OLD | NEW |
(Empty) | |
| 1 #include <cstdio> |
| 2 |
| 3 #include "ActiveFilter.h" |
| 4 #include "StringScanner.h" |
| 5 |
| 6 namespace |
| 7 { |
| 8 OwnedString to_string(unsigned int i) |
| 9 { |
| 10 char buffer[11]; |
| 11 int len = sprintf(buffer, "%u", i); |
| 12 |
| 13 OwnedString result(len); |
| 14 for (String::size_type i = 0; i < len; i++) |
| 15 result[i] = buffer[i]; |
| 16 return result; |
| 17 } |
| 18 } |
| 19 |
| 20 ActiveFilter::ActiveFilter(const String& text, bool ignoreTrailingDot) |
| 21 : Filter(text), mDisabled(false), mHitCount(0), mLastHit(0), |
| 22 mIgnoreTrailingDot(ignoreTrailingDot) |
| 23 { |
| 24 } |
| 25 |
| 26 void ActiveFilter::ToLower(DependentString&& str) |
| 27 { |
| 28 for (String::size_type i = 0; i < str.length(); ++i) |
| 29 { |
| 30 String::value_type currChar = str[i]; |
| 31 |
| 32 // This should be more efficient with a lookup table but I couldn't measure |
| 33 // any performance difference. |
| 34 if (currChar >= u'A' && currChar <= u'Z') |
| 35 str[i] = currChar + u'a' - u'A'; |
| 36 else if (currChar >= 128) |
| 37 { |
| 38 // It seems that calling JS it the easiest solution for lowercasing |
| 39 // Unicode characters. |
| 40 str[i] = EM_ASM_INT({ |
| 41 return String.fromCharCode($0).toLowerCase().charCodeAt(0); |
| 42 }, currChar); |
| 43 } |
| 44 } |
| 45 } |
| 46 |
| 47 ActiveFilter::DomainMap* ActiveFilter::GetDomains() const |
| 48 { |
| 49 return mDomains.get(); |
| 50 } |
| 51 |
| 52 ActiveFilter::SitekeySet* ActiveFilter::GetSitekeys() const |
| 53 { |
| 54 return mSitekeys.get(); |
| 55 } |
| 56 |
| 57 void ActiveFilter::ParseDomains(const String& domains, |
| 58 String::value_type separator) const |
| 59 { |
| 60 DomainMap::size_type count = 2; |
| 61 for (String::size_type i = 0; i < domains.length(); i++) |
| 62 if (domains[i] == separator) |
| 63 count++; |
| 64 |
| 65 mDomains.reset(new DomainMap(count)); |
| 66 annotate_address(mDomains.get(), "DomainMap"); |
| 67 |
| 68 StringScanner scanner(domains, 0, separator); |
| 69 String::size_type start = 0; |
| 70 bool reverse = false; |
| 71 bool hasIncludes = false; |
| 72 bool done = false; |
| 73 while (!done) |
| 74 { |
| 75 done = scanner.done(); |
| 76 String::value_type currChar = scanner.next(); |
| 77 if (currChar == u'~' && scanner.position() == start) |
| 78 { |
| 79 start++; |
| 80 reverse = true; |
| 81 } |
| 82 else if (currChar == separator) |
| 83 { |
| 84 String::size_type len = scanner.position() - start; |
| 85 if (len > 0 && mIgnoreTrailingDot && domains[start + len - 1] == '.') |
| 86 len--; |
| 87 if (len > 0) |
| 88 { |
| 89 enter_context("Adding to ActiveFilter.mDomains"); |
| 90 (*mDomains)[DependentString(domains, start, len)] = !reverse; |
| 91 exit_context(); |
| 92 |
| 93 if (!reverse) |
| 94 hasIncludes = true; |
| 95 } |
| 96 start = scanner.position() + 1; |
| 97 reverse = false; |
| 98 } |
| 99 } |
| 100 enter_context("Adding to ActiveFilter.mDomains"); |
| 101 (*mDomains)[u""_str] = !hasIncludes; |
| 102 exit_context(); |
| 103 } |
| 104 |
| 105 void ActiveFilter::AddSitekey(const String& sitekey) const |
| 106 { |
| 107 if (!mSitekeys) |
| 108 { |
