python3-parser 1.0.0

data/ext/python3-parser/antlr4-upstream/runtime/Cpp/runtime/src/UnbufferedCharStream.h

@@ -0,0 +1,123 @@
+/* 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 "CharStream.h"
+
+namespace antlr4 {
+
+  /// Do not buffer up the entire char stream. It does keep a small buffer
+  /// for efficiency and also buffers while a mark exists (set by the
+  /// lookahead prediction in parser). "Unbuffered" here refers to fact
+  /// that it doesn't buffer all data, not that's it's on demand loading of char.
+  class ANTLR4CPP_PUBLIC UnbufferedCharStream : public CharStream {
+  public:
+    /// The name or source of this char stream.
+    std::string name;
+
+    UnbufferedCharStream(std::wistream &input);
+
+    virtual void consume() override;
+    virtual size_t LA(ssize_t i) override;
+
+    /// <summary>
+    /// Return a marker that we can release later.
+    /// <p/>
+    /// The specific marker value used for this class allows for some level of
+    /// protection against misuse where {@code seek()} is called on a mark or
+    /// {@code release()} is called in the wrong order.
+    /// </summary>
+    virtual ssize_t mark() override;
+
+    /// <summary>
+    /// Decrement number of markers, resetting buffer if we hit 0. </summary>
+    /// <param name="marker"> </param>
+    virtual void release(ssize_t marker) override;
+    virtual size_t index() override;
+
+    /// <summary>
+    /// Seek to absolute character index, which might not be in the current
+    ///  sliding window.  Move {@code p} to {@code index-bufferStartIndex}.
+    /// </summary>
+    virtual void seek(size_t index) override;
+    virtual size_t size() override;
+    virtual std::string getSourceName() const override;
+    virtual std::string getText(const misc::Interval &interval) override;
+
+  protected:
+    /// A moving window buffer of the data being scanned. While there's a marker,
+    /// we keep adding to buffer. Otherwise, <seealso cref="#consume consume()"/> resets so
+    /// we start filling at index 0 again.
+    // UTF-32 encoded.
+#if defined(_MSC_VER) && _MSC_VER == 1900
+    i32string _data; // Custom type for VS 2015.
+    typedef __int32 storage_type;
+#else
+    std::u32string _data;
+    typedef char32_t storage_type;
+#endif
+
+    /// <summary>
+    /// 0..n-1 index into <seealso cref="#data data"/> of next character.
+    /// <p/>
+    /// The {@code LA(1)} character is {@code data[p]}. If {@code p == n}, we are
+    /// out of buffered characters.
+    /// </summary>
+    size_t _p;
+
+    /// <summary>
+    /// Count up with <seealso cref="#mark mark()"/> and down with
+    /// <seealso cref="#release release()"/>. When we {@code release()} the last mark,
+    /// {@code numMarkers} reaches 0 and we reset the buffer. Copy
+    /// {@code data[p]..data[n-1]} to {@code data[0]..data[(n-1)-p]}.
+    /// </summary>
+    size_t _numMarkers;
+
+    /// This is the {@code LA(-1)} character for the current position.
+    size_t _lastChar; // UTF-32
+
+    /// <summary>
+    /// When {@code numMarkers > 0}, this is the {@code LA(-1)} character for the
+    /// first character in <seealso cref="#data data"/>. Otherwise, this is unspecified.
+    /// </summary>
+    size_t _lastCharBufferStart; // UTF-32
+
+    /// <summary>
+    /// Absolute character index. It's the index of the character about to be
+    /// read via {@code LA(1)}. Goes from 0 to the number of characters in the
+    /// entire stream, although the stream size is unknown before the end is
+    /// reached.
+    /// </summary>
+    size_t _currentCharIndex;
+
+    std::wistream &_input;
+
+    /// <summary>
+    /// Make sure we have 'want' elements from current position <seealso cref="#p p"/>.
+    /// Last valid {@code p} index is {@code data.length-1}. {@code p+need-1} is
+    /// the char index 'need' elements ahead. If we need 1 element,
+    /// {@code (p+1-1)==p} must be less than {@code data.length}.
+    /// </summary>
+    virtual void sync(size_t want);
+
+    /// <summary>
+    /// Add {@code n} characters to the buffer. Returns the number of characters
+    /// actually added to the buffer. If the return value is less than {@code n},
+    /// then EOF was reached before {@code n} characters could be added.
+    /// </summary>
+    virtual size_t fill(size_t n);
+
+    /// Override to provide different source of characters than
+    /// <seealso cref="#input input"/>.
