-
Notifications
You must be signed in to change notification settings - Fork 1
Commit
This commit does not belong to any branch on this repository, and may belong to a fork outside of the repository.
Merge pull request #4 from xxxsen/xxxsen/feature/add_scrape_source
Xxxsen/feature/add scrape source
- Loading branch information
Showing
13 changed files
with
179 additions
and
26 deletions.
There are no files selected for viewing
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Original file line number | Diff line number | Diff line change |
---|---|---|
@@ -0,0 +1,45 @@ | ||
package parser | ||
|
||
import ( | ||
"context" | ||
"math" | ||
"strconv" | ||
"strings" | ||
"yamdc/searcher/decoder" | ||
"yamdc/searcher/utils" | ||
|
||
"github.com/xxxsen/common/logutil" | ||
"go.uber.org/zap" | ||
) | ||
|
||
func DefaultHHMMSSDurationParser(ctx context.Context) decoder.NumberParseFunc { | ||
return func(v string) int64 { | ||
res := strings.Split(v, ":") | ||
if len(res) > 3 { | ||
logutil.GetLogger(ctx).Error("invalid time format", zap.String("data", v)) | ||
return 0 | ||
} | ||
var sec int64 | ||
for i := 0; i < len(res); i++ { | ||
item := strings.TrimSpace(res[len(res)-i-1]) | ||
val, err := strconv.ParseInt(item, 10, 60) | ||
if err != nil { | ||
logutil.GetLogger(ctx).Error("invalid time format", zap.String("data", v)) | ||
return 0 | ||
} | ||
sec += val * int64(math.Pow(60, float64(i))) | ||
} | ||
return sec | ||
} | ||
} | ||
|
||
func DefaultDurationParser(ctx context.Context) decoder.NumberParseFunc { | ||
return func(v string) int64 { | ||
val, err := utils.ToDuration(v) | ||
if err != nil { | ||
logutil.GetLogger(ctx).Error("decode duration failed", zap.Error(err), zap.String("data", v)) | ||
return 0 | ||
} | ||
return val | ||
} | ||
} |
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Original file line number | Diff line number | Diff line change |
---|---|---|
@@ -0,0 +1,24 @@ | ||
package parser | ||
|
||
import ( | ||
"context" | ||
"testing" | ||
|
||
"github.com/stretchr/testify/assert" | ||
) | ||
|
||
type testPair struct { | ||
in string | ||
sec int64 | ||
} | ||
|
||
func TestHHMMSS(t *testing.T) { | ||
tests := []testPair{ | ||
{in: "01 :01: 01", sec: 1*3600 + 60 + 1}, | ||
{in: "02: 05", sec: 2*60 + 5}, | ||
} | ||
for _, tst := range tests { | ||
out := DefaultHHMMSSDurationParser(context.Background())(tst.in) | ||
assert.Equal(t, tst.sec, out) | ||
} | ||
} |
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Original file line number | Diff line number | Diff line change |
---|---|---|
|
@@ -11,4 +11,5 @@ const ( | |
SSFreeJavBt = "freejavbt" | ||
SSJavDB = "javdb" | ||
SS18AV = "18av" | ||
SSTKTube = "tktube" | ||
) |
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Original file line number | Diff line number | Diff line change |
---|---|---|
@@ -0,0 +1,86 @@ | ||
package plugin | ||
|
||
import ( | ||
"fmt" | ||
"net/http" | ||
"strings" | ||
"yamdc/model" | ||
"yamdc/number" | ||
"yamdc/searcher/decoder" | ||
"yamdc/searcher/parser" | ||
) | ||
|
||
type tktube struct { | ||
DefaultPlugin | ||
} | ||
|
||
func (p *tktube) OnPrecheckRequest(ctx *PluginContext, n *number.Number) (bool, error) { | ||
return number.IsFc2(n.GetNumberID()), nil | ||
} | ||
|
||
func (p *tktube) OnMakeHTTPRequest(ctx *PluginContext, n *number.Number) (*http.Request, error) { | ||
nid := strings.ReplaceAll(n.GetNumberID(), "-", "--") | ||
ctx.SetKey("number", n.GetNumberID()) | ||
uri := fmt.Sprintf("https://tktube.com/zh/search/%s/", nid) | ||
return http.NewRequest(http.MethodGet, uri, nil) | ||
} | ||
|
||
func (p *tktube) OnHandleHTTPRequest(ctx *PluginContext, invoker HTTPInvoker, req *http.Request) (*http.Response, error) { | ||
numberId := strings.ToUpper(ctx.GetKeyOrDefault("number", "").(string)) | ||
return HandleXPathTwoStepSearch(ctx, invoker, req, &XPathTwoStepContext{ | ||
Ps: []*XPathPair{ | ||
{ | ||
Name: "links", | ||
XPath: `//div[@id="list_videos_videos_list_search_result_items"]/div/a/@href`, | ||
}, | ||
{ | ||
Name: "names", | ||
XPath: `//div[@id="list_videos_videos_list_search_result_items"]/div/a/strong[@class="title"]/text()`, | ||
}, | ||
}, | ||
LinkSelector: func(ps []*XPathPair) (string, bool, error) { | ||
links := ps[0].Result | ||
names := ps[1].Result | ||
for i := 0; i < len(links); i++ { | ||
if strings.Contains(strings.ToUpper(names[i]), numberId) { | ||
return links[i], true, nil | ||
} | ||
} | ||
return "", false, nil | ||
}, | ||
ValidStatusCode: []int{http.StatusOK}, | ||
CheckResultCountMatch: true, | ||
LinkPrefix: "", | ||
}) | ||
} | ||
|
||
func (p *tktube) OnDecodeHTTPData(ctx *PluginContext, data []byte) (*model.AvMeta, bool, error) { | ||
dec := decoder.XPathHtmlDecoder{ | ||
TitleExpr: `//div[@class="headline"]/h1/text()`, | ||
PlotExpr: "", | ||
ActorListExpr: `//div[contains(text(), "女優:")]/a[contains(@href, "models")]/text()`, | ||
ReleaseDateExpr: `//div[@class="item"]/span[contains(text(), "加入日期:")]/em/text()`, | ||
DurationExpr: `//div[@class="item"]/span[contains(text(), "時長:")]/em/text()`, | ||
StudioExpr: "", | ||
LabelExpr: "", | ||
DirectorExpr: "", | ||
SeriesExpr: "", | ||
GenreListExpr: `//div[contains(text(), "標籤:")]/a[contains(@href, "tags")]/text()`, | ||
CoverExpr: `//meta[@property="og:image"]/@content`, | ||
PosterExpr: "", | ||
SampleImageListExpr: "", | ||
} | ||
meta, err := dec.DecodeHTML(data, | ||
decoder.WithDurationParser(parser.DefaultHHMMSSDurationParser(ctx.GetContext())), | ||
decoder.WithReleaseDateParser(parser.DefaultReleaseDateParser(ctx.GetContext())), | ||
) | ||
if err != nil { | ||
return nil, false, err | ||
} | ||
meta.Number = ctx.GetKeyOrDefault("number", "").(string) | ||
return meta, true, nil | ||
} | ||
|
||
func init() { | ||
Register(SSTKTube, PluginToCreator(&tktube{})) | ||
} |