diff --git a/.gitignore b/.gitignore index b292a69..0808890 100644 --- a/.gitignore +++ b/.gitignore @@ -2,6 +2,8 @@ src/magneticod/vendor src/magneticod/Gopkg.lock src/magneticow/vendor src/magneticow/Gopkg.lock +src/persistence/vendor +src/persistence/Gopkg.lock # Created by https://www.gitignore.io/api/linux,python,pycharm diff --git a/.travis.yml b/.travis.yml index ab96d01..fc167ea 100644 --- a/.travis.yml +++ b/.travis.yml @@ -1,15 +1,16 @@ sudo: enabled dist: xenial -language: python -python: - - "3.5" -# versions 3.6.0 and 3.6.1 have bugs that affect magneticod +language: go +go: + - 1.8.3 before_install: - "sudo apt-get update -qq" - "sudo apt-get install python3-dev" - "pip3 install mypy pylint" + - "" + install: - "pip3 install ./magneticod" - "pip3 install ./magneticow" diff --git a/magneticow/magneticow/static/styles/homepage.css b/magneticow/magneticow/static/styles/homepage.css deleted file mode 100644 index 384d922..0000000 --- a/magneticow/magneticow/static/styles/homepage.css +++ /dev/null @@ -1,27 +0,0 @@ -main { - display: flex; - align-items: center; - align-content: center; - - height: calc(100vh - 2 * 3em); - width: calc(100vw - 2 * 3em); -} - -main form { - max-width: 600px; - width: 100%; - - margin-left: 0.5em; -} - -main form input { - width: 100%; -} - -main > div { - margin-right: 0.5em; -} - -footer { - margin-top: 0.833em; -} diff --git a/magneticow/magneticow/templates/homepage.html b/magneticow/magneticow/templates/homepage.html deleted file mode 100644 index b6d19d8..0000000 --- a/magneticow/magneticow/templates/homepage.html +++ /dev/null @@ -1,23 +0,0 @@ - - - - - magneticow - - - - - - -
-
magneticow​(pre-alpha)
-
- -
-
- - - - diff --git a/magneticow/magneticow/templates/torrents.html b/magneticow/magneticow/templates/torrents.html deleted file mode 100644 index 8dc543a..0000000 --- a/magneticow/magneticow/templates/torrents.html +++ /dev/null @@ -1,90 +0,0 @@ - - - - - {% if search %}"{{search}}"{% else %}Most recent torrents{% endif %} - magneticow - - - - - - -
-
magneticow​(pre-alpha)
-
- -
-
- feed icon subscribe -
-
-
- - - - - - - - - - - {% for torrent in torrents %} - - - - - - - {% endfor %} - -
- {% if sorted_by == "name ASC" %} - Name ▲ - {% elif sorted_by == "name DESC" %} - Name ▼ - {% else %} - Name - {% endif %} - - {% if sorted_by == "total_size ASC" %} - Size ▲ - {% elif sorted_by == "total_size DESC" %} - Size ▼ - {% else %} - Size - {% endif %} - - {% if sorted_by == "discovered_on ASC" %} - Discovered on ▲ - {% elif sorted_by == "discovered_on DESC" %} - Discovered on ▼ - {% else %} - Discovered on - {% endif %} -
- Magnet link{{ torrent.name }}{{ torrent.size }}{{ torrent.discovered_on }}
-
- - - \ No newline at end of file diff --git a/src/magneticod/bittorrent/operations.go b/src/magneticod/bittorrent/operations.go index c9bcf78..845658c 100644 --- a/src/magneticod/bittorrent/operations.go +++ b/src/magneticod/bittorrent/operations.go @@ -8,6 +8,9 @@ import ( "github.com/anacrolix/torrent" "github.com/anacrolix/torrent/metainfo" "go.uber.org/zap" + "os" + "path" + "persistence" ) @@ -40,48 +43,46 @@ func (ms *MetadataSink) awaitMetadata(infoHash metainfo.Hash, peer torrent.Peer) return } - var files []metainfo.FileInfo - if len(info.Files) == 0 { - if strings.ContainsRune(info.Name, '/') { - // A single file torrent cannot have any '/' characters in its name. We treat it as - // illegal. - zap.L().Sugar().Debugf("!!!! illegal character in name! \"%s\"", info.Name) - return - } - files = []metainfo.FileInfo{{Length: info.Length, Path:[]string{info.Name}}} - } else { - // TODO: We have to make sure that anacrolix/torrent checks for '/' character in file paths - // before concatenating them. This is currently assumed here. We should write a test for it. - files = info.Files + var files []persistence.File + for _, file := range info.Files { + files = append(files, persistence.File{ + Size: file.Length, + Path: file.DisplayPath(info), + }) } var totalSize uint64 for _, file := range files { - if file.Length < 0 { + if file.Size < 0 { // All files' sizes must be greater than or equal to zero, otherwise treat them as // illegal and ignore. - zap.L().Sugar().Debugf("!!!! file size zero or less! \"%s\" (%d)", file.Path, file.Length) + zap.L().Sugar().Debugf("!!!! file size zero or less! \"%s\" (%d)", file.Path, file.Size) return } - totalSize += uint64(file.Length) + totalSize += uint64(file.Size) } ms.flush(Metadata{ - InfoHash: infoHash[:], - Name: info.Name, - TotalSize: totalSize, + InfoHash: infoHash[:], + Name: info.Name, + TotalSize: totalSize, DiscoveredOn: time.Now().Unix(), - Files: files, + Files: files, + Peers: nil, }) } -func (fs *FileSink) awaitFile(infoHash []byte, filePath string, peer *torrent.Peer) { +func (fs *FileSink) awaitFile(req *FileRequest) { + // Remove the download directory of the torrent after the operation is completed. + // TODO: what if RemoveAll() returns error, do we care, and if we do, how to handle it? + defer os.RemoveAll(path.Join(fs.baseDownloadDir, string(req.InfoHash))) + var infoHash_ [20]byte - copy(infoHash_[:], infoHash) + copy(infoHash_[:], req.InfoHash) t, isNew := fs.client.AddTorrentInfoHash(infoHash_) - if peer != nil { - t.AddPeers([]torrent.Peer{*peer}) + if len(req.Peers) > 0 { + t.AddPeers(req.Peers) } if !isNew { // Return immediately if we are trying to await on an ongoing file-downloading operation. @@ -90,7 +91,7 @@ func (fs *FileSink) awaitFile(infoHash []byte, filePath string, peer *torrent.Pe } // Setup & start the timeout timer. - timeout := time.After(fs.timeout) + timeout := time.After(fs.timeoutDuration) // Once we return from this function, drop the torrent from the client. // TODO: Check if dropping a torrent also cancels any outstanding read operations? @@ -105,7 +106,7 @@ func (fs *FileSink) awaitFile(infoHash []byte, filePath string, peer *torrent.Pe var match *torrent.File for _, file := range t.Files() { - if file.Path() == filePath { + if file.Path() == req.Path { match = &file } else { file.Cancel() @@ -117,13 +118,12 @@ func (fs *FileSink) awaitFile(infoHash []byte, filePath string, peer *torrent.Pe zap.L().Warn( "The leech (FileSink) has been requested to download a file which does not exist!", - zap.ByteString("torrent", infoHash), - zap.String("requestedFile", filePath), + zap.ByteString("torrent", req.InfoHash), + zap.String("requestedFile", req.Path), zap.Strings("allFiles", filePaths), ) } - reader := t.NewReader() defer reader.Close() @@ -133,18 +133,17 @@ func (fs *FileSink) awaitFile(infoHash []byte, filePath string, peer *torrent.Pe select { case fileData := <-fileDataChan: if fileData != nil { - fs.flush(File{ - torrentInfoHash: infoHash, - path: match.Path(), - data: fileData, + fs.flush(FileResult{ + Request: req, + FileData: fileData, }) } case <- timeout: zap.L().Debug( "Timeout while downloading a file!", - zap.ByteString("torrent", infoHash), - zap.String("file", filePath), + zap.ByteString("torrent", req.InfoHash), + zap.String("file", req.Path), ) } } @@ -156,9 +155,10 @@ func downloadFile(file torrent.File, reader *torrent.Reader, fileDataChan chan<- fileData := make([]byte, file.Length()) n, err := readSeeker.Read(fileData) if int64(n) != file.Length() { + infoHash := file.Torrent().InfoHash() zap.L().Debug( "Not all of a file could be read!", - zap.ByteString("torrent", file.Torrent().InfoHash()[:]), + zap.ByteString("torrent", infoHash[:]), zap.String("file", file.Path()), zap.Int64("fileLength", file.Length()), zap.Int("n", n), @@ -167,10 +167,11 @@ func downloadFile(file torrent.File, reader *torrent.Reader, fileDataChan chan<- return } if err != nil { + infoHash := file.Torrent().InfoHash() zap.L().Debug( "Error while downloading a file!", zap.Error(err), - zap.ByteString("torrent", file.Torrent().InfoHash()[:]), + zap.ByteString("torrent", infoHash[:]), zap.String("file", file.Path()), zap.Int64("fileLength", file.Length()), zap.Int("n", n), diff --git a/src/magneticod/bittorrent/sinkFile.go b/src/magneticod/bittorrent/sinkFile.go index 2592fea..5d75c0c 100644 --- a/src/magneticod/bittorrent/sinkFile.go +++ b/src/magneticod/bittorrent/sinkFile.go @@ -12,28 +12,33 @@ import ( "go.uber.org/zap" ) - -type File struct{ - torrentInfoHash []byte - path string - data []byte +type FileRequest struct { + InfoHash []byte + Path string + Peers []torrent.Peer } +type FileResult struct { + // Request field is the original Request + Request *FileRequest + FileData []byte +} type FileSink struct { + baseDownloadDir string client *torrent.Client - drain chan File + drain chan FileResult terminated bool termination chan interface{} - timeout time.Duration + timeoutDuration time.Duration } // NewFileSink creates a new FileSink. // // cAddr : client address // mlAddr: mainline DHT node address -func NewFileSink(cAddr, mlAddr string, timeout time.Duration) *FileSink { +func NewFileSink(cAddr, mlAddr string, timeoutDuration time.Duration) *FileSink { fs := new(FileSink) mlUDPAddr, err := net.ResolveUDPAddr("udp", mlAddr) @@ -49,6 +54,11 @@ func NewFileSink(cAddr, mlAddr string, timeout time.Duration) *FileSink { return nil } + fs.baseDownloadDir = path.Join( + appdirs.UserCacheDir("magneticod", "", "", true), + "downloads", + ) + fs.client, err = torrent.NewClient(&torrent.Config{ ListenAddr: cAddr, DisableTrackers: true, @@ -57,10 +67,7 @@ func NewFileSink(cAddr, mlAddr string, timeout time.Duration) *FileSink { Passive: true, NoSecurity: true, }, - DefaultStorage: storage.NewFileByInfoHash(path.Join( - appdirs.UserCacheDir("magneticod", "", "", true), - "downloads", - )), + DefaultStorage: storage.NewFileByInfoHash(fs.baseDownloadDir), }) if