templ wip 2

api/app-admin-routes.go

@@ -174,7 +174,10 @@ func (api *API) appGetAdminLogs(c *gin.Context) {
 	rAdminLogs.Filter = strings.TrimSpace(rAdminLogs.Filter)
 
 	var jqFilter *gojq.Code
-	if rAdminLogs.Filter != "" {
+	var basicFilter string
+	if strings.HasPrefix(rAdminLogs.Filter, "\"") && strings.HasSuffix(rAdminLogs.Filter, "\"") {
+		basicFilter = rAdminLogs.Filter[1 : len(rAdminLogs.Filter)-1]
+	} else if rAdminLogs.Filter != "" {
 		parsed, err := gojq.Parse(rAdminLogs.Filter)
 		if err != nil {
 			log.Error("Unable to parse JQ filter")
@@ -220,12 +223,17 @@ func (api *API) appGetAdminLogs(c *gin.Context) {
 			continue
 		}
 
-		// No Filter
-		if jqFilter == nil {
+		// Basic Filter
+		if basicFilter != "" && strings.Contains(string(rawData), basicFilter) {
 			logLines = append(logLines, string(rawData))
 			continue
 		}
 
+		// No JQ Filter
+		if jqFilter == nil {
+			continue
+		}
+
 		// Error or nil
 		result, _ := jqFilter.Run(jsonMap).Next()
 		if _, ok := result.(error); ok {

api/app-routes.go

@@ -26,7 +26,6 @@ import (
 	"reichard.io/antholume/ngtemplates/common"
 	"reichard.io/antholume/ngtemplates/pages"
 	"reichard.io/antholume/search"
-	"reichard.io/antholume/utils"
 )
 
 type backupType string
@@ -747,57 +746,50 @@ func (api *API) appSaveNewDocument(c *gin.Context) {
 	}
 
 	// Send Message
-	sendDownloadMessage("Downloading document...", gin.H{"Progress": 10})
+	sendDownloadMessage("Downloading document...", gin.H{"Progress": 1})
+
+	// Scaled Download Function
+	lastTime := time.Now()
+	downloadFunc := func(p float32) {
+		nowTime := time.Now()
+		if nowTime.Before(lastTime.Add(time.Millisecond * 500)) {
+			return
+		}
+		scaledProgress := int((p * 95 / 100) + 2)
+		sendDownloadMessage("Downloading document...", gin.H{"Progress": scaledProgress})
+		lastTime = nowTime
+	}
 
 	// Save Book
-	tempFilePath, err := search.SaveBook(rDocAdd.ID, rDocAdd.Source)
+	tempFilePath, metadata, err := search.SaveBook(rDocAdd.ID, rDocAdd.Source, downloadFunc)
 	if err != nil {
-		log.Warn("Temp File Error: ", err)
+		log.Warn("Save Book Error: ", err)
 		sendDownloadMessage("Unable to download file", gin.H{"Error": true})
 		return
 	}
 
 	// Send Message
-	sendDownloadMessage("Calculating partial MD5...", gin.H{"Progress": 60})
+	sendDownloadMessage("Saving document...", gin.H{"Progress": 98})
 
-	// Calculate Partial MD5 ID
-	partialMD5, err := utils.CalculatePartialMD5(tempFilePath)
-	if err != nil {
-		log.Warn("Partial MD5 Error: ", err)
-		sendDownloadMessage("Unable to calculate partial MD5", gin.H{"Error": true})
+	// Derive Author / Title
+	docAuthor := "Unknown"
+	if *metadata.Author != "" {
+		docAuthor = *metadata.Author
+	} else if *rDocAdd.Author != "" {
+		docAuthor = *rDocAdd.Author
 	}
 
-	// Send Message
-	sendDownloadMessage("Saving file...", gin.H{"Progress": 60})
-
-	// Derive Extension on MIME
-	fileMime, err := mimetype.DetectFile(tempFilePath)
-	if err != nil {
-		log.Warn("MIME Detect Error: ", err)
-		sendDownloadMessage("Unable to download file", gin.H{"Error": true})
-		return
-	}
-	fileExtension := fileMime.Extension()
-
-	// Derive Filename
-	var fileName string
-	if *rDocAdd.Author != "" {
-		fileName = fileName + *rDocAdd.Author
-	} else {
-		fileName = fileName + "Unknown"
+	docTitle := "Unknown"
+	if *metadata.Title != "" {
+		docTitle = *metadata.Title
+	} else if *rDocAdd.Title != "" {
+		docTitle = *rDocAdd.Title
 	}
 
-	if *rDocAdd.Title != "" {
-		fileName = fileName + " - " + *rDocAdd.Title
-	} else {
-		fileName = fileName + " - Unknown"
-	}
-
-	// Remove Slashes
+	// Remove Slashes & Sanitize File Name
+	fileName := fmt.Sprintf("%s - %s", docAuthor, docTitle)
 	fileName = strings.ReplaceAll(fileName, "/", "")
-
-	// Derive & Sanitize File Name
-	fileName = "." + filepath.Clean(fmt.Sprintf("/%s [%s]%s", fileName, *partialMD5, fileExtension))
+	fileName = "." + filepath.Clean(fmt.Sprintf("/%s [%s]%s", fileName, *metadata.PartialMD5, metadata.Type))
 
 	// Open Source File
 	sourceFile, err := os.Open(tempFilePath)
@@ -829,37 +821,15 @@ func (api *API) appSaveNewDocument(c *gin.Context) {
 	}
 
 	// Send Message
-	sendDownloadMessage("Calculating MD5...", gin.H{"Progress": 70})
-
-	// Get MD5 Hash
-	fileHash, err := getFileMD5(safePath)
-	if err != nil {
-		log.Error("Hash Failure: ", err)
-		sendDownloadMessage("Unable to calculate MD5", gin.H{"Error": true})
-		return
-	}
-
-	// Send Message
-	sendDownloadMessage("Calculating word count...", gin.H{"Progress": 80})
-
-	// Get Word Count
-	wordCount, err := metadata.GetWordCount(safePath)
-	if err != nil {
-		log.Error("Word Count Failure: ", err)
-		sendDownloadMessage("Unable to calculate word count", gin.H{"Error": true})
-		return
-	}
-
-	// Send Message
-	sendDownloadMessage("Saving to database...", gin.H{"Progress": 90})
+	sendDownloadMessage("Saving to database...", gin.H{"Progress": 99})
 
 	// Upsert Document
 	if _, err = api.db.Queries.UpsertDocument(api.db.Ctx, database.UpsertDocumentParams{
-		ID:       *partialMD5,
-		Title:    rDocAdd.Title,
-		Author:   rDocAdd.Author,
-		Md5:      fileHash,
-		Words:    wordCount,
+		ID:       *metadata.PartialMD5,
+		Title:    &docTitle,
+		Author:   &docAuthor,
+		Md5:      metadata.MD5,
+		Words:    metadata.WordCount,
 		Filepath: &fileName,
 		Basepath: &basePath,
 	}); err != nil {
@@ -872,7 +842,7 @@ func (api *API) appSaveNewDocument(c *gin.Context) {
 	sendDownloadMessage("Download Success", gin.H{
 		"Progress":   100,
 		"ButtonText": "Go to Book",
-		"ButtonHref": fmt.Sprintf("./documents/%s", *partialMD5),
+		"ButtonHref": fmt.Sprintf("./documents/%s", *metadata.PartialMD5),
 	})
 }
 

database/db.go

@@ -1,6 +1,6 @@
 // Code generated by sqlc. DO NOT EDIT.
 // versions:
-//   sqlc v1.26.0
+//   sqlc v1.27.0
 