| 109 mSitekeys.reset(new SitekeySet()); |
| 110 annotate_address(mSitekeys.get(), "SitekeySet"); |
| 111 } |
| 112 |
| 113 enter_context("Adding to ActiveFilter.mSitekeys"); |
| 114 mSitekeys->insert(sitekey); |
| 115 exit_context(); |
| 116 } |
| 117 |
| 118 bool ActiveFilter::IsActiveOnDomain(DependentString& docDomain, const String& si
tekey) const |
| 119 { |
| 120 auto sitekeys = GetSitekeys(); |
| 121 if (sitekeys && !sitekeys->find(sitekey)) |
| 122 return false; |
| 123 |
| 124 // If no domains are set the rule matches everywhere |
| 125 auto domains = GetDomains(); |
| 126 if (!domains) |
| 127 return true; |
| 128 |
| 129 // If the document has no host name, match only if the filter isn't restricted |
| 130 // to specific domains |
| 131 if (docDomain.empty()) |
| 132 return (*domains)[u""_str]; |
| 133 |
| 134 ToLower(DependentString(docDomain)); |
| 135 |
| 136 String::size_type len = docDomain.length(); |
| 137 if (len > 0 && mIgnoreTrailingDot && docDomain[len - 1] == '.') |
| 138 docDomain.reset(docDomain, 0, len - 1); |
| 139 while (true) |
| 140 { |
| 141 auto it = domains->find(docDomain); |
| 142 if (it) |
| 143 return it->second; |
| 144 |
| 145 String::size_type nextDot = docDomain.find(u'.'); |
| 146 if (nextDot == docDomain.npos) |
| 147 break; |
| 148 docDomain.reset(docDomain, nextDot + 1); |
| 149 } |
| 150 return (*domains)[u""_str]; |
| 151 } |
| 152 |
| 153 bool ActiveFilter::IsActiveOnlyOnDomain(DependentString& docDomain) const |
| 154 { |
| 155 auto domains = GetDomains(); |
| 156 if (!domains || docDomain.empty() || (*domains)[u""_str]) |
| 157 return false; |
| 158 |
| 159 ToLower(DependentString(docDomain)); |
| 160 |
| 161 String::size_type len = docDomain.length(); |
| 162 if (len > 0 && mIgnoreTrailingDot && docDomain[len - 1] == '.') |
| 163 docDomain.reset(docDomain, 0, len - 1); |
| 164 for (auto it = domains->begin(); it != domains->end(); ++it) |
| 165 { |
| 166 if (!it->second || it->first.equals(docDomain)) |
| 167 continue; |
| 168 |
| 169 size_t len1 = it->first.length(); |
| 170 size_t len2 = docDomain.length(); |
| 171 if (len1 > len2 && |
| 172 DependentString(it->first, len1 - len2).equals(docDomain) && |
| 173 it->first[len1 - len2 - 1] == u'.') |
| 174 { |
| 175 continue; |
| 176 } |
| 177 |
| 178 return false; |
| 179 } |
| 180 return true; |
| 181 } |
| 182 |
| 183 bool ActiveFilter::IsGeneric() const |
| 184 { |
| 185 auto sitekeys = GetSitekeys(); |
| 186 auto domains = GetDomains(); |
| 187 return !sitekeys && (!domains || (*domains)[u""_str]); |
| 188 } |
| 189 |
| 190 OwnedString ActiveFilter::Serialize() const |
| 191 { |
| 192 /* TODO this is very inefficient */ |
| 193 OwnedString result(Filter::Serialize()); |
| 194 if (mDisabled) |
| 195 result.append(u"disabled=true\n"_str); |
| 196 if (mHitCount) |
| 197 { |
| 198 result.append(u"hitCount="_str); |
| 199 result.append(to_string(mHitCount)); |
| 200 result.append(u'\n'); |
| 201 } |
| 202 if (mLastHit) |
| 203 { |
| 204 result.append(u"lastHit="_str); |
| 205 result.append(to_string(mLastHit)); |
| 206 result.append(u'\n'); |
| 207 } |
| 208 return result; |
| 209 } |
OLD | NEW |