Issue 4125 - [emscripten] Convert filter classes to C++

compiled/RegExpFilter.cpp

Index: compiled/RegExpFilter.cpp
===================================================================
new file mode 100644
--- /dev/null
+++ b/compiled/RegExpFilter.cpp
@@ -0,0 +1,307 @@
+#include <climits>
+
+#include <emscripten.h>
+
+#include "RegExpFilter.h"
+#include "WhitelistFilter.h"
+#include "InvalidFilter.h"
+#include "StringScanner.h"
+#include "StringMap.h"
+
+namespace
+{
+  enum
+  {
+    TYPE_OTHER = 0x1,
+    TYPE_SCRIPT = 0x2,
+    TYPE_IMAGE = 0x4,
+    TYPE_STYLESHEET = 0x8,
+    TYPE_OBJECT = 0x10,
+    TYPE_SUBDOCUMENT = 0x20,
+    TYPE_DOCUMENT = 0x40,
+    TYPE_PING = 0x400,
+    TYPE_XMLHTTPREQUEST = 0x800,
+    TYPE_OBJECT_SUBREQUEST = 0x1000,
+    TYPE_MEDIA = 0x4000,
+    TYPE_FONT = 0x8000,
+    TYPE_POPUP = 0x8000000,
+    TYPE_GENERICBLOCK = 0x10000000,
+    TYPE_GENERICHIDE = 0x20000000,
+    TYPE_ELEMHIDE = 0x40000000,
+  };
+
+  StringMap<int> typeMap {
+    {u"other"_str, TYPE_OTHER},
+    {u"script"_str, TYPE_SCRIPT},
+    {u"image"_str, TYPE_IMAGE},
+    {u"stylesheet"_str, TYPE_STYLESHEET},
+    {u"object"_str, TYPE_OBJECT},
+    {u"subdocument"_str, TYPE_SUBDOCUMENT},
+    {u"document"_str, TYPE_DOCUMENT},
+    {u"xbl"_str, TYPE_OTHER},          // Backwards compat
+    {u"ping"_str, TYPE_PING},
+    {u"xmlhttprequest"_str, TYPE_XMLHTTPREQUEST},
+    {u"object-subrequest"_str, TYPE_OBJECT_SUBREQUEST},
+    {u"dtd"_str, TYPE_OTHER},          // Backwards compat
+    {u"media"_str, TYPE_MEDIA},
+    {u"font"_str, TYPE_FONT},
+    {u"background"_str, TYPE_IMAGE},   // Backwards compat
+
+    {u"popup"_str, TYPE_POPUP},
+    {u"genericblock"_str, TYPE_GENERICBLOCK},
+    {u"generichide"_str, TYPE_GENERICHIDE},
+    {u"elemhide"_str, TYPE_ELEMHIDE},
+  };
+
+  int defaultTypeMask = INT_MAX & ~(TYPE_DOCUMENT | TYPE_ELEMHIDE | TYPE_POPUP |
+      TYPE_GENERICBLOCK | TYPE_GENERICHIDE);
+
+  int GenerateRegExp(const String& regexp, bool matchCase)
+  {
+    return EM_ASM_INT(return regexps.create($0, $1), &regexp, matchCase);
+  }
+}
+
+RegExpFilter::RegExpFilter(const String& text,
+    String::size_type patternStart, String::size_type patternEnd)
+    : ActiveFilter(text, true), mRegexpId(0),
+      mRegexpSource(String(mText, patternStart, patternEnd - patternStart)),
+      mContentType(-1), mMatchCase(false), mThirdParty(TrippleState::ANY)
+{
+  String options(mText, patternEnd + 1);
+  StringScanner scanner(options, u',');
+  int optionStart = 0;
+  int optionEnd = -1;
+  int valueStart = -1;
+  bool done = false;
+  while (!done)
+  {
+    done = scanner.done();
+    switch (scanner.next())
+    {
+      case u'=':
+        if (optionEnd < 0)
+        {
+          optionEnd = scanner.position();
+          valueStart = optionEnd + 1;
+        }
+        break;
+      case u',':
+        if (optionEnd < 0)
+          optionEnd = scanner.position();
+        ProcessOption(options, optionStart, optionEnd, valueStart, scanner.position());
+        optionStart = scanner.position() + 1;
+        optionEnd = -1;
+        valueStart = -1;
+        break;
+    }
+  }
+  if (mContentType < 0)
+    mContentType = defaultTypeMask;
+
+  size_t len = mRegexpSource.length();
+  if (len >= 2 && mRegexpSource[0] == u'/' && mRegexpSource[len - 1] == u'/')
+  {
+    mRegexpSource.reset(mRegexpSource, 1 , len - 2);
+    mRegexpId = GenerateRegExp(mRegexpSource, mMatchCase);
+
+    int errorLength = EM_ASM_INT(return regexps.getErrorLength($0), mRegexpId);
+    if (errorLength >= 0)
+    {
+      String error(errorLength);
+      EM_ASM_ARGS(regexps.getError($0, $1), mRegexpId, error.data());
+      throw error;
+    }
+  }
+}
+
+RegExpFilter::~RegExpFilter()
+{
+  if (mRegexpId)
+    EM_ASM_ARGS(regexps.delete($0), mRegexpId);
+}
+
+void RegExpFilter::ProcessOption(String& options, int optionStart,
+    int optionEnd, int valueStart, int valueEnd)
+{
+  if (optionEnd <= optionStart)
+    return;
+
+  bool reverse = false;
+  if (options[optionStart] == u'~')
+  {
+    reverse = true;
+    optionStart++;
+  }
+
+  String name(options, optionStart, optionEnd - optionStart);
+  for (size_t i = 0; i < name.length(); ++i)
+  {
+    char16_t currChar = name[i];
+    if (currChar >= u'A' && currChar <= u'Z')
+      name[i] = currChar + u'a' - u'A';
+    else if (currChar == u'_')
+      name[i] = u'-';
+  }
+
+  auto it = typeMap.find(name);
+  if (it != typeMap.end())
+  {
+    if (mContentType < 0)
+      mContentType = reverse ? defaultTypeMask : 0;
+    if (reverse)