 package database
 

database/document_user_statistics.sql

@@ -0,0 +1,228 @@
+WITH new_activity AS (
+    SELECT
+        document_id,
+        user_id
+    FROM activity
+    WHERE
+        created_at > COALESCE(
+            (SELECT MAX(last_seen) FROM document_user_statistics),
+            '1970-01-01T00:00:00Z'
+        )
+    GROUP BY user_id, document_id
+),
+
+intermediate_ga AS (
+    SELECT
+        ga.id AS row_id,
+        ga.user_id,
+        ga.document_id,
+        ga.duration,
+        ga.start_time,
+        ga.start_percentage,
+        ga.end_percentage,
+        ga.created_at,
+
+        -- Find Overlapping Events (Assign Unique ID)
+        (
+            SELECT MIN(id)
+            FROM activity AS overlap
+            WHERE
+                ga.document_id = overlap.document_id
+                AND ga.user_id = overlap.user_id
+                AND ga.start_percentage <= overlap.end_percentage
+                AND ga.end_percentage >= overlap.start_percentage
+        ) AS group_leader
+    FROM activity AS ga
+    INNER JOIN new_activity AS na
+    WHERE na.user_id = ga.user_id AND na.document_id = ga.document_id
+),
+
+grouped_activity AS (
+    SELECT
+        user_id,
+        document_id,
+        MAX(created_at) AS created_at,
+        MAX(start_time) AS start_time,
+        MIN(start_percentage) AS start_percentage,
+        MAX(end_percentage) AS end_percentage,
+        MAX(end_percentage) - MIN(start_percentage) AS read_percentage,
+        SUM(duration) AS duration
+    FROM intermediate_ga
+    GROUP BY group_leader
+),
+
+current_progress AS (
+    SELECT
+        user_id,
+        document_id,
+        COALESCE((
+            SELECT percentage
+            FROM document_progress AS dp
+            WHERE
+                dp.user_id = iga.user_id
+                AND dp.document_id = iga.document_id
+            ORDER BY created_at DESC
+            LIMIT 1
+        ), end_percentage) AS percentage
+    FROM intermediate_ga AS iga
+    GROUP BY user_id, document_id
+    HAVING MAX(start_time)
+)
+
+INSERT INTO document_user_statistics
+SELECT
+    ga.document_id,
+    ga.user_id,
+    cp.percentage,
+    MAX(ga.start_time) AS last_read,
+    MAX(ga.created_at) AS last_seen,
+    SUM(ga.read_percentage) AS read_percentage,
+
+    -- All Time WPM
+    SUM(ga.duration) AS total_time_seconds,
+    (CAST(COALESCE(d.words, 0.0) AS REAL) * SUM(read_percentage))
+        AS total_words_read,
+    (CAST(COALESCE(d.words, 0.0) AS REAL) * SUM(read_percentage))
+    / (SUM(ga.duration) / 60.0) AS total_wpm,
+
+    -- Yearly WPM
+    SUM(
+        CASE
+            WHEN
+                ga.start_time >= DATE('now', '-1 year')
+                THEN ga.duration
+            ELSE 0
+        END
+    )
+        AS yearly_time_seconds,
+    (
+        CAST(COALESCE(d.words, 0.0) AS REAL)
+        * SUM(
+            CASE
+                WHEN
+                    ga.start_time >= DATE('now', '-1 year')
+                    THEN read_percentage
+                ELSE 0
+            END
+        )
+    )
+        AS yearly_words_read,
+    COALESCE((
+        CAST(COALESCE(d.words, 0.0) AS REAL)
+        * SUM(
+            CASE
+                WHEN
+                    ga.start_time >= DATE('now', '-1 year')
+                    THEN read_percentage
+            END
+        )
+    )
+    / (
+        SUM(
+            CASE
+                WHEN
+                    ga.start_time >= DATE('now', '-1 year')
+                    THEN ga.duration
+            END
+        )
+        / 60.0
+    ), 0.0)
+        AS yearly_wpm,
+
+        -- Monthly WPM
+    SUM(
+        CASE
+            WHEN
+                ga.start_time >= DATE('now', '-1 month')
+                THEN ga.duration
+            ELSE 0
+        END
+    )
+        AS monthly_time_seconds,
+    (
+        CAST(COALESCE(d.words, 0.0) AS REAL)
+        * SUM(
+            CASE
+                WHEN
+                    ga.start_time >= DATE('now', '-1 month')
+                    THEN read_percentage
+                ELSE 0
+            END
+        )
+    )
+        AS monthly_words_read,
+    COALESCE((
+        CAST(COALESCE(d.words, 0.0) AS REAL)
+        * SUM(
+            CASE
+                WHEN
+                    ga.start_time >= DATE('now', '-1 month')
+                    THEN read_percentage
+            END
+        )
+    )
+    / (
+        SUM(
+            CASE
+                WHEN
+                    ga.start_time >= DATE('now', '-1 month')
+                    THEN ga.duration
+            END
+        )
+        / 60.0
+    ), 0.0)
+        AS monthly_wpm,
+
+        -- Weekly WPM
+    SUM(
+        CASE
+            WHEN
+                ga.start_time >= DATE('now', '-7 days')
+                THEN ga.duration
+            ELSE 0
+        END
+    )
+        AS weekly_time_seconds,
+    (
+        CAST(COALESCE(d.words, 0.0) AS REAL)
+        * SUM(
+            CASE
+                WHEN
+                    ga.start_time >= DATE('now', '-7 days')
+                    THEN read_percentage
+                ELSE 0
+            END
+        )
+    )
+        AS weekly_words_read,
+    COALESCE((
+        CAST(COALESCE(d.words, 0.0) AS REAL)
+        * SUM(
+            CASE
+                WHEN
+                    ga.start_time >= DATE('now', '-7 days')
+                    THEN read_percentage
+            END
+        )
+    )
+    / (
+        SUM(
+            CASE
+                WHEN
+                    ga.start_time >= DATE('now', '-7 days')
+                    THEN ga.duration
+            END
+        )
+        / 60.0
+    ), 0.0)
+        AS weekly_wpm
+
+FROM grouped_activity AS ga
+INNER JOIN
+    current_progress AS cp
+    ON ga.user_id = cp.user_id AND ga.document_id = cp.document_id
+INNER JOIN
+    documents AS d
+    ON ga.document_id = d.id
+GROUP BY ga.document_id, ga.user_id
+ORDER BY total_wpm DESC;

database/manager.go

@@ -28,8 +28,11 @@ type DBManager struct {
 //go:embed schema.sql
 var ddl string
 
-//go:embed views.sql
-var views string
+//go:embed user_streaks.sql
+var user_streaks string
+
+//go:embed document_user_statistics.sql
+var document_user_statistics string
 
 //go:embed migrations/*
 var migrations embed.FS
@@ -41,6 +44,11 @@ func init() {
 		Deterministic: true,
 		Scalar:        localTime,
 	})
+	sqlite.MustRegisterFunction("LOCAL_DATE", &sqlite.FunctionImpl{
+		NArgs:         2,
+		Deterministic: true,
+		Scalar:        localDate,
+	})
 }
 
 // NewMgr Returns an initialized manager
@@ -104,12 +112,6 @@ func (dbm *DBManager) init() error {
 		return err
 	}
 
-	// Execute views
-	if _, err := dbm.DB.Exec(views, nil); err != nil {
-		log.Panicf("Error executing views: %v", err)
-		return err
-	}
-
 	// Update settings
 	err = dbm.updateSettings()
 	if err != nil {
@@ -144,21 +146,13 @@ func (dbm *DBManager) Reload() error {
 // CacheTempTables clears existing statistics and recalculates
 func (dbm *DBManager) CacheTempTables() error {
 	start := time.Now()
-	user_streaks_sql := `
-	  DELETE FROM user_streaks;
-	  INSERT INTO user_streaks SELECT * FROM view_user_streaks;
-	`
-	if _, err := dbm.DB.ExecContext(dbm.Ctx, user_streaks_sql); err != nil {
+	if _, err := dbm.DB.ExecContext(dbm.Ctx, user_streaks); err != nil {
 		return err
 	}
 	log.Debug("Cached 'user_streaks' in: ", time.Since(start))
 
 	start = time.Now()
-	document_statistics_sql := `
-	  DELETE FROM document_user_statistics;
-	  INSERT INTO document_user_statistics SELECT * FROM view_document_user_statistics;
-	`
-	if _, err := dbm.DB.ExecContext(dbm.Ctx, document_statistics_sql); err != nil {
+	if _, err := dbm.DB.ExecContext(dbm.Ctx, document_user_statistics); err != nil {
 		return err
 	}
 	log.Debug("Cached 'document_user_statistics' in: ", time.Since(start))
@@ -240,5 +234,30 @@ func localTime(ctx *sqlite.FunctionContext, args []driver.Value) (driver.Value,
 		return nil, errors.New("unable to parse time")
 	}
 
