diff --git a/go.mod b/go.mod index f3b86e6..05c2633 100755 --- a/go.mod +++ b/go.mod @@ -14,4 +14,5 @@ require ( github.com/pkg/errors v0.8.1 github.com/sirupsen/logrus v1.4.2 github.com/urfave/cli/v2 v2.0.0 + golang.org/x/net v0.0.0-20210520170846-37e1c6afe023 ) diff --git a/go.sum b/go.sum new file mode 100644 index 0000000..7e1d7ca --- /dev/null +++ b/go.sum @@ -0,0 +1,100 @@ +github.com/BurntSushi/toml v0.3.1 h1:WXkYYl6Yr3qBf1K79EBnL4mak0OimBfB0XUf9Vl28OQ= +github.com/BurntSushi/toml v0.3.1/go.mod h1:xHWCNGjB5oqiDr8zfno3MHue2Ht5sIBksp03qcyfWMU= +github.com/chromedp/cdproto v0.0.0-20191114225735-6626966fbae4 h1:QD3KxSJ59L2lxG6MXBjNHxiQO2RmxTQ3XcK+wO44WOg= +github.com/chromedp/cdproto v0.0.0-20191114225735-6626966fbae4/go.mod h1:PfAWWKJqjlGFYJEidUM6aVIWPr0EpobeyVWEEmplX7g= +github.com/chromedp/chromedp v0.5.2 h1:W8xBXQuUnd2dZK0SN/lyVwsQM7KgW+kY5HGnntms194= +github.com/chromedp/chromedp v0.5.2/go.mod h1:rsTo/xRo23KZZwFmWk2Ui79rBaVRRATCjLzNQlOFSiA= +github.com/clbanning/mxj v1.8.5-0.20200714211355-ff02cfb8ea28 h1:LdXxtjzvZYhhUaonAaAKArG3pyC67kGL3YY+6hGG8G4= +github.com/clbanning/mxj v1.8.5-0.20200714211355-ff02cfb8ea28/go.mod h1:BVjHeAH+rl9rs6f+QIpeRl0tfu10SXn1pUSa5PVGJng= +github.com/cpuguy83/go-md2man/v2 v2.0.0-20190314233015-f79a8a8ca69d/go.mod h1:maD7wRr/U5Z6m/iR4s+kqSMx2CaBsrgA7czyZG/E6dU= +github.com/cpuguy83/go-md2man/v2 v2.0.0 h1:EoUDS0afbrsXAZ9YQ9jdu/mZ2sXgT1/2yyNng4PGlyM= +github.com/cpuguy83/go-md2man/v2 v2.0.0/go.mod h1:maD7wRr/U5Z6m/iR4s+kqSMx2CaBsrgA7czyZG/E6dU= +github.com/davecgh/go-spew v1.1.0/go.mod h1:J7Y8YcW2NihsgmVo/mv3lAwl/skON4iLHjSsI+c5H38= +github.com/davecgh/go-spew v1.1.1 h1:vj9j/u1bqnvCEfJOwUhtlOARqs3+rkHYY13jYWTU97c= +github.com/davecgh/go-spew v1.1.1/go.mod h1:J7Y8YcW2NihsgmVo/mv3lAwl/skON4iLHjSsI+c5H38= +github.com/deckarep/golang-set v1.7.1 h1:SCQV0S6gTtp6itiFrTqI+pfmJ4LN85S1YzhDf9rTHJQ= +github.com/deckarep/golang-set v1.7.1/go.mod h1:93vsz/8Wt4joVM7c2AVqh+YRMiUSc14yDtF28KmMOgQ= +github.com/fatih/color v1.12.0 h1:mRhaKNwANqRgUBGKmnI5ZxEk7QXmjQeCcuYFMX2bfcc= +github.com/fatih/color v1.12.0/go.mod h1:ELkj/draVOlAH/xkhN6mQ50Qd0MPOk5AAr3maGEBuJM= +github.com/fsnotify/fsnotify v1.4.9 h1:hsms1Qyu0jgnwNXIxa+/V/PDsU6CfLf6CNO8H7IWoS4= +github.com/fsnotify/fsnotify v1.4.9/go.mod h1:znqG4EE+3YCdAaPaxE2ZRY/06pZUdp0tY4IgpuI1SZQ= +github.com/go-sql-driver/mysql v1.6.0 h1:BCTh4TKNUYmOmMUcQ3IipzF5prigylS7XXjEkfCHuOE= +github.com/go-sql-driver/mysql v1.6.0/go.mod