* Add migration to create studio aliases table
* Refactor studioQueryBuilder.Query to use filterBuilder
* Expand GraphQL API with aliases support for studio
* Add aliases support for studios to the UI
* List aliases in details panel
* Allow editing aliases in edit panel
* Add 'aliases' filter when searching
* Find studios by alias in filter / select
* Add auto-tagging based on studio aliases
* Support studio aliases for filename parsing
* Support importing and exporting of studio aliases
* Search for studio alias as well during scraping
* Add migration script for tag relations table
* Expand hierarchical filter features
Expand the features of the hierarchical multi input filter with support
for using a relations table, which only has parent_id and child_id
columns, and support adding an additional intermediate table to join on,
for example for scenes and tags which are linked by the scenes_tags
table as well.
* Add hierarchical filtering for tags
* Add hierarchical tags support to scene markers
Refactor filtering of scene markers to filterBuilder and in the process
add support for hierarchical tags as well.
* List parent and child tags on tag details page
* Support setting parent and child tags
Add support for setting parent and child tags during tag creation and
tag updates.
* Validate no loops are created in tags hierarchy
* Update tag merging to support tag hierarcy
* Add unit tests for tags.EnsureUniqueHierarchy
* Fix applying recursive to with clause
The SQL `RECURSIVE` of a `WITH` clause only needs to be applied once,
imediately after the `WITH`. So this fixes the query building to do just
that, automatically applying the `RECURSIVE` keyword when any added with
clause is added as recursive.
* Rename hierarchical root id column
* Rewrite hierarchical filtering for performance
Completely rewrite the hierarchical filtering to optimize for
performance. Doing the recursive query in combination with a complex
query seems to break SQLite optimizing some things which means that the
recursive part might be 2,5 second slower than adding a static
`VALUES()` list. This is mostly noticable in case of the tag hierarchy
where setting an exclusion with any depth (or depth: all) being applied
has this performance impact of 2,5 second. "Include" also suffered this
issue, but some rewritten query by joining in the *_tags table in one
pass and applying a `WHERE x IS NOT NULL` filter did seem to optimize
that case. But that optimization isn't applied to the `IS NULL` filter
of "exclude". Running a simple query beforehand to get all (recursive)
items and then applying them to the query doesn't have this performance
penalty.
* Remove UI references to child studios and tags
* Add parents to tag export
* Support importing of parent relationship for tags
* Assign stable ids to parent / child badges
* Silence Apollo warning on parents/children fields on tags
Silence warning triggered by Apollo GraphQL by explicitly instructing it
to use the incoming parents/children values. By default it already does
this, but it triggers a warning as it might be unintended that it uses
the incoming values (instead of for example merging both arrays).
Setting merge to false still applies the same behaviour (use only
incoming values) but silences the warning as it's explicitly configured
to work like this.
* Rework detecting unique tag hierarchy
Completely rework the unique tag hierarchy to detect invalid hierarchies
for which a tag is "added in the middle". So when there are tags A <- B
and A <- C, you could previously edit tag B and add tag C as a sub tag
without it being noticed as parent A being applied twice (to tag C).
While afterwards saving tag C would fail as tag A was applied as parent
twice. The updated code correctly detects this scenario as well.
Furthermore the error messaging has been reworked a bit and the message
now mentions both the direct parent / sub tag as well as the tag which
would results in the error. So in aboves example it would now show the
message that tag C can't be applied because tag A already is a parent.
* Update relations on cached tags when needed
Update the relations on cached tags when a tag is created / updated /
deleted so these always reflect the correct state. Otherwise (re)opening
a tag might still show the old relations untill the page is fully
reloaded or the list is navigated. But this obviously is strange when
you for example have tag A, create or update tag B to have a relation to
tag A, and from tags B page click through to tag A and it doesn't show
that it is linked to tag B.
The strings.Replace function counts the number of replacements. If 0,
the original string is returned. Hence, there is no need to check if a
replacement will happen before doing the work.
* Remove stuff which isn't being used
Some fields, functions and structs aren't in use by the project. Remove
them for janitorial reasons.
* Remove more unused code
All of these functions are currently not in use. Clean up the code by
removal, since the version control has the code if need be.
