SoXX
3be16a9277
First implementation of the new algorithm that got proposed in issue #5 Reviewed-on: anthrove/e621-to-graph#20 Reviewed-by: Lennard Brinkhaus <lennard.brinkhaus@noreply.localhost> Reviewed-by: daskadse <daskadse@noreply.localhost> Co-authored-by: SoXX <soxx@fenpa.ws> Co-committed-by: SoXX <soxx@fenpa.ws>
300 lines
8.4 KiB
Go
300 lines
8.4 KiB
Go
package service
|
|
|
|
import (
|
|
"context"
|
|
"git.dragse.it/anthrove/e621-sdk-go/pkg/e621"
|
|
"git.dragse.it/anthrove/e621-sdk-go/pkg/e621/model"
|
|
"git.dragse.it/anthrove/e621-to-graph/pkg/logic"
|
|
"git.dragse.it/anthrove/e621-to-graph/pkg/util"
|
|
log "github.com/sirupsen/logrus"
|
|
"time"
|
|
)
|
|
|
|
func ScrapeUser(ctx context.Context, graphConnection logic.GraphConnection, client *e621.Client, username string) error {
|
|
var err error
|
|
|
|
scrapeTime := time.Now()
|
|
|
|
e621User, err := client.GetUserByName(username).Execute()
|
|
if err != nil {
|
|
return err
|
|
}
|
|
|
|
if e621User.IsBanned {
|
|
log.WithFields(log.Fields{
|
|
"e621_username": e621User.Name,
|
|
"e621_user_id": e621User.ID,
|
|
"e621_user_bann": e621User.IsBanned,
|
|
}).Info("service: user is Banned")
|
|
return nil
|
|
}
|
|
|
|
err = graphConnection.UploadUser(ctx, e621User)
|
|
if err != nil {
|
|
return err
|
|
}
|
|
|
|
currentDBFavCount, err := graphConnection.GetUserFavoriteCount(ctx, e621User.ID)
|
|
if err != nil {
|
|
return err
|
|
}
|
|
|
|
favoriteBuilder, err := client.GetFavoritesForUser(e621User.Name)
|
|
if err != nil {
|
|
return err
|
|
}
|
|
|
|
if currentDBFavCount > e621User.FavoriteCount {
|
|
//TODO: IMPLEMENT USER MARKED FOR DELETED FAVS
|
|
log.WithFields(log.Fields{
|
|
"e621_username": e621User.Name,
|
|
"e621_user_id": e621User.ID,
|
|
"e621_current_db_favorite_count": currentDBFavCount,
|
|
"e621_user_favorite_count": e621User.FavoriteCount,
|
|
}).Debug("service: user has favorites deleted")
|
|
}
|
|
|
|
var pageIndex = 1
|
|
for currentDBFavCount < e621User.FavoriteCount {
|
|
|
|
favorites, err := favoriteBuilder.Page(pageIndex).Execute()
|
|
if err != nil {
|
|
return err
|
|
}
|
|
|
|
if len(favorites) <= 0 {
|
|
return nil
|
|
}
|
|
|
|
for _, favorite := range favorites {
|
|
|
|
if currentDBFavCount == e621User.FavoriteCount {
|
|
break
|
|
}
|
|
|
|
isFaved, err := graphConnection.CheckUserToPostLink(ctx, favorite.ID, e621User.ID)
|
|
if err != nil {
|
|
return err
|
|
}
|
|
|
|
if !isFaved {
|
|
err = uploadDataToDB(ctx, graphConnection, favorite, e621User)
|
|
if err != nil {
|
|
return err
|
|
}
|
|
currentDBFavCount++
|
|
}
|
|
|
|
}
|
|
pageIndex++
|
|
}
|
|
|
|
log.WithFields(log.Fields{
|
|
"e621_username": e621User.Name,
|
|
"e621_user_id": e621User.ID,
|
|
"post_amount": e621User.FavoriteCount,
|
|
"scrape_time": time.Since(scrapeTime),
|
|
}).Info("service: finished processing favorites")
|
|
|
|
return nil
|
|
}
|
|
|
|
func uploadDataToDB(ctx context.Context, graphConnection logic.GraphConnection, favorite model.Post, e621User model.User) error {
|
|
start := time.Now()
|
|
err := uploadNodes(ctx, graphConnection, favorite)
|
|
if err != nil {
|
|
return err
|
|
}
|
|
log.WithFields(log.Fields{
|
|
"e621_username": e621User.Name,
|
|
"e621_user_id": e621User.ID,