h1:DCzpHaOWr8IXmIStZouvnhqoel9Qv2LBy8hT2VhHyBg= +github.com/gobwas/httphead v0.0.0-20180130184737-2c6c146eadee h1:s+21KNqlpePfkah2I+gwHF8xmJWRjooY+5248k6m4A0= +github.com/gobwas/httphead v0.0.0-20180130184737-2c6c146eadee/go.mod h1:L0fX3K22YWvt/FAX9NnzrNzcI4wNYi9Yku4O0LKYflo= +github.com/gobwas/pool v0.2.0 h1:QEmUOlnSjWtnpRGHF3SauEiOsy82Cup83Vf2LcMlnc8= +github.com/gobwas/pool v0.2.0/go.mod h1:q8bcK0KcYlCgd9e7WYLm9LpyS+YeLd8JVDW6WezmKEw= +github.com/gobwas/ws v1.0.2 h1:CoAavW/wd/kulfZmSIBt6p24n4j7tHgNVCjsfHVNUbo= +github.com/gobwas/ws v1.0.2/go.mod h1:szmBTxLgaFppYjEmNtny/v3w89xOydFnnZMcgRRu/EM= +github.com/gogf/gf v1.16.6 h1:Yp5YfwnGz41d1tiVqxcWXiPXyuzjTb7ax4SnPSXxDE8= +github.com/gogf/gf v1.16.6/go.mod h1:4LoHfEBl2jbVmZpVx+qk2La3zWr1V315FtF2PVZuyQ8= +github.com/gomodule/redigo v1.8.5 h1:nRAxCa+SVsyjSBrtZmG/cqb6VbTmuRzpg/PoTFlpumc= +github.com/gomodule/redigo v1.8.5/go.mod h1:P9dn9mFrCBvWhGE1wpxx6fgq7BAeLBk+UUUzlpkBYO0= +github.com/google/go-cmp v0.5.6 h1:BKbKCqvP6I+rmFHt06ZmyQtvB8xAkWdhFyr0ZUNZcxQ= +github.com/google/go-cmp v0.5.6/go.mod h1:v8dTdLbMG2kIc/vJvl+f65V22dbkXbowE6jgT/gNBxE= +github.com/gorilla/websocket v1.4.2 h1:+/TMaTYc4QFitKJxsQ7Yye35DkWvkdLcvGKqM+x0Ufc= +github.com/gorilla/websocket v1.4.2/go.mod h1:YR8l580nyteQvAITg2hZ9XVh4b55+EU/adAjf1fMHhE= +github.com/grokify/html-strip-tags-go v0.0.0-20190921062105-daaa06bf1aaf h1:wIOAyJMMen0ELGiFzlmqxdcV1yGbkyHBAB6PolcNbLA= +github.com/grokify/html-strip-tags-go v0.0.0-20190921062105-daaa06bf1aaf/go.mod h1:2Su6romC5/1VXOQMaWL2yb618ARB8iVo6/DR99A6d78= +github.com/knq/sysutil v0.0.0-20191005231841-15668db23d08 h1:V0an7KRw92wmJysvFvtqtKMAPmvS5O0jtB0nYo6t+gs= +github.com/knq/sysutil v0.0.0-20191005231841-15668db23d08/go.mod h1:dFWs1zEqDjFtnBXsd1vPOZaLsESovai349994nHx3e0= +github.com/konsorten/go-windows-terminal-sequences v1.0.1 h1:mweAR1A6xJ3oS2pRaGiHgQ4OO8tzTaLawm8vnODuwDk= +github.com/konsorten/go-windows-terminal-sequences v1.0.1/go.mod h1:T0+1ngSBFLxvqU3pZ+m/2kptfBszLMUkC4ZK/EgS/cQ= +github.com/mailru/easyjson v0.7.0 h1:aizVhC/NAAcKWb+5QsU1iNOZb4Yws5UO2I+aIprQITM= +github.com/mailru/easyjson v0.7.0/go.mod h1:KAzv3t3aY1NaHWoQz1+4F1ccyAH66Jk7yos7ldAVICs= +github.com/mattn/go-colorable v0.1.8 h1:c1ghPdyEDarC70ftn0y+A/Ee++9zz8ljHG1b13eJ0s8= +github.com/mattn/go-colorable v0.1.8/go.mod h1:u6P/XSegPjTcexA+o6vUJrdnUu04hMope9wVRipJSqc= +github.com/mattn/go-isatty v0.0.12 h1:wuysRhFDzyxgEmMf5xjvJ2M9dZoWAXNNr5LSBS7uHXY= +github.com/mattn/go-isatty