* Remove unused functions
There's a large set of unused functions and variables in the code base.
Remove these, so it clearer what code to support going forward.
Dead code has been eliminated.
Where applicable, comment const-sections in tests, so reserved
identifiers are still known.
* Fix use-def of tsURL
The first def of tsURL doesn't matter because there's no use before
we hit the 2nd def.
* Remove dead code assignment
Setting logFile = "" is effectively dead code, because there's no use
of it later.
* Comment out found
The variable 'found' is dead in the function (because no post-process
action is following it). Comment it for now.
* Comment dead code in tests
These might provide hints as to what isn't covered at the moment.
* Dead code removal
In the case of constants where iota is involved, move the iota so it
matches the current key values.
This avoids problems with persistently stored key IDs.
* Bump Go to 1.17, refactor build/x86_64 Dockerfile to make better use of multi-stage
* Bump to 1.17 from 1.16
* Bump packr version, provide needed legacy env var
* Add apple silicon support, fix macos build chain
* Update unused travis ci
* Fix error string capitalization
Error strings often follow another string. Hence, they should not be
capitalized, unless referencing a name.
* Uncapitalize more error strings
While here, use %v on the error directly, which makes it easier to wrap
the error later with %w if need be.
* Uncapitalize more error strings
While here, rename Url to URL as a nitpick.
* When stopping, close the database
This patch is likely to cause errornous behavior in the application.
This is due to the fact that the application doesn't gracefully shut
down, but is forcefully terminated.
However, the purpose is to uncover what needs to be done, to make
it a more graceful shutdown.
The call to p.ExecuteTask happens in a separate go-routine. It writes,
under m.mutex, into the job's details. However, the test goroutine
itself reads j.Details which is a read race.
Protect the reads in the test by the lock.
This makes `package job` pass `go test -race`
* Unify scraped types
* Make name fields optional
* Unify single scrape queries
* Change UI to use new interfaces
* Add multi scrape interfaces
* Use images instead of image
* Add script offset / delay to Handy support.
Further work on #1376.
Offsets are added to the current video position, so a positive value leads to earlier motion. (The most common setting.)
This is needed because most script times have a consistent delay when compared to the video. (Delay from the API calls to the server should be handled by the server offset calculation.)
* Rename scriptOffset to funscriptOffset
* Correct localisation keys
Co-authored-by: WithoutPants <53250216+WithoutPants@users.noreply.github.com>
* Add sorting on image and gallery count to performers
* Make performer table headers translatable
* Add image and gallery count to performers table
* Make sure list table container fits the table
Make the table container minimally as wide as the table. This fixes the
table "overflowing" to the left and right and the left not being
accessible.
* Remove unnecessary truncation in tables
IMO there is no need to truncate the title in the scenes table and the
name in the performers table. This because both tables also contain an
image which means that multiple lines should be possible without really
extending the height of the row. Furthermore both tables contain other
values which might be way longer and also aren't wrapped (like tags and
performers for scenes, and aliases for performers).
* Fix: config race conditions with RWMutex
Added RWMutex to config.Instance which read or write locks
all instances where viper is used.
Refactored checksum manager to only use config and not
viper directly anymore.
All stash viper operations are now "behind" the config.Instance
and thus mutex "protected".
* Fix hierarchical criteria performance issue
Don't apply recursive clause to hierarchical criteria when the depth is
set to 0 (i.e.: no recursion is needed).
This undoes the current performance penalty on for example the studios
page. This as reported in #1519, using a database of 4M images, 30K
scenes and 500 studios. Without this fix loading the studios overview,
with the default of 40 items per page, takes 6 to 7 seconds. With this
fix it only takes 0,07 seconds reverting the performance back to the
pre-hierarchical filtering performance (tested against 508f7b84 which
was the last commit before #1397 was merged).
* Add (not) between modifiers for number criterion
* Extract list filters into dedicated components
Extract the filters from the AddFiltersDialog into custom components.
This allows for further refactorring where components will be bound to
criterions.
* Add placeholders to number and duration criterions
* Add backwards compatibility for saved filters
Co-authored-by: WithoutPants <53250216+WithoutPants@users.noreply.github.com>
* Add config option for scraper tag exclusion patterns
Add a config option for exclusing tags / tag patterns from the scraper
results.