-	return formattedTime.In(timeZone).Format("2006-01-02 15:04:05.000"), nil
+	return formattedTime.In(timeZone).Format(time.RFC3339), nil
+}
+
+// localDate is a custom SQL function that is registered as LOCAL_DATE in the init function
+func localDate(ctx *sqlite.FunctionContext, args []driver.Value) (driver.Value, error) {
+	timeStr, ok := args[0].(string)
+	if !ok {
+		return nil, errors.New("both arguments to TZTime must be strings")
+	}
+
+	timeZoneStr, ok := args[1].(string)
+	if !ok {
+		return nil, errors.New("both arguments to TZTime must be strings")
+	}
+
+	timeZone, err := time.LoadLocation(timeZoneStr)
+	if err != nil {
+		return nil, errors.New("unable to parse timezone")
+	}
+
+	formattedTime, err := time.ParseInLocation(time.RFC3339, timeStr, time.UTC)
+	if err != nil {
+		return nil, errors.New("unable to parse time")
+	}
+
+	return formattedTime.In(timeZone).Format("2006-01-02"), nil
 }

database/models.go

@@ -1,6 +1,6 @@
 // Code generated by sqlc. DO NOT EDIT.
 // versions:
-//   sqlc v1.26.0
+//   sqlc v1.27.0
 
 package database
 
@@ -62,6 +62,7 @@ type DocumentUserStatistic struct {
 	UserID             string  `json:"user_id"`
 	Percentage         float64 `json:"percentage"`
 	LastRead           string  `json:"last_read"`
+	LastSeen           string  `json:"last_seen"`
 	ReadPercentage     float64 `json:"read_percentage"`
 	TotalTimeSeconds   int64   `json:"total_time_seconds"`
 	TotalWordsRead     int64   `json:"total_words_read"`
@@ -115,4 +116,8 @@ type UserStreak struct {
 	CurrentStreak          int64  `json:"current_streak"`
 	CurrentStreakStartDate string `json:"current_streak_start_date"`
 	CurrentStreakEndDate   string `json:"current_streak_end_date"`
+	LastTimezone           string `json:"last_timezone"`
+	LastSeen               string `json:"last_seen"`
+	LastRecord             string `json:"last_record"`
+	LastCalculated         string `json:"last_calculated"`
 }

database/query.sql

@@ -67,7 +67,7 @@ WITH filtered_activity AS (
 SELECT
     document_id,
     device_id,
-    CAST(STRFTIME('%Y-%m-%d %H:%M:%S', LOCAL_TIME(activity.start_time, users.timezone)) AS TEXT) AS start_time,
+    LOCAL_TIME(activity.start_time, users.timezone) AS start_time,
     title,
     author,
     duration,
@@ -80,7 +80,7 @@ LEFT JOIN users ON users.id = activity.user_id;
 
 -- name: GetDailyReadStats :many
 WITH RECURSIVE last_30_days AS (
-    SELECT DATE(LOCAL_TIME(STRFTIME('%Y-%m-%dT%H:%M:%SZ', 'now'), timezone)) AS date
+    SELECT LOCAL_DATE(STRFTIME('%Y-%m-%dT%H:%M:%SZ', 'now'), timezone) AS date
     FROM users WHERE users.id = $user_id
     UNION ALL
     SELECT DATE(date, '-1 days')
@@ -99,7 +99,7 @@ filtered_activity AS (
 activity_days AS (
     SELECT
         SUM(duration) AS seconds_read,
-        DATE(LOCAL_TIME(start_time, timezone)) AS day
+        LOCAL_DATE(start_time, timezone) AS day
     FROM filtered_activity AS activity
     LEFT JOIN users ON users.id = activity.user_id
     GROUP BY day
@@ -138,8 +138,8 @@ WHERE id = $device_id LIMIT 1;
 SELECT
     devices.id,
     devices.device_name,
-    CAST(STRFTIME('%Y-%m-%d %H:%M:%S', LOCAL_TIME(devices.created_at, users.timezone)) AS TEXT) AS created_at,
-    CAST(STRFTIME('%Y-%m-%d %H:%M:%S', LOCAL_TIME(devices.last_synced, users.timezone)) AS TEXT) AS last_synced
+    LOCAL_TIME(devices.created_at, users.timezone) AS created_at,
+    LOCAL_TIME(devices.last_synced, users.timezone) AS last_synced
 FROM devices
 JOIN users ON users.id = devices.user_id
 WHERE users.id = $user_id
@@ -283,7 +283,7 @@ SELECT
     ROUND(CAST(progress.percentage AS REAL) * 100, 2) AS percentage,
     progress.document_id,
     progress.user_id,
-    CAST(STRFTIME('%Y-%m-%d %H:%M:%S', LOCAL_TIME(progress.created_at, users.timezone)) AS TEXT) AS created_at
+    LOCAL_TIME(progress.created_at, users.timezone) AS created_at
 FROM document_progress AS progress
 LEFT JOIN users ON progress.user_id = users.id
 LEFT JOIN devices ON progress.device_id = devices.id

database/query.sql.go

@@ -1,6 +1,6 @@
 // Code generated by sqlc. DO NOT EDIT.
 // versions:
-//   sqlc v1.26.0
+//   sqlc v1.27.0
 // source: query.sql
 
 package database
@@ -193,7 +193,7 @@ WITH filtered_activity AS (
 SELECT
     document_id,
     device_id,
-    CAST(STRFTIME('%Y-%m-%d %H:%M:%S', LOCAL_TIME(activity.start_time, users.timezone)) AS TEXT) AS start_time,
+    LOCAL_TIME(activity.start_time, users.timezone) AS start_time,
     title,
     author,
     duration,
@@ -214,15 +214,15 @@ type GetActivityParams struct {
 }
 
 type GetActivityRow struct {
-	DocumentID      string  `json:"document_id"`
-	DeviceID        string  `json:"device_id"`
-	StartTime       string  `json:"start_time"`
-	Title           *string `json:"title"`
-	Author          *string `json:"author"`
-	Duration        int64   `json:"duration"`
-	StartPercentage float64 `json:"start_percentage"`
-	EndPercentage   float64 `json:"end_percentage"`
-	ReadPercentage  float64 `json:"read_percentage"`
+	DocumentID      string      `json:"document_id"`
+	DeviceID        string      `json:"device_id"`
+	StartTime       interface{} `json:"start_time"`
+	Title           *string     `json:"title"`
+	Author          *string     `json:"author"`
+	Duration        int64       `json:"duration"`
+	StartPercentage float64     `json:"start_percentage"`
+	EndPercentage   float64     `json:"end_percentage"`
+	ReadPercentage  float64     `json:"read_percentage"`
 }
 
 func (q *Queries) GetActivity(ctx context.Context, arg GetActivityParams) ([]GetActivityRow, error) {
@@ -266,7 +266,7 @@ func (q *Queries) GetActivity(ctx context.Context, arg GetActivityParams) ([]Get
 