v0.0.12/go.mod h1:cbi8OIDigv2wuxKPP5vlRcQ1OAZbq2CE4Kysco4FUpU= +github.com/mattn/go-runewidth v0.0.9 h1:Lm995f3rfxdpd6TSmuVCHVb/QhupuXlYr8sCI/QdE+0= +github.com/mattn/go-runewidth v0.0.9/go.mod h1:H031xJmbD/WCDINGzjvQ9THkh0rPKHF+m2gUSrubnMI= +github.com/olekukonko/tablewriter v0.0.5 h1:P2Ga83D34wi1o9J6Wh1mRuqd4mF/x/lgBS7N7AbDhec= +github.com/olekukonko/tablewriter v0.0.5/go.mod h1:hPp6KlRPjbx+hW8ykQs1w3UBbZlj6HuIJcUGPhkA7kY= +github.com/panjf2000/ants/v2 v2.2.2 h1:TWzusBjq/IflXhy+/S6u5wmMLCBdJnB9tPIx9Zmhvok= +github.com/panjf2000/ants/v2 v2.2.2/go.mod h1:1GFm8bV8nyCQvU5K4WvBCTG1/YBFOD2VzjffD8fV55A= +github.com/pkg/errors v0.8.1 h1:iURUrRGxPUNPdy5/HRSm+Yj6okJ6UtLINN0Q9M4+h3I= +github.com/pkg/errors v0.8.1/go.mod h1:bwawxfHBFNV+L2hUp1rHADufV3IMtnDRdf1r5NINEl0= +github.com/pmezard/go-difflib v1.0.0 h1:4DBwDE0NGyQoBHbLQYPwSUPoCMWR5BEzIk/f1lZbAQM= +github.com/pmezard/go-difflib v1.0.0/go.mod h1:iKH77koFhYxTK1pcRnkKkqfTogsbg7gZNVY4sRDYZ/4= +github.com/russross/blackfriday/v2 v2.0.1 h1:lPqVAte+HuHNfhJ/0LC98ESWRz8afy9tM/0RK8m9o+Q= +github.com/russross/blackfriday/v2 v2.0.1/go.mod h1:+Rmxgy9KzJVeS9/2gXHxylqXiyQDYRxCVz55jmeOWTM= +github.com/shurcooL/sanitized_anchor_name v1.0.0 h1:PdmoCO6wvbs+7yrJyMORt4/BmY5IYyJwS/kOiWx8mHo= +github.com/shurcooL/sanitized_anchor_name v1.0.0/go.mod h1:1NzhyTcUVG4SuEtjjoZeVRXNmyL/1OwPU0+IJeTBvfc= +github.com/sirupsen/logrus v1.4.2 h1:SPIRibHv4MatM3XXNO2BJeFLZwZ2LvZgfQ5+UNI2im4= +github.com/sirupsen/logrus v1.4.2/go.mod h1:tLMulIdttU9McNUspp0xgXVQah82FyeX6MwdIuYE2rE= +github.com/stretchr/objx v0.1.0/go.mod h1:HFkY916IF+rwdDfMAkV7OtwuqBVzrE8GR6GFx+wExME= +github.com/stretchr/objx v0.1.1/go.mod h1:HFkY916IF+rwdDfMAkV7OtwuqBVzrE8GR6GFx+wExME= +github.com/stretchr/testify v1.2.2/go.mod h1:a8OnRcib4nhh0OaRAV+Yts87kKdq0PP7pXfy6kDkUVs= +github.com/stretchr/testify v1.5.1/go.mod h1:5W2xD1RspED5o8YsWQXVCued0rvSQ+mT+I5cxcmMvtA= +github.com/stretchr/testify v1.7.0 h1:nwc3DEeHmmLAfoZucVR881uASk0Mfjw8xYJ99tb5CcY= +github.com/stretchr/testify v1.7.0/go.mod h1:6Fq8oRcR53rry900zMqJjRRixrwX3KX962/h/Wwjteg= +github.com/urfave/cli/v2 v2.0.0 h1:+HU9SCbu8GnEUFtIBfuUNXN39ofWViIEJIp6SURMpCg= +github.com/urfave/cli/v2 v2.0.0/go.mod h1:SE9GqnLQmjVa0iPEY0f1w3ygNIYcIJ0OKPMoW2caLfQ= +go.opentelemetry.io/otel v1.0.0-RC2 h1:SHhxSjB+omnGZPgGlKe+QMp3MyazcOHdQ8qwo89oKbg= +go.opentelemetry.io/otel v1.0.0-RC2/go.mod h1:w1thVQ7qbAy8MHb0IFj8a5Q2QU0l2ksf8u/CN8m3NOM= +go.opentelemetry.io/otel/oteltest