// Copright 2014 PDFium Authors. All rights reserved.
// Use of this source code is governed by a BSD-style license that can be
// found in the LICENSE file.

// Original code copyright 2014 Foxit Software Inc. http://www.foxitsoftware.com

#include "xfa/fxfa/fm2js/cxfa_fmlexer.h"

#include <algorithm>

#include "core/fxcrt/fx_extension.h"
#include "third_party/base/ptr_util.h"
#include "third_party/icu/source/common/unicode/uchar.h"

namespace {

bool IsFormCalcCharacter(wchar_t c) {
  return (c >= 0x09 && c <= 0x0D) || (c >= 0x20 && c <= 0xd7FF) ||
         (c >= 0xE000 && c <= 0xFFFD);
}

bool IsIdentifierCharacter(wchar_t c) {
  return u_isalnum(c) || c == 0x005F ||  // '_'
         c == 0x0024;                    // '$'
}

bool IsInitialIdentifierCharacter(wchar_t c) {
  return u_isalpha(c) || c == 0x005F ||  // '_'
         c == 0x0024 ||                  // '$'
         c == 0x0021;                    // '!'
}

bool IsWhitespaceCharacter(wchar_t c) {
  return c == 0x0009 ||  // Horizontal tab
         c == 0x000B ||  // Vertical tab
         c == 0x000C ||  // Form feed
         c == 0x0020;    // Space
}

const XFA_FMKeyword keyWords[] = {
    {TOKdo, L"do"},
    {TOKkseq, L"eq"},
    {TOKksge, L"ge"},
    {TOKksgt, L"gt"},
    {TOKif, L"if"},
    {TOKin, L"in"},
    {TOKksle, L"le"},
    {TOKkslt, L"lt"},
    {TOKksne, L"ne"},
    {TOKksor, L"or"},
    {TOKnull, L"null"},
    {TOKbreak, L"break"},
    {TOKksand, L"and"},
    {TOKend, L"end"},
    {TOKeof, L"eof"},
    {TOKfor, L"for"},
    {TOKnan, L"nan"},
    {TOKksnot, L"not"},
    {TOKvar, L"var"},
    {TOKthen, L"then"},
    {TOKelse, L"else"},
    {TOKexit, L"exit"},
    {TOKdownto, L"downto"},
    {TOKreturn, L"return"},
    {TOKinfinity, L"infinity"},
    {TOKendwhile, L"endwhile"},
    {TOKforeach, L"foreach"},
    {TOKendfunc, L"endfunc"},
    {TOKelseif, L"elseif"},
    {TOKwhile, L"while"},
    {TOKendfor, L"endfor"},
    {TOKthrow, L"throw"},
    {TOKstep, L"step"},
    {TOKupto, L"upto"},
    {TOKcontinue, L"continue"},
    {TOKfunc, L"func"},
    {TOKendif, L"endif"},
};

#ifndef NDEBUG
const wchar_t* const tokenStrings[] = {
    L"TOKand",        L"TOKlparen",     L"TOKrparen",   L"TOKmul",
    L"TOKplus",       L"TOKcomma",      L"TOKminus",    L"TOKdot",
    L"TOKdiv",        L"TOKlt",         L"TOKassign",   L"TOKgt",
    L"TOKlbracket",   L"TOKrbracket",   L"TOKor",       L"TOKdotscream",
    L"TOKdotstar",    L"TOKdotdot",     L"TOKle",       L"TOKne",
    L"TOKeq",         L"TOKge",         L"TOKdo",       L"TOKkseq",
    L"TOKksge",       L"TOKksgt",       L"TOKif",       L"TOKin",
    L"TOKksle",       L"TOKkslt",       L"TOKksne",     L"TOKksor",
    L"TOKnull",       L"TOKbreak",      L"TOKksand",    L"TOKend",
    L"TOKeof",        L"TOKfor",        L"TOKnan",      L"TOKksnot",
    L"TOKvar",        L"TOKthen",       L"TOKelse",     L"TOKexit",
    L"TOKdownto",     L"TOKreturn",     L"TOKinfinity", L"TOKendwhile",
    L"TOKforeach",    L"TOKendfunc",    L"TOKelseif",   L"TOKwhile",
    L"TOKendfor",     L"TOKthrow",      L"TOKstep",     L"TOKupto",
    L"TOKcontinue",   L"TOKfunc",       L"TOKendif",    L"TOKstar",
    L"TOKidentifier", L"TOKunderscore", L"TOKdollar",   L"TOKexclamation",
    L"TOKcall",       L"TOKstring",     L"TOKnumber",   L"TOKreserver",
};
#endif  // NDEBUG