 const getDailyReadStats = `-- name: GetDailyReadStats :many
 WITH RECURSIVE last_30_days AS (
-    SELECT DATE(LOCAL_TIME(STRFTIME('%Y-%m-%dT%H:%M:%SZ', 'now'), timezone)) AS date
+    SELECT LOCAL_DATE(STRFTIME('%Y-%m-%dT%H:%M:%SZ', 'now'), timezone) AS date
     FROM users WHERE users.id = ?1
     UNION ALL
     SELECT DATE(date, '-1 days')
@@ -285,7 +285,7 @@ filtered_activity AS (
 activity_days AS (
     SELECT
         SUM(duration) AS seconds_read,
-        DATE(LOCAL_TIME(start_time, timezone)) AS day
+        LOCAL_DATE(start_time, timezone) AS day
     FROM filtered_activity AS activity
     LEFT JOIN users ON users.id = activity.user_id
     GROUP BY day
@@ -422,8 +422,8 @@ const getDevices = `-- name: GetDevices :many
 SELECT
     devices.id,
     devices.device_name,
-    CAST(STRFTIME('%Y-%m-%d %H:%M:%S', LOCAL_TIME(devices.created_at, users.timezone)) AS TEXT) AS created_at,
-    CAST(STRFTIME('%Y-%m-%d %H:%M:%S', LOCAL_TIME(devices.last_synced, users.timezone)) AS TEXT) AS last_synced
+    LOCAL_TIME(devices.created_at, users.timezone) AS created_at,
+    LOCAL_TIME(devices.last_synced, users.timezone) AS last_synced
 FROM devices
 JOIN users ON users.id = devices.user_id
 WHERE users.id = ?1
@@ -431,10 +431,10 @@ ORDER BY devices.last_synced DESC
 `
 
 type GetDevicesRow struct {
-	ID         string `json:"id"`
-	DeviceName string `json:"device_name"`
-	CreatedAt  string `json:"created_at"`
-	LastSynced string `json:"last_synced"`
+	ID         string      `json:"id"`
+	DeviceName string      `json:"device_name"`
+	CreatedAt  interface{} `json:"created_at"`
+	LastSynced interface{} `json:"last_synced"`
 }
 
 func (q *Queries) GetDevices(ctx context.Context, userID string) ([]GetDevicesRow, error) {
@@ -902,7 +902,7 @@ SELECT
     ROUND(CAST(progress.percentage AS REAL) * 100, 2) AS percentage,
     progress.document_id,
     progress.user_id,
-    CAST(STRFTIME('%Y-%m-%d %H:%M:%S', LOCAL_TIME(progress.created_at, users.timezone)) AS TEXT) AS created_at
+    LOCAL_TIME(progress.created_at, users.timezone) AS created_at
 FROM document_progress AS progress
 LEFT JOIN users ON progress.user_id = users.id
 LEFT JOIN devices ON progress.device_id = devices.id
@@ -929,13 +929,13 @@ type GetProgressParams struct {
 }
 
 type GetProgressRow struct {
-	Title      *string `json:"title"`
-	Author     *string `json:"author"`
-	DeviceName string  `json:"device_name"`
-	Percentage float64 `json:"percentage"`
-	DocumentID string  `json:"document_id"`
-	UserID     string  `json:"user_id"`
-	CreatedAt  string  `json:"created_at"`
+	Title      *string     `json:"title"`
+	Author     *string     `json:"author"`
+	DeviceName string      `json:"device_name"`
+	Percentage float64     `json:"percentage"`
+	DocumentID string      `json:"document_id"`
+	UserID     string      `json:"user_id"`
+	CreatedAt  interface{} `json:"created_at"`
 }
 
 func (q *Queries) GetProgress(ctx context.Context, arg GetProgressParams) ([]GetProgressRow, error) {
@@ -1078,7 +1078,7 @@ func (q *Queries) GetUserStatistics(ctx context.Context) ([]GetUserStatisticsRow
 }
 
 const getUserStreaks = `-- name: GetUserStreaks :many
-SELECT user_id, "window", max_streak, max_streak_start_date, max_streak_end_date, current_streak, current_streak_start_date, current_streak_end_date FROM user_streaks
+SELECT user_id, "window", max_streak, max_streak_start_date, max_streak_end_date, current_streak, current_streak_start_date, current_streak_end_date, last_timezone, last_seen, last_record, last_calculated FROM user_streaks
 WHERE user_id = ?1
 `
 
@@ -1100,6 +1100,10 @@ func (q *Queries) GetUserStreaks(ctx context.Context, userID string) ([]UserStre
 			&i.CurrentStreak,
 			&i.CurrentStreakStartDate,
 			&i.CurrentStreakEndDate,
+			&i.LastTimezone,
+			&i.LastSeen,
+			&i.LastRecord,
+			&i.LastCalculated,
 		); err != nil {
 			return nil, err
 		}

database/schema.sql

@@ -118,30 +118,13 @@ CREATE TABLE IF NOT EXISTS settings (
     created_at DATETIME NOT NULL DEFAULT (STRFTIME('%Y-%m-%dT%H:%M:%SZ', 'now'))
 );
 
----------------------------------------------------------------
------------------------ Temporary Tables ----------------------
----------------------------------------------------------------
-
--- Temporary User Streaks Table (Cached from View)
-CREATE TEMPORARY TABLE IF NOT EXISTS user_streaks (
-    user_id TEXT NOT NULL,
-    window TEXT NOT NULL,
-
-    max_streak INTEGER NOT NULL,
-    max_streak_start_date TEXT NOT NULL,
-    max_streak_end_date TEXT NOT NULL,
-
-    current_streak INTEGER NOT NULL,
-    current_streak_start_date TEXT NOT NULL,
-    current_streak_end_date TEXT NOT NULL
-);
-
--- Temporary Document User Statistics Table (Cached from View)
-CREATE TEMPORARY TABLE IF NOT EXISTS document_user_statistics (
+-- Document User Statistics Table
+CREATE TABLE IF NOT EXISTS document_user_statistics (
     document_id TEXT NOT NULL,
     user_id TEXT NOT NULL,
     percentage REAL NOT NULL,
-    last_read TEXT NOT NULL,
+    last_read DATETIME NOT NULL,
+    last_seen DATETIME NOT NULL,
     read_percentage REAL NOT NULL,
 
     total_time_seconds INTEGER NOT NULL,
@@ -163,21 +146,40 @@ CREATE TEMPORARY TABLE IF NOT EXISTS document_user_statistics (
     UNIQUE(document_id, user_id) ON CONFLICT REPLACE
 );
 
+-- User Streaks Table
+DROP TABLE IF EXISTS user_streaks;
+CREATE TABLE IF NOT EXISTS user_streaks (
+    user_id TEXT NOT NULL,
+    window TEXT NOT NULL,
+
+    max_streak INTEGER NOT NULL,
+    max_streak_start_date TEXT NOT NULL,
+    max_streak_end_date TEXT NOT NULL,
+
+    current_streak INTEGER NOT NULL,
+    current_streak_start_date TEXT NOT NULL,
+    current_streak_end_date TEXT NOT NULL,
+
+    last_timezone TEXT NOT NULL,
+    last_seen TEXT NOT NULL,
+    last_record TEXT NOT NULL,
+    last_calculated TEXT NOT NULL,
+
+    UNIQUE(user_id, window) ON CONFLICT REPLACE
+);
 
 ---------------------------------------------------------------
 --------------------------- Indexes ---------------------------
 ---------------------------------------------------------------
 
 CREATE INDEX IF NOT EXISTS activity_start_time ON activity (start_time);
+CREATE INDEX IF NOT EXISTS activity_created_at ON activity (created_at);
 CREATE INDEX IF NOT EXISTS activity_user_id ON activity (user_id);
 CREATE INDEX IF NOT EXISTS activity_user_id_document_id ON activity (
     user_id,
     document_id
 );
 
