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

compiled/ActiveFilter.cpp

+#include "ActiveFilter.h"
+#include "StringScanner.h"
+
+namespace
+{
+  const std::u16string ProcessDomain(const std::u16string& docDomain,
+      bool ignoreTrailingDot)
+  {
+    size_t len = docDomain.length();
+    if (ignoreTrailingDot)
+    {
+      for (; len > 0; --len)
+      {
+        if (docDomain[len - 1] != u'.')
+          break;
+      }
+    }
+
+    std::u16string domain(docDomain.substr(0, len));
+    for (size_t i = 0, l = domain.length(); i < l; ++i)
+    {
+      char16_t currChar = domain[i];
+      // TODO: This needs to work for non-ASCII as well
+      if (currChar >= u'a' && currChar <= u'z')
+        domain[i] = currChar + u'A' - u'a';
+    }
+    return domain;
+  }
+
+  const std::u16string to_u16string(unsigned int i)
+  {
+    std::string str(std::to_string(i));
+    std::u16string result(str.length(), u'\0');
+
+    for (size_t i = 0, l = str.length(); i < l; ++i)
+      result[i] = str[i];
+    return result;
+  }
+}
+
+ActiveFilter::ActiveFilter(const std::u16string& text, bool ignoreTrailingDot)
+    : Filter(text), disabled(false), hitCount(0), lastHit(0),
+      ignoreTrailingDot(ignoreTrailingDot)
+{
+}
+
+void ActiveFilter::ParseDomains(const std::u16string& str, char16_t separator)
+{
+  StringScanner scanner(str + separator);
+  size_t start = 0;
+  bool reverse = false;
+  bool hasIncludes = false;
+  while (!scanner.done())
+  {
+    char16_t currChar = scanner.next();
+    if (currChar == u'~' && scanner.position() == start)
+    {
+      start++;
+      reverse = true;
+    }
+    else if (currChar == separator)
+    {
+      if (scanner.position() > start)
+      {
+        std::u16string domain(
+          ProcessDomain(str.substr(start, scanner.position() - start),
+              ignoreTrailingDot)
+        );
+        domains[domain] = !reverse;
+        if (!reverse)
+          hasIncludes = true;
+      }
+      start = scanner.position() + 1;
+      reverse = false;
+    }
+  }
+  domains[u""] = !hasIncludes;
+}
+
+
+bool ActiveFilter::IsActiveOnDomain(const std::u16string& docDomain,
+    const std::u16string& sitekey)
+{
+  if (!sitekeys.empty() && sitekeys.find(sitekey) == sitekeys.end())
+    return false;
+
+  // If no domains are set the rule matches everywhere
+  if (domains.empty())
+    return true;
+
+  // If the document has no host name, match only if the filter isn't restricted
+  // to specific domains
+  if (docDomain.empty())
+    return domains[u""];
+
+  std::u16string domain(ProcessDomain(docDomain, ignoreTrailingDot));
+  while (true)
+  {
+    auto it = domains.find(domain);
+    if (it != domains.end())
+      return it->second;
+
+    size_t nextDot = domain.find(u'.');
+    if (nextDot == domain.npos)
+      break;
+    domain = domain.substr(nextDot + 1);
+  }
+  return domains[u""];
+}
+
+bool ActiveFilter::IsActiveOnlyOnDomain(const std::u16string& docDomain)
+{
+  if (domains.empty() || docDomain.empty() || domains[u""])
+    return false;
+
+  std::u16string domain(ProcessDomain(docDomain, ignoreTrailingDot));
+  for (auto it = domains.begin(); it != domains.end(); ++it)
+  {
+    if (!it->second || it->first == domain)
+      continue;
+
+    size_t len1 = it->first.length();
+    size_t len2 = domain.length();
+    if (len1 > len2 && it->first.rfind(u"." + domain) == len1 - len2 - 1)
+      continue;
+
+    return false;
+  }
+  return true;
+}
+
+bool ActiveFilter::IsGeneric()
+{
+  return (sitekeys.empty() && (domains.empty() || domains[u""]));
+}
+
+const std::u16string ActiveFilter::Serialize()
+{
+  std::u16string result(Filter::Serialize());
+  if (disabled)
+    result += u"disabled=true\n";
+  if (hitCount)
+    result += u"hitCount=" + to_u16string(hitCount) + u"\n";
+  if (lastHit)
+    result += u"lastHit=" + to_u16string(lastHit) + u"\n";
+  return result;
+}