XFA_FM_TOKEN TokenizeIdentifier(const WideStringView& str) {
  const XFA_FMKeyword* result = std::find_if(
      std::begin(keyWords), std::end(keyWords),
      [&str](const XFA_FMKeyword& iter) { return iter.m_keyword == str; });
  if (result != std::end(keyWords) && result->m_keyword == str)
    return result->m_type;
  return TOKidentifier;
}

}  // namespace

CXFA_FMToken::CXFA_FMToken(XFA_FM_TOKEN token) : m_type(token) {}

CXFA_FMToken::CXFA_FMToken() : CXFA_FMToken(TOKreserver) {}

CXFA_FMToken::CXFA_FMToken(const CXFA_FMToken&) = default;

CXFA_FMToken::~CXFA_FMToken() = default;

#ifndef NDEBUG
WideString CXFA_FMToken::ToDebugString() const {
  WideString str(L"type = ");
  str += tokenStrings[m_type];
  str += L", string = ";
  str += m_string;
  return str;
}
#endif  // NDEBUG

CXFA_FMLexer::CXFA_FMLexer(const WideStringView& wsFormCalc)
    : m_cursor(wsFormCalc.unterminated_c_str()),
      m_end(m_cursor + wsFormCalc.GetLength() - 1),
      m_lexer_error(false) {}

CXFA_FMLexer::~CXFA_FMLexer() {}

CXFA_FMToken CXFA_FMLexer::NextToken() {
  if (m_lexer_error)
    return CXFA_FMToken();

  while (m_cursor <= m_end && *m_cursor) {
    if (!IsFormCalcCharacter(*m_cursor)) {
      RaiseError();
      return CXFA_FMToken();
    }

    switch (*m_cursor) {
      case '\n':
        ++m_cursor;
        break;
      case '\r':
        ++m_cursor;
        break;
      case ';':
        AdvanceForComment();
        break;
      case '"':
        return AdvanceForString();
      case '0':
      case '1':
      case '2':
      case '3':
      case '4':
      case '5':
      case '6':
      case '7':
      case '8':
      case '9':
        return AdvanceForNumber();
      case '=':
        ++m_cursor;
        if (m_cursor > m_end)
          return CXFA_FMToken(TOKassign);

        if (!IsFormCalcCharacter(*m_cursor)) {
          RaiseError();
          return CXFA_FMToken();
        }
        if (*m_cursor == '=') {
          ++m_cursor;
          return CXFA_FMToken(TOKeq);
        }
        return CXFA_FMToken(TOKassign);
      case '<':
        ++m_cursor;
        if (m_cursor > m_end)
          return CXFA_FMToken(TOKlt);

        if (!IsFormCalcCharacter(*m_cursor)) {
          RaiseError();
          return CXFA_FMToken();
        }
        if (*m_cursor == '=') {
          ++m_cursor;
          return CXFA_FMToken(TOKle);
        }
        if (*m_cursor == '>') {
          ++m_cursor;
          return CXFA_FMToken(TOKne);
        }
        return CXFA_FMToken(TOKlt);
      case '>':
        ++m_cursor;
        if (m_cursor > m_end)
          return CXFA_FMToken(TOKgt);

        if (!IsFormCalcCharacter(*m_cursor)) {
          RaiseError();
          return CXFA_FMToken();
        }
        if (*m_cursor == '=') {
          ++m_cursor;
          return CXFA_FMToken(TOKge);
        }
        return CXFA_FMToken(TOKgt);
      case ',':
        ++m_cursor;
        return CXFA_FMToken(TOKcomma);
      case '(':
        ++m_cursor;
        return CXFA_FMToken(TOKlparen);
      case ')':
        ++m_cursor;
        return CXFA_FMToken(TOKrparen);
      case '[':
        ++m_cursor;
        return CXFA_FMToken(TOKlbracket);
      case ']':
        ++m_cursor;
        return CXFA_FMToken(TOKrbracket);
      case '&':
        ++m_cursor;
        return CXFA_FMToken(TOKand);
      case '|':
        ++m_cursor;
        return CXFA_FMToken(TOKor);
      case '+':
        ++m_cursor;
        return CXFA_FMToken(TOKplus);
      case '-':
        ++m_cursor;
        return CXFA_FMToken(TOKminus);
      case '*':
        ++m_cursor;
        return CXFA_FMToken(TOKmul);
      case '/': {
        ++m_cursor;
        if (m_cursor > m_end)
          return CXFA_FMToken(TOKdiv);

        if (!IsFormCalcCharacter(*m_cursor)) {
          RaiseError();
          return CXFA_FMToken();
        }
        if (*m_cursor != '/')
          return CXFA_FMToken(TOKdiv);