-DROP VIEW IF EXISTS view_user_streaks;
-DROP VIEW IF EXISTS view_document_user_statistics;
-
 ---------------------------------------------------------------
 --------------------------- Triggers --------------------------
 ---------------------------------------------------------------

database/user_streaks.sql

@@ -0,0 +1,154 @@
+WITH updated_users AS (
+    SELECT a.user_id
+    FROM activity AS a
+    LEFT JOIN users AS u ON u.id = a.user_id
+    LEFT JOIN user_streaks AS s ON a.user_id = s.user_id AND s.window = 'DAY'
+    WHERE
+        a.created_at > COALESCE(s.last_seen, '1970-01-01')
+        AND LOCAL_DATE(s.last_record, u.timezone) != LOCAL_DATE(a.start_time, u.timezone)
+    GROUP BY a.user_id
+),
+
+outdated_users AS (
+    SELECT
+        a.user_id,
+        u.timezone AS last_timezone,
+        MAX(a.created_at) AS last_seen,
+        MAX(a.start_time) AS last_record,
+        STRFTIME('%Y-%m-%dT%H:%M:%SZ', 'now') AS last_calculated
+    FROM activity AS a
+    LEFT JOIN users AS u ON u.id = a.user_id
+    LEFT JOIN user_streaks AS s ON a.user_id = s.user_id AND s.window = 'DAY'
+    GROUP BY a.user_id
+    HAVING
+        -- User Changed Timezones
+        s.last_timezone != u.timezone
+
+        -- Users Date Changed
+        OR LOCAL_DATE(COALESCE(s.last_calculated, '1970-01-01T00:00:00Z'), u.timezone) !=
+            LOCAL_DATE(STRFTIME('%Y-%m-%dT%H:%M:%SZ', 'now'), u.timezone)
+
+        -- User Added New Data
+        OR a.user_id IN updated_users
+),
+
+document_windows AS (
+    SELECT
+        activity.user_id,
+        users.timezone,
+        DATE(
+            LOCAL_DATE(activity.start_time, users.timezone),
+            'weekday 0', '-7 day'
+        ) AS weekly_read,
+        LOCAL_DATE(activity.start_time, users.timezone) AS daily_read
+    FROM activity
+    INNER JOIN outdated_users ON outdated_users.user_id = activity.user_id
+    LEFT JOIN users ON users.id = activity.user_id
+    GROUP BY activity.user_id, weekly_read, daily_read
+),
+
+weekly_partitions AS (
+    SELECT
+        user_id,
+        timezone,
+        'WEEK' AS "window",
+        weekly_read AS read_window,
+        ROW_NUMBER() OVER (
+            PARTITION BY user_id ORDER BY weekly_read DESC
+        ) AS seqnum
+    FROM document_windows
+    GROUP BY user_id, weekly_read
+),
+
+daily_partitions AS (
+    SELECT
+        user_id,
+        timezone,
+        'DAY' AS "window",
+        daily_read AS read_window,
+        ROW_NUMBER() OVER (
+            PARTITION BY user_id ORDER BY daily_read DESC
+        ) AS seqnum
+    FROM document_windows
+    GROUP BY user_id, daily_read
+),
+
+streaks AS (
+    SELECT
+        COUNT(*) AS streak,
+        MIN(read_window) AS start_date,
+        MAX(read_window) AS end_date,
+        window,
+        user_id,
+        timezone
+    FROM daily_partitions
+    GROUP BY
+        timezone,
+        user_id,
+        DATE(read_window, '+' || seqnum || ' day')
+
+    UNION ALL
+
+    SELECT
+        COUNT(*) AS streak,
+        MIN(read_window) AS start_date,
+        MAX(read_window) AS end_date,
+        window,
+        user_id,
+        timezone
+    FROM weekly_partitions
+    GROUP BY
+        timezone,
+        user_id,
+        DATE(read_window, '+' || (seqnum * 7) || ' day')
+),
+
+max_streak AS (
+    SELECT
+        MAX(streak) AS max_streak,
+        start_date AS max_streak_start_date,
+        end_date AS max_streak_end_date,
+        window,
+        user_id
+    FROM streaks
+    GROUP BY user_id, window
+),
+
+current_streak AS (
+    SELECT
+        streak AS current_streak,
+        start_date AS current_streak_start_date,
+        end_date AS current_streak_end_date,
+        window,
+        user_id
+    FROM streaks
+    WHERE CASE
+      WHEN window = "WEEK" THEN
+          DATE(LOCAL_DATE(STRFTIME('%Y-%m-%dT%H:%M:%SZ', 'now'), timezone), 'weekday 0', '-14 day') = current_streak_end_date
+          OR DATE(LOCAL_DATE(STRFTIME('%Y-%m-%dT%H:%M:%SZ', 'now'), timezone), 'weekday 0', '-7 day') = current_streak_end_date
+      WHEN window = "DAY" THEN
+          DATE(LOCAL_DATE(STRFTIME('%Y-%m-%dT%H:%M:%SZ', 'now'), timezone), '-1 day') = current_streak_end_date
+          OR DATE(LOCAL_DATE(STRFTIME('%Y-%m-%dT%H:%M:%SZ', 'now'), timezone)) = current_streak_end_date
+    END
+    GROUP BY user_id, window
+)
+
+INSERT INTO user_streaks
+SELECT
+    max_streak.user_id,
+    max_streak.window,
+    IFNULL(max_streak, 0) AS max_streak,
+    IFNULL(max_streak_start_date, "N/A") AS max_streak_start_date,
+    IFNULL(max_streak_end_date, "N/A") AS max_streak_end_date,
+    IFNULL(current_streak.current_streak, 0) AS current_streak,
+    IFNULL(current_streak.current_streak_start_date, "N/A") AS current_streak_start_date,
+    IFNULL(current_streak.current_streak_end_date, "N/A") AS current_streak_end_date,
+    outdated_users.last_timezone AS last_timezone,
+    outdated_users.last_seen AS last_seen,
+    outdated_users.last_record AS last_record,
+    outdated_users.last_calculated AS last_calculated
+FROM max_streak
+JOIN outdated_users ON max_streak.user_id = outdated_users.user_id
+LEFT JOIN current_streak ON
+    current_streak.user_id = max_streak.user_id
+    AND current_streak.window = max_streak.window;

