From 7adf8c1fa8d519b8b57292763eb7703e7d3cc580 Mon Sep 17 00:00:00 2001 From: David Kalnischkies Date: Mon, 9 May 2022 11:42:48 +0200 Subject: Look at non by-hash paths in copy and file methods Ideally copy and file mirrors would support by-hash as well, but its harder to setup and maintain especially if you want to cache an online mirror who has by-hash enabled. We can avoid unneeded roundtrips (in the best case) and entire cache misses (in the usual worst case) by "just" telling methods that the URI we passed it has the requested file perhaps also in other paths. This is done in pseudo-relative paths as we would otherwise need to teach redirection code to rewrite those URIs as well. A method like http can easily ignore this value and await explicit instructions to look at that file, but inspecting the path in local sources via file or copy is (comparatively) free, so we just do it immediately. If that ends up being the wrong version of the file as by-hash would have protected us from we are in this feature branch now falling back to other mirrors which are like the ones online and in support of by-hash. --- methods/file.cc | 111 +++++++++++++++++++++++++++++++------------------------- 1 file changed, 62 insertions(+), 49 deletions(-) (limited to 'methods/file.cc') diff --git a/methods/file.cc b/methods/file.cc index ff93f83d0..279711195 100644 --- a/methods/file.cc +++ b/methods/file.cc @@ -29,7 +29,7 @@ class FileMethod : public aptMethod { - virtual bool Fetch(FetchItem *Itm) APT_OVERRIDE; + bool URIAcquire(std::string const &Message, FetchItem *Itm) APT_OVERRIDE; public: FileMethod() : aptMethod("file", "1.0", SingleInstance | SendConfig | LocalOnly | SendURIEncoded) @@ -41,88 +41,101 @@ class FileMethod : public aptMethod // FileMethod::Fetch - Fetch a file /*{{{*/ // --------------------------------------------------------------------- /* */ -bool FileMethod::Fetch(FetchItem *Itm) +bool FileMethod::URIAcquire(std::string const &Message, FetchItem *Itm) { URI Get(Itm->Uri); - std::string const File = DecodeSendURI(Get.Path); - FetchResult Res; if (Get.Host.empty() == false) return _error->Error(_("Invalid URI, local URIS must not start with //")); - struct stat Buf; + std::vector Files; + Files.emplace_back(DecodeSendURI(Get.Path)); + for (auto const &AltPath : VectorizeString(LookupTag(Message, "Alternate-Paths"), '\n')) + Files.emplace_back(CombineWithAlternatePath(flNotFile(Files[0]), DecodeSendURI(AltPath))); + + FetchResult Res; // deal with destination files which might linger around - if (lstat(Itm->DestFile.c_str(), &Buf) == 0) + struct stat Buf; + if (lstat(Itm->DestFile.c_str(), &Buf) == 0 && S_ISLNK(Buf.st_mode) && Buf.st_size > 0) + { + char name[Buf.st_size + 1]; + if (ssize_t const sp = readlink(Itm->DestFile.c_str(), name, Buf.st_size); sp == -1) + { + Itm->LastModified = 0; + RemoveFile("file", Itm->DestFile); + } + } + + int olderrno = 0; + // See if the file exists + for (auto const &File : Files) { - if ((Buf.st_mode & S_IFREG) != 0) + if (stat(File.c_str(), &Buf) == 0) { + Res.Size = Buf.st_size; + Res.Filename = File; + Res.LastModified = Buf.st_mtime; + Res.IMSHit = false; if (Itm->LastModified == Buf.st_mtime && Itm->LastModified != 0) { - if (Itm->ExpectedHashes.VerifyFile(File)) - { - Res.Filename = Itm->DestFile; + auto const filesize = Itm->ExpectedHashes.FileSize(); + if (filesize != 0 && filesize == Res.Size) Res.IMSHit = true; - } } + break; } + if (olderrno == 0) + olderrno = errno; } - if (Res.IMSHit != true) - RemoveFile("file", Itm->DestFile); - int olderrno = 0; - // See if the file exists - if (stat(File.c_str(),&Buf) == 0) + if (not Res.IMSHit) { - Res.Size = Buf.st_size; - Res.Filename = File; - Res.LastModified = Buf.st_mtime; - Res.IMSHit = false; - if (Itm->LastModified == Buf.st_mtime && Itm->LastModified != 0) + RemoveFile("file", Itm->DestFile); + if (not Res.Filename.empty()) { - unsigned long long const filesize = Itm->ExpectedHashes.FileSize(); - if (filesize != 0 && filesize == Res.Size) - Res.IMSHit = true; + URIStart(Res); + CalculateHashes(Itm, Res); } - - CalculateHashes(Itm, Res); } - else - olderrno = errno; - if (Res.IMSHit == false) - URIStart(Res); // See if the uncompressed file exists and reuse it FetchResult AltRes; - AltRes.Filename.clear(); - std::vector extensions = APT::Configuration::getCompressorExtensions(); - for (std::vector::const_iterator ext = extensions.begin(); ext != extensions.end(); ++ext) + for (auto const &File : Files) { - if (APT::String::Endswith(File, *ext) == true) + for (const auto &ext : APT::Configuration::getCompressorExtensions()) { - std::string const unfile = File.substr(0, File.length() - ext->length()); - if (stat(unfile.c_str(),&Buf) == 0) + if (APT::String::Endswith(File, ext)) { - AltRes.Size = Buf.st_size; - AltRes.Filename = unfile; - AltRes.LastModified = Buf.st_mtime; - AltRes.IMSHit = false; - if (Itm->LastModified == Buf.st_mtime && Itm->LastModified != 0) - AltRes.IMSHit = true; - if (Res.Filename.empty()) - CalculateHashes(Itm, AltRes); - break; + std::string const unfile = File.substr(0, File.length() - ext.length()); + if (stat(unfile.c_str(), &Buf) == 0) + { + AltRes.Size = Buf.st_size; + AltRes.Filename = unfile; + AltRes.LastModified = Buf.st_mtime; + AltRes.IMSHit = false; + if (Itm->LastModified == Buf.st_mtime && Itm->LastModified != 0) + AltRes.IMSHit = true; + if (Res.Filename.empty()) + { + URIStart(Res); + CalculateHashes(Itm, AltRes); + } + break; + } + // no break here as we could have situations similar to '.gz' vs '.tar.gz' here } - // no break here as we could have situations similar to '.gz' vs '.tar.gz' here } + if (not AltRes.Filename.empty()) + break; } - if (AltRes.Filename.empty() == false) + if (not AltRes.Filename.empty()) URIDone(Res,&AltRes); - else if (Res.Filename.empty() == false) + else if (not Res.Filename.empty()) URIDone(Res); else { errno = olderrno; - return _error->Errno(File.c_str(), _("File not found")); + return _error->Errno(Files[0].c_str(), _("File not found")); } return true; -- cgit v1.2.3-70-g09d2