Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Remove unused code (search update) #1581

Merged
merged 1 commit into from
Jun 5, 2022
Merged

Remove unused code (search update) #1581

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
1 commit
Select commit
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
11 changes: 3 additions & 8 deletions modules/backend/src/main/scala/docspell/backend/BackendApp.scala
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -47,7 +47,6 @@ trait BackendApp[F[_]] {
def userTask: OUserTask[F]
def folder: OFolder[F]
def customFields: OCustomFields[F]
def simpleSearch: OSimpleSearch[F]
def clientSettings: OClientSettings[F]
def totp: OTotp[F]
def share: OShare[F]
Expand Down Expand Up @@ -99,8 +98,6 @@ object BackendApp {
itemImpl <- OItem(store, ftsClient, createIndex, schedulerModule.jobs)
itemSearchImpl <- OItemSearch(store)
fulltextImpl <- OFulltext(
itemSearchImpl,
ftsClient,
store,
schedulerModule.jobs
)
Expand All @@ -112,15 +109,15 @@ object BackendApp {
)
folderImpl <- OFolder(store)
customFieldsImpl <- OCustomFields(store)
simpleSearchImpl = OSimpleSearch(fulltextImpl, itemSearchImpl)
clientSettingsImpl <- OClientSettings(store)
searchImpl <- Resource.pure(OSearch(store, ftsClient))
shareImpl <- Resource.pure(
OShare(store, itemSearchImpl, simpleSearchImpl, javaEmil)
OShare(store, itemSearchImpl, searchImpl, javaEmil)
)
notifyImpl <- ONotification(store, notificationMod)
bookmarksImpl <- OQueryBookmarks(store)
fileRepoImpl <- OFileRepository(store, schedulerModule.jobs)
itemLinkImpl <- Resource.pure(OItemLink(store, itemSearchImpl))
itemLinkImpl <- Resource.pure(OItemLink(store, searchImpl))
downloadAllImpl <- Resource.pure(ODownloadAll(store, jobImpl, schedulerModule.jobs))
attachImpl <- Resource.pure(OAttachment(store, ftsClient, schedulerModule.jobs))
addonsImpl <- Resource.pure(
Expand All @@ -132,7 +129,6 @@ object BackendApp {
joexImpl
)
)
searchImpl <- Resource.pure(OSearch(store, ftsClient))
} yield new BackendApp[F] {
val pubSub = pubSubT
val login = loginImpl
Expand All @@ -153,7 +149,6 @@ object BackendApp {
val userTask = userTaskImpl
val folder = folderImpl
val customFields = customFieldsImpl
val simpleSearch = simpleSearchImpl
val clientSettings = clientSettingsImpl
val totp = totpImpl
val share = shareImpl
Expand Down
281 changes: 1 addition & 280 deletions modules/backend/src/main/scala/docspell/backend/ops/OFulltext.scala
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -6,46 +6,17 @@

package docspell.backend.ops

import cats.data.NonEmptyList
import cats.effect._
import cats.implicits._
import fs2.Stream

import docspell.backend.JobFactory
import docspell.backend.ops.OItemSearch._
import docspell.common._
import docspell.ftsclient._
import docspell.query.ItemQuery._
import docspell.query.ItemQueryDsl._
import docspell.scheduler.JobStore
import docspell.store.queries.{QFolder, QItem, SelectedItem}
import docspell.store.Store
import docspell.store.records.RJob
import docspell.store.{Store, qb}

trait OFulltext[F[_]] {

def findItems(maxNoteLen: Int)(
q: Query,
fts: OFulltext.FtsInput,
batch: qb.Batch
): F[Vector[OFulltext.FtsItem]]

/** Same as `findItems` but does more queries per item to find all tags. */
def findItemsWithTags(maxNoteLen: Int)(
q: Query,
fts: OFulltext.FtsInput,
batch: qb.Batch
): F[Vector[OFulltext.FtsItemWithTags]]

def findIndexOnly(maxNoteLen: Int)(
fts: OFulltext.FtsInput,
account: AccountId,
batch: qb.Batch
): F[Vector[OFulltext.FtsItemWithTags]]

def findIndexOnlySummary(account: AccountId, fts: OFulltext.FtsInput): F[SearchSummary]
def findItemsSummary(q: Query, fts: OFulltext.FtsInput): F[SearchSummary]

