expressir 0.2.4 → 0.2.5

data/ext/express-parser/antlr4-upstream/runtime/Cpp/runtime/src/tree/pattern/TextChunk.cpp

@@ -0,0 +1,28 @@
+/* Copyright (c) 2012-2017 The ANTLR Project. All rights reserved.
+ * Use of this file is governed by the BSD 3-clause license that
+ * can be found in the LICENSE.txt file in the project root.
+ */
+
+#include "Exceptions.h"
+
+#include "tree/pattern/TextChunk.h"
+
+using namespace antlr4::tree::pattern;
+
+TextChunk::TextChunk(const std::string &text) : text(text) {
+  if (text == "") {
+    throw IllegalArgumentException("text cannot be nul");
+  }
+
+}
+
+TextChunk::~TextChunk() {
+}
+
+std::string TextChunk::getText() {
+  return text;
+}
+
+std::string TextChunk::toString() {
+  return std::string("'") + text + std::string("'");
+}

data/ext/express-parser/antlr4-upstream/runtime/Cpp/runtime/src/tree/pattern/TextChunk.h

@@ -0,0 +1,51 @@
+/* Copyright (c) 2012-2017 The ANTLR Project. All rights reserved.
+ * Use of this file is governed by the BSD 3-clause license that
+ * can be found in the LICENSE.txt file in the project root.
+ */
+
+#pragma once
+
+#include "Chunk.h"
+
+namespace antlr4 {
+namespace tree {
+namespace pattern {
+
+  /// <summary>
+  /// Represents a span of raw text (concrete syntax) between tags in a tree
+  /// pattern string.
+  /// </summary>
+  class ANTLR4CPP_PUBLIC TextChunk : public Chunk {
+  private:
+    /// <summary>
+    /// This is the backing field for <seealso cref="#getText"/>.
+    /// </summary>
+    const std::string text;
+
+    /// <summary>
+    /// Constructs a new instance of <seealso cref="TextChunk"/> with the specified text.
+    /// </summary>
+    /// <param name="text"> The text of this chunk. </param>
+    /// <exception cref="IllegalArgumentException"> if {@code text} is {@code null}. </exception>
+  public:
+    TextChunk(const std::string &text);
+    virtual ~TextChunk();
+
+    /// <summary>
+    /// Gets the raw text of this chunk.
+    /// </summary>
+    /// <returns> The text of the chunk. </returns>
+    std::string getText();
+
+    /// <summary>
+    /// {@inheritDoc}
+    /// <p/>
+    /// The implementation for <seealso cref="TextChunk"/> returns the result of
+    /// <seealso cref="#getText()"/> in single quotes.
+    /// </summary>
+    virtual std::string toString() override;
+  };
+
+} // namespace pattern
+} // namespace tree
+} // namespace antlr4

data/ext/express-parser/antlr4-upstream/runtime/Cpp/runtime/src/tree/pattern/TokenTagToken.cpp

@@ -0,0 +1,36 @@
+/* Copyright (c) 2012-2017 The ANTLR Project. All rights reserved.
+ * Use of this file is governed by the BSD 3-clause license that
+ * can be found in the LICENSE.txt file in the project root.
+ */
+
+#include "tree/pattern/TokenTagToken.h"
+
+using namespace antlr4::tree::pattern;
+
+TokenTagToken::TokenTagToken(const std::string &/*tokenName*/, int type)
+  : CommonToken(type), tokenName(""), label("") {
+}
+
+TokenTagToken::TokenTagToken(const std::string &tokenName, int type, const std::string &label)
+  : CommonToken(type), tokenName(tokenName), label(label) {
+}
+
+std::string TokenTagToken::getTokenName() const {
+  return tokenName;
+}
+
+std::string TokenTagToken::getLabel() const {
+  return label;
+}
+
+std::string TokenTagToken::getText() const {
+  if (!label.empty()) {
+    return "<" + label + ":" + tokenName + ">";
+  }
+
+  return "<" + tokenName + ">";
+}
+
+std::string TokenTagToken::toString() const {
+  return tokenName + ":" + std::to_string(_type);
+}

data/ext/express-parser/antlr4-upstream/runtime/Cpp/runtime/src/tree/pattern/TokenTagToken.h