|
|
"post_id": favorite.ID,
|
|
"upload_time": time.Since(start),
|
|
}).Debug("service: uploaded post")
|
|
|
|
start = time.Now()
|
|
err = uploadPostToUserRelationship(ctx, graphConnection, favorite, e621User)
|
|
if err != nil {
|
|
log.Fatal(err)
|
|
return err
|
|
}
|
|
|
|
err = uploadSourceTagRelationship(ctx, graphConnection, favorite)
|
|
if err != nil {
|
|
log.Fatal(err)
|
|
return err
|
|
}
|
|
|
|
err = uploadGeneralTagRelationship(ctx, graphConnection, favorite)
|
|
if err != nil {
|
|
log.Fatal(err)
|
|
return err
|
|
}
|
|
|
|
err = uploadCharacterTagtRelationship(ctx, graphConnection, favorite)
|
|
if err != nil {
|
|
log.Fatal(err)
|
|
return err
|
|
}
|
|
|
|
err = uploadCopyrightTagRelationship(ctx, graphConnection, favorite)
|
|
if err != nil {
|
|
log.Fatal(err)
|
|
return err
|
|
}
|
|
|
|
err = uploadArtistTagRelationship(ctx, graphConnection, favorite)
|
|
if err != nil {
|
|
log.Fatal(err)
|
|
return err
|
|
}
|
|
log.WithFields(log.Fields{
|
|
"e621_username": e621User.Name,
|
|
"e621_user_id": e621User.ID,
|
|
"post_id": favorite.ID,
|
|
"upload_time": time.Since(start),
|
|
}).Debug("service: made relationship")
|
|
return nil
|
|
}
|
|
|
|
// uploadNodes uploads the post to the database and creates the nodes
|
|
func uploadNodes(ctx context.Context, graphConnection logic.GraphConnection, post model.Post) error {
|
|
|
|
uniqueGeneralTags := make([]string, 0)
|
|
uniqueCharacterTags := make([]string, 0)
|
|
uniqueCopyrightTags := make([]string, 0)
|
|
uniqueArtistTags := make([]string, 0)
|
|
|
|
allGeneralTags := make([]string, 0)
|
|
allCharacterTags := make([]string, 0)
|
|
allCopyrightTags := make([]string, 0)
|
|
allArtistTags := make([]string, 0)
|
|
|
|
allGeneralTags = append(allGeneralTags, post.Tags.General...)
|
|
allCharacterTags = append(allCharacterTags, post.Tags.Character...)
|
|
allCopyrightTags = append(allCopyrightTags, post.Tags.Copyright...)
|
|
allArtistTags = append(allArtistTags, post.Tags.Artist...)
|
|
|
|
uniqueGeneralTags = util.UniqueNonEmptyElementsOf(allGeneralTags)
|
|
uniqueCharacterTags = util.UniqueNonEmptyElementsOf(allCharacterTags)
|
|
uniqueCopyrightTags = util.UniqueNonEmptyElementsOf(allCopyrightTags)
|
|
uniqueArtistTags = util.UniqueNonEmptyElementsOf(allArtistTags)
|
|
|
|
err := graphConnection.UploadPost(ctx, post.ID)
|
|
if err != nil {
|
|
return err
|
|
}
|
|
|
|
// Uploads the source to the database
|
|
for _, source := range post.Sources {
|
|
err := graphConnection.UploadSource(ctx, source)
|
|
if err != nil {
|
|
return err
|
|
}
|
|
}
|
|
|
|
for _, uniqueGeneralTag := range uniqueGeneralTags {
|
|
err := graphConnection.UploadTag(ctx, uniqueGeneralTag, "general")
|
|
if err != nil {
|
|
return err
|
|
}
|
|
}
|
|
for _, uniqueCharacterTag := range uniqueCharacterTags {
|
|
err := graphConnection.UploadTag(ctx, uniqueCharacterTag, "character")
|
|
if err != nil {
|
|
return err
|
|
}
|
|
}
|
|
|
|
for _, uniqueCopyrightTag := range uniqueCopyrightTags {
|
|
err := graphConnection.UploadTag(ctx, uniqueCopyrightTag, "copyright")
|
|
if err != nil {
|
|
return err
|
|
}
|
|
}
|
|
|
|
for _, uniqueArtistTag := range uniqueArtistTags {
|
|