/** Clears the full-text index completely and launches a task that indexes all data. */
def reindexAll: F[Unit]

Expand All @@ -56,30 +27,7 @@ trait OFulltext[F[_]] {
}

object OFulltext {

case class FtsInput(
query: String,
highlightPre: String = "***",
highlightPost: String = "***"
)

case class FtsDataItem(
score: Double,
matchData: FtsResult.MatchData,
context: List[String]
)
case class FtsData(
maxScore: Double,
count: Int,
qtime: Duration,
items: List[FtsDataItem]
)
case class FtsItem(item: ListItem, ftsData: FtsData)
case class FtsItemWithTags(item: ListItemWithTags, ftsData: FtsData)

def apply[F[_]: Async](
itemSearch: OItemSearch[F],
fts: FtsClient[F],
store: Store[F],
jobStore: JobStore[F]
): Resource[F, OFulltext[F]] =
Expand All @@ -103,232 +51,5 @@ object OFulltext {
if (exist.isDefined) ().pure[F]
else jobStore.insertIfNew(job.encode)
} yield ()

def findIndexOnly(maxNoteLen: Int)(
ftsQ: OFulltext.FtsInput,
account: AccountId,
batch: qb.Batch
): F[Vector[OFulltext.FtsItemWithTags]] = {
val fq = FtsQuery(
ftsQ.query,
account.collective,
Set.empty,
Set.empty,
batch.limit,
batch.offset,
FtsQuery.HighlightSetting(ftsQ.highlightPre, ftsQ.highlightPost)
)
for {
_ <- logger.trace(s"Find index only: ${ftsQ.query}/$batch")
folders <- store.transact(QFolder.getMemberFolders(account))
ftsR <- fts.search(fq.withFolders(folders))
ftsItems = ftsR.results.groupBy(_.itemId)
select =
ftsItems.values
.map(_.minBy(-_.score))
.map(r => SelectedItem(r.itemId, r.score))
.toSet
now <- Timestamp.current[F]
itemsWithTags <-
store
.transact(
QItem.findItemsWithTags(
account.collective,
QItem.findSelectedItems(
Query.all(account),
now.toUtcDate,
maxNoteLen,
select
)
)
)
.take(batch.limit.toLong)
.compile
.toVector
res =
itemsWithTags
.collect(convertFtsData(ftsR, ftsItems))
.map { case (li, fd) => FtsItemWithTags(li, fd) }
} yield res
}

def findIndexOnlySummary(
account: AccountId,
ftsQ: OFulltext.FtsInput
): F[SearchSummary] = {
val fq = FtsQuery(
ftsQ.query,
account.collective,
Set.empty,
Set.empty,
500,
0,
FtsQuery.HighlightSetting.default
)

for {
folder <- store.transact(QFolder.getMemberFolders(account))
now <- Timestamp.current[F]
itemIds <- fts
.searchAll(fq.withFolders(folder))
.flatMap(r => Stream.emits(r.results.map(_.itemId)))
.compile
.to(Set)
itemIdsQuery = NonEmptyList
.fromList(itemIds.toList)
.map(ids => Attr.ItemId.in(ids.map(_.id)))
.getOrElse(Attr.ItemId.notExists)
q = Query
.all(account)
.withFix(_.copy(query = itemIdsQuery.some))
res <- store.transact(QItem.searchStats(now.toUtcDate, None)(q))
} yield res
}

def findItems(
maxNoteLen: Int
)(q: Query, ftsQ: FtsInput, batch: qb.Batch): F[Vector[FtsItem]] =
findItemsFts(
q,
ftsQ,
batch.first,
itemSearch.findItems(maxNoteLen),
convertFtsData[ListItem]
)
.drop(batch.offset.toLong)
.take(batch.limit.toLong)
.map { case (li, fd) => FtsItem(li, fd) }
.compile
.toVector

def findItemsWithTags(maxNoteLen: Int)(
q: Query,
ftsQ: FtsInput,
batch: qb.Batch
): F[Vector[FtsItemWithTags]] =
findItemsFts(
q,
ftsQ,
batch.first,
itemSearch.findItemsWithTags(maxNoteLen),
convertFtsData[ListItemWithTags]
)
.drop(batch.offset.toLong)
.take(batch.limit.toLong)
.map { case (li, fd) => FtsItemWithTags(li, fd) }
.compile
.toVector