+    virtual char32_t nextChar();
+    virtual void add(char32_t c);
+    size_t getBufferStartIndex() const;
+
+  private:
+    void InitializeInstanceFields();
+  };
+
+} // namespace antlr4

data/ext/python3-parser/antlr4-upstream/runtime/Cpp/runtime/src/UnbufferedTokenStream.cpp

@@ -0,0 +1,270 @@
+/* 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 "Token.h"
+#include "Exceptions.h"
+#include "assert.h"
+#include "TokenSource.h"
+#include "support/Arrays.h"
+#include "misc/Interval.h"
+#include "RuleContext.h"
+#include "WritableToken.h"
+
+#include "UnbufferedTokenStream.h"
+
+using namespace antlr4;
+
+UnbufferedTokenStream::UnbufferedTokenStream(TokenSource *tokenSource) : UnbufferedTokenStream(tokenSource, 256) {
+}
+
+UnbufferedTokenStream::UnbufferedTokenStream(TokenSource *tokenSource, int /*bufferSize*/)
+  : _tokenSource(tokenSource), _lastToken(nullptr), _lastTokenBufferStart(nullptr)
+{
+  InitializeInstanceFields();
+  fill(1); // prime the pump
+}
+
+UnbufferedTokenStream::~UnbufferedTokenStream() {
+}
+
+Token* UnbufferedTokenStream::get(size_t i) const
+{ // get absolute index
+  size_t bufferStartIndex = getBufferStartIndex();
+  if (i < bufferStartIndex || i >= bufferStartIndex + _tokens.size()) {
+    throw IndexOutOfBoundsException(std::string("get(") + std::to_string(i) + std::string(") outside buffer: ")
+      + std::to_string(bufferStartIndex) + std::string("..") + std::to_string(bufferStartIndex + _tokens.size()));
+  }
+  return _tokens[i - bufferStartIndex].get();
+}
+
+Token* UnbufferedTokenStream::LT(ssize_t i)
+{
+  if (i == -1) {
+    return _lastToken;
+  }
+
+  sync(i);
+  ssize_t index = static_cast<ssize_t>(_p) + i - 1;
+  if (index < 0) {
+    throw IndexOutOfBoundsException(std::string("LT(") + std::to_string(i) + std::string(") gives negative index"));
+  }
+
+  if (index >= static_cast<ssize_t>(_tokens.size())) {
+    assert(_tokens.size() > 0 && _tokens.back()->getType() == EOF);
+    return _tokens.back().get();
+  }
+
+  return _tokens[static_cast<size_t>(index)].get();
+}
+
+size_t UnbufferedTokenStream::LA(ssize_t i)
+{
+  return LT(i)->getType();
+}
+
+TokenSource* UnbufferedTokenStream::getTokenSource() const
+{
+  return _tokenSource;
+}
+
+std::string UnbufferedTokenStream::getText()
+{
+  return "";
+}
+
+std::string UnbufferedTokenStream::getText(RuleContext* ctx)
+{
+  return getText(ctx->getSourceInterval());
+}
+
+std::string UnbufferedTokenStream::getText(Token *start, Token *stop)
+{
+  return getText(misc::Interval(start->getTokenIndex(), stop->getTokenIndex()));
+}
+
+void UnbufferedTokenStream::consume()
+{
+  if (LA(1) == EOF) {
+    throw IllegalStateException("cannot consume EOF");
+  }
+
+  // buf always has at least tokens[p==0] in this method due to ctor
+  _lastToken = _tokens[_p].get(); // track last token for LT(-1)
+
+  // if we're at last token and no markers, opportunity to flush buffer
+  if (_p == _tokens.size() - 1 && _numMarkers == 0) {
+    _tokens.clear();
+    _p = 0;
+    _lastTokenBufferStart = _lastToken;
+  } else {
+    ++_p;
+  }
+
+  ++_currentTokenIndex;
+  sync(1);
+}
+
+/// <summary>
+/// Make sure we have 'need' elements from current position <seealso cref="#p p"/>. Last valid
+///  {@code p} index is {@code tokens.length-1}.  {@code p+need-1} is the tokens index 'need' elements
+///  ahead.  If we need 1 element, {@code (p+1-1)==p} must be less than {@code tokens.length}.
+/// </summary>
+void UnbufferedTokenStream::sync(ssize_t want)
+{
+  ssize_t need = (static_cast<ssize_t>(_p) + want - 1) - static_cast<ssize_t>(_tokens.size()) + 1; // how many more elements we need?
+  if (need > 0) {
+    fill(static_cast<size_t>(need));
+  }
+}
+
+/// <summary>
+/// Add {@code n} elements to the buffer. Returns the number of tokens
+/// actually added to the buffer. If the return value is less than {@code n},
+/// then EOF was reached before {@code n} tokens could be added.