@@ -0,0 +1,80 @@
+/* Copyright (c) 2012-2017 The ANTLR Project. All rights reserved.
+ * Use of this file is governed by the BSD 3-clause license that
+ * can be found in the LICENSE.txt file in the project root.
+ */
+
+#pragma once
+
+#include "CommonToken.h"
+
+namespace antlr4 {
+namespace tree {
+namespace pattern {
+
+  /// <summary>
+  /// A <seealso cref="Token"/> object representing a token of a particular type; e.g.,
+  /// {@code <ID>}. These tokens are created for <seealso cref="TagChunk"/> chunks where the
+  /// tag corresponds to a lexer rule or token type.
+  /// </summary>
+  class ANTLR4CPP_PUBLIC TokenTagToken : public CommonToken {
+    /// <summary>
+    /// This is the backing field for <seealso cref="#getTokenName"/>.
+    /// </summary>
+  private:
+    const std::string tokenName;
+    /// <summary>
+    /// This is the backing field for <seealso cref="#getLabe"/>.
+    /// </summary>
+    const std::string label;
+
+    /// <summary>
+    /// Constructs a new instance of <seealso cref="TokenTagToken"/> for an unlabeled tag
+    /// with the specified token name and type.
+    /// </summary>
+    /// <param name="tokenName"> The token name. </param>
+    /// <param name="type"> The token type. </param>
+  public:
+    TokenTagToken(const std::string &tokenName, int type); //this(tokenName, type, nullptr);
+
+    /// <summary>
+    /// Constructs a new instance of <seealso cref="TokenTagToken"/> with the specified
+    /// token name, type, and label.
+    /// </summary>
+    /// <param name="tokenName"> The token name. </param>
+    /// <param name="type"> The token type. </param>
+    /// <param name="label"> The label associated with the token tag, or {@code null} if
+    /// the token tag is unlabeled. </param>
+    TokenTagToken(const std::string &tokenName, int type, const std::string &label);
+
+    /// <summary>
+    /// Gets the token name. </summary>
+    /// <returns> The token name. </returns>
+    std::string getTokenName() const;
+
+    /// <summary>
+    /// Gets the label associated with the rule tag.
+    /// </summary>
+    /// <returns> The name of the label associated with the rule tag, or
+    /// {@code null} if this is an unlabeled rule tag. </returns>
+    std::string getLabel() const;
+
+    /// <summary>
+    /// {@inheritDoc}
+    /// <p/>
+    /// The implementation for <seealso cref="TokenTagToken"/> returns the token tag
+    /// formatted with {@code <} and {@code >} delimiters.
+    /// </summary>
+    virtual std::string getText() const override;
+
+    /// <summary>
+    /// {@inheritDoc}
+    /// <p/>
+    /// The implementation for <seealso cref="TokenTagToken"/> returns a string of the form
+    /// {@code tokenName:type}.
+    /// </summary>
+    virtual std::string toString() const override;
+  };
+
+} // namespace pattern
+} // namespace tree
+} // namespace antlr4

data/ext/express-parser/antlr4-upstream/runtime/Cpp/runtime/src/tree/xpath/XPath.cpp