def findItemsSummary(q: Query, ftsQ: OFulltext.FtsInput): F[SearchSummary] =
for {
search <- itemSearch.findItems(0)(q, Batch.all)
fq = FtsQuery(
ftsQ.query,
q.fix.account.collective,
search.map(_.id).toSet,
Set.empty,
500,
0,
FtsQuery.HighlightSetting.default
)
items <- fts
.searchAll(fq)
.flatMap(r => Stream.emits(r.results.map(_.itemId)))
.compile
.to(Set)
itemIdsQuery = NonEmptyList
.fromList(items.toList)
.map(ids => Attr.ItemId.in(ids.map(_.id)))
.getOrElse(Attr.ItemId.notExists)
qnext = q.withFix(_.copy(query = itemIdsQuery.some))
now <- Timestamp.current[F]
res <- store.transact(QItem.searchStats(now.toUtcDate, None)(qnext))
} yield res

// Helper

private def findItemsFts[A: ItemId, B](
q: Query,
ftsQ: FtsInput,
batch: qb.Batch,
search: (Query, qb.Batch) => F[Vector[A]],
convert: (
FtsResult,
Map[Ident, List[FtsResult.ItemMatch]]
) => PartialFunction[A, (A, FtsData)]
): Stream[F, (A, FtsData)] =
findItemsFts0(q, ftsQ, batch, search, convert)
.takeThrough(_._1 >= batch.limit)
.flatMap(x => Stream.emits(x._2))

private def findItemsFts0[A: ItemId, B](
q: Query,
ftsQ: FtsInput,
batch: qb.Batch,
search: (Query, qb.Batch) => F[Vector[A]],
convert: (
FtsResult,
Map[Ident, List[FtsResult.ItemMatch]]
) => PartialFunction[A, (A, FtsData)]
): Stream[F, (Int, Vector[(A, FtsData)])] = {
val sqlResult = search(q, batch)
val fq = FtsQuery(
ftsQ.query,
q.fix.account.collective,
Set.empty,
Set.empty,
0,
0,
FtsQuery.HighlightSetting(ftsQ.highlightPre, ftsQ.highlightPost)
)

val qres =
for {
items <- sqlResult
ids = items.map(a => ItemId[A].itemId(a))
idsNel = NonEmptyList.fromFoldable(ids)
// must find all index results involving the items.
// Currently there is one result per item + one result per
// attachment
limit <- idsNel
.map(itemIds => store.transact(QItem.countAttachmentsAndItems(itemIds)))
.getOrElse(0.pure[F])
ftsQ = fq.copy(items = ids.toSet, limit = limit)
ftsR <- fts.search(ftsQ)
ftsItems = ftsR.results.groupBy(_.itemId)
res = items.collect(convert(ftsR, ftsItems))
} yield (items.size, res)

Stream.eval(qres) ++ findItemsFts0(q, ftsQ, batch.next, search, convert)
}

private def convertFtsData[A: ItemId](
ftr: FtsResult,
ftrItems: Map[Ident, List[FtsResult.ItemMatch]]
): PartialFunction[A, (A, FtsData)] = {
case a if ftrItems.contains(ItemId[A].itemId(a)) =>
val ftsDataItems = ftrItems
.getOrElse(ItemId[A].itemId(a), Nil)
.map(im =>
FtsDataItem(im.score, im.data, ftr.highlight.getOrElse(im.id, Nil))
)
(a, FtsData(ftr.maxScore, ftr.count, ftr.qtime, ftsDataItems))
}
})

trait ItemId[A] {
def itemId(a: A): Ident
}
object ItemId {
def apply[A](implicit ev: ItemId[A]): ItemId[A] = ev

def from[A](f: A => Ident): ItemId[A] =
(a: A) => f(a)

implicit val listItemId: ItemId[ListItem] =
ItemId.from(_.id)

implicit val listItemWithTagsId: ItemId[ListItemWithTags] =
ItemId.from(_.item.id)
}
}
Loading