        AdvanceForComment();
        break;
      }
      case '.':
        ++m_cursor;
        if (m_cursor > m_end)
          return CXFA_FMToken(TOKdot);

        if (!IsFormCalcCharacter(*m_cursor)) {
          RaiseError();
          return CXFA_FMToken();
        }

        if (*m_cursor == '.') {
          ++m_cursor;
          return CXFA_FMToken(TOKdotdot);
        }
        if (*m_cursor == '*') {
          ++m_cursor;
          return CXFA_FMToken(TOKdotstar);
        }
        if (*m_cursor == '#') {
          ++m_cursor;
          return CXFA_FMToken(TOKdotscream);
        }
        if (*m_cursor <= '9' && *m_cursor >= '0') {
          --m_cursor;
          return AdvanceForNumber();
        }
        return CXFA_FMToken(TOKdot);
      default:
        if (IsWhitespaceCharacter(*m_cursor)) {
          ++m_cursor;
          break;
        }
        if (!IsInitialIdentifierCharacter(*m_cursor)) {
          RaiseError();
          return CXFA_FMToken();
        }
        return AdvanceForIdentifier();
    }
  }
  return CXFA_FMToken(TOKeof);
}

CXFA_FMToken CXFA_FMLexer::AdvanceForNumber() {
  // This will set end to the character after the end of the number.
  wchar_t* end = nullptr;
  if (m_cursor)
    wcstod(const_cast<wchar_t*>(m_cursor), &end);
  if (!end || FXSYS_iswalpha(*end)) {
    RaiseError();
    return CXFA_FMToken();
  }

  CXFA_FMToken token(TOKnumber);
  token.m_string =
      WideStringView(m_cursor, static_cast<size_t>(end - m_cursor));
  m_cursor = end;
  return token;
}

CXFA_FMToken CXFA_FMLexer::AdvanceForString() {
  CXFA_FMToken token(TOKstring);

  const wchar_t* start = m_cursor;
  ++m_cursor;
  while (m_cursor <= m_end && *m_cursor) {
    if (!IsFormCalcCharacter(*m_cursor))
      break;

    if (*m_cursor == '"') {
      // Check for escaped "s, i.e. "".
      ++m_cursor;
      // If the end of the input has been reached it was not escaped.
      if (m_cursor > m_end) {
        token.m_string =
            WideStringView(start, static_cast<size_t>(m_cursor - start));
        return token;
      }
      // If the next character is not a " then the end of the string has been
      // found.
      if (*m_cursor != '"') {
        if (!IsFormCalcCharacter(*m_cursor))
          break;

        token.m_string = WideStringView(start, (m_cursor - start));
        return token;
      }
    }
    ++m_cursor;
  }

  // Didn't find the end of the string.
  RaiseError();
  return CXFA_FMToken();
}

CXFA_FMToken CXFA_FMLexer::AdvanceForIdentifier() {
  const wchar_t* start = m_cursor;
  ++m_cursor;
  while (m_cursor <= m_end && *m_cursor) {
    if (!IsFormCalcCharacter(*m_cursor)) {
      RaiseError();
      return CXFA_FMToken();
    }
    if (!IsIdentifierCharacter(*m_cursor))
      break;

    ++m_cursor;
  }

  WideStringView str =
      WideStringView(start, static_cast<size_t>(m_cursor - start));
  CXFA_FMToken token(TokenizeIdentifier(str));
  token.m_string = str;
  return token;
}

void CXFA_FMLexer::AdvanceForComment() {
  m_cursor++;
  while (m_cursor <= m_end && *m_cursor) {
    if (!IsFormCalcCharacter(*m_cursor)) {
      RaiseError();
      return;
    }
    if (*m_cursor == L'\r') {
      ++m_cursor;
      return;
    }
    if (*m_cursor == L'\n') {
      ++m_cursor;
      return;
    }
    ++m_cursor;
  }
}