err != nil { zap.L().Fatal("Leech could NOT create a new torrent client!", zap.Error(err)) @@ -68,21 +75,26 @@ func NewFileSink(cAddr, mlAddr string, timeout time.Duration) *FileSink { return nil } - fs.drain = make(chan File) + fs.drain = make(chan FileResult) fs.termination = make(chan interface{}) - fs.timeout = timeout + fs.timeoutDuration = timeoutDuration return fs } - -// peer might be nil -func (fs *FileSink) Sink(infoHash []byte, filePath string, peer *torrent.Peer) { - go fs.awaitFile(infoHash, filePath, peer) +// peer field is optional and might be nil. +func (fs *FileSink) Sink(infoHash []byte, path string, peers []torrent.Peer) { + if fs.terminated { + zap.L().Panic("Trying to Sink() an already closed FileSink!") + } + go fs.awaitFile(&FileRequest{ + InfoHash: infoHash, + Path: path, + Peers: peers, + }) } - -func (fs *FileSink) Drain() <-chan File { +func (fs *FileSink) Drain() <-chan FileResult { if fs.terminated { zap.L().Panic("Trying to Drain() an already closed FileSink!") } @@ -98,7 +110,7 @@ func (fs *FileSink) Terminate() { } -func (fs *FileSink) flush(result File) { +func (fs *FileSink) flush(result FileResult) { if !fs.terminated { fs.drain <- result } diff --git a/src/magneticod/bittorrent/sinkMetadata.go b/src/magneticod/bittorrent/sinkMetadata.go index a01c9fe..a5761d9 100644 --- a/src/magneticod/bittorrent/sinkMetadata.go +++ b/src/magneticod/bittorrent/sinkMetadata.go @@ -1,13 +1,11 @@ package bittorrent import ( - "net" - "go.uber.org/zap" "github.com/anacrolix/torrent" - "github.com/anacrolix/torrent/metainfo" "magneticod/dht/mainline" + "persistence" ) @@ -19,7 +17,13 @@ type Metadata struct { TotalSize uint64 DiscoveredOn int64 // Files must be populated for both single-file and multi-file torrents! - Files []metainfo.FileInfo + Files []persistence.File + // Peers is the list of the "active" peers at the time of fetching metadata. Currently, it's + // always nil as anacrolix/torrent does not support returning list of peers for a given torrent, + // but in the future, this information can be useful for the CompletingCoordinator which can use + // those Peers to download the README file (if any found). + Peers []torrent.Peer + } @@ -31,11 +35,11 @@ type MetadataSink struct { } -func NewMetadataSink(laddr net.TCPAddr) *MetadataSink { +func NewMetadataSink(laddr string) *MetadataSink { ms := new(MetadataSink) var err error ms.client, err = torrent.NewClient(&torrent.Config{ - ListenAddr: laddr.String(), + ListenAddr: laddr, DisableTrackers: true, DisablePEX: true, // TODO: Should we disable DHT to force the client to use the peers we supplied only, or not? diff --git a/src/magneticod/coordinators.go b/src/magneticod/coordinators.go new file mode 100644 index 0000000..e4e59a9 --- /dev/null +++ b/src/magneticod/coordinators.go @@ -0,0 +1,111 @@ +package main + +import ( + "regexp" + "sync" + "time" + + "github.com/anacrolix/torrent" + + "persistence" + + "magneticod/bittorrent" + +) + +type completionRequest struct { + infoHash []byte + path string + peers []torrent.Peer + time time.Time +} + +type completionResult struct { + InfoHash []byte + Path string + Data []byte +} + +type CompletingCoordinator struct { + database persistence.Database + maxReadmeSize uint + sink *bittorrent.FileSink + queue chan completionRequest + queueMutex sync.Mutex + outputChan chan completionResult + readmeRegex *regexp.Regexp + terminated bool + termination chan interface{} +} + +type CompletingCoordinatorOpFlags struct { + LeechClAddr string + LeechMlAddr string + LeechTimeout time.Duration + ReadmeMaxSize uint + ReadmeRegex *regexp.Regexp +} + +func NewCompletingCoordinator(database persistence.Database, opFlags CompletingCoordinatorOpFlags) (cc *CompletingCoordinator) { + cc = new(CompletingCoordinator) + cc.database = database + cc.maxReadmeSize = opFlags.ReadmeMaxSize + cc.sink = bittorrent.NewFileSink(opFlags.LeechClAddr, opFlags.LeechMlAddr, opFlags.LeechTimeout) + cc.queue = make(chan completionRequest, 100) + cc.readmeRegex = opFlags.ReadmeRegex + cc.termination = make(chan interface{}) + return +} + +func (cc *CompletingCoordinator) Request(infoHash []byte, path string, peers []torrent.Peer) { + cc.queueMutex.Lock() + defer cc.queueMutex.Unlock() + + // If queue is full discard the oldest request as it is more likely to be outdated. + if len(cc.queue) == cap(cc.queue) { + <- cc.queue + } + + // Imagine, if this function [Request()] was called by another goroutine right when we were + // here: the moment where we removed the oldest entry in the queue to free a single space for + // the newest one. Imagine, now, that the second Request() call manages to add its own entry + // to the queue, making the current goroutine wait until the cc.queue channel is available. + // + // Hence to prevent that we use cc.queueMutex + + cc.queue <- completionRequest{ + infoHash: infoHash, + path: path, + peers: peers, + time: time.Now(), + } +} + +func (cc *CompletingCoordinator) Start() { + go cc.complete() +} + +func (cc *CompletingCoordinator) Output() <-chan completionResult { + return cc.outputChan +} + +func (cc *CompletingCoordinator) complete() { + for { + select { + case request := <-cc.queue: + // Discard requests older than 2 minutes. + // TODO: Instead of settling on 2 minutes as an arbitrary value, do some research to + // learn average peer lifetime in the BitTorrent network. + if time.Now().Sub(request.time) > 2 * time.Minute { + continue + } + cc.sink.Sink(request.infoHash, request.path, request.peers) + + case <-cc.termination: + break + + default: + cc.database.FindAnIncompleteTorrent(cc.readmeRegex, cc.maxReadmeSize) + } + } +} diff --git a/src/magneticod/dht/mainline/bloomFilter.go b/src/magneticod/dht/mainline/bloomFilter.go new file mode 100644 index 0000000..29fa90c --- /dev/null +++ b/src/magneticod/dht/mainline/bloomFilter.go @@ -0,0 +1,59 @@ +package mainline + +import ( + "net" + "go.uber.org/zap" + "crypto/sha1" + "math/bits" + "math" +) + +const ( + k uint32 = 2 + m uint32 = 256 * 8 +) + +type BloomFilter struct { + filter [m/8]byte +} + +func (bf *BloomFilter) InsertIP(ip net.IP) { + if !(len(ip) == net.IPv4len || len(ip) == net.IPv6len) { + zap.S().Panicf("Attempted to insert an invalid IP to the bloom filter! %d", len(ip)) + } + + hash := sha1.Sum(ip) + + var index1, index2 uint32 + index1 = uint32(hash[0]) | uint32(hash[1]) << 8 + index2 = uint32(hash[2]) | uint32(hash[3]) << 8 + + // truncate index to m (11 bits required) + index1 %= m + index2 %= m + + // set bits at index1 and index2 + bf.filter[index1 / 8] |= 0x01 << (index1 % 8) + bf.filter[index2 / 8] |= 0x01 << (index2 % 8) +} + +func (bf *BloomFilter) Estimate() float64 { + // TODO: make it faster? + var nZeroes uint32 = 0 + for _, b := range bf.filter { + nZeroes += 8 - uint32(bits.OnesCount8(uint8(b))) + } + + var c uint32 + if m - 1 < nZeroes { + c = m - 1 + } else { + c = nZeroes + } + return math.Log(float64(c) / float64(m)) / (float64(k) * math.Log(1 - 1/float64(m))) +} + +func (bf *BloomFilter) Filter() (filterCopy [m/8]byte) { + copy(filterCopy[:], bf.filter[:]) + return filterCopy +} diff --git a/src/magneticod/dht/mainline/bloomFilter_test.go b/src/magneticod/dht/mainline/bloomFilter_test.go new file mode 100644 index 0000000..6df9fd6 --- /dev/null +++ b/src/magneticod/dht/mainline/bloomFilter_test.go @@ -0,0 +1,64 @@ +package mainline + +import ( + "bytes" + "testing" + "encoding/hex" + "strings" + "fmt" +) + +func TestBEP33Filter(t *testing.T) { + bf := new(BloomFilter) + populateForBEP33(bf) + + resultingFilter := bf.Filter() + var expectedFilter [256]byte + hex.Decode(expectedFilter[:], []byte(strings.Replace( + "F6C3F5EA A07FFD91 BDE89F77 7F26FB2B FF37BDB8 FB2BBAA2 FD3DDDE7 BACFFF75 EE7CCBAE" + + "FE5EEDB1 FBFAFF67 F6ABFF5E 43DDBCA3 FD9B9FFD F4FFD3E9 DFF12D1B DF59DB53 DBE9FA5B" + + "7FF3B8FD FCDE1AFB 8BEDD7BE 2F3EE71E BBBFE93B CDEEFE14 8246C2BC 5DBFF7E7 EFDCF24F" + + "D8DC7ADF FD8FFFDF DDFFF7A4 BBEEDF5C B95CE81F C7FCFF1F F4FFFFDF E5F7FDCB B7FD79B3" + + "FA1FC77B FE07FFF9 05B7B7FF C7FEFEFF E0B8370B B0CD3F5B 7F2BD93F EB4386CF DD6F7FD5" + + "BFAF2E9E BFFFFEEC D67ADBF7 C67F17EF D5D75EBA 6FFEBA7F FF47A91E B1BFBB53 E8ABFB57" + + "62ABE8FF 237279BF EFBFEEF5 FFC5FEBF DFE5ADFF ADFEE1FB 737FFFFB FD9F6AEF FEEE76B6" + + "FD8F72EF", + " ", "", -1))) + if !bytes.Equal(resultingFilter[:], expectedFilter[:]) { + t.Fail() + } +} + +func TestBEP33Estimation(t *testing.T) { + bf := new(BloomFilter) + populateForBEP33(bf) + + // Because Go lacks a truncate function for floats... + if fmt.Sprintf("%.5f", bf.Estimate())[:9] != "1224.9308" { + t.Errorf("Expected 1224.9308 got %f instead!", bf.Estimate()) + } +} + +func populateForBEP33(bf *BloomFilter) { + // 192.0.2.0 to 192.0.2.255 (both ranges inclusive) + addr := []byte{192, 0, 2, 0} + for i := 0; i <= 255; i++ { + addr[3] = uint8(i) + bf.InsertIP(addr) + } + + // 2001:DB8:: to 2001:DB8::3E7 (both ranges inclusive) + addr = []byte{32, 1, 13, 184, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0} + for i := 0; i <= 2; i++ { + addr[14] = uint8(i) + for e := 0; e <= 255; e++ { + addr[15] = uint8(e) + bf.InsertIP(addr) + } + } + addr[14] = 3 + for e := 0; e <= 231; e++ { + addr[15] = uint8(e) + bf.InsertIP(addr) + } +} diff --git a/src/magneticod/dht/mainline/codec.go b/src/magneticod/dht/mainline/codec.go index ee66ed6..bd559c5 100644 --- a/src/magneticod/dht/mainline/codec.go +++ b/src/magneticod/dht/mainline/codec.go @@ -14,6 +14,7 @@ import ( "github.com/anacrolix/torrent/bencode" "github.com/anacrolix/missinggo/iter" "regexp" + "github.com/willf/bloom" ) @@ -46,6 +47,21 @@ type QueryArguments struct { Port int `bencode:"port,omitempty"` // Use senders apparent DHT port ImpliedPort int `bencode:"implied_port,omitempty"` + + // Indicates whether the querying node is seeding the torrent it announces. + // Defined in BEP 33 "DHT Scrapes" for `announce_peer` queries. + Seed int `bencode:"seed,omitempty"` + + // If 1, then the responding node should try to fill the `values` list with non-seed items on a + // best-effort basis." + // Defined in BEP 33 "DHT Scrapes" for `get_peers` queries. + NoSeed int `bencode:"noseed,omitempty"` + // If 1, then the responding node should add two fields to the "r" dictionary in the response: + // - `BFsd`: Bloom Filter (256 bytes) representing all stored seeds for that infohash + // - `BFpe`: Bloom Filter (256 bytes) representing all stored peers (leeches) for that + // infohash + // Defined in BEP 33 "DHT Scrapes" for `get_peers` queries. + Scrape int `bencode:"noseed,omitempty"` } @@ -58,6 +74,15 @@ type ResponseValues struct { Token []byte `bencode:"token,omitempty"` // Torrent peers Values []CompactPeer `bencode:"values,omitempty"` + + // If `scrape` is set to 1 in the `get_peers` query then the responding node should add the + // below two fields to the "r" dictionary in the response: + // Defined in BEP 33 "DHT Scrapes" for responses to `get_peers` queries. + // Bloom Filter (256 bytes) representing all stored seeds for that infohash: + BFsd *bloom.BloomFilter `bencode:"BFsd,omitempty"` + // Bloom Filter (256 bytes) representing all stored peers (leeches) for that infohash: + BFpe *bloom.BloomFilter `bencode:"BFpe,omitempty"` + // TODO: write marshallers for those fields above ^^ } @@ -204,7 +229,7 @@ func (cni *CompactNodeInfo) UnmarshalBinary(b []byte) error { func (cnis CompactNodeInfos) MarshalBencode() ([]byte, error) { - ret := make([]byte, 0) // TODO: this doesn't look idiomatic at all, is this the right way? + var ret []byte for _, cni := range cnis { ret = append(ret, cni.MarshalBinary()...) diff --git a/src/magneticod/dht/mainline/protocol.go b/src/magneticod/dht/mainline/protocol.go index 42249ca..c2dd74a 100644 --- a/src/magneticod/dht/mainline/protocol.go +++ b/src/magneticod/dht/mainline/protocol.go @@ -31,7 +31,7 @@ type ProtocolEventHandlers struct { } -func NewProtocol(laddr net.UDPAddr, eventHandlers ProtocolEventHandlers) (p *Protocol) { +func NewProtocol(laddr string, eventHandlers ProtocolEventHandlers) (p *Protocol) { p = new(Protocol) p.transport = NewTransport(laddr, p.onMessage) p.eventHandlers = eventHandlers diff --git a/src/magneticod/dht/mainline/service.go b/src/magneticod/dht/mainline/service.go index 305371b..c01e5c1 100644 --- a/src/magneticod/dht/mainline/service.go +++ b/src/magneticod/dht/mainline/service.go @@ -38,7 +38,7 @@ type TrawlingServiceEventHandlers struct { } -func NewTrawlingService(laddr net.UDPAddr, eventHandlers TrawlingServiceEventHandlers) *TrawlingService { +func NewTrawlingService(laddr string, eventHandlers TrawlingServiceEventHandlers) *TrawlingService { service := new(TrawlingService) service.protocol = NewProtocol( laddr, diff --git a/src/magneticod/dht/mainline/transport.go b/src/magneticod/dht/mainline/transport.go index 064f52b..eaa2f4b 100644 --- a/src/magneticod/dht/mainline/transport.go +++ b/src/magneticod/dht/mainline/transport.go @@ -10,8 +10,8 @@ import ( type Transport struct { - conn *net.UDPConn - laddr net.UDPAddr + conn *net.UDPConn + laddr *net.UDPAddr started bool // OnMessage is the function that will be called when Transport receives a packet that is @@ -21,10 +21,13 @@ type Transport struct { } -func NewTransport(laddr net.UDPAddr, onMessage func(*Message, net.Addr)) (*Transport) { +func NewTransport(laddr string, onMessage func(*Message, net.Addr)) (*Transport) { transport := new(Transport) transport.onMessage = onMessage - transport.laddr = laddr + var err error; transport.laddr, err = net.ResolveUDPAddr("udp", laddr) + if err != nil { + zap.L().Panic("Could not resolve the UDP address for the trawler!", zap.Error(err)) + } return transport } @@ -45,7 +48,7 @@ func (t *Transport) Start() { t.started = true var err error - t.conn, err = net.ListenUDP("udp", &t.laddr) + t.conn, err = net.ListenUDP("udp", t.laddr) if err != nil { zap.L().Fatal("Could NOT create a UDP socket!", zap.Error(err)) } diff --git a/src/magneticod/dht/managers.go b/src/magneticod/dht/managers.go index 4ffabf6..1dbbba6 100644 --- a/src/magneticod/dht/managers.go +++ b/src/magneticod/dht/managers.go @@ -1,10 +1,6 @@ package dht -import ( - "magneticod/dht/mainline" - "net" - "github.com/bradfitz/iter" -) +import "magneticod/dht/mainline" type TrawlingManager struct { @@ -14,29 +10,20 @@ type TrawlingManager struct { } -func NewTrawlingManager(mlAddrs []net.UDPAddr) *TrawlingManager { +func NewTrawlingManager(mlAddrs []string) *TrawlingManager { manager := new(TrawlingManager) manager.output = make(chan mainline.TrawlingResult) - if mlAddrs != nil { - for _, addr := range mlAddrs { - manager.services = append(manager.services, mainline.NewTrawlingService( - addr, - mainline.TrawlingServiceEventHandlers{ - OnResult: manager.onResult, - }, - )) - } - } else { - addr := net.UDPAddr{IP: []byte("\x00\x00\x00\x00"), Port: 0} - for range iter.N(1) { - manager.services = append(manager.services, mainline.NewTrawlingService( - addr, - mainline.TrawlingServiceEventHandlers{ - OnResult: manager.onResult, - }, - )) - } + if mlAddrs == nil { + mlAddrs = []string{"0.0.0.0:0"} + } + for _, addr := range mlAddrs { + manager.services = append(manager.services, mainline.NewTrawlingService( + addr, + mainline.TrawlingServiceEventHandlers{ + OnResult: manager.onResult, + }, + )) } for _, service := range manager.services { diff --git a/src/magneticod/main.go b/src/magneticod/main.go index c7ab3f3..1d8885f 100644 --- a/src/magneticod/main.go +++ b/src/magneticod/main.go @@ -1,20 +1,23 @@ package main import ( + "fmt" "net" "os" "os/signal" "regexp" + "time" "github.com/jessevdk/go-flags" "github.com/pkg/profile" "go.uber.org/zap" "go.uber.org/zap/zapcore" + "persistence" + "magneticod/bittorrent" "magneticod/dht" - "fmt" - "time" + "github.com/anacrolix/torrent/metainfo" ) type cmdFlags struct { @@ -35,14 +38,13 @@ type cmdFlags struct { LeechMlAddr string `long:"leech-ml-addr" descrition:"Address to be used by the mainline DHT node for fetching README files." default:"0.0.0.0:0"` LeechTimeout uint `long:"leech-timeout" description:"Number of integer seconds to pass before a leech timeouts." default:"300"` ReadmeMaxSize uint `long:"readme-max-size" description:"Maximum size -which must be greater than zero- of a description file in bytes." default:"20480"` - ReadmeRegexes []string `long:"readme-regex" description:"Regular expression(s) which will be tested against the name of the README files, in the supplied order."` + ReadmeRegex string `long:"readme-regex" description:"Regular expression(s) which will be tested against the name of the README files, in the supplied order."` Verbose []bool `short:"v" long:"verbose" description:"Increases verbosity."` Profile string `long:"profile" description:"Enable profiling." default:""` - // ==== Deprecated Flags ==== - // TODO: don't even support deprecated flags! + // ==== OLD Flags ==== // DatabaseFile is akin to Database flag, except that it was used when SQLite was the only // persistence backend ever conceived, so it's the path* to the database file, which was -by @@ -53,7 +55,6 @@ type cmdFlags struct { // On BSDs? : TODO? // On anywhere else: TODO? // TODO: Is the path* absolute or can be relative as well? - // DatabaseFile string } const ( @@ -70,19 +71,17 @@ type opFlags struct { TrawlerMlAddrs []string TrawlerMlInterval time.Duration - // TODO: is this even supported by anacrolix/torrent? FetcherAddr string FetcherTimeout time.Duration StatistMlAddrs []string StatistMlTimeout time.Duration - // TODO: is this even supported by anacrolix/torrent? LeechClAddr string LeechMlAddr string LeechTimeout time.Duration ReadmeMaxSize uint - ReadmeRegexes []*regexp.Regexp + ReadmeRegex *regexp.Regexp Verbosity int @@ -90,12 +89,12 @@ type opFlags struct { } func main() { - atom := zap.NewAtomicLevel() + loggerLevel := zap.NewAtomicLevel() // Logging levels: ("debug", "info", "warn", "error", "dpanic", "panic", and "fatal"). logger := zap.New(zapcore.NewCore( zapcore.NewConsoleEncoder(zap.NewDevelopmentEncoderConfig()), zapcore.Lock(os.Stderr), - atom, + loggerLevel, )) defer logger.Sync() zap.ReplaceGlobals(logger) @@ -111,81 +110,70 @@ func main() { switch opFlags.Verbosity { case 0: - atom.SetLevel(zap.WarnLevel) + loggerLevel.SetLevel(zap.WarnLevel) case 1: - atom.SetLevel(zap.InfoLevel) - // Default: i.e. in case of 2 or more. + loggerLevel.SetLevel(zap.InfoLevel) + // Default: i.e. in case of 2 or more. default: - atom.SetLevel(zap.DebugLevel) + loggerLevel.SetLevel(zap.DebugLevel) } zap.ReplaceGlobals(logger) - /* - updating_manager := nil - statistics_sink := nil - completing_manager := nil - file_sink := nil - */ // Handle Ctrl-C gracefully. - interrupt_chan := make(chan os.Signal) - signal.Notify(interrupt_chan, os.Interrupt) + interruptChan := make(chan os.Signal) + signal.Notify(interruptChan, os.Interrupt) - database, err := NewDatabase(opFlags.Database) + database, err := persistence.MakeDatabase(opFlags.DatabaseURL) if err != nil { - logger.Sugar().Fatalf("Could not open the database at `%s`: %s", opFlags.Database, err.Error()) + logger.Sugar().Fatalf("Could not open the database at `%s`: %s", opFlags.DatabaseURL, err.Error()) } - trawlingManager := dht.NewTrawlingManager(opFlags.MlTrawlerAddrs) - metadataSink := bittorrent.NewMetadataSink(opFlags.FetcherAddr) - fileSink := bittorrent.NewFileSink() - - go func() { - for { - select { - case result := <-trawlingManager.Output(): - logger.Debug("result: ", zap.String("hash", result.InfoHash.String())) - if !database.DoesExist(result.InfoHash[:]) { - metadataSink.Sink(result) - } - - case metadata := <-metadataSink.Drain(): - logger.Sugar().Infof("D I S C O V E R E D: `%s` %x", - metadata.Name, metadata.InfoHash) - if err := database.AddNewTorrent(metadata); err != nil { - logger.Sugar().Fatalf("Could not add new torrent %x to the database: %s", - metadata.InfoHash, err.Error()) - } - - case <-interrupt_chan: - trawlingManager.Terminate() - break - } - } - }() - - go func() { - - }() - - go func() { - - }() - + trawlingManager := dht.NewTrawlingManager(opFlags.TrawlerMlAddrs) + metadataSink := bittorrent.NewMetadataSink(opFlags.FetcherAddr) + completingCoordinator := NewCompletingCoordinator(database, CompletingCoordinatorOpFlags{ + LeechClAddr: opFlags.LeechClAddr, + LeechMlAddr: opFlags.LeechMlAddr, + LeechTimeout: opFlags.LeechTimeout, + ReadmeMaxSize: opFlags.ReadmeMaxSize, + ReadmeRegex: opFlags.ReadmeRegex, + }) /* - for { - select { + refreshingCoordinator := NewRefreshingCoordinator(database, RefreshingCoordinatorOpFlags{ - case updating_manager.Output(): - - case statistics_sink.Sink(): - - case completing_manager.Output(): - - case file_sink.Sink(): + }) */ - <-interrupt_chan + for { + select { + case result := <-trawlingManager.Output(): + logger.Debug("result: ", zap.String("hash", result.InfoHash.String())) + exists, err := database.DoesTorrentExist(result.InfoHash[:]) + if err != nil { + zap.L().Fatal("Could not check whether torrent exists!", zap.Error(err)) + } else if !exists { + metadataSink.Sink(result) + } + + case metadata := <-metadataSink.Drain(): + if err := database.AddNewTorrent(metadata.InfoHash, metadata.Name, metadata.Files); err != nil { + logger.Sugar().Fatalf("Could not add new torrent %x to the database: %s", + metadata.InfoHash, err.Error()) + } + logger.Sugar().Infof("D I S C O V E R E D: `%s` %x", metadata.Name, metadata.InfoHash) + + if readmePath := findReadme(opFlags.ReadmeRegex, metadata.Files); readmePath != nil { + completingCoordinator.Request(metadata.InfoHash, *readmePath, metadata.Peers) + } + + case result := <-completingCoordinator.Output(): + database.AddReadme(result.InfoHash, result.Path, result.Data) + + case <-interruptChan: + trawlingManager.Terminate() + break + } + } } func parseFlags() (opF opFlags) { @@ -237,13 +225,13 @@ func parseFlags() (opF opFlags) { } if err = checkAddrs([]string{cmdF.LeechClAddr}); err != nil { - zap.S().Fatal("Of argument `leech-cl-addr` %s", err.Error()) + zap.S().Fatalf("Of argument `leech-cl-addr` %s", err.Error()) } else { opF.LeechClAddr = cmdF.LeechClAddr } if err = checkAddrs([]string{cmdF.LeechMlAddr}); err != nil { - zap.S().Fatal("Of argument `leech-ml-addr` %s", err.Error()) + zap.S().Fatalf("Of argument `leech-ml-addr` %s", err.Error()) } else { opF.LeechMlAddr = cmdF.LeechMlAddr } @@ -260,13 +248,9 @@ func parseFlags() (opF opFlags) { opF.ReadmeMaxSize = cmdF.ReadmeMaxSize } - for i, s := range cmdF.ReadmeRegexes { - regex, err := regexp.Compile(s) - if err != nil { - zap.S().Fatalf("Of argument `readme-regex` with %d(th) regex `%s`: %s", i + 1, s, err.Error()) - } else { - opF.ReadmeRegexes = append(opF.ReadmeRegexes, regex) - } + opF.ReadmeRegex, err = regexp.Compile(cmdF.ReadmeRegex) + if err != nil { + zap.S().Fatalf("Argument `readme-regex` is not a valid regex: %s", err.Error()) } opF.Verbosity = len(cmdF.Verbose) @@ -286,3 +270,15 @@ func checkAddrs(addrs []string) error { } return nil } + +// findReadme looks for a possible Readme file whose path is matched by the pathRegex. +// If there are multiple matches, the first one is returned. +// If there are no matches, nil returned. +func findReadme(pathRegex *regexp.Regexp, files []persistence.File) *string { + for _, file := range files { + if pathRegex.MatchString(file.Path) { + return &file.Path + } + } + return nil +} diff --git a/src/magneticod/persistence.go b/src/magneticod/persistence/database.go similarity index 69% rename from src/magneticod/persistence.go rename to src/magneticod/persistence/database.go index 1a938fa..ff866b1 100644 --- a/src/magneticod/persistence.go +++ b/src/magneticod/persistence/database.go @@ -1,4 +1,4 @@ -package main +package persistence import ( "bytes" @@ -13,6 +13,7 @@ import ( "go.uber.org/zap" "magneticod/bittorrent" + "regexp" ) type engineType uint8 @@ -78,7 +79,7 @@ func NewDatabase(rawurl string) (*Database, error) { } -func (db *Database) DoesExist(infoHash []byte) bool { +func (db *Database) DoesTorrentExist(infoHash []byte) bool { for _, torrent := range db.newTorrents { if bytes.Equal(infoHash, torrent.InfoHash) { return true; @@ -96,6 +97,30 @@ func (db *Database) DoesExist(infoHash []byte) bool { return rows.Next() } +func (db *Database) FindAnIncompleteTorrent(pathRegex *regexp.Regexp, maxSize uint) error { + switch db.engine { + case SQLITE: + return db.findAnIncompleteTorrent_SQLite(pathRegex, maxSize) + + default: + zap.L().Fatal("Unknown database engine!", zap.Uint8("engine", uint8(db.engine))) + return nil + } +} + +func (db *Database) findAnIncompleteTorrent_SQLite(pathRegex *regexp.Regexp, maxSize uint) error { + // TODO: Prefer torrents with most seeders & leechs (i.e. most popular) + _, err := db.database.Query(` + SELECT torrents.info_hash, files.path FROM files WHERE files.path REGEXP ? + INNER JOIN torrents ON files.torrent_id = torrents.id LIMIT 1; + `) + if err != nil { + return err + } + + return nil +} + // AddNewTorrent adds a new torrent to the *queue* to be flushed to the persistent database. func (db *Database) AddNewTorrent(torrent bittorrent.Metadata) error { @@ -107,7 +132,7 @@ func (db *Database) AddNewTorrent(torrent bittorrent.Metadata) error { // that it doesn't exists there, add it to the sink. // Hence check for the last time whether the torrent exists in the database, and only if not, // add it. - if db.DoesExist(torrent.InfoHash) { + if db.DoesTorrentExist(torrent.InfoHash) { return nil; } @@ -125,8 +150,14 @@ func (db *Database) AddNewTorrent(torrent bittorrent.Metadata) error { } +func (db *Database) AddReadme(infoHash []byte, path string, data []byte) error { + // TODO + return nil +} + + func (db *Database) commitNewTorrents() error { - tx, err := db.database.Begin() + tx, err := db.database.Begin() if err != nil { return fmt.Errorf("sql.DB.Begin()! %s", err.Error()) } @@ -216,11 +247,11 @@ func setupSqliteDatabase(database *sql.DB) error { // // Enable foreign key constraints in SQLite which are crucial to prevent programmer errors on // our side. - _, err := database.Exec( - `PRAGMA journal_mode=WAL; + _, err := database.Exec(` + PRAGMA journal_mode=WAL; PRAGMA temp_store=1; - PRAGMA foreign_keys=ON;`, - ) + PRAGMA foreign_keys=ON; + `) if err != nil { return err } @@ -231,33 +262,28 @@ func setupSqliteDatabase(database *sql.DB) error { } // Essential, and valid for all user_version`s: - _, err = tx.Exec( - `CREATE TABLE IF NOT EXISTS torrents ( + // TODO: "torrent_id" column of the "files" table can be NULL, how can we fix this in a new schema? + _, err = tx.Exec(` + CREATE TABLE IF NOT EXISTS torrents ( id INTEGER PRIMARY KEY, info_hash BLOB NOT NULL UNIQUE, name TEXT NOT NULL, total_size INTEGER NOT NULL CHECK(total_size > 0), discovered_on INTEGER NOT NULL CHECK(discovered_on > 0) ); - - CREATE INDEX IF NOT EXISTS info_hash_index ON torrents (info_hash); - CREATE TABLE IF NOT EXISTS files ( id INTEGER PRIMARY KEY, torrent_id INTEGER REFERENCES torrents ON DELETE CASCADE ON UPDATE RESTRICT, size INTEGER NOT NULL, path TEXT NOT NULL ); - `, - ) + `) if err != nil { return err } // Get the user_version: - res, err := tx.Query( - `PRAGMA user_version;`, - ) + res, err := tx.Query(`PRAGMA user_version;`) if err != nil { return err } @@ -265,18 +291,57 @@ func setupSqliteDatabase(database *sql.DB) error { res.Next() res.Scan(&userVersion) - // Upgrade to the latest schema: switch userVersion { // Upgrade from user_version 0 to 1 + // The Change: + // * `info_hash_index` is recreated as UNIQUE. case 0: - _, err = tx.Exec( - `ALTER TABLE torrents ADD COLUMN readme TEXT; - PRAGMA user_version = 1;`, - ) + zap.S().Warnf("Updating database schema from 0 to 1... (this might take a while)") + _, err = tx.Exec(` + DROP INDEX info_hash_index; + CREATE UNIQUE INDEX info_hash_index ON torrents (info_hash); + PRAGMA user_version = 1; + `) + if err != nil { + return err + } + fallthrough + // Upgrade from user_version 1 to 2 + // The Change: + // * Added `is_readme` and `content` columns to the `files` table, and the constraints & the + // the indices they entail. + // * Added unique index `readme_index` on `files` table. + case 1: + zap.S().Warnf("Updating database schema from 1 to 2... (this might take a while)") + // We introduce two new columns here: content BLOB, and is_readme INTEGER which we treat as + // a bool (hence the CHECK). + // The reason for the change is that as we introduce the new "readme" feature which + // downloads a readme file as a torrent descriptor, we needed to store it somewhere in the + // database with the following conditions: + // + // 1. There can be one and only one readme (content) for a given torrent; hence the + // UNIQUE INDEX on (torrent_id, is_description) (remember that SQLite treats each NULL + // value as distinct [UNIQUE], see https://sqlite.org/nulls.html). + // 2. We would like to keep the readme (content) associated with the file it came from; + // hence we modify the files table instead