v1.0.0-RC2 h1:xNKqMhlZYkASSyvF4JwObZFMq0jhFN3c3SP+2rCzVPk= +go.opentelemetry.io/otel/oteltest v1.0.0-RC2/go.mod h1:kiQ4tw5tAL4JLTbcOYwK1CWI1HkT5aiLzHovgOVnz/A= +go.opentelemetry.io/otel/trace v1.0.0-RC2 h1:dunAP0qDULMIT82atj34m5RgvsIK6LcsXf1c/MsYg1w= +go.opentelemetry.io/otel/trace v1.0.0-RC2/go.mod h1:JPQ+z6nNw9mqEGT8o3eoPTdnNI+Aj5JcxEsVGREIAy4= +golang.org/x/net v0.0.0-20210520170846-37e1c6afe023 h1:ADo5wSpq2gqaCGQWzk7S5vd//0iyyLeAratkEoG5dLE= +golang.org/x/net v0.0.0-20210520170846-37e1c6afe023/go.mod h1:9nx3DQGgdP8bBQD5qxJ1jj9UTztislL4KSBs9R2vV5Y= +golang.org/x/sys v0.0.0-20190422165155-953cdadca894/go.mod h1:h1NjWce9XRLGQEsW7wpKNCjG9DtNlClVuFLEZdDNbEs= +golang.org/x/sys v0.0.0-20191005200804-aed5e4c7ecf9/go.mod h1:h1NjWce9XRLGQEsW7wpKNCjG9DtNlClVuFLEZdDNbEs= +golang.org/x/sys v0.0.0-20191113165036-4c7a9d0fe056/go.mod h1:h1NjWce9XRLGQEsW7wpKNCjG9DtNlClVuFLEZdDNbEs= +golang.org/x/sys v0.0.0-20200116001909-b77594299b42/go.mod h1:h1NjWce9XRLGQEsW7wpKNCjG9DtNlClVuFLEZdDNbEs= +golang.org/x/sys v0.0.0-20200223170610-d5e6a3e2c0ae/go.mod h1:h1NjWce9XRLGQEsW7wpKNCjG9DtNlClVuFLEZdDNbEs= +golang.org/x/sys v0.0.0-20201119102817-f84b799fce68/go.mod h1:h1NjWce9XRLGQEsW7wpKNCjG9DtNlClVuFLEZdDNbEs= +golang.org/x/sys v0.0.0-20210423082822-04245dca01da h1:b3NXsE2LusjYGGjL5bxEVZZORm/YEFFrWFjR8eFrw/c= +golang.org/x/sys v0.0.0-20210423082822-04245dca01da/go.mod h1:h1NjWce9XRLGQEsW7wpKNCjG9DtNlClVuFLEZdDNbEs= +golang.org/x/term v0.0.0-20201126162022-7de9c90e9dd1/go.mod h1:bj7SfCRtBDWHUb9snDiAeCFNEtKQo2Wmx5Cou7ajbmo= +golang.org/x/text v0.3.6 h1:aRYxNxv6iGQlyVaZmk6ZgYEDa+Jg18DxebPSrd6bg1M= +golang.org/x/text v0.3.6/go.mod h1:5Zoc/QRtKVWzQhOtBMvqHzDpF6irO9z98xDceosuGiQ= +golang.org/x/tools v0.0.0-20180917221912-90fa682c2a6e/go.mod h1:n7NCudcB/nEzxVGmLbDWY5pfWTLqBcC2KZ6jyYvM4mQ= +golang.org/x/xerrors v0.0.0-20191204190536-9bdfabe68543 h1:E7g+9GITq07hpfrRu66IVDexMakfv52eLZ2CXBWiKr4= +golang.org/x/xerrors v0.0.0-20191204190536-9bdfabe68543/go.mod h1:I/5z698sn9Ka8TeJc9MKroUUfqBBauWjQqLJ2OPfmY0= +gopkg.in/check.v1 v0.0.0-20161208181325-20d25e280405 h1:yhCVgyC4o1eVCa2tZl7eS0r+SDo693bJlVdllGtEeKM= +gopkg.in/check.v1 v0.0.0-20161208181325-20d25e280405/go.mod h1:Co6ibVJAznAaIkqp8huTwlJQCZ016jof/cbN4VW5Yz0= +gopkg.in/yaml.v2 v2.2.2/go.mod h1:hI93XBmqTisBFMUTm0b8Fm+jr3Dg1NNxqwp+5A1VGuI= +gopkg.in/yaml.v3 v3.0.0-20200313102051-9f266ea9e77c/go.mod h1:K4uyk7z7BCEPqu6E+C64Yfv1cQ7kz7rIZviUmN+EgEM= +gopkg.in/yaml.v3 