+/// </summary>
+size_t UnbufferedTokenStream::fill(size_t n)
+{
+  for (size_t i = 0; i < n; i++) {
+    if (_tokens.size() > 0 && _tokens.back()->getType() == EOF) {
+      return i;
+    }
+
+    add(_tokenSource->nextToken());
+  }
+
+  return n;
+}
+
+void UnbufferedTokenStream::add(std::unique_ptr<Token> t)
+{
+  WritableToken *writable = dynamic_cast<WritableToken *>(t.get());
+  if (writable != nullptr) {
+    writable->setTokenIndex(int(getBufferStartIndex() + _tokens.size()));
+  }
+
+  _tokens.push_back(std::move(t));
+}
+
+/// <summary>
+/// Return a marker that we can release later.
+/// <p/>
+/// The specific marker value used for this class allows for some level of
+/// protection against misuse where {@code seek()} is called on a mark or
+/// {@code release()} is called in the wrong order.
+/// </summary>
+ssize_t UnbufferedTokenStream::mark()
+{
+  if (_numMarkers == 0) {
+    _lastTokenBufferStart = _lastToken;
+  }
+
+  int mark = -_numMarkers - 1;
+  _numMarkers++;
+  return mark;
+}
+
+void UnbufferedTokenStream::release(ssize_t marker)
+{
+  ssize_t expectedMark = -_numMarkers;
+  if (marker != expectedMark) {
+    throw IllegalStateException("release() called with an invalid marker.");
+  }
+
+  _numMarkers--;
+  if (_numMarkers == 0) { // can we release buffer?
+    if (_p > 0) {
+      // Copy tokens[p]..tokens[n-1] to tokens[0]..tokens[(n-1)-p], reset ptrs
+      // p is last valid token; move nothing if p==n as we have no valid char
+      _tokens.erase(_tokens.begin(), _tokens.begin() + static_cast<ssize_t>(_p));
+      _p = 0;
+    }
+
+    _lastTokenBufferStart = _lastToken;
+  }
+}
+
+size_t UnbufferedTokenStream::index()
+{
+  return _currentTokenIndex;
+}
+
+void UnbufferedTokenStream::seek(size_t index)
+{ // seek to absolute index
+  if (index == _currentTokenIndex) {
+    return;
+  }
+
+  if (index > _currentTokenIndex) {
+    sync(ssize_t(index - _currentTokenIndex));
+    index = std::min(index, getBufferStartIndex() + _tokens.size() - 1);
+  }
+
+  size_t bufferStartIndex = getBufferStartIndex();
+  if (bufferStartIndex > index) {
+    throw IllegalArgumentException(std::string("cannot seek to negative index ") + std::to_string(index));
+  }
+
+  size_t i = index - bufferStartIndex;
+  if (i >= _tokens.size()) {
+    throw UnsupportedOperationException(std::string("seek to index outside buffer: ") + std::to_string(index) +
+      " not in " + std::to_string(bufferStartIndex) + ".." + std::to_string(bufferStartIndex + _tokens.size()));
+  }
+
+  _p = i;
+  _currentTokenIndex = index;
+  if (_p == 0) {
+    _lastToken = _lastTokenBufferStart;
+  } else {
+    _lastToken = _tokens[_p - 1].get();
+  }
+}
+
+size_t UnbufferedTokenStream::size()
+{
+  throw UnsupportedOperationException("Unbuffered stream cannot know its size");
+}
+
+std::string UnbufferedTokenStream::getSourceName() const
+{
+  return _tokenSource->getSourceName();
+}
+
+std::string UnbufferedTokenStream::getText(const misc::Interval &interval)
+{
+  size_t bufferStartIndex = getBufferStartIndex();
+  size_t bufferStopIndex = bufferStartIndex + _tokens.size() - 1;
+
+  size_t start = interval.a;
+  size_t stop = interval.b;
+  if (start < bufferStartIndex || stop > bufferStopIndex) {
+    throw UnsupportedOperationException(std::string("interval ") + interval.toString() +
+      " not in token buffer window: " + std::to_string(bufferStartIndex) + ".." + std::to_string(bufferStopIndex));
+  }
+
+  size_t a = start - bufferStartIndex;
+  size_t b = stop - bufferStartIndex;
+
+  std::stringstream ss;
+  for (size_t i = a; i <= b; i++) {
+    Token *t = _tokens[i].get();
+    if (i > 0)
+      ss << ", ";
+    ss << t->getText();
+  }
+
+  return ss.str();
+}
+
+size_t UnbufferedTokenStream::getBufferStartIndex() const
+{
+  return _currentTokenIndex - _p;
+}
+
+void UnbufferedTokenStream::InitializeInstanceFields()
+{
+  _p = 0;
+  _numMarkers = 0;
+  _currentTokenIndex = 0;
+}

data/ext/python3-parser/antlr4-upstream/runtime/Cpp/runtime/src/UnbufferedTokenStream.h

@@ -0,0 +1,115 @@
+/* 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 "TokenStream.h"