@@ -0,0 +1,154 @@
+/* Copyright (c) 2012-2017 The ANTLR Project. All rights reserved.
+ * Use of this file is governed by the BSD 3-clause license that
+ * can be found in the LICENSE.txt file in the project root.
+ */
+
+#include "XPathLexer.h"
+#include "XPathLexerErrorListener.h"
+#include "XPathElement.h"
+#include "XPathWildcardAnywhereElement.h"
+#include "XPathWildcardElement.h"
+#include "XPathTokenAnywhereElement.h"
+#include "XPathTokenElement.h"
+#include "XPathRuleAnywhereElement.h"
+#include "XPathRuleElement.h"
+
+#include "XPath.h"
+
+using namespace antlr4;
+using namespace antlr4::tree;
+using namespace antlr4::tree::xpath;
+
+const std::string XPath::WILDCARD = "*";
+const std::string XPath::NOT = "!";
+
+XPath::XPath(Parser *parser, const std::string &path) {
+  _parser = parser;
+  _path = path;
+}
+
+std::vector<std::unique_ptr<XPathElement>> XPath::split(const std::string &path) {
+  ANTLRInputStream in(path);
+  XPathLexer lexer(&in);
+  lexer.removeErrorListeners();
+  XPathLexerErrorListener listener;
+  lexer.addErrorListener(&listener);
+  CommonTokenStream tokenStream(&lexer);
+  try {
+    tokenStream.fill();
+  } catch (LexerNoViableAltException &) {
+    size_t pos = lexer.getCharPositionInLine();
+    std::string msg = "Invalid tokens or characters at index " + std::to_string(pos) + " in path '" + path + "'";
+    throw IllegalArgumentException(msg);
+  }
+
+  std::vector<Token *> tokens = tokenStream.getTokens();
+  std::vector<std::unique_ptr<XPathElement>> elements;
+  size_t n = tokens.size();
+  size_t i = 0;
+  bool done = false;
+  while (!done && i < n) {
+    Token *el = tokens[i];
+    Token *next = nullptr;
+    switch (el->getType()) {
+      case XPathLexer::ROOT:
+      case XPathLexer::ANYWHERE: {
+        bool anywhere = el->getType() == XPathLexer::ANYWHERE;
+        i++;
+        next = tokens[i];
+        bool invert = next->getType() == XPathLexer::BANG;
+        if (invert) {
+          i++;
+          next = tokens[i];
+        }
+        std::unique_ptr<XPathElement> pathElement = getXPathElement(next, anywhere);
+        pathElement->setInvert(invert);
+        elements.push_back(std::move(pathElement));
+        i++;
+        break;
+
+      }
+      case XPathLexer::TOKEN_REF:
+      case XPathLexer::RULE_REF:
+      case XPathLexer::WILDCARD:
+        elements.push_back(getXPathElement(el, false));
+        i++;
+        break;
+
+      case Token::EOF:
+        done = true;
+        break;
+
+      default :
+        throw IllegalArgumentException("Unknown path element " + el->toString());
+    }
+  }
+
+  return elements;
+}
+
+std::unique_ptr<XPathElement> XPath::getXPathElement(Token *wordToken, bool anywhere) {
+  if (wordToken->getType() == Token::EOF) {
+    throw IllegalArgumentException("Missing path element at end of path");
+  }
+
+  std::string word = wordToken->getText();
+  size_t ttype = _parser->getTokenType(word);
+  ssize_t ruleIndex = _parser->getRuleIndex(word);
+  switch (wordToken->getType()) {
+    case XPathLexer::WILDCARD :
+      if (anywhere)
+        return std::unique_ptr<XPathWildcardAnywhereElement>(new XPathWildcardAnywhereElement());
+      return std::unique_ptr<XPathWildcardElement>(new XPathWildcardElement());
+
+    case XPathLexer::TOKEN_REF:
+    case XPathLexer::STRING :
+      if (ttype == Token::INVALID_TYPE) {
+        throw IllegalArgumentException(word + " at index " + std::to_string(wordToken->getStartIndex()) + " isn't a valid token name");
+      }
+      if (anywhere)
+        return std::unique_ptr<XPathTokenAnywhereElement>(new XPathTokenAnywhereElement(word, (int)ttype));
+      return std::unique_ptr<XPathTokenElement>(new XPathTokenElement(word, (int)ttype));
+
+    default :
+      if (ruleIndex == -1) {
+        throw IllegalArgumentException(word + " at index " + std::to_string(wordToken->getStartIndex()) + " isn't a valid rule name");
+      }
+      if (anywhere)
+        return std::unique_ptr<XPathRuleAnywhereElement>(new XPathRuleAnywhereElement(word, (int)ruleIndex));
+      return std::unique_ptr<XPathRuleElement>(new XPathRuleElement(word, (int)ruleIndex));
+  }
+}
+
+static ParserRuleContext dummyRoot;
+
+std::vector<ParseTree *> XPath::findAll(ParseTree *tree, std::string const& xpath, Parser *parser) {
+  XPath p(parser, xpath);
+  return p.evaluate(tree);
+}
+
+std::vector<ParseTree *> XPath::evaluate(ParseTree *t) {
+  dummyRoot.children = { t }; // don't set t's parent.
+
+  std::vector<ParseTree *> work = { &dummyRoot };
+
+  size_t i = 0;
+  std::vector<std::unique_ptr<XPathElement>> elements = split(_path);
+
+  while (i < elements.size()) {
+    std::vector<ParseTree *> next;
+    for (auto node : work) {
+      if (!node->children.empty()) {
+        // only try to match next element if it has children
+        // e.g., //func/*/stat might have a token node for which
+        // we can't go looking for stat nodes.
+        auto matching = elements[i]->evaluate(node);
+        next.insert(next.end(), matching.begin(), matching.end());
+      }
+    }
+    i++;
+    work = next;
+  }
+
+  return work;
+}

data/ext/express-parser/antlr4-upstream/runtime/Cpp/runtime/src/tree/xpath/XPath.h