v3.0.0-20210107192922-496545a6307b h1:h8qDotaEPuJATrMmW04NCwg7v22aHH28wwpauUhK9Oo= +gopkg.in/yaml.v3 v3.0.0-20210107192922-496545a6307b/go.mod h1:K4uyk7z7BCEPqu6E+C64Yfv1cQ7kz7rIZviUmN+EgEM= diff --git a/pkg/model/url.go b/pkg/model/url.go index ca74a84..af1764a 100755 --- a/pkg/model/url.go +++ b/pkg/model/url.go @@ -1,12 +1,15 @@ package model import ( - "crawlergo/pkg/tools/requests" "errors" "fmt" "net/url" "regexp" "strings" + + "golang.org/x/net/publicsuffix" + + "crawlergo/pkg/tools/requests" ) type URL struct { @@ -122,17 +125,20 @@ func (u *URL) NavigationUrl() string { */ func (u *URL) RootDomain() string { domain := u.Hostname() - if strings.Count(domain, ".") == 1 { - return domain + suffix, icann := publicsuffix.PublicSuffix(strings.ToLower(domain)) + // 如果不是 icann 的域名，返回空字符串 + if !icann { + return "" } - - parts := strings.Split(domain, ".") - if len(parts) >= 2 { - parts = parts[len(parts)-2:] - return strings.Join(parts, ".") - } else { + i := len(domain) - len(suffix) - 1 + // 如果域名错误 + if i <= 0 { + return "" + } + if domain[i] != '.' { return "" } + return domain[1+strings.LastIndex(domain[:i], "."):] } /** diff --git a/pkg/model/url_test.go b/pkg/model/url_test.go new file mode 100644 index 0000000..8c775a4 --- /dev/null +++ b/pkg/model/url_test.go @@ -0,0 +1,45 @@ +package model + +import ( + "net/url" + "testing" + + "golang.org/x/net/publicsuffix" +) + +var ( + rootDomainTestCases = []struct { + domain string + rootDomain string + wantICANN bool + }{ + {"www.amazon.co.uk", "amazon.co.uk", true}, + {"www.baidu.com", "baidu.com", true}, + {"www.baidu.com.cn", "baidu.com.cn", true}, + {"www.pku.edu.cn", "pku.edu.cn", true}, + {"www.example1.debian.org", "debian.org", true}, + {"www.golang.dev", "golang.dev", true}, + // 以下都是一些特殊的 case，主要包括一些特殊的域名和私有域名，一般情况遇不到 + // error domains + {"com.cn", "", true}, + // not an icann domain + {"www.example0.debian.net", "", false}, + {"s3.cn-north-1.amazonaws.com.cn", "", false}, + {"www.0emm.com", "", false}, + {"there.is.no.such-tld", "", false}, + } +) + +func TestRootDomain(t *testing.T) { + for _, tc := range rootDomainTestCases { + u := &URL{url.URL{Host: tc.domain}} + rootDomain := u.RootDomain() + _, icann := publicsuffix.PublicSuffix(u.Hostname()) + if rootDomain != tc.rootDomain { + t.Errorf("%s parse root domain failed", tc.domain) + } + if icann != tc.wantICANN { + t.Errorf("%s not an icann domain", tc.domain) + } + } +}