database/views.sql

@@ -1,296 +0,0 @@
----------------------------------------------------------------
----------------------------- Views ----------------------------
----------------------------------------------------------------
-
---------------------------------
---------- User Streaks ---------
---------------------------------
-
-CREATE VIEW view_user_streaks AS
-
-WITH document_windows AS (
-    SELECT
-        activity.user_id,
-        users.timezone,
-        DATE(
-            LOCAL_TIME(activity.start_time, users.timezone),
-            'weekday 0', '-7 day'
-        ) AS weekly_read,
-        DATE(LOCAL_TIME(activity.start_time, users.timezone)) AS daily_read
-    FROM activity
-    LEFT JOIN users ON users.id = activity.user_id
-    GROUP BY activity.user_id, weekly_read, daily_read
-),
-weekly_partitions AS (
-    SELECT
-        user_id,
-        timezone,
-        'WEEK' AS "window",
-        weekly_read AS read_window,
-        row_number() OVER (
-            PARTITION BY user_id ORDER BY weekly_read DESC
-        ) AS seqnum
-    FROM document_windows
-    GROUP BY user_id, weekly_read
-),
-daily_partitions AS (
-    SELECT
-        user_id,
-        timezone,
-        'DAY' AS "window",
-        daily_read AS read_window,
-        row_number() OVER (
-            PARTITION BY user_id ORDER BY daily_read DESC
-        ) AS seqnum
-    FROM document_windows
-    GROUP BY user_id, daily_read
-),
-streaks AS (
-    SELECT
-        COUNT(*) AS streak,
-        MIN(read_window) AS start_date,
-        MAX(read_window) AS end_date,
-        window,
-        user_id,
-        timezone
-    FROM daily_partitions
-    GROUP BY
-        timezone,
-        user_id,
-        DATE(read_window, '+' || seqnum || ' day')
-
-    UNION ALL
-
-    SELECT
-        COUNT(*) AS streak,
-        MIN(read_window) AS start_date,
-        MAX(read_window) AS end_date,
-        window,
-        user_id,
-        timezone
-    FROM weekly_partitions
-    GROUP BY
-        timezone,
-        user_id,
-        DATE(read_window, '+' || (seqnum * 7) || ' day')
-),
-max_streak AS (
-    SELECT
-        MAX(streak) AS max_streak,
-        start_date AS max_streak_start_date,
-        end_date AS max_streak_end_date,
-        window,
-        user_id
-    FROM streaks
-    GROUP BY user_id, window
-),
-current_streak AS (
-    SELECT
-        streak AS current_streak,
-        start_date AS current_streak_start_date,
-        end_date AS current_streak_end_date,
-        window,
-        user_id
-    FROM streaks
-    WHERE CASE
-      WHEN window = "WEEK" THEN
-          DATE(LOCAL_TIME(STRFTIME('%Y-%m-%dT%H:%M:%SZ', 'now'), timezone), 'weekday 0', '-14 day') = current_streak_end_date
-          OR DATE(LOCAL_TIME(STRFTIME('%Y-%m-%dT%H:%M:%SZ', 'now'), timezone), 'weekday 0', '-7 day') = current_streak_end_date
-      WHEN window = "DAY" THEN
-          DATE(LOCAL_TIME(STRFTIME('%Y-%m-%dT%H:%M:%SZ', 'now'), timezone), '-1 day') = current_streak_end_date
-          OR DATE(LOCAL_TIME(STRFTIME('%Y-%m-%dT%H:%M:%SZ', 'now'), timezone)) = current_streak_end_date
-    END
-    GROUP BY user_id, window
-)
-SELECT
-    max_streak.user_id,
-    max_streak.window,
-    IFNULL(max_streak, 0) AS max_streak,
-    IFNULL(max_streak_start_date, "N/A") AS max_streak_start_date,
-    IFNULL(max_streak_end_date, "N/A") AS max_streak_end_date,
-    IFNULL(current_streak, 0) AS current_streak,
-    IFNULL(current_streak_start_date, "N/A") AS current_streak_start_date,
-    IFNULL(current_streak_end_date, "N/A") AS current_streak_end_date
-FROM max_streak
-LEFT JOIN current_streak ON
-    current_streak.user_id = max_streak.user_id
-    AND current_streak.window = max_streak.window;
-
---------------------------------
-------- Document Stats ---------
---------------------------------
-
-CREATE VIEW view_document_user_statistics AS
-
-WITH intermediate_ga AS (
-    SELECT
-        ga1.id AS row_id,
-        ga1.user_id,
-        ga1.document_id,
-        ga1.duration,
-        ga1.start_time,
-        ga1.start_percentage,
-        ga1.end_percentage,
-
-        -- Find Overlapping Events (Assign Unique ID)
-        (
-            SELECT MIN(id)
-            FROM activity AS ga2
-            WHERE
-                ga1.document_id = ga2.document_id
-                AND ga1.user_id = ga2.user_id
-                AND ga1.start_percentage <= ga2.end_percentage
-                AND ga1.end_percentage >= ga2.start_percentage
-        ) AS group_leader
-    FROM activity AS ga1
-),
-
-grouped_activity AS (
-    SELECT
-        user_id,
-        document_id,
-        MAX(start_time) AS start_time,
-        MIN(start_percentage) AS start_percentage,
-        MAX(end_percentage) AS end_percentage,
-        MAX(end_percentage) - MIN(start_percentage) AS read_percentage,
-        SUM(duration) AS duration
-    FROM intermediate_ga
-    GROUP BY group_leader
-),
-
-current_progress AS (
-    SELECT
-        user_id,
-        document_id,
-        COALESCE((
-            SELECT percentage
-            FROM document_progress AS dp
-            WHERE
-                dp.user_id = iga.user_id
-                AND dp.document_id = iga.document_id
-            ORDER BY created_at DESC
-            LIMIT 1
-        ), end_percentage) AS percentage
-    FROM intermediate_ga AS iga
-    GROUP BY user_id, document_id
-    HAVING MAX(start_time)
-)
-
-SELECT
-    ga.document_id,
-    ga.user_id,
-    cp.percentage,
-    MAX(start_time) AS last_read,
-    SUM(read_percentage) AS read_percentage,
-
-    -- All Time WPM
-    SUM(duration) AS total_time_seconds,
-    (CAST(COALESCE(d.words, 0.0) AS REAL) * SUM(read_percentage))
-        AS total_words_read,
-    (CAST(COALESCE(d.words, 0.0) AS REAL) * SUM(read_percentage))
-    / (SUM(duration) / 60.0) AS total_wpm,
-
-    -- Yearly WPM
-    SUM(CASE WHEN start_time >= DATE('now', '-1 year') THEN duration ELSE 0 END)
-        AS yearly_time_seconds,
-    (
-        CAST(COALESCE(d.words, 0.0) AS REAL)
-        * SUM(
-            CASE
-                WHEN start_time >= DATE('now', '-1 year') THEN read_percentage
-                ELSE 0
-            END
-        )
-    )
-        AS yearly_words_read,
-    COALESCE((
-        CAST(COALESCE(d.words, 0.0) AS REAL)
-        * SUM(
-            CASE
-                WHEN start_time >= DATE('now', '-1 year') THEN read_percentage
-            END
-        )
-    )
-    / (
-        SUM(
-            CASE
-                WHEN start_time >= DATE('now', '-1 year') THEN duration
-            END
-        )
-        / 60.0
-    ), 0.0)
-        AS yearly_wpm,
-
-    -- Monthly WPM
-    SUM(
-        CASE WHEN start_time >= DATE('now', '-1 month') THEN duration ELSE 0 END
-    )
-        AS monthly_time_seconds,
-    (
-        CAST(COALESCE(d.words, 0.0) AS REAL)
-        * SUM(
-            CASE
-                WHEN start_time >= DATE('now', '-1 month') THEN read_percentage
-                ELSE 0
-            END
-        )
-    )
-        AS monthly_words_read,
-    COALESCE((
-        CAST(COALESCE(d.words, 0.0) AS REAL)
-        * SUM(
-            CASE
-                WHEN start_time >= DATE('now', '-1 month') THEN read_percentage
-            END
-        )
-    )
-    / (
-        SUM(
-            CASE
-                WHEN start_time >= DATE('now', '-1 month') THEN duration
-            END
-        )
-        / 60.0
-    ), 0.0)
-        AS monthly_wpm,
-
-    -- Weekly WPM
-    SUM(CASE WHEN start_time >= DATE('now', '-7 days') THEN duration ELSE 0 END)
-        AS weekly_time_seconds,
-    (
-        CAST(COALESCE(d.words, 0.0) AS REAL)
-        * SUM(
-            CASE
-                WHEN start_time >= DATE('now', '-7 days') THEN read_percentage
-                ELSE 0
-            END
-        )
-    )
-        AS weekly_words_read,
-    COALESCE((
-        CAST(COALESCE(d.words, 0.0) AS REAL)
-        * SUM(
-            CASE
-                WHEN start_time >= DATE('now', '-7 days') THEN read_percentage
-            END
-        )
-    )
-    / (
-        SUM(
-            CASE
-                WHEN start_time >= DATE('now', '-7 days') THEN duration
-            END
-        )
-        / 60.0
-    ), 0.0)
-        AS weekly_wpm
-
-FROM grouped_activity AS ga
-INNER JOIN
-    current_progress AS cp
-    ON ga.user_id = cp.user_id AND ga.document_id = cp.document_id
-INNER JOIN
-    documents AS d
-    ON ga.document_id = d.id
-GROUP BY ga.document_id, ga.user_id
-ORDER BY total_wpm DESC;

search/anna.go

@@ -3,6 +3,7 @@ package search
 import (
 	"fmt"
 	"io"
+	"net/url"
 	"regexp"
 	"strings"
 
@@ -11,50 +12,14 @@ import (
 
 var commentRE = regexp.MustCompile(`(?s)<!--(.*?)-->`)
 