@@ -0,0 +1,86 @@
+/* Copyright (c) 2012-2017 The ANTLR Project. All rights reserved.
+ * Use of this file is governed by the BSD 3-clause license that
+ * can be found in the LICENSE.txt file in the project root.
+ */
+
+#pragma once
+
+#include "antlr4-common.h"
+
+namespace antlr4 {
+namespace tree {
+namespace xpath {
+
+  /// Represent a subset of XPath XML path syntax for use in identifying nodes in
+  /// parse trees.
+  ///
+  /// <para>
+  /// Split path into words and separators {@code /} and {@code //} via ANTLR
+  /// itself then walk path elements from left to right. At each separator-word
+  /// pair, find set of nodes. Next stage uses those as work list.</para>
+  ///
+  /// <para>
+  /// The basic interface is
+  /// <seealso cref="XPath#findAll ParseTree.findAll"/>{@code (tree, pathString, parser)}.
+  /// But that is just shorthand for:</para>
+  ///
+  /// <pre>
+  /// <seealso cref="XPath"/> p = new <seealso cref="XPath#XPath XPath"/>(parser, pathString);
+  /// return p.<seealso cref="#evaluate evaluate"/>(tree);
+  /// </pre>
+  ///
+  /// <para>
+  /// See {@code org.antlr.v4.test.TestXPath} for descriptions. In short, this
+  /// allows operators:</para>
+  ///
+  /// <dl>
+  /// <dt>/</dt> <dd>root</dd>
+  /// <dt>//</dt> <dd>anywhere</dd>
+  /// <dt>!</dt> <dd>invert; this must appear directly after root or anywhere
+  /// operator</dd>
+  /// </dl>
+  ///
+  /// <para>
+  /// and path elements:</para>
+  ///
+  /// <dl>
+  /// <dt>ID</dt> <dd>token name</dd>
+  /// <dt>'string'</dt> <dd>any string literal token from the grammar</dd>
+  /// <dt>expr</dt> <dd>rule name</dd>
+  /// <dt>*</dt> <dd>wildcard matching any node</dd>
+  /// </dl>
+  ///
+  /// <para>
+  /// Whitespace is not allowed.</para>
+
+  class ANTLR4CPP_PUBLIC XPath {
+  public:
+    static const std::string WILDCARD; // word not operator/separator
+    static const std::string NOT; // word for invert operator
+
+    XPath(Parser *parser, const std::string &path);
+    virtual ~XPath() {}
+
+    // TODO: check for invalid token/rule names, bad syntax
+    virtual std::vector<std::unique_ptr<XPathElement>> split(const std::string &path);
+
+    static std::vector<ParseTree *> findAll(ParseTree *tree, std::string const& xpath, Parser *parser);
+
+    /// Return a list of all nodes starting at {@code t} as root that satisfy the
+    /// path. The root {@code /} is relative to the node passed to
+    /// <seealso cref="#evaluate"/>.
+    virtual std::vector<ParseTree *> evaluate(ParseTree *t);
+
+  protected:
+    std::string _path;
+    Parser *_parser;
+
+    /// Convert word like {@code *} or {@code ID} or {@code expr} to a path
+    /// element. {@code anywhere} is {@code true} if {@code //} precedes the
+    /// word.
+    virtual std::unique_ptr<XPathElement> getXPathElement(Token *wordToken, bool anywhere);
+  };
+
+} // namespace xpath
+} // namespace tree
+} // namespace antlr4

data/ext/express-parser/antlr4-upstream/runtime/Cpp/runtime/src/tree/xpath/XPathElement.cpp

@@ -0,0 +1,31 @@
+/* Copyright (c) 2012-2017 The ANTLR Project. All rights reserved.
+ * Use of this file is governed by the BSD 3-clause license that
+ * can be found in the LICENSE.txt file in the project root.
+ */
+
+#include "support/CPPUtils.h"
+
+#include "XPathElement.h"
+
+using namespace antlr4::tree;
+using namespace antlr4::tree::xpath;
+
+XPathElement::XPathElement(const std::string &nodeName) {
+  _nodeName = nodeName;
+}
+
+XPathElement::~XPathElement() {
+}
+
+std::vector<ParseTree *> XPathElement::evaluate(ParseTree * /*t*/) {
+  return {};
+}
+
+std::string XPathElement::toString() const {
+  std::string inv = _invert ? "!" : "";
+  return antlrcpp::toString(*this) + "[" + inv + _nodeName + "]";
+}
+
+void XPathElement::setInvert(bool value) {
+  _invert = value;
+}