+
+namespace antlr4 {
+
+  class ANTLR4CPP_PUBLIC UnbufferedTokenStream : public TokenStream {
+  public:
+    UnbufferedTokenStream(TokenSource *tokenSource);
+    UnbufferedTokenStream(TokenSource *tokenSource, int bufferSize);
+    UnbufferedTokenStream(const UnbufferedTokenStream& other) = delete;
+    virtual ~UnbufferedTokenStream();
+
+    UnbufferedTokenStream& operator = (const UnbufferedTokenStream& other) = delete;
+
+    virtual Token* get(size_t i) const override;
+    virtual Token* LT(ssize_t i) override;
+    virtual size_t LA(ssize_t i) override;
+
+    virtual TokenSource* getTokenSource() const override;
+
+    virtual std::string getText(const misc::Interval &interval) override;
+    virtual std::string getText() override;
+    virtual std::string getText(RuleContext *ctx) override;
+    virtual std::string getText(Token *start, Token *stop) override;
+
+    virtual void consume() override;
+
+    /// <summary>
+    /// Return a marker that we can release later.
+    /// <p/>
+    /// The specific marker value used for this class allows for some level of
+    /// protection against misuse where {@code seek()} is called on a mark or
+    /// {@code release()} is called in the wrong order.
+    /// </summary>
+    virtual ssize_t mark() override;
+    virtual void release(ssize_t marker) override;
+    virtual size_t index() override;
+    virtual void seek(size_t index) override;
+    virtual size_t size() override;
+    virtual std::string getSourceName() const override;
+
+  protected:
+    /// Make sure we have 'need' elements from current position p. Last valid
+    /// p index is tokens.length - 1.  p + need - 1 is the tokens index 'need' elements
+    /// ahead.  If we need 1 element, (p+1-1)==p must be less than tokens.length.
+    TokenSource *_tokenSource;
+
+    /// <summary>
+    /// A moving window buffer of the data being scanned. While there's a marker,
+    /// we keep adding to buffer. Otherwise, <seealso cref="#consume consume()"/> resets so
+    /// we start filling at index 0 again.
+    /// </summary>
+
+    std::vector<std::unique_ptr<Token>> _tokens;
+
+    /// <summary>
+    /// 0..n-1 index into <seealso cref="#tokens tokens"/> of next token.
+    /// <p/>
+    /// The {@code LT(1)} token is {@code tokens[p]}. If {@code p == n}, we are
+    /// out of buffered tokens.
+    /// </summary>
+    size_t _p;
+
+    /// <summary>
+    /// Count up with <seealso cref="#mark mark()"/> and down with
+    /// <seealso cref="#release release()"/>. When we {@code release()} the last mark,
+    /// {@code numMarkers} reaches 0 and we reset the buffer. Copy
+    /// {@code tokens[p]..tokens[n-1]} to {@code tokens[0]..tokens[(n-1)-p]}.
+    /// </summary>
+    int _numMarkers;
+
+    /// <summary>
+    /// This is the {@code LT(-1)} token for the current position.
+    /// </summary>
+    Token *_lastToken;
+
+    /// <summary>
+    /// When {@code numMarkers > 0}, this is the {@code LT(-1)} token for the
+    /// first token in <seealso cref="#tokens"/>. Otherwise, this is {@code null}.
+    /// </summary>
+    Token *_lastTokenBufferStart;
+
+    /// <summary>
+    /// Absolute token index. It's the index of the token about to be read via
+    /// {@code LT(1)}. Goes from 0 to the number of tokens in the entire stream,
+    /// although the stream size is unknown before the end is reached.
+    /// <p/>
+    /// This value is used to set the token indexes if the stream provides tokens
+    /// that implement <seealso cref="WritableToken"/>.
+    /// </summary>
+    size_t _currentTokenIndex;
+
+    virtual void sync(ssize_t want);
+
+    /// <summary>
+    /// Add {@code n} elements to the buffer. Returns the number of tokens
+    /// actually added to the buffer. If the return value is less than {@code n},
+    /// then EOF was reached before {@code n} tokens could be added.
+    /// </summary>
+    virtual size_t fill(size_t n);
+    virtual void add(std::unique_ptr<Token> t);
+
+    size_t getBufferStartIndex() const;
+
+  private:
+    void InitializeInstanceFields();
+  };
+
+} // namespace antlr4