-func parseAnnasArchiveDownloadURL(body io.ReadCloser) (string, error) {
-	// Parse
-	defer body.Close()
-	doc, _ := goquery.NewDocumentFromReader(body)
-
-	// Return Download URL
-	downloadPath, exists := doc.Find("body > table > tbody > tr > td > a").Attr("href")
-	if !exists {
-		return "", fmt.Errorf("Download URL not found")
-	}
-
-	// Possible Funky URL
-	downloadPath = strings.ReplaceAll(downloadPath, "\\", "/")
-
-	return fmt.Sprintf("http://libgen.li/%s", downloadPath), nil
-}
-
-// getAnnasArchiveBookSelection parses potentially commented out HTML. For some reason
-// Annas Archive comments out blocks "below the fold". They aren't rendered until you
-// scroll. This attempts to parse the commented out HTML.
-func getAnnasArchiveBookSelection(rawBook *goquery.Selection) *goquery.Selection {
-	rawHTML, err := rawBook.Html()
+func searchAnnasArchive(query string) ([]SearchItem, error) {
+	searchURL := "https://annas-archive.org/search?index=&q=%s&ext=epub&sort=&lang=en"
+	url := fmt.Sprintf(searchURL, url.QueryEscape(query))
+	body, err := getPage(url)
 	if err != nil {
-		return rawBook
+		return nil, err
 	}
-
-	strippedHTML := strings.TrimSpace(rawHTML)
-	if !strings.HasPrefix(strippedHTML, "<!--") || !strings.HasSuffix(strippedHTML, "-->") {
-		return rawBook
-	}
-
-	allMatches := commentRE.FindAllStringSubmatch(strippedHTML, -1)
-	if len(allMatches) != 1 || len(allMatches[0]) != 2 {
-		return rawBook
-	}
-
-	captureGroup := allMatches[0][1]
-	docReader := strings.NewReader(captureGroup)
-	doc, err := goquery.NewDocumentFromReader(docReader)
-	if err != nil {
-		return rawBook
-	}
-
-	return doc.Selection
+	return parseAnnasArchive(body)
 }
 
 func parseAnnasArchive(body io.ReadCloser) ([]SearchItem, error) {
@@ -107,3 +72,32 @@ func parseAnnasArchive(body io.ReadCloser) ([]SearchItem, error) {
 	// Return Results
 	return allEntries, nil
 }
+
+// getAnnasArchiveBookSelection parses potentially commented out HTML. For some reason
+// Annas Archive comments out blocks "below the fold". They aren't rendered until you
+// scroll. This attempts to parse the commented out HTML.
+func getAnnasArchiveBookSelection(rawBook *goquery.Selection) *goquery.Selection {
+	rawHTML, err := rawBook.Html()
+	if err != nil {
+		return rawBook
+	}
+
+	strippedHTML := strings.TrimSpace(rawHTML)
+	if !strings.HasPrefix(strippedHTML, "<!--") || !strings.HasSuffix(strippedHTML, "-->") {
+		return rawBook
+	}
+
+	allMatches := commentRE.FindAllStringSubmatch(strippedHTML, -1)
+	if len(allMatches) != 1 || len(allMatches[0]) != 2 {
+		return rawBook
+	}
+
+	captureGroup := allMatches[0][1]
+	docReader := strings.NewReader(captureGroup)
+	doc, err := goquery.NewDocumentFromReader(docReader)
+	if err != nil {
+		return rawBook
+	}
+
+	return doc.Selection
+}

search/downloaders.go

@@ -0,0 +1,69 @@
+package search
+
+import (
+	"errors"
+	"fmt"
+	"strings"
+
+	"github.com/PuerkitoBio/goquery"
+)
+
+func getLibGenDownloadURL(md5 string, _ Source) ([]string, error) {
+	// Get Page
+	body, err := getPage("http://libgen.li/ads.php?md5=" + md5)
+	if err != nil {
+		return nil, err
+	}
+	defer body.Close()
+
+	// Parse
+	doc, err := goquery.NewDocumentFromReader(body)
+	if err != nil {
+		return nil, err
+	}
+
+	// Return Download URL
+	downloadPath, exists := doc.Find("body > table > tbody > tr > td > a").Attr("href")
+	if !exists {
+		return nil, fmt.Errorf("Download URL not found")
+	}
+
+	// Possible Funky URL
+	downloadPath = strings.ReplaceAll(downloadPath, "\\", "/")
+	return []string{fmt.Sprintf("http://libgen.li/%s", downloadPath)}, nil
+}
+
+func getLibraryDownloadURL(md5 string, source Source) ([]string, error) {
+	// Derive Info URL
+	var infoURL string
+	switch source {
+	case SOURCE_LIBGEN_FICTION, SOURCE_ANNAS_ARCHIVE:
+		infoURL = "http://library.lol/fiction/" + md5
+	case SOURCE_LIBGEN_NON_FICTION:
+		infoURL = "http://library.lol/main/" + md5
+	default:
+		return nil, errors.New("invalid source")
+	}
+
+	// Get Page
+	body, err := getPage(infoURL)
+	if err != nil {
+		return nil, err
+	}
+	defer body.Close()
+
+	// Parse
+	doc, err := goquery.NewDocumentFromReader(body)
+	if err != nil {
+		return nil, err
+	}
+
+	// Return Download URL
+	// downloadURL, _ := doc.Find("#download [href*=cloudflare]").Attr("href")
+	downloadURL, exists := doc.Find("#download h2 a").Attr("href")
+	if !exists {
+		return nil, errors.New("Download URL not found")
+	}
+
+	return []string{downloadURL}, nil
+}

search/libgen.go

@@ -3,12 +3,23 @@ package search
 import (
 	"fmt"
 	"io"
+	"net/url"
 	"strings"
 	"time"
 
 	"github.com/PuerkitoBio/goquery"
 )
 
+func searchLibGenFiction(query string) ([]SearchItem, error) {
+	searchURL := "https://libgen.is/fiction/?q=%s&language=English&format=epub"
+	url := fmt.Sprintf(searchURL, url.QueryEscape(query))
+	body, err := getPage(url)
+	if err != nil {
+		return nil, err
+	}
+	return parseLibGenFiction(body)
+}
+
 func parseLibGenFiction(body io.ReadCloser) ([]SearchItem, error) {
 	// Parse
 	defer body.Close()
@@ -62,6 +73,16 @@ func parseLibGenFiction(body io.ReadCloser) ([]SearchItem, error) {
 	return allEntries, nil
 }
 
+func searchLibGenNonFiction(query string) ([]SearchItem, error) {
+	searchURL := "https://libgen.is/search.php?req=%s"
+	url := fmt.Sprintf(searchURL, url.QueryEscape(query))
+	body, err := getPage(url)
+	if err != nil {
+		return nil, err
+	}
+	return parseLibGenNonFiction(body)
+}
+
 func parseLibGenNonFiction(body io.ReadCloser) ([]SearchItem, error) {
 	// Parse
 	defer body.Close()
@@ -106,18 +127,3 @@ func parseLibGenNonFiction(body io.ReadCloser) ([]SearchItem, error) {
 	// Return Results
 	return allEntries, nil
 }
-
-func parseLibGenDownloadURL(body io.ReadCloser) (string, error) {
-	// Parse
-	defer body.Close()
-	doc, _ := goquery.NewDocumentFromReader(body)
-
-	// Return Download URL
-	// downloadURL, _ := doc.Find("#download [href*=cloudflare]").Attr("href")
-	downloadURL, exists := doc.Find("#download h2 a").Attr("href")
-	if !exists {
-		return "", fmt.Errorf("Download URL not found")
-	}
-
-	return downloadURL, nil
-}

search/progress.go

@@ -0,0 +1,22 @@
+package search
+
+type writeCounter struct {
+	Total            int64
+	Current          int64
+	ProgressFunction func(float32)
+}
+
+func (wc *writeCounter) Write(p []byte) (int, error) {
+	n := len(p)
+	wc.Current += int64(n)
+	wc.flushProgress()
+	return n, nil
+}
+
+func (wc *writeCounter) flushProgress() {
+	if wc.ProgressFunction == nil || wc.Total < 100000 {
+		return
+	}
+	percentage := float32(wc.Current) * 100 / float32(wc.Total)
+	wc.ProgressFunction(percentage)
+}

search/search.go

@@ -2,17 +2,18 @@ package search
 
 import (
 	"crypto/tls"
+	"errors"
 	"fmt"
 	"io"
 	"net/http"
-	"net/url"
 	"os"
 	"time"
 
 	log "github.com/sirupsen/logrus"
+	"reichard.io/antholume/metadata"
 )
 
