponent(TripleComponent::Iri::iriref(p_._iri));
+            : TripleComponent(TripleComponent::Iri::fromIriref(p_._iri));
     return {s_, p, o_, additionalScanColumns_};
   }
 };