of the torrents table. + // + // Regarding the implementation details, following constraints arise: + // + // 1. The column is_readme is either NULL or 1, and if it is 1, then content column cannot + // be NULL (but might be an empty BLOB). Vice versa, if content column of a row is, + // NULL then is_readme must be NULL. + // + // This is to prevent unused content fields filling up the database, and to catch + // programmers' errors. + _, err = tx.Exec(` + ALTER TABLE files ADD COLUMN is_readme INTEGER CHECK (is_readme IS NULL OR is_readme=1); + ALTER TABLE files ADD COLUMN content BLOB CHECK((content IS NULL AND is_readme IS NULL) OR (content IS NOT NULL AND is_readme=1)); + CREATE UNIQUE INDEX readme_index ON files (torrent_id, is_readme); + PRAGMA user_version = 2; + `) if err != nil { return err } - // Add `fallthrough`s as needed to keep upgrading... } if err = tx.Commit(); err != nil { diff --git a/src/magneticod/persistence_test.go b/src/magneticod/persistence/database_test.go similarity index 97% rename from src/magneticod/persistence_test.go rename to src/magneticod/persistence/database_test.go index 716f216..98ffafe 100644 --- a/src/magneticod/persistence_test.go +++ b/src/magneticod/persistence/database_test.go @@ -1,4 +1,4 @@ -package main +package persistence import ( "path" diff --git a/src/magneticow/bindata.go b/src/magneticow/bindata.go new file mode 100644 index 0000000..24f10a3 --- /dev/null +++ b/src/magneticow/bindata.go @@ -0,0 +1,262 @@ +package main + +import ( + "fmt" + "io/ioutil" +) + +// bindata_read reads the given file from disk. It returns +// an error on failure. +func bindata_read(path, name string) ([]byte, error) { + buf, err := ioutil.ReadFile(path) + if err != nil { + err = fmt.Errorf("Error reading asset %s at %s: %v", name, path, err) + } + return buf, err +} + + +// templates_torrent_html reads file data from disk. +// It panics if something went wrong in the process. +func templates_torrent_html() ([]byte, error) { + return bindata_read( + "/home/bora/labs/magnetico/src/magneticow/data/templates/torrent.html", + "templates/torrent.html", + ) +} + +// templates_feed_xml reads file data from disk. +// It panics if something went wrong in the process. +func templates_feed_xml() ([]byte, error) { + return bindata_read( + "/home/bora/labs/magnetico/src/magneticow/data/templates/feed.xml", + "templates/feed.xml", + ) +} + +// templates_homepage_html reads file data from disk. +// It panics if something went wrong in the process. +func templates_homepage_html() ([]byte, error) { + return bindata_read( + "/home/bora/labs/magnetico/src/magneticow/data/templates/homepage.html", + "templates/homepage.html", + ) +} + +// templates_statistics_html reads file data from disk. +// It panics if something went wrong in the process. +func templates_statistics_html() ([]byte, error) { + return bindata_read( + "/home/bora/labs/magnetico/src/magneticow/data/templates/statistics.html", + "templates/statistics.html", + ) +} + +// templates_torrents_html reads file data from disk. +// It panics if something went wrong in the process. +func templates_torrents_html() ([]byte, error) { + return bindata_read( + "/home/bora/labs/magnetico/src/magneticow/data/templates/torrents.html", + "templates/torrents.html", + ) +} + +// static_scripts_plotly_v1_26_1_min_js reads file data from disk. +// It panics if something went wrong in the process. +func static_scripts_plotly_v1_26_1_min_js() ([]byte, error) { + return bindata_read( + "/home/bora/labs/magnetico/src/magneticow/data/static/scripts/plotly-v1.26.1.min.js", + "static/scripts/plotly-v1.26.1.min.js", + ) +} + +// static_scripts_statistics_js reads file data from disk. +// It panics if something went wrong in the process. +func static_scripts_statistics_js() ([]byte, error) { + return bindata_read( + "/home/bora/labs/magnetico/src/magneticow/data/static/scripts/statistics.js", + "static/scripts/statistics.js", + ) +} + +// static_scripts_torrent_js reads file data from disk. +// It panics if something went wrong in the process. +func static_scripts_torrent_js() ([]byte, error) { + return bindata_read( + "/home/bora/labs/magnetico/src/magneticow/data/static/scripts/torrent.js", + "static/scripts/torrent.js", + ) +} + +// static_styles_reset_css reads file data from disk. +// It panics if something went wrong in the process. +func static_styles_reset_css() ([]byte, error) { + return bindata_read( + "/home/bora/labs/magnetico/src/magneticow/data/static/styles/reset.css", + "static/styles/reset.css", + ) +} + +// static_styles_statistics_css reads file data from disk. +// It panics if something went wrong in the process. +func static_styles_statistics_css() ([]byte, error) { + return bindata_read( + "/home/bora/labs/magnetico/src/magneticow/data/static/styles/statistics.css", + "static/styles/statistics.css", + ) +} + +// static_styles_torrent_css reads file data from disk. +// It panics if something went wrong in the process. +func static_styles_torrent_css() ([]byte, error) { + return bindata_read( + "/home/bora/labs/magnetico/src/magneticow/data/static/styles/torrent.css", + "static/styles/torrent.css", + ) +} + +// static_styles_torrents_css reads file data from disk. +// It panics if something went wrong in the process. +func static_styles_torrents_css() ([]byte, error) { + return bindata_read( + "/home/bora/labs/magnetico/src/magneticow/data/static/styles/torrents.css", + "static/styles/torrents.css", + ) +} + +// static_styles_homepage_css reads file data from disk. +// It panics if something went wrong in the process. +func static_styles_homepage_css() ([]byte, error) { + return bindata_read( + "/home/bora/labs/magnetico/src/magneticow/data/static/styles/homepage.css", + "static/styles/homepage.css", + ) +} + +// static_styles_essential_css reads file data from disk. +// It panics if something went wrong in the process. +func static_styles_essential_css() ([]byte, error) { + return bindata_read( + "/home/bora/labs/magnetico/src/magneticow/data/static/styles/essential.css", + "static/styles/essential.css", + ) +} + +// static_assets_magnet_gif reads file data from disk. +// It panics if something went wrong in the process. +func static_assets_magnet_gif() ([]byte, error) { + return bindata_read( + "/home/bora/labs/magnetico/src/magneticow/data/static/assets/magnet.gif", + "static/assets/magnet.gif", + ) +} + +// static_assets_feed_png reads file data from disk. +// It panics if something went wrong in the process. +func static_assets_feed_png() ([]byte, error) { + return bindata_read( + "/home/bora/labs/magnetico/src/magneticow/data/static/assets/feed.png", + "static/assets/feed.png", + ) +} + +// static_fonts_notomono_license_ofl_txt reads file data from disk. +// It panics if something went wrong in the process. +func static_fonts_notomono_license_ofl_txt() ([]byte, error) { + return bindata_read( + "/home/bora/labs/magnetico/src/magneticow/data/static/fonts/NotoMono/LICENSE_OFL.txt", + "static/fonts/NotoMono/LICENSE_OFL.txt", + ) +} + +// static_fonts_notomono_regular_ttf reads file data from disk. +// It panics if something went wrong in the process. +func static_fonts_notomono_regular_ttf() ([]byte, error) { + return bindata_read( + "/home/bora/labs/magnetico/src/magneticow/data/static/fonts/NotoMono/Regular.ttf", + "static/fonts/NotoMono/Regular.ttf", + ) +} + +// static_fonts_notosansui_license_ofl_txt reads file data from disk. +// It panics if something went wrong in the process. +func static_fonts_notosansui_license_ofl_txt() ([]byte, error) { + return bindata_read( + "/home/bora/labs/magnetico/src/magneticow/data/static/fonts/NotoSansUI/LICENSE_OFL.txt", + "static/fonts/NotoSansUI/LICENSE_OFL.txt", + ) +} + +// static_fonts_notosansui_bold_ttf reads file data from disk. +// It panics if something went wrong in the process. +func static_fonts_notosansui_bold_ttf() ([]byte, error) { + return bindata_read( + "/home/bora/labs/magnetico/src/magneticow/data/static/fonts/NotoSansUI/Bold.ttf", + "static/fonts/NotoSansUI/Bold.ttf", + ) +} + +// static_fonts_notosansui_bolditalic_ttf reads file data from disk. +// It panics if something went wrong in the process. +func static_fonts_notosansui_bolditalic_ttf() ([]byte, error) { + return bindata_read( + "/home/bora/labs/magnetico/src/magneticow/data/static/fonts/NotoSansUI/BoldItalic.ttf", + "static/fonts/NotoSansUI/BoldItalic.ttf", + ) +} + +// static_fonts_notosansui_italic_ttf reads file data from disk. +// It panics if something went wrong in the process. +func static_fonts_notosansui_italic_ttf() ([]byte, error) { + return bindata_read( + "/home/bora/labs/magnetico/src/magneticow/data/static/fonts/NotoSansUI/Italic.ttf", + "static/fonts/NotoSansUI/Italic.ttf", + ) +} + +// static_fonts_notosansui_regular_ttf reads file data from disk. +// It panics if something went wrong in the process. +func static_fonts_notosansui_regular_ttf() ([]byte, error) { + return bindata_read( + "/home/bora/labs/magnetico/src/magneticow/data/static/fonts/NotoSansUI/Regular.ttf", + "static/fonts/NotoSansUI/Regular.ttf", + ) +} + +// Asset loads and returns the asset for the given name. +// It returns an error if the asset could not be found or +// could not be loaded. +func Asset(name string) ([]byte, error) { + if f, ok := _bindata[name]; ok { + return f() + } + return nil, fmt.Errorf("Asset %s not found", name) +} + +// _bindata is a table, holding each asset generator, mapped to its name. +var _bindata = map[string] func() ([]byte, error) { + "templates/torrent.html": templates_torrent_html, + "templates/feed.xml": templates_feed_xml, + "templates/homepage.html": templates_homepage_html, + "templates/statistics.html": templates_statistics_html, + "templates/torrents.html": templates_torrents_html, + "static/scripts/plotly-v1.26.1.min.js": static_scripts_plotly_v1_26_1_min_js, + "static/scripts/statistics.js": static_scripts_statistics_js, + "static/scripts/torrent.js": static_scripts_torrent_js, + "static/styles/reset.css": static_styles_reset_css, + "static/styles/statistics.css": static_styles_statistics_css, + "static/styles/torrent.css": static_styles_torrent_css, + "static/styles/torrents.css": static_styles_torrents_css, + "static/styles/homepage.css": static_styles_homepage_css, + "static/styles/essential.css": static_styles_essential_css, + "static/assets/magnet.gif": static_assets_magnet_gif, + "static/assets/feed.png": static_assets_feed_png, + "static/fonts/NotoMono/LICENSE_OFL.txt": static_fonts_notomono_license_ofl_txt, + "static/fonts/NotoMono/Regular.ttf": static_fonts_notomono_regular_ttf, + "static/fonts/NotoSansUI/LICENSE_OFL.txt": static_fonts_notosansui_license_ofl_txt, + "static/fonts/NotoSansUI/Bold.ttf": static_fonts_notosansui_bold_ttf, + "static/fonts/NotoSansUI/BoldItalic.ttf": static_fonts_notosansui_bolditalic_ttf, + "static/fonts/NotoSansUI/Italic.ttf": static_fonts_notosansui_italic_ttf, + "static/fonts/NotoSansUI/Regular.ttf": static_fonts_notosansui_regular_ttf, + +} diff --git a/magneticow/magneticow/static/assets/feed.png b/src/magneticow/data/static/assets/feed.png similarity index 100% rename from magneticow/magneticow/static/assets/feed.png rename to src/magneticow/data/static/assets/feed.png diff --git a/magneticow/magneticow/static/assets/magnet.gif b/src/magneticow/data/static/assets/magnet.gif similarity index 100% rename from magneticow/magneticow/static/assets/magnet.gif rename to src/magneticow/data/static/assets/magnet.gif diff --git a/magneticow/magneticow/static/fonts/NotoMono/LICENSE_OFL.txt b/src/magneticow/data/static/fonts/NotoMono/LICENSE_OFL.txt similarity index 100% rename from magneticow/magneticow/static/fonts/NotoMono/LICENSE_OFL.txt rename to src/magneticow/data/static/fonts/NotoMono/LICENSE_OFL.txt diff --git a/magneticow/magneticow/static/fonts/NotoMono/Regular.ttf b/src/magneticow/data/static/fonts/NotoMono/Regular.ttf similarity index 100% rename from magneticow/magneticow/static/fonts/NotoMono/Regular.ttf rename to src/magneticow/data/static/fonts/NotoMono/Regular.ttf diff --git a/magneticow/magneticow/static/fonts/NotoSansUI/Bold.ttf b/src/magneticow/data/static/fonts/NotoSansUI/Bold.ttf similarity index 100% rename from magneticow/magneticow/static/fonts/NotoSansUI/Bold.ttf rename to src/magneticow/data/static/fonts/NotoSansUI/Bold.ttf diff --git a/magneticow/magneticow/static/fonts/NotoSansUI/BoldItalic.ttf b/src/magneticow/data/static/fonts/NotoSansUI/BoldItalic.ttf similarity index 100% rename from magneticow/magneticow/static/fonts/NotoSansUI/BoldItalic.ttf rename to src/magneticow/data/static/fonts/NotoSansUI/BoldItalic.ttf diff --git a/magneticow/magneticow/static/fonts/NotoSansUI/Italic.ttf b/src/magneticow/data/static/fonts/NotoSansUI/Italic.ttf similarity index 100% rename from magneticow/magneticow/static/fonts/NotoSansUI/Italic.ttf rename to src/magneticow/data/static/fonts/NotoSansUI/Italic.ttf diff --git a/magneticow/magneticow/static/fonts/NotoSansUI/LICENSE_OFL.txt b/src/magneticow/data/static/fonts/NotoSansUI/LICENSE_OFL.txt similarity index 100% rename from magneticow/magneticow/static/fonts/NotoSansUI/LICENSE_OFL.txt rename to src/magneticow/data/static/fonts/NotoSansUI/LICENSE_OFL.txt diff --git a/magneticow/magneticow/static/fonts/NotoSansUI/Regular.ttf b/src/magneticow/data/static/fonts/NotoSansUI/Regular.ttf similarity index 100% rename from magneticow/magneticow/static/fonts/NotoSansUI/Regular.ttf rename to src/magneticow/data/static/fonts/NotoSansUI/Regular.ttf diff --git a/magneticow/magneticow/static/scripts/plotly-v1.26.1.min.js b/src/magneticow/data/static/scripts/plotly-v1.26.1.min.js similarity index 100% rename from magneticow/magneticow/static/scripts/plotly-v1.26.1.min.js rename to src/magneticow/data/static/scripts/plotly-v1.26.1.min.js diff --git a/magneticow/magneticow/static/scripts/statistics.js b/src/magneticow/data/static/scripts/statistics.js similarity index 100% rename from magneticow/magneticow/static/scripts/statistics.js rename to src/magneticow/data/static/scripts/statistics.js diff --git a/magneticow/magneticow/static/scripts/torrent.js b/src/magneticow/data/static/scripts/torrent.js similarity index 100% rename from magneticow/magneticow/static/scripts/torrent.js rename to src/magneticow/data/static/scripts/torrent.js diff --git a/magneticow/magneticow/static/styles/essential.css b/src/magneticow/data/static/styles/essential.css similarity index 94% rename from magneticow/magneticow/static/styles/essential.css rename to src/magneticow/data/static/styles/essential.css index 6c940cb..6a92c4e 100644 --- a/magneticow/magneticow/static/styles/essential.css +++ b/src/magneticow/data/static/styles/essential.css @@ -39,7 +39,7 @@ html { } pre { - font-family: 'Noto Mono'; + font-family: 'Noto Mono', monospace; line-height: 1.2em; } @@ -49,6 +49,12 @@ body { line-height: 1.45; } +@media (max-width: 616px) { + body { + padding: 1em 8px 1em 8px; + } +} + b { font-weight: bold; } diff --git a/src/magneticow/data/static/styles/homepage.css b/src/magneticow/data/static/styles/homepage.css new file mode 100644 index 0000000..dcca09c --- /dev/null +++ b/src/magneticow/data/static/styles/homepage.css @@ -0,0 +1,38 @@ +main { + display: flex; + align-items: center; + align-content: center; + + height: calc(100vh - 2*16px - 0.833em - 23px); /* 100vh - body's padding(s) - footer margin - footer height */ + width: 100%; +} + +@media (max-width: 616px) { + main { + flex-direction: column; + justify-content: center; + align-items: flex-start; + } +} + +main div#magneticow { + white-space: nowrap; + margin: 0 0.5em 0.5em 0; +} + +main form { + max-width: 600px; + width: 100%; +} + +main form input { + width: 100%; +} + +main > div { + margin-right: 0.5em; +} + +footer { + margin-top: 0.833em; +} diff --git a/magneticow/magneticow/static/styles/reset.css b/src/magneticow/data/static/styles/reset.css similarity index 100% rename from magneticow/magneticow/static/styles/reset.css rename to src/magneticow/data/static/styles/reset.css diff --git a/magneticow/magneticow/static/styles/statistics.css b/src/magneticow/data/static/styles/statistics.css similarity index 100% rename from magneticow/magneticow/static/styles/statistics.css rename to src/magneticow/data/static/styles/statistics.css diff --git a/magneticow/magneticow/static/styles/torrent.css b/src/magneticow/data/static/styles/torrent.css similarity index 100% rename from magneticow/magneticow/static/styles/torrent.css rename to src/magneticow/data/static/styles/torrent.css diff --git a/magneticow/magneticow/static/styles/torrents.css b/src/magneticow/data/static/styles/torrents.css similarity index 100% rename from magneticow/magneticow/static/styles/torrents.css rename to src/magneticow/data/static/styles/torrents.css diff --git a/magneticow/magneticow/templates/feed.xml b/src/magneticow/data/templates/feed.xml similarity index 66% rename from magneticow/magneticow/templates/feed.xml rename to src/magneticow/data/templates/feed.xml index 997210c..6a3c230 100644 --- a/magneticow/magneticow/templates/feed.xml +++ b/src/magneticow/data/templates/feed.xml @@ -1,12 +1,15 @@ - {{ title }} + {{ .Title }} + {% for item in items %} {{ item.title }} + {{ item.DiscoveredOn }} {{ item.info_hash }} + {% endfor %} diff --git a/src/magneticow/data/templates/homepage.html b/src/magneticow/data/templates/homepage.html new file mode 100644 index 0000000..cbe3872 --- /dev/null +++ b/src/magneticow/data/templates/homepage.html @@ -0,0 +1,24 @@ + + + + + + magneticow + + + + + + +
+
magneticow​(pre-alpha)
+
+ +
+
+ + + + diff --git a/magneticow/magneticow/templates/statistics.html b/src/magneticow/data/templates/statistics.html similarity index 100% rename from magneticow/magneticow/templates/statistics.html rename to src/magneticow/data/templates/statistics.html diff --git a/magneticow/magneticow/templates/torrent.html b/src/magneticow/data/templates/torrent.html similarity index 100% rename from magneticow/magneticow/templates/torrent.html rename to src/magneticow/data/templates/torrent.html diff --git a/src/magneticow/data/templates/torrents.html b/src/magneticow/data/templates/torrents.html new file mode 100644 index 0000000..4caa44d --- /dev/null +++ b/src/magneticow/data/templates/torrents.html @@ -0,0 +1,66 @@ + + + + + {% if .search %}"{{.search}}"{% else %}Most recent torrents{% endif %} - magneticow + + + + + + +
+
magneticow​(pre-alpha)
+
+ +
+
+ feed icon subscribe +
+
+
+ + + + + + + + + + + {% for torrent in torrents %} + + + + + + + {% endfor %} + +
NameSizeDiscovered on
+ Magnet link{{ torrent.name }}{{ torrent.size }}{{ torrent.discovered_on }}
+