-const userAgent string = "Mozilla/5.0 (Macintosh; Intel Mac OS X 10.15; rv:127.0) Gecko/20100101 Firefox/127.0"
+const userAgent string = "Mozilla/5.0 (Macintosh; Intel Mac OS X 10_15_7) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/131.0.0.0 Safari/537.36"
 
 type Cadence string
 
@@ -21,13 +22,6 @@ const (
 	CADENCE_TOP_MONTH Cadence = "m"
 )
 
-type BookType int
-
-const (
-	BOOK_FICTION BookType = iota
-	BOOK_NON_FICTION
-)
-
 type Source string
 
 const (
@@ -47,108 +41,58 @@ type SearchItem struct {
 	UploadDate string
 }
 
-type sourceDef struct {
-	searchURL         string
-	downloadURL       string
-	parseSearchFunc   func(io.ReadCloser) ([]SearchItem, error)
-	parseDownloadFunc func(io.ReadCloser) (string, error)
+type searchFunc func(query string) (searchResults []SearchItem, err error)
+type downloadFunc func(md5 string, source Source) (downloadURL []string, err error)
+
+var searchDefs = map[Source]searchFunc{
+	SOURCE_ANNAS_ARCHIVE:      searchAnnasArchive,
+	SOURCE_LIBGEN_FICTION:     searchLibGenFiction,
+	SOURCE_LIBGEN_NON_FICTION: searchLibGenNonFiction,
 }
 
-var sourceDefs = map[Source]sourceDef{
-	SOURCE_ANNAS_ARCHIVE: {
-		searchURL:         "https://annas-archive.org/search?index=&q=%s&ext=epub&sort=&lang=en",
-		downloadURL:       "http://libgen.li/ads.php?md5=%s",
-		parseSearchFunc:   parseAnnasArchive,
-		parseDownloadFunc: parseAnnasArchiveDownloadURL,
-	},
-	SOURCE_LIBGEN_FICTION: {
-		searchURL:         "https://libgen.is/fiction/?q=%s&language=English&format=epub",
-		downloadURL:       "http://libgen.li/ads.php?md5=%s",
-		parseSearchFunc:   parseLibGenFiction,
-		parseDownloadFunc: parseAnnasArchiveDownloadURL,
-	},
-	SOURCE_LIBGEN_NON_FICTION: {
-		searchURL:         "https://libgen.is/search.php?req=%s",
-		downloadURL:       "http://libgen.li/ads.php?md5=%s",
-		parseSearchFunc:   parseLibGenNonFiction,
-		parseDownloadFunc: parseAnnasArchiveDownloadURL,
-	},
+var downloadFuncs = []downloadFunc{
+	getLibGenDownloadURL,
+	getLibraryDownloadURL,
 }
 
 func SearchBook(query string, source Source) ([]SearchItem, error) {
-	def := sourceDefs[source]
-	log.Debug("Source: ", def)
-	url := fmt.Sprintf(def.searchURL, url.QueryEscape(query))
-	body, err := getPage(url)
-	if err != nil {
-		return nil, err
+	searchFunc, found := searchDefs[source]
+	if !found {
+		return nil, fmt.Errorf("invalid source: %s", source)
 	}
-	return def.parseSearchFunc(body)
+	log.Debug("Source: ", source)
+	return searchFunc(query)
 }
 
-func SaveBook(id string, source Source) (string, error) {
-	def := sourceDefs[source]
-	log.Debug("Source: ", def)
-	url := fmt.Sprintf(def.downloadURL, id)
+func SaveBook(md5 string, source Source, progressFunc func(float32)) (string, *metadata.MetadataInfo, error) {
+	for _, f := range downloadFuncs {
+		downloadURLs, err := f(md5, source)
+		if err != nil {
+			log.Error("failed to acquire download urls")
+			continue
+		}
 
-	body, err := getPage(url)
-	if err != nil {
-		return "", err
+		for _, bookURL := range downloadURLs {
+			// Download File
+			log.Info("Downloading Book: ", bookURL)
+			fileName, err := downloadBook(bookURL, progressFunc)
+			if err != nil {
+				log.Error("Book URL API Failure: ", err)
+				continue
+			}
+
+			// Get Metadata
+			metadata, err := metadata.GetMetadata(fileName)
+			if err != nil {
+				log.Error("Book Metadata Failure: ", err)
+				continue
+			}
+
+			return fileName, metadata, nil
+		}
 	}
 
-	bookURL, err := def.parseDownloadFunc(body)
-	if err != nil {
-		log.Error("Parse Download URL Error: ", err)
-		return "", fmt.Errorf("Download Failure")
-	}
-
-	// Create File
-	tempFile, err := os.CreateTemp("", "book")
-	if err != nil {
-		log.Error("File Create Error: ", err)
-		return "", fmt.Errorf("File Failure")
-	}
-	defer tempFile.Close()
-
-	// Download File
-	log.Info("Downloading Book: ", bookURL)
-	resp, err := downloadBook(bookURL)
-	if err != nil {
-		os.Remove(tempFile.Name())
-		log.Error("Book URL API Failure: ", err)
-		return "", fmt.Errorf("API Failure")
-	}
-	defer resp.Body.Close()
-
-	// Copy File to Disk
-	log.Info("Saving Book")
-	_, err = io.Copy(tempFile, resp.Body)
-	if err != nil {
-		os.Remove(tempFile.Name())
-		log.Error("File Copy Error: ", err)
-		return "", fmt.Errorf("File Failure")
-	}
-
-	return tempFile.Name(), nil
-}
-
-func GetBookURL(id string, bookType BookType) (string, error) {
-	// Derive Info URL
-	var infoURL string
-	if bookType == BOOK_FICTION {
-		infoURL = "http://library.lol/fiction/" + id
-	} else if bookType == BOOK_NON_FICTION {
-		infoURL = "http://library.lol/main/" + id
-	}
-
-	// Parse & Derive Download URL
-	body, err := getPage(infoURL)
-	if err != nil {
-		return "", err
-	}
-
-	// downloadURL := parseLibGenDownloadURL(body)
-	return parseLibGenDownloadURL(body)
+	return "", nil, errors.New("failed to download book")
 }
 
 func getPage(page string) (io.ReadCloser, error) {
@@ -162,8 +106,6 @@ func getPage(page string) (io.ReadCloser, error) {
 	if err != nil {
 		return nil, err
 	}
-
-	// Set User-Agent
 	req.Header.Set("User-Agent", userAgent)
 
 	// Do Request
@@ -176,7 +118,7 @@ func getPage(page string) (io.ReadCloser, error) {
 	return resp.Body, err
 }
 
-func downloadBook(bookURL string) (*http.Response, error) {
+func downloadBook(bookURL string, progressFunc func(float32)) (string, error) {
 	log.Debug("URL: ", bookURL)
 
 	// Allow Insecure
@@ -189,11 +131,33 @@ func downloadBook(bookURL string) (*http.Response, error) {
 	// Start Request
 	req, err := http.NewRequest("GET", bookURL, nil)
 	if err != nil {
-		return nil, err
+		return "", err
 	}
-
-	// Set User-Agent
 	req.Header.Set("User-Agent", userAgent)
 
-	return client.Do(req)
+	// Perform API Request
+	resp, err := client.Do(req)
+	if err != nil {
+		return "", err
+	}
+
+	// Create File
+	tempFile, err := os.CreateTemp("", "book")
+	if err != nil {
+		log.Error("File Create Error: ", err)
+		return "", fmt.Errorf("failed to create temp file: %w", err)
+	}
+	defer tempFile.Close()
+
+	// Copy File to Disk
+	log.Info("Saving Book")
+	counter := &writeCounter{Total: resp.ContentLength, ProgressFunction: progressFunc}
+	_, err = io.Copy(tempFile, io.TeeReader(resp.Body, counter))
+	if err != nil {
+		os.Remove(tempFile.Name())
+		log.Error("File Copy Error: ", err)
+		return "", fmt.Errorf("failed to copy response to temp file: %w", err)
+	}
+
+	return tempFile.Name(), nil
 }