/* -*- Mode: C++; tab-width: 4; indent-tabs-mode: nil; c-basic-offset: 4 -*- */ /* * This file is part of the LibreOffice project. * * This Source Code Form is subject to the terms of the Mozilla Public * License, v. 2.0. If a copy of the MPL was not distributed with this * file, You can obtain one at http://mozilla.org/MPL/2.0/. * * This file incorporates work covered by the following license notice: * * Licensed to the Apache Software Foundation (ASF) under one or more * contributor license agreements. See the NOTICE file distributed * with this work for additional information regarding copyright * ownership. The ASF licenses this file to you under the Apache * License, Version 2.0 (the "License"); you may not use this file * except in compliance with the License. You may obtain a copy of * the License at http://www.apache.org/licenses/LICENSE-2.0 . */ #include #include #include #include #include #include #include #include #include #include #include #include #include #include #include #include #include #include #include #include #include #include #include #include #include #include #include #include #include #include using namespace com::sun::star; OUString URIHelper::SmartRel2Abs(INetURLObject const & rTheBaseURIRef, OUString const & rTheRelURIRef, Link const & rMaybeFileHdl, bool bCheckFileExists, bool bIgnoreFragment, INetURLObject::EncodeMechanism eEncodeMechanism, INetURLObject::DecodeMechanism eDecodeMechanism, rtl_TextEncoding eCharset, FSysStyle eStyle) { // Backwards compatibility: if( rTheRelURIRef.startsWith("#") ) return rTheRelURIRef; INetURLObject aAbsURIRef; if (rTheBaseURIRef.HasError()) aAbsURIRef. SetSmartURL(rTheRelURIRef, eEncodeMechanism, eCharset, eStyle); else { bool bWasAbsolute; aAbsURIRef = rTheBaseURIRef.smartRel2Abs(rTheRelURIRef, bWasAbsolute, bIgnoreFragment, eEncodeMechanism, eCharset, false/*bRelativeNonURIs*/, eStyle); if (bCheckFileExists && !bWasAbsolute && (aAbsURIRef.GetProtocol() == INetProtocol::File)) { INetURLObject aNonFileURIRef; aNonFileURIRef.SetSmartURL(rTheRelURIRef, eEncodeMechanism, eCharset, eStyle); if (!aNonFileURIRef.HasError() && aNonFileURIRef.GetProtocol() != INetProtocol::File) { bool bMaybeFile = false; if (rMaybeFileHdl.IsSet()) { OUString aFilePath(rTheRelURIRef); bMaybeFile = rMaybeFileHdl.Call(&aFilePath); } if (!bMaybeFile) aAbsURIRef = aNonFileURIRef; } } } return aAbsURIRef.GetMainURL(eDecodeMechanism, eCharset); } namespace { Link gMaybeFileHdl; } void URIHelper::SetMaybeFileHdl(Link const & rTheMaybeFileHdl) { gMaybeFileHdl = rTheMaybeFileHdl; } Link const & URIHelper::GetMaybeFileHdl() { return gMaybeFileHdl; } namespace { bool isAbsoluteHierarchicalUriReference( css::uno::Reference< css::uri::XUriReference > const & uriReference) { return uriReference.is() && uriReference->isAbsolute() && !uriReference->hasRelativePath(); } // To improve performance, assume that if for any prefix URL of a given // hierarchical URL either a UCB content cannot be created, or the UCB content // does not support the getCasePreservingURL command, then this will hold for // any other prefix URL of the given URL, too: enum Result { Success, GeneralFailure, SpecificFailure }; Result normalizePrefix( css::uno::Reference< css::ucb::XUniversalContentBroker > const & broker, OUString const & uri, OUString * normalized) { OSL_ASSERT(broker.is() && normalized != nullptr); css::uno::Reference< css::ucb::XContent > content; try { content = broker->queryContent(broker->createContentIdentifier(uri)); } catch (css::ucb::IllegalIdentifierException &) {} if (!content.is()) { return GeneralFailure; } try { bool ok = (css::uno::Reference< css::ucb::XCommandProcessor >( content, css::uno::UNO_QUERY_THROW)->execute( css::ucb::Command("getCasePreservingURL", -1, css::uno::Any()), 0, css::uno::Reference< css::ucb::XCommandEnvironment >()) >>= *normalized); OSL_ASSERT(ok); } catch (css::uno::RuntimeException &) { throw; } catch (css::ucb::UnsupportedCommandException &) { return GeneralFailure; } catch (css::uno::Exception &) { return SpecificFailure; } return Success; } OUString normalize( css::uno::Reference< css::ucb::XUniversalContentBroker > const & broker, css::uno::Reference< css::uri::XUriReferenceFactory > const & uriFactory, OUString const & uriReference) { // normalizePrefix can potentially fail (a typically example being a file // URL that denotes a non-existing resource); in such a case, try to // normalize as long a prefix of the given URL as possible (i.e., normalize // all the existing directories within the path): OUString normalized; sal_Int32 n = uriReference.indexOf('#'); normalized = n == -1 ? uriReference : uriReference.copy(0, n); switch (normalizePrefix(broker, normalized, &normalized)) { case Success: return n == -1 ? normalized : normalized + uriReference.subView(n); case GeneralFailure: return uriReference; case SpecificFailure: default: break; } css::uno::Reference< css::uri::XUriReference > ref( uriFactory->parse(uriReference)); if (!isAbsoluteHierarchicalUriReference(ref)) { return uriReference; } sal_Int32 count = ref->getPathSegmentCount(); if (count < 2) { return uriReference; } OUStringBuffer head(ref->getScheme()); head.append(':'); if (ref->hasAuthority()) { head.append("//"); head.append(ref->getAuthority()); } for (sal_Int32 i = count - 1; i > 0; --i) { OUStringBuffer buf(head); for (sal_Int32 j = 0; j < i; ++j) { buf.append('/'); buf.append(ref->getPathSegment(j)); } normalized = buf.makeStringAndClear(); if (normalizePrefix(broker, normalized, &normalized) != SpecificFailure) { buf.append(normalized); css::uno::Reference< css::uri::XUriReference > preRef( uriFactory->parse(normalized)); if (!isAbsoluteHierarchicalUriReference(preRef)) { // This could only happen if something is inconsistent: break; } sal_Int32 preCount = preRef->getPathSegmentCount(); // normalizePrefix may have added or removed a final slash: if (preCount != i) { if (preCount == i - 1) { buf.append('/'); } else if (preCount - 1 == i && !buf.isEmpty() && buf[buf.getLength() - 1] == '/') { buf.setLength(buf.getLength() - 1); } else { // This could only happen if something is inconsistent: break; } } for (sal_Int32 j = i; j < count; ++j) { buf.append('/'); buf.append(ref->getPathSegment(j)); } if (ref->hasQuery()) { buf.append('?'); buf.append(ref->getQuery()); } if (ref->hasFragment()) { buf.append('#'); buf.append(ref->getFragment()); } return buf.makeStringAndClear(); } } return uriReference; } } css::uno::Reference< css::uri::XUriReference > URIHelper::normalizedMakeRelative( css::uno::Reference< css::uno::XComponentContext > const & context, OUString const & baseUriReference, OUString const & uriReference) { OSL_ASSERT(context.is()); css::uno::Reference< css::ucb::XUniversalContentBroker > broker( css::ucb::UniversalContentBroker::create(context)); css::uno::Reference< css::uri::XUriReferenceFactory > uriFactory( css::uri::UriReferenceFactory::create(context)); return uriFactory->makeRelative( uriFactory->parse(normalize(broker, uriFactory, baseUriReference)), uriFactory->parse(normalize(broker, uriFactory, uriReference)), true, true, false); } OUString URIHelper::simpleNormalizedMakeRelative( OUString const & baseUriReference, OUString const & uriReference) { css::uno::Reference< css::uri::XUriReference > rel( URIHelper::normalizedMakeRelative( comphelper::getProcessComponentContext(), baseUriReference, uriReference)); return rel.is() ? rel->getUriReference() : uriReference; } // FindFirstURLInText namespace { sal_Int32 nextChar(std::u16string_view rStr, sal_Int32 nPos) { return rtl::isHighSurrogate(rStr[nPos]) && rStr.size() - nPos >= 2 && rtl::isLowSurrogate(rStr[nPos + 1]) ? nPos + 2 : nPos + 1; } bool isBoundary1(CharClass const & rCharClass, OUString const & rStr, sal_Int32 nPos, sal_Int32 nEnd) { if (nPos == nEnd) return true; if (rCharClass.isLetterNumeric(rStr, nPos)) return false; switch (rStr[nPos]) { case '$': case '%': case '&': case '-': case '/': case '@': case '\\': return false; default: return true; } } bool isBoundary2(CharClass const & rCharClass, OUString const & rStr, sal_Int32 nPos, sal_Int32 nEnd) { if (nPos == nEnd) return true; if (rCharClass.isLetterNumeric(rStr, nPos)) return false; switch (rStr[nPos]) { case '!': case '#': case '$': case '%': case '&': case '\'': case '*': case '+': case '-': case '/': case '=': case '?': case '@': case '^': case '_': case '`': case '{': case '|': case '}': case '~': return false; default: return true; } } // tdf#145381 Added MatchingBracketDepth counter to detect matching closing // brackets that are part of the uri bool checkWChar(CharClass const & rCharClass, OUString const & rStr, sal_Int32 * pPos, sal_Int32 * pEnd, sal_Int32 * pMatchingBracketDepth = nullptr, bool bBackslash = false, bool bPipe = false) { sal_Unicode c = rStr[*pPos]; if (rtl::isAscii(c)) { static sal_uInt8 const aMap[128] = { 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 0, 0, 4, 4, 4, 1, // !"#$%&' 5, 6, 1, 1, 1, 4, 1, 4, // ()*+,-./ 4, 4, 4, 4, 4, 4, 4, 4, // 01234567 4, 4, 1, 1, 0, 1, 0, 1, // 89:;<=>? 4, 4, 4, 4, 4, 4, 4, 4, // @ABCDEFG 4, 4, 4, 4, 4, 4, 4, 4, // HIJKLMNO 4, 4, 4, 4, 4, 4, 4, 4, // PQRSTUVW 4, 4, 4, 1, 2, 1, 0, 1, // XYZ[\]^_ 0, 4, 4, 4, 4, 4, 4, 4, // `abcdefg 4, 4, 4, 4, 4, 4, 4, 4, // hijklmno 4, 4, 4, 4, 4, 4, 4, 4, // pqrstuvw 4, 4, 4, 0, 3, 0, 1, 0 }; // xyz{|}~ switch (aMap[c]) { default: // not uric return false; case 1: // uric ++(*pPos); return true; case 2: // "\" if (bBackslash) { *pEnd = ++(*pPos); return true; } else return false; case 3: // "|" if (bPipe) { *pEnd = ++(*pPos); return true; } else return false; case 4: // alpha, digit, "$", "%", "&", "-", "/", "@" (see // isBoundary1) *pEnd = ++(*pPos); return true; case 5: // opening bracket ++(*pPos); if(nullptr != pMatchingBracketDepth) ++(*pMatchingBracketDepth); return true; case 6: // closing bracket ++(*pPos); if(nullptr != pMatchingBracketDepth && *pMatchingBracketDepth > 0) { --(*pMatchingBracketDepth); // tdf#145381 When there was an opening bracket, detect this closing bracket // as part of the uri *pEnd = *pPos; } return