err := graphConnection.UploadTag(ctx, uniqueArtistTag, "artist")
|
|
if err != nil {
|
|
return err
|
|
}
|
|
}
|
|
|
|
return nil
|
|
}
|
|
|
|
// uploadPostToUserRelationship creates a relationship between the user and the post
|
|
func uploadPostToUserRelationship(ctx context.Context, graphConnection logic.GraphConnection, post model.Post, e621User model.User) error {
|
|
err := graphConnection.EstablishUserToPostLink(ctx, post.ID, e621User.ID)
|
|
if err != nil {
|
|
return err
|
|
}
|
|
// log.Printf("Created UserToPostRelationship for user: %s to post: %d", e621User.Name, post.ID)
|
|
return nil
|
|
}
|
|
|
|
// uploadSourceTagRelationship creates a relationship between the post and the source
|
|
func uploadSourceTagRelationship(ctx context.Context, graphConnection logic.GraphConnection, post model.Post) error {
|
|
for _, source := range post.Sources {
|
|
err := graphConnection.EstablishPostToSourceLink(ctx, post.ID, source)
|
|
if err != nil {
|
|
return err
|
|
}
|
|
// log.Printf("Created PostToSourceRelationship for Post: %d to source: %s", post.ID, source)
|
|
|
|
}
|
|
return nil
|
|
}
|
|
|
|
// uploadGeneralTagRelationship creates a relationship between the post and the general tag
|
|
func uploadGeneralTagRelationship(ctx context.Context, graphConnection logic.GraphConnection, post model.Post) error {
|
|
for _, generalTag := range post.Tags.General {
|
|
err := graphConnection.EstablishPostToTagLink(ctx, post.ID, generalTag)
|
|
if err != nil {
|
|
return err
|
|
}
|
|
// log.Printf("Created PostToTagRelationship for post: %d to general tag: %s", post.ID, generalTag)
|
|
|
|
}
|
|
return nil
|
|
}
|
|
|
|
// uploadCharacterTagtRelationship creates a relationship between the post and the character tag
|
|
func uploadCharacterTagtRelationship(ctx context.Context, graphConnection logic.GraphConnection, post model.Post) error {
|
|
for _, characterTag := range post.Tags.Character {
|
|
err := graphConnection.EstablishPostToTagLink(ctx, post.ID, characterTag)
|
|
if err != nil {
|
|
return err
|
|
}
|
|
// log.Printf("Created PostToTagRelationship for post: %d to character tag: %s", post.ID, characterTag)
|
|
|
|
}
|
|
return nil
|
|
}
|
|
|
|
// uploadCopyrightTagRelationship creates a relationship between the post and the copyright tag
|
|
func uploadCopyrightTagRelationship(ctx context.Context, graphConnection logic.GraphConnection, post model.Post) error {
|
|
for _, copyrightTag := range post.Tags.Copyright {
|
|
err := graphConnection.EstablishPostToTagLink(ctx, post.ID, copyrightTag)
|
|
if err != nil {
|
|
return err
|
|
}
|
|
// log.Printf("Created PostToTagRelationship for post: %d to copyright tag: %s", post.ID, copyrightTag)
|
|
|
|
}
|
|
return nil
|
|
}
|
|
|
|
// uploadArtistTagRelationship creates a relationship between the post and the artist tag
|
|
func uploadArtistTagRelationship(ctx context.Context, graphConnection logic.GraphConnection, post model.Post) error {
|
|
for _, artistTag := range post.Tags.Artist {
|
|
err := graphConnection.EstablishPostToTagLink(ctx, post.ID, artistTag)
|
|
if err != nil {
|
|
return err
|
|
}
|
|
// log.Printf("Created PostToTagRelationship for post: %d to artist tag: %s", post.ID, artistTag)
|
|
|
|
}
|
|
return nil
|
|
}
|