+      mContentType &= ~it->second;
+    else
+      mContentType |= it->second;
+  }
+  else if (name.equals(u"domain"_str))
+  {
+    if (valueStart >= 0 && valueEnd > valueStart)
+      ParseDomains(String(options, valueStart, valueEnd - valueStart), u'|');
+  }
+  else if (name.equals(u"sitekey"_str))
+  {
+    if (valueStart >= 0 && valueEnd > valueStart)
+    {
+      StringScanner scanner(String(options, valueStart, valueEnd - valueStart), u'|');
+      size_t start = 0;
+      bool done = false;
+      while (!done)
+      {
+        done = scanner.done();
+        if (scanner.next() == u'|')
+        {
+          if (scanner.position() > start)
+            AddSitekey(String(options, valueStart + start, scanner.position() - start));
+          start = scanner.position() + 1;
+        }
+      }
+    }
+  }
+  else if (name.equals(u"match-case"_str))
+    mMatchCase = !reverse;
+  else if (name.equals(u"third-party"_str))
+    mThirdParty = reverse ? TrippleState::NO : TrippleState::YES;
+  else if (name.equals(u"collapse"_str))
+    mCollapse = reverse ? TrippleState::NO : TrippleState::YES;
+  else
+  {
+    String error(u"Unknown option "_str);
+    error.append(name);
+    throw std::move(error);
+  }
+}
+
+Filter* RegExpFilter::Create(const String& text)
+{
+  bool blocking = true;
+  String::size_type patternStart = 0;
+  if (text.length() >= 2 && text[0] == u'@' && text[1] == u'@')
+  {
+    blocking = false;
+    patternStart = 2;
+  }
+
+  String::size_type patternEnd = text.find(u'$', patternStart);
+  if (patternEnd == text.npos)
+    patternEnd = text.length();
+
+  try
+  {
+    if (blocking)
+      return new RegExpFilter(text, patternStart, patternEnd);
+    else
+      return new WhitelistFilter(text, patternStart, patternEnd);
+  }
+  catch (const String& reason)
+  {
+    return new InvalidFilter(text, reason);
+  }
+}
+
+void RegExpFilter::InitJSTypes()
+{
+  EM_ASM(exports.RegExpFilter.typeMap = {};);
+  for (auto it = typeMap.begin(); it != typeMap.end(); ++it)
+    EM_ASM_ARGS(exports.RegExpFilter.typeMap[getStringData($0).replace("-", "_").toUpperCase()] = $1, &(it->first), it->second);
+}
+
+String RegExpFilter::RegExpFromSource(const String& source)
+{
+  /* TODO: this is very inefficient */
+
+  // Note: This doesn't remove trailing wildcards, otherwise the result should
+  // be identical to Filter.toRegExp().
+  String result;
+  String::value_type prevChar = u'*';
+  for (String::size_type i = 0; i < source.length(); ++i)
+  {
+    String::value_type currChar = source[i];
+    switch (currChar)
+    {
+      case u'*':
+        if (prevChar != u'*')
+          result.append(u".*"_str);
+        break;
+      case u'^':
+        result.append(u"(?:[\\x00-\\x24\\x26-\\x2C\\x2F\\x3A-\\x40\\x5B-\\x5E\\x60\\x7B-\\x7F]|$)"_str);
+        break;
+      case u'|':
+        if (i == 0)
+        {
+          // Anchor at expression start, maybe extended anchor?
+          if (i + 1 < source.length() && source[i + 1] == u'|')
+          {
+            result.append(u"^[\\w\\-]+:\\/+(?!\\/)(?:[^\\/]+\\.)?"_str);
+            ++i;
+          }
+          else
+            result.append(u'^');
+        }
+        else if (i == source.length() - 1)
+        {
+          // Anchor at expression end, ignore if following separator placeholder
+          if (prevChar != u'^')
+            result.append(u'$');
+        }
+        else
+        {
+          // Not actually an anchor, escape it
+          result.append(u"\\|"_str);
+        }
+        break;
+      default:
+        if (!(currChar >= u'a' && currChar <= u'z') &&
+            !(currChar >= u'A' && currChar <= u'Z') &&
+            !(currChar >= u'0' && currChar <= u'9') &&
+            currChar < 128)
+        {
+          result.append(u'\\');
+        }
+        result.append(currChar);
+    }
+    prevChar = currChar;
+  }
+  return std::move(result);
+}
+
+Filter::Type RegExpFilter::GetType() const
+{
+  return Type::BLOCKING;
+}
+
+bool RegExpFilter::Matches(const String& location, int typeMask,
+    String& docDomain, bool thirdParty, const String& sitekey) const
+{
+  if (!(mContentType & typeMask) ||
+      (mThirdParty == TrippleState::YES && !thirdParty) ||
+      (mThirdParty == TrippleState::NO && thirdParty) ||
+      !IsActiveOnDomain(docDomain, sitekey))
+  {
+    return false;
+  }
+
+  if (!mRegexpId)
+    mRegexpId = GenerateRegExp(RegExpFromSource(mRegexpSource), mMatchCase);
+  return EM_ASM_INT(return regexps.test($0, $1), mRegexpId, &location);
+}