+ + + \ No newline at end of file diff --git a/src/magneticow/main.go b/src/magneticow/main.go index 046b183..2bc09d8 100644 --- a/src/magneticow/main.go +++ b/src/magneticow/main.go @@ -1,25 +1,20 @@ -// magneticow - Lightweight web interface for magnetico. -// Copyright (C) 2017 Mert Bora ALPER -// Dedicated to Cemile Binay, in whose hands I thrived. -// -// This program is free software: you can redistribute it and/or modify it under the terms of the -// GNU General Public License as published by the Free Software Foundation, either version 3 of the -// License, or (at your option) any later version. -// -// This program is distributed in the hope that it will be useful, but WITHOUT ANY WARRANTY; without -// even the implied warranty of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU -// General Public License for more details. -// -// You should have received a copy of the GNU General Public License along with this program. If -// not, see . package main import ( + "html/template" + "log" "net/http" + "strings" "github.com/gorilla/mux" + + "persistence" ) +const N_TORRENTS = 20 + +var templates map[string]*template.Template +var database persistence.Database func main() { router := mux.NewRouter() @@ -28,23 +23,48 @@ func main() { router.HandleFunc("/torrents/{infohash}", torrentsInfohashHandler) router.HandleFunc("/torrents/{infohash}/{name}", torrentsInfohashNameHandler) router.HandleFunc("/statistics", statisticsHandler) + router.PathPrefix("/static").HandlerFunc(staticHandler) + router.HandleFunc("/feed", feedHandler) + + templates = make(map[string]*template.Template) + templates["feed"] = template.Must(template.New("feed").Parse(string(mustAsset("templates/feed.xml")))) + templates["homepage"] = template.Must(template.New("homepage").Parse(string(mustAsset("templates/homepage.html")))) + templates["statistics"] = template.Must(template.New("statistics").Parse(string(mustAsset("templates/statistics.html")))) + templates["torrent"] = template.Must(template.New("torrent").Parse(string(mustAsset("templates/torrent.html")))) + templates["torrents"] = template.Must(template.New("torrents").Parse(string(mustAsset("templates/torrents.html")))) + + var err error + database, err = persistence.MakeDatabase("sqlite3:///home/bora/.local/share/magneticod/database.sqlite3") + if err != nil { + panic(err.Error()) + } + http.ListenAndServe(":8080", router) } func rootHandler(w http.ResponseWriter, r *http.Request) { - + count, err := database.GetNumberOfTorrents() + if err != nil { + panic(err.Error()) + } + templates["homepage"].Execute(w, count) } func torrentsHandler(w http.ResponseWriter, r *http.Request) { - + /* + newestTorrents, err := database.NewestTorrents(N_TORRENTS) + if err != nil { + panic(err.Error()) + } + templates["torrents"].Execute(w, nil) + */ } - func torrentsInfohashHandler(w http.ResponseWriter, r *http.Request) { - + // redirect to torrents/{infohash}/name } @@ -61,3 +81,28 @@ func statisticsHandler(w http.ResponseWriter, r *http.Request) { func feedHandler(w http.ResponseWriter, r *http.Request) { } + +func staticHandler(w http.ResponseWriter, r *http.Request) { + data, err := Asset(r.URL.Path[1:]) + if err != nil { + http.NotFound(w, r) + return + } + + var contentType string + if strings.HasSuffix(r.URL.Path, ".css") { + contentType = "text/css; charset=utf-8" + } else { // fallback option + contentType = http.DetectContentType(data) + } + w.Header().Set("Content-Type", contentType) + w.Write(data) +} + +func mustAsset(name string) []byte { + data, err := Asset(name) + if err != nil { + log.Panicf("Could NOT access the requested resource `%s`: %s", name, err.Error()) + } + return data +} diff --git a/src/persistence/Gopkg.toml b/src/persistence/Gopkg.toml new file mode 100644 index 0000000..403c7cd --- /dev/null +++ b/src/persistence/Gopkg.toml @@ -0,0 +1,26 @@ + +# Gopkg.toml example +# +# Refer to https://github.com/golang/dep/blob/master/docs/Gopkg.toml.md +# for detailed Gopkg.toml documentation. +# +# required = ["github.com/user/thing/cmd/thing"] +# ignored = ["github.com/user/project/pkgX", "bitbucket.org/user/project/pkgA/pkgY"] +# +# [[constraint]] +# name = "github.com/user/project" +# version = "1.0.0" +# +# [[constraint]] +# name = "github.com/user/project2" +# branch = "dev" +# source = "github.com/myfork/project2" +# +# [[override]] +# name = "github.com/x/y" +# version = "2.4.0" + + +[[constraint]] + name = "go.uber.org/zap" + version = "1.6.0" diff --git a/src/persistence/interface.go b/src/persistence/interface.go new file mode 100644 index 0000000..5dc3eae --- /dev/null +++ b/src/persistence/interface.go @@ -0,0 +1,116 @@ +package persistence + +import ( + "fmt" + "regexp" + "net/url" +) + +type Database interface { + Engine() databaseEngine + DoesTorrentExist(infoHash []byte) (bool, error) + // GiveAnIncompleteTorrentByInfoHash returns (*gives*) an incomplete -i.e. one that doesn't have + // readme downloaded yet- torrent from the database. + // GiveAnIncompleteTorrent might return a nil slice for infoHash, a nil string, and a nil err, + // meaning that no incomplete torrent could be found in the database (congrats!). + GiveAnIncompleteTorrent(pathRegex *regexp.Regexp, maxSize uint) (infoHash []byte, path string, err error) + GiveAStaleTorrent() (infoHash []byte, err error) + AddNewTorrent(infoHash []byte, name string, files []File) error + AddReadme(infoHash []byte, path string, content string) error + Close() error + + // GetNumberOfTorrents returns the number of torrents saved in the database. Might be an + // approximation. + GetNumberOfTorrents() (uint, error) + NewestTorrents(n uint) ([]TorrentMetadata, error) + SearchTorrents(query string, orderBy orderingCriteria, descending bool, mustHaveReadme bool) ([]TorrentMetadata, error) + // GetTorrents returns the TorrentExtMetadata for the torrent of the given infoHash. Might return + // nil, nil if the torrent does not exist in the database. + GetTorrent(infoHash []byte) (*TorrentMetadata, error) + GetFiles(infoHash []byte) ([]File, error) + GetReadme(infoHash []byte) (string, error) + GetStatistics(from ISO8601, period uint) (*Statistics, error) +} + +type orderingCriteria uint8 + +const ( + BY_NAME orderingCriteria = 1 + BY_SIZE = 2 + BY_DISCOVERED_ON = 3 + BY_N_FILES = 4 + BY_N_SEEDERS = 5 + BY_N_LEECHERS = 6 + BY_UPDATED_ON = 7 + BY_N_SEEDERS_TO_N_LEECHERS_RATIO = 8 + BY_N_SEEDERS_PLUS_N_LEECHERS = 9 +) + + +type statisticsGranularity uint8 +type ISO8601 string + +const ( + MINUTELY_STATISTICS statisticsGranularity = 1 + HOURLY_STATISTICS = 2 + DAILY_STATISTICS = 3 + WEEKLY_STATISTICS = 4 + MONTHLY_STATISTICS = 5 + YEARLY_STATISTICS = 6 +) + +type databaseEngine uint8 + +const ( + SQLITE3_ENGINE databaseEngine = 1 +) + +type Statistics struct { + Granularity statisticsGranularity + From ISO8601 + Period uint + + // All these slices below have the exact length equal to the Period. + NTorrentsDiscovered []uint + NFilesDiscovered []uint + NReadmesDownloaded []uint + NTorrentsUpdated []uint +} + +type File struct { + Size int64 + Path string +} + +type TorrentMetadata struct { + infoHash []byte + name string + size uint64 + discoveredOn int64 + hasReadme bool + nFiles uint + // values below 0 indicates that no data is available: + nSeeders int + nLeechers int + updatedOn int +} + +func MakeDatabase(rawURL string) (Database, error) { + url_, err := url.Parse(rawURL) + if err != nil { + return nil, err + } + + switch url_.Scheme { + case "sqlite3": + return makeSqlite3Database(url_) + + case "postgresql": + return nil, fmt.Errorf("postgresql is not yet supported!") + + case "mysql": + return nil, fmt.Errorf("mysql is not yet supported!") + } + + return nil, fmt.Errorf("unknown URI scheme (database engine)!") +} diff --git a/src/persistence/mysql.go b/src/persistence/mysql.go new file mode 100644 index 0000000..dc7cf83 --- /dev/null +++ b/src/persistence/mysql.go @@ -0,0 +1 @@ +package persistence diff --git a/src/persistence/postgresql.go b/src/persistence/postgresql.go new file mode 100644 index 0000000..dc7cf83 --- /dev/null +++ b/src/persistence/postgresql.go @@ -0,0 +1 @@ +package persistence diff --git a/src/persistence/sqlite3.go b/src/persistence/sqlite3.go new file mode 100644 index 0000000..6fed455 --- /dev/null +++ b/src/persistence/sqlite3.go @@ -0,0 +1,427 @@ +package persistence + +import ( + "net/url" + "path" + "os" + "fmt" + "database/sql" + "regexp" + + "go.uber.org/zap" + "time" +) + +type sqlite3Database struct { + conn *sql.DB +} + +func (db *sqlite3Database) Engine() databaseEngine { + return SQLITE3_ENGINE +} + +func makeSqlite3Database(url_ *url.URL) (Database, error) { + db := new(sqlite3Database) + + dbDir, _ := path.Split(url_.Path) + if err := os.MkdirAll(dbDir, 0755); err != nil { + return nil, fmt.Errorf("for directory `%s`: %s", dbDir, err.Error()) + } + + var err error + db.conn, err = sql.Open("sqlite3", url_.Path) + if err != nil { + return nil, err + } + + // > Open may just validate its arguments without creating a connection to the database. To + // > verify that the data source name is valid, call Ping. + // https://golang.org/pkg/database/sql/#Open + if err = db.conn.Ping(); err != nil { + return nil, err + } + + if err := db.setupDatabase(); err != nil { + return nil, err + } + + return db, nil +} + +func (db *sqlite3Database) DoesTorrentExist(infoHash []byte) (bool, error) { + rows, err := db.conn.Query("SELECT 1 FROM torrents WHERE info_hash = ?;", infoHash) + if err != nil { + return false, err; + } + + // If rows.Next() returns true, meaning that the torrent is in the database, return true; else + // return false. + exists := rows.Next() + + if err = rows.Close(); err != nil { + return false, err; + } + + return exists, nil; +} + +func (db *sqlite3Database) GiveAnIncompleteTorrent(pathRegex *regexp.Regexp, maxSize uint) (infoHash []byte, path string, err error) { + rows, err := db.conn.Query("SELECT info_hash FROM torrents WHERE has_readme = 0;") + if err != nil { + return nil, "", err + } + + if rows.Next() != true { + return nil, "", nil + } + + if err = rows.Scan(&infoHash); err != nil { + return nil, "", err + } + + if err = rows.Close(); err != nil { + return nil, "", err + } + + // TODO + return infoHash, "", nil +} + +func (db *sqlite3Database) GiveAStaleTorrent() (infoHash []byte, err error) { + // TODO + return nil, nil +} + +func (db *sqlite3Database) AddNewTorrent(infoHash []byte, name string, files []File) error { + // Although we check whether the torrent exists in the database before asking MetadataSink to + // fetch its metadata, the torrent can also exists in the Sink before that. Now, if a torrent in + // the sink is still being fetched, that's still not a problem as we just add the new peer for + // the torrent and exit, but if the torrent is complete (i.e. its metadata) and if its waiting + // in the channel to be received, a race condition arises when we query the database and seeing + // that it doesn't exists there, add it to the sink. + // Hence check for the last time whether the torrent exists in the database, and only if not, + // add it. + exists, err := db.DoesTorrentExist(infoHash) + if err != nil { + return err; + } else if exists { + return nil; + } + + tx, err := db.conn.Begin() + if err != nil { + return err + } + // If everything goes as planned and no error occurs, we will commit the transaction before + // returning from the function so the tx.Rollback() call will fail, trying to rollback a + // committed transaction. BUT, if an error occurs, we'll get our transaction rollback'ed, which + // is nice. + defer tx.Rollback() + + var total_size int64 = 0 + for _, file := range files { + total_size += file.Size + } + + res, err := tx.Exec(` + INSERT INTO torrents ( + info_hash, + name, + total_size, + discovered_on, + n_files, + ) VALUES (?, ?, ?, ?, ?, ?, ?, ?); + `, infoHash, name, total_size, time.Now().Unix(), len(files)) + if err != nil { + return err + } + + var lastInsertId int64 + if lastInsertId, err = res.LastInsertId(); err != nil { + return fmt.Errorf("sql.Result.LastInsertId()! %s", err.Error()) + } + + for _, file := range files { + _, err = tx.Exec("INSERT INTO files (torrent_id, size, path) VALUES (?, ?, ?);", + lastInsertId, file.Size, file.Path, + ) + if err != nil { + return err + } + } + + err = tx.Commit() + if err != nil { + return err + } + + return nil +} + +func (db *sqlite3Database) AddReadme(infoHash []byte, path string, content string) error { + _, err := db.conn.Exec( + `UPDATE files SET is_readme = 1, content = ? + WHERE path = ? AND (SELECT id FROM torrents WHERE info_hash = ?) = torrent_id;`, + content, path, infoHash, + ) + if err != nil { + return err + } + + return nil +} + +func (db *sqlite3Database) Close() error { + return db.conn.Close() +} + +func (db *sqlite3Database) GetNumberOfTorrents() (uint, error) { + // COUNT(ROWID) is much more inefficient since it scans the whole table, so use MAX(ROWID) + rows, err := db.conn.Query("SELECT MAX(ROWID) FROM torrents;") + if err != nil { + return 0, err + } + + if rows.Next() != true { + fmt.Errorf("No rows returned from `SELECT MAX(ROWID)`!") + } + + var n uint + if err = rows.Scan(&n); err != nil { + return 0, err + } + + if err = rows.Close(); err != nil { + return 0, err + } + + return n, nil +} + +func (db *sqlite3Database) NewestTorrents(n uint) ([]TorrentMetadata, error) { + rows, err := db.conn.Query(` + SELECT + info_hash, + name, + total_size, + discovered_on, + has_readme, + n_files, + n_seeders, + n_leechers, + updated_on + FROM torrents + ORDER BY discovered_on DESC LIMIT ?; + `, n, + ) + if err != nil { + return nil, err + } + + var torrents []TorrentMetadata + for rows.Next() { + tm := new(TorrentMetadata) + rows.Scan( + &tm.infoHash, &tm.name, &tm.discoveredOn, &tm.hasReadme, &tm.nFiles, &tm.nSeeders, + &tm.nLeechers, &tm.updatedOn, + ) + torrents = append(torrents, *tm) + } + + if err = rows.Close(); err != nil { + return nil, err + } + + return torrents, nil +} + +func (db *sqlite3Database) SearchTorrents(query string, orderBy orderingCriteria, descending bool, mustHaveReadme bool) ([]TorrentMetadata, error) { // TODO + // TODO: + return nil, nil +} + +func (db *sqlite3Database) GetTorrent(infoHash []byte) (*TorrentMetadata, error) { + rows, err := db.conn.Query( + `SELECT + info_hash, + name, + size, + discovered_on, + has_readme, + n_files, + n_seeders, + n_leechers, + updated_on + FROM torrents + WHERE info_hash = ?`, + infoHash, + ) + if err != nil { + return nil, err + } + + if rows.Next() != true { + return nil, nil + } + + tm := new(TorrentMetadata) + rows.Scan( + &tm.infoHash, &tm.name, &tm.discoveredOn, &tm.hasReadme, &tm.nFiles, &tm.nSeeders, + &tm.nLeechers, &tm.updatedOn, + ) + + if err = rows.Close(); err != nil { + return nil, err + } + + return tm, nil +} + +func (db *sqlite3Database) GetFiles(infoHash []byte) ([]File, error) { + // TODO + return nil, nil +} + +func (db *sqlite3Database) GetReadme(infoHash []byte) (string, error) { + // TODO + return "", nil +} + + +func (db *sqlite3Database) GetStatistics(from ISO8601, period uint) (*Statistics, error) { + // TODO + return nil, nil +} + +func (db *sqlite3Database) commitQueuedTorrents() error { + return nil +} + +func (db *sqlite3Database) setupDatabase() error { + // Enable Write-Ahead Logging for SQLite as "WAL provides more concurrency as readers do not + // block writers and a writer does not block readers. Reading and writing can proceed + // concurrently." + // Caveats: + // * Might be unsupported by OSes other than Windows and UNIXes. + // * Does not work over a network filesystem. + // * Transactions that involve changes against multiple ATTACHed databases are not atomic + // across all databases as a set. + // See: https://www.sqlite.org/wal.html + // + // Force SQLite to use disk, instead of memory, for all temporary files to reduce the memory + // footprint. + // + // Enable foreign key constraints in SQLite which are crucial to prevent programmer errors on + // our side. + _, err := db.conn.Exec(` + PRAGMA journal_mode=WAL; + PRAGMA temp_store=1; + PRAGMA foreign_keys=ON; + `) + if err != nil { + return err + } + + tx, err := db.conn.Begin() + if err != nil { + return err + } + // If everything goes as planned and no error occurs, we will commit the transaction before + // returning from the function so the tx.Rollback() call will fail, trying to rollback a + // committed transaction. BUT, if an error occurs, we'll get our transaction rollback'ed, which + // is nice. + defer tx.Rollback() + + // Essential, and valid for all user_version`s: + // TODO: "torrent_id" column of the "files" table can be NULL, how can we fix this in a new schema? + _, err = tx.Exec(` + CREATE TABLE IF NOT EXISTS torrents ( + id INTEGER PRIMARY KEY, + info_hash BLOB NOT NULL UNIQUE, + name TEXT NOT NULL, + total_size INTEGER NOT NULL CHECK(total_size > 0), + discovered_on INTEGER NOT NULL CHECK(discovered_on > 0) + ); + CREATE TABLE IF NOT EXISTS files ( + id INTEGER PRIMARY KEY, + torrent_id INTEGER REFERENCES torrents ON DELETE CASCADE ON UPDATE RESTRICT, + size INTEGER NOT NULL, + path TEXT NOT NULL + ); + `) + if err != nil { + return err + } + + // Get the user_version: + res, err := tx.Query("PRAGMA user_version;") + if err != nil { + return err + } + var userVersion int; + if res.Next() != true { + return fmt.Errorf("PRAGMA user_version did not return any rows!") + } + if err = res.Scan(&userVersion); err != nil { + return err + } + + switch userVersion { + // Upgrade from user_version 0 to 1 + // The Change: + // * `info_hash_index` is recreated as UNIQUE. + case 0: + zap.S().Warnf("Updating database schema from 0 to 1... (this might take a while)") + _, err = tx.Exec(` + DROP INDEX info_hash_index; + CREATE UNIQUE INDEX info_hash_index ON torrents (info_hash); + PRAGMA user_version = 1; + `) + if err != nil { + return err + } + fallthrough + // Upgrade from user_version 1 to 2 + // The Change: + // * Added `is_readme` and `content` columns to the `files` table, and the constraints & the + // the indices they entail. + // * Added unique index `readme_index` on `files` table. + case 1: + zap.S().Warnf("Updating database schema from 1 to 2... (this might take a while)") + // We introduce two new columns here: content BLOB, and is_readme INTEGER which we treat as + // a bool (hence the CHECK). + // The reason for the change is that as we introduce the new "readme" feature which + // downloads a readme file as a torrent descriptor, we needed to store it somewhere in the + // database with the following conditions: + // + // 1. There can be one and only one readme (content) for a given torrent; hence the + // UNIQUE INDEX on (torrent_id, is_description) (remember that SQLite treats each NULL + // value as distinct [UNIQUE], see https://sqlite.org/nulls.html). + // 2. We would like to keep the readme (content) associated with the file it came from; + // hence we modify the files table instead of the torrents table. + // + // Regarding the implementation details, following constraints arise: + // + // 1. The column is_readme is either NULL or 1, and if it is 1, then content column cannot + // be NULL (but might be an empty BLOB). Vice versa, if content column of a row is, + // NULL then is_readme must be NULL. + // + // This is to prevent unused content fields filling up the database, and to catch + // programmers' errors. + _, err = tx.Exec(` + ALTER TABLE files ADD COLUMN is_readme INTEGER CHECK (is_readme IS NULL OR is_readme=1) DEFAULT NULL; + ALTER TABLE files ADD COLUMN content BLOB CHECK((content IS NULL AND is_readme IS NULL) OR (content IS NOT NULL AND is_readme=1)) DEFAULT NULL; + CREATE UNIQUE INDEX readme_index ON files (torrent_id, is_readme); + PRAGMA user_version = 2; + `) + if err != nil { + return err + } + } + + if err = tx.Commit(); err != nil { + return err + } + + return nil +} \ No newline at end of file