true; } } else if (rCharClass.isLetterNumeric(rStr, *pPos)) { *pEnd = *pPos = nextChar(rStr, *pPos); return true; } else return false; } sal_uInt32 scanDomain(OUString const & rStr, sal_Int32 * pPos, sal_Int32 nEnd) { sal_Unicode const * pBuffer = rStr.getStr(); sal_Unicode const * p = pBuffer + *pPos; sal_uInt32 nLabels = INetURLObject::scanDomain(p, pBuffer + nEnd, false); *pPos = sal::static_int_cast< sal_Int32 >(p - pBuffer); return nLabels; } } OUString URIHelper::FindFirstURLInText(OUString const & rText, sal_Int32 & rBegin, sal_Int32 & rEnd, CharClass const & rCharClass, INetURLObject::EncodeMechanism eMechanism, rtl_TextEncoding eCharset) { if (rBegin > rEnd || rEnd > rText.getLength()) return OUString(); // Search for the first substring of [rBegin..rEnd[ that matches any of the // following productions (for which the appropriate style bit is set in // eStyle, if applicable). // 1st Production (known scheme): // \B1 ":" 1*wchar ["#" 1*wchar] // \B1 // 2nd Production (file): // \B1 "FILE:" 1*(wchar / "\" / "|") ["#" 1*wchar] \B1 // 3rd Production (ftp): // \B1 "FTP" 2*("." label) ["/" *wchar] ["#" 1*wchar] \B1 // 4th Production (http): // \B1 "WWW" 2*("." label) ["/" *wchar] ["#" 1*wchar] \B1 // 5th Production (mailto): // \B2 local-part "@" domain \B1 // 6th Production (UNC file): // \B1 "\\" domain "\" *(wchar / "\") \B1 // 7th Production (DOS file): // \B1 ALPHA ":\" *(wchar / "\") \B1 // 8th Production (Unix-like DOS file): // \B1 ALPHA ":/" *(wchar / "\") \B1 // The productions use the following auxiliary rules. // local-part = atom *("." atom) // atom = 1*(alphanum / "!" / "#" / "$" / "%" / "&" / "'" / "*" / "+" // / "-" / "/" / "=" / "?" / "^" / "_" / "`" / "{" / "|" / "}" // / "~") // domain = label *("." label) // label = alphanum [*(alphanum / "-") alphanum] // alphanum = ALPHA / DIGIT // wchar = // "\B1" (boundary 1) stands for the beginning or end of the block of text, // or a character that is neither (a) a letter or digit (according to // rCharClass), nor (b) any of "$", "%", "&", "-", "/", "@", or "\". // (FIXME: What was the rationale for this set of punctuation characters?) // "\B2" (boundary 2) stands for the beginning or end of the block of text, // or a character that is neither (a) a letter or digit (according to // rCharClass), nor (b) any of "!", "#", "$", "%", "&", "'", "*", "+", "-", // "/", "=", "?", "@", "^", "_", "`", "{", "|", "}", or "~" (i.e., an RFC // 822 character, or "@" from \B1's set above). // Productions 1--4, and 6--8 try to find a maximum-length match, but they // stop at the first character that is a "\B1" character which is // only followed by "\B1" characters (taking "\" and "|" characters into // account appropriately). Production 5 simply tries to find a maximum- // length match. // Productions 1--4 use the given eMechanism and eCharset. Productions 5--9 // use EncodeMechanism::All. // Productions 6--9 are only applicable if the FSysStyle::Dos bit is set in // eStyle. // tdf#145381: In addition to the productions I added a mechanism to detect // matching brackets. The task presents the case of an url that ends on a // closing bracket. This needs to be detected as part of the uri in the case // that a matching opening bracket exists. bool bBoundary1 = true; bool bBoundary2 = true; for (sal_Int32 nPos = rBegin; nPos != rEnd; nPos = nextChar(rText, nPos)) { sal_Unicode c = rText[nPos]; if (bBoundary1) { if (rtl::isAsciiAlpha(c)) { sal_Int32 i = nPos; INetProtocol eScheme = INetURLObject::CompareProtocolScheme(rText.subView(i, rEnd - i)); if (eScheme == INetProtocol::File) // 2nd { while (rText[i++] != ':') ; sal_Int32 nPrefixEnd = i; sal_Int32 nUriEnd = i; while (i != rEnd && checkWChar(rCharClass, rText, &i, &nUriEnd, nullptr, true, true)) ; if (i != nPrefixEnd && i != rEnd && rText[i] == '#') { ++i; while (i != rEnd && checkWChar(rCharClass, rText, &i, &nUriEnd)) ; } if (nUriEnd != nPrefixEnd && isBoundary1(rCharClass, rText, nUriEnd, rEnd)) { INetURLObject aUri(rText.subView(nPos, nUriEnd - nPos), INetProtocol::File, eMechanism, eCharset, FSysStyle::Detect); if (!aUri.HasError()) { rBegin = nPos; rEnd = nUriEnd; return aUri.GetMainURL(INetURLObject::DecodeMechanism::ToIUri); } } } else if (eScheme != INetProtocol::NotValid) // 1st { while (rText[i++] != ':') ; sal_Int32 nPrefixEnd = i; sal_Int32 nUriEnd = i; sal_Int32 nMatchingBracketDepth = 0; while (i != rEnd && checkWChar(rCharClass, rText, &i, &nUriEnd, &nMatchingBracketDepth)) ; if (i != nPrefixEnd && i != rEnd && rText[i] == '#') { ++i; while (i != rEnd && checkWChar(rCharClass, rText, &i, &nUriEnd)) ; } if (nUriEnd != nPrefixEnd && (isBoundary1(rCharClass, rText, nUriEnd, rEnd) || rText[nUriEnd] == '\\')) { INetURLObject aUri(rText.subView(nPos, nUriEnd - nPos), INetProtocol::Http, eMechanism, eCharset); if (!aUri.HasError()) { rBegin = nPos; rEnd = nUriEnd; return aUri.GetMainURL(INetURLObject::DecodeMechanism::ToIUri); } } } // 3rd, 4th: i = nPos; sal_uInt32 nLabels = scanDomain(rText, &i, rEnd); if (nLabels >= 3 && rText[nPos + 3] == '.' && (((rText[nPos] == 'w' || rText[nPos] == 'W') && (rText[nPos + 1] == 'w' || rText[nPos + 1] == 'W') && (rText[nPos + 2] == 'w' || rText[nPos + 2] == 'W')) || ((rText[nPos] == 'f' || rText[nPos] == 'F') && (rText[nPos + 1] == 't' || rText[nPos + 1] == 'T') && (rText[nPos + 2] == 'p' || rText[nPos + 2] == 'P')))) // (note that rText.GetChar(nPos + 3) is guaranteed to be // valid) { sal_Int32 nUriEnd = i; if (i != rEnd && rText[i] == '/') { nUriEnd = ++i; while (i != rEnd && checkWChar(rCharClass, rText, &i, &nUriEnd)) ; } if (i != rEnd && rText[i] == '#') { ++i; while (i != rEnd && checkWChar(rCharClass, rText, &i, &nUriEnd)) ; } if (isBoundary1(rCharClass, rText, nUriEnd, rEnd) || rText[nUriEnd] == '\\') { INetURLObject aUri(rText.subView(nPos, nUriEnd - nPos), INetProtocol::Http, eMechanism, eCharset); if (!aUri.HasError()) { rBegin = nPos; rEnd = nUriEnd; return aUri.GetMainURL(INetURLObject::DecodeMechanism::ToIUri); } } } if (rEnd - nPos >= 3 && rText[nPos + 1] == ':' && (rText[nPos + 2] == '/' || rText[nPos + 2] == '\\')) // 7th, 8th { i = nPos + 3; sal_Int32 nUriEnd = i; while (i != rEnd && checkWChar(rCharClass, rText, &i, &nUriEnd)) ; if (isBoundary1(rCharClass, rText, nUriEnd, rEnd)) { INetURLObject aUri(rText.subView(nPos, nUriEnd - nPos), INetProtocol::File, INetURLObject::EncodeMechanism::All, RTL_TEXTENCODING_UTF8, FSysStyle::Dos); if (!aUri.HasError()) { rBegin = nPos; rEnd = nUriEnd; return aUri.GetMainURL(INetURLObject::DecodeMechanism::ToIUri); } } } } else if (rEnd - nPos >= 2 && rText[nPos] == '\\' && rText[nPos + 1] == '\\') // 6th { sal_Int32 i = nPos + 2; sal_uInt32 nLabels = scanDomain(rText, &i, rEnd); if (nLabels >= 1 && i != rEnd && rText[i] == '\\') { sal_Int32 nUriEnd = ++i; while (i != rEnd && checkWChar(rCharClass, rText, &i, &nUriEnd, nullptr, true)) ; if (isBoundary1(rCharClass, rText, nUriEnd, rEnd)) { INetURLObject aUri(rText.subView(nPos, nUriEnd - nPos), INetProtocol::File, INetURLObject::EncodeMechanism::All, RTL_TEXTENCODING_UTF8, FSysStyle::Dos); if (!aUri.HasError()) { rBegin = nPos; rEnd = nUriEnd; return aUri.GetMainURL(INetURLObject::DecodeMechanism::ToIUri); } } } } } if (bBoundary2 && INetMIME::isAtomChar(c)) // 5th { bool bDot = false; for (sal_Int32 i = nPos + 1; i != rEnd; ++i) { sal_Unicode c2 = rText[i]; if (INetMIME::isAtomChar(c2)) bDot = false; else if (bDot) break; else if (c2 == '.') bDot = true; else { if (c2 == '@') { ++i; sal_uInt32 nLabels = scanDomain(rText, &i, rEnd); if (nLabels >= 1 && isBoundary1(rCharClass, rText, i, rEnd)) { INetURLObject aUri(rText.subView(nPos, i - nPos), INetProtocol::Mailto, INetURLObject::EncodeMechanism::All); if (!aUri.HasError()) { rBegin = nPos; rEnd = i; return aUri.GetMainURL( INetURLObject::DecodeMechanism::ToIUri); } } } break; } } } bBoundary1 = isBoundary1(rCharClass, rText, nPos, rEnd); bBoundary2 = isBoundary2(rCharClass, rText, nPos, rEnd); } rBegin = rEnd; return OUString(); } OUString URIHelper::removePassword(OUString const & rURI, INetURLObject::EncodeMechanism eEncodeMechanism, INetURLObject::DecodeMechanism eDecodeMechanism, rtl_TextEncoding eCharset) { INetURLObject aObj(rURI, eEncodeMechanism, eCharset); return aObj.HasError() ? rURI : aObj.GetURLNoPass(eDecodeMechanism, eCharset); } OUString URIHelper::resolveIdnaHost(OUString const & url) { css::uno::Reference uri( css::uri::UriReferenceFactory::create( comphelper::getProcessComponentContext()) ->parse(url)); if (!(uri.is() && uri->hasAuthority())) { return url; } auto auth(uri->getAuthority()); if (auth.isEmpty()) return url; sal_Int32 hostStart = auth.indexOf('@') + 1; sal_Int32 hostEnd = auth.getLength(); while (hostEnd > hostStart && rtl::isAsciiDigit(auth[hostEnd - 1])) { --hostEnd; } if (hostEnd > hostStart && auth[hostEnd - 1] == ':') { --hostEnd; } else { hostEnd = auth.getLength(); } auto asciiOnly = true; for (auto i = hostStart; i != hostEnd; ++i) { if (!rtl::isAscii(auth[i])) { asciiOnly = false; break; } } if (asciiOnly) { // Avoid icu::IDNA case normalization in purely non-IDNA domain names: return url; } UErrorCode e = U_ZERO_ERROR; std::unique_ptr idna( icu::IDNA::createUTS46Instance( (UIDNA_USE_STD3_RULES | UIDNA_CHECK_BIDI | UIDNA_CHECK_CONTEXTJ #if U_ICU_VERSION_MAJOR_NUM >= 49 | UIDNA_CHECK_CONTEXTO #endif ), e)); if (U_FAILURE(e)) { SAL_WARN("vcl.gdi", "icu::IDNA::createUTS46Instance " << e); return url; } icu::UnicodeString ascii; icu::IDNAInfo info; idna->nameToASCII( icu::UnicodeString( reinterpret_cast(auth.getStr() + hostStart), hostEnd - hostStart), ascii, info, e); if (U_FAILURE(e) || info.hasErrors()) { return url; } OUStringBuffer buf(uri->getScheme()); buf.append(OUString::Concat("://") + auth.subView(0, hostStart)); buf.append( reinterpret_cast(ascii.getBuffer()), ascii.length()); buf.append(auth.subView(hostEnd) + uri->getPath()); if (uri->hasQuery()) { buf.append("?" + uri->getQuery()); } if (uri->hasFragment()) { buf.append("#" + uri->getFragment()); } return buf.makeStringAndClear(); } /* vim:set shiftwidth=4 softtabstop=4 expandtab: */