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

compiled/Filter.cpp

+#include "Filter.h"
+#include "CommentFilter.h"
+#include "RegExpFilter.h"
+#include "ElemHideFilter.h"
+#include "ElemHideException.h"
+#include "StringMap.h"
+
+namespace
+{
+  StringMap<Filter*> knownFilters(8192);
+
+  void trim_spaces(String& str)
+  {
+    String::size_type pos;
+
+    // Remove leading whitespace
+    for (pos = 0; pos < str.length(); ++pos)
+      if (str[pos] != u' ')
+        break;
+    str.reset(str, pos);
+
+    // Remove trailing whitespace
+    for (pos = str.length(); pos > 0; --pos)
+      if (str[pos - 1] != u' ')
+        break;
+    str.reset(str, 0, pos);
+  }
+
+  void remove_spaces(String& str)
+  {
+    String::size_type pos;
+
+    for (String::size_type i = 0; i < str.length(); ++i)
+      if (str[i] != u' ')
+        str[pos++] = str[i];
+
+    str.reset(str, 0, pos);
+  }
+}
+
+Filter::Filter(const String& text)
+    : ref_counted(), mText(text)
+{
+  annotate_address(this, "Filter");
+  mText.ensure_own_buffer();
+}
+
+Filter::~Filter()
+{
+  // TODO: This should be removing from knownFilters
+}
+
+String Filter::Serialize() const
+{
+  String result(u"[Filter]\ntext="_str);
+  result.append(mText);
+  result.append(u'\n');
+  return std::move(result.ensure_own_buffer());
+}
+
+Filter* Filter::FromText(const String& text)
+{
+  auto it = knownFilters.find(text);
+  if (it != knownFilters.end())
+    return it->second;
+
+  FilterPtr filter(CommentFilter::Create(text));
+  if (!filter)
+    filter.reset(ElemHideBase::Create(text));
+  if (!filter)
+    filter.reset(RegExpFilter::Create(text));
+
+  enter_context("Adding to known filters");
+  knownFilters[filter->mText] = filter.get();
+  exit_context();
+
+  // TODO: We intentionally leak the filter here - currently it won't be used
+  // for anything and would be deleted immediately.
+  filter->AddRef();
+
+  return filter;
+}
+
+String Filter::Normalize(String& text)
+{
+  // Removing special characters like line breaks
+  String::size_type delta = 0;
+  for (String::size_type i = 0; i < text.length(); ++i)
+  {
+    if (text[i] >= u' ')
+      text[i - delta] = text[i];
+    else
+      ++delta;
+  }
+  text.reset(text, 0, text.length() - delta);
+
+  trim_spaces(text);
+
+  {
+    String::size_type domainsEnd;
+    String::size_type selectorStart;
+    Filter::Type type = ElemHideBase::Parse(text, &domainsEnd, &selectorStart);
+    if (type != Filter::Type::UNKNOWN)
+    {
+      String domains(text, 0, domainsEnd);
+      String selector(text, selectorStart);
+      remove_spaces(domains);
+      trim_spaces(selector);
+
+      String::size_type domainsDelta = domainsEnd - domains.length();
+      String::size_type selectorDelta = text.length() - selectorStart -
+          selector.length();
+
+      if (domainsDelta)
+        for (String::size_type i = domainsEnd; i < selectorStart; ++i)
+          text[i - domainsDelta] = text[i];
+
+      if (domainsDelta + selectorDelta)
+        for (String::size_type i = 0; i < selector.length(); ++i)
+          text[selectorStart - domainsDelta + i] = selector[i];
+
+      text.reset(text, 0, text.length() - domainsDelta - selectorDelta);
+      return text;
+    }
+  }
+
+  if (CommentFilter::Parse(text) == Filter::Type::UNKNOWN)
+    remove_spaces(text);
+  return text;
+}