* Handle tag exclusion patterns during scraping
* Actually implement TagFilter.marker_count
The marker_count filter/criterion as defined in TagFilterType isn't
actually implemented. This adds an implementation for it.
Do note this implementation _might_ have performance issues because of
using OR (in the join). Another implentation would be to remove both
joins and use:
```SQL
COUNT(
SELECT id FROM scene_markers WHERE primary_tag_id = tags.id
UNION
SELECT scene_marker_id FROM scene_markers_tags WHERE tag_id = tags.id
)
```
Note this doesn't require a DISTINCT as UNION already removes any
duplicate records.
* Restore marker count filter and sorting
Co-authored-by: WithoutPants <53250216+WithoutPants@users.noreply.github.com>
* Change scrapers overview into collapsible per type
* Move scraping configuration to (renamed) scrapers tab
Rename the Scrapers tab to Scraping and move the scraping configuration
to this tab.
* Fix width in database test setup
* Added more filters on resolution field
* added test to verify resolution range is defined for every resolution
* Refactor UI code
Co-authored-by: WithoutPants <53250216+WithoutPants@users.noreply.github.com>
* Add API to merge tags
Add new API endpoint, `tagsMerge(source, destination)` to merge multiple
tags into a single one. The "sources" must be provided as a list of ids
and the destination as a single id. All usages of the source tags
(scenes, markers (primary and additional), images, galleries and
performers) will be updated to the destination tag, all aliases of the
source tags will be updated to the destination, and the name of the
source will be added as alias to the destination as well.
* Add merge tag UI
* Add unit tests
* Update test mocks
* Update internationalisation
* Add changelog entry
Co-authored-by: gitgiggety <gitgiggety@outlook.com>
* Add basic support for hierarchical filters
Add a new `hierarchicalMultiCriterionHandlerBuilder` filter type which
can / will be used for filtering hierarchical things like the
parent/child relation of the studios.
On the frontend side a new IHierarchicalLabeledIdCriterion criterion
type has been added to accompany this new filter type.
* Refactor movieQueryBuilder to use filterBuilder
Refactor the movieQueryBuilder to use the filterBuilder just as scene,
image and gallery as well.
* Support specifying depth for studios filter
Add an optional depth field to the studios filter for scenes, images,
galleries and movies. When specified that number of included
(grant)children are shown as well. In other words: this adds support for
showing scenes set to child studios when searching on the parent studio.
Co-authored-by: WithoutPants <53250216+WithoutPants@users.noreply.github.com>
* Add Tag Update/UpdateFull
* Tag alias implementation
* Refactor tag page
* Add aliases in UI
* Include tag aliases in q filter
* Include aliases in tag select
* Add aliases to auto-tagger
* Use aliases in scraper
* Add tag aliases for filename parser
* utils: vtt: add tests
In lieu of documentation.
* utils: vtt: rewrite for correctness and simplicity
Now handles fractional seconds and negative values correctly.
* cleanup: remove dead code
removing some code that does nothing
* cleanup: fixing usage of deprecated gqlgen/graphql api in api/changeset_translator
* cleanup: changing to recommended comparison methods
Changing byte and case-insensitive string comparison to the recommended methods.
* cleanup: making staticcheck happy
* Add funscript route to scenes
Adds a /scene/:id/funscript route which serves a funscript file, if present.
Current convention is that these are files stored with the same path, but with the extension ".funscript".
* Look for funscript during scan
This is stored in the Scene record and used to drive UI changes for funscript support.
Currently, that's limited to a funscript link in the Scene's file info.
* Add filtering and sorting for interactive
* Add Handy connection key to interface config
* Add Handy client and placeholder component.
Uses defucilis/thehandy, but not thehandy-react as I had difficulty integrating the context with the existing components.
Instead, the expensive calculation for the server time offset is put in localStorage for reuse.
A debounce was added when scrubbing the video, as otherwise it spammed the Handy API with updates to the current offset.
* Apply all post processors to performer
Scraping a performer by fragment doesn't correctly work with tags.
When tags are returned to the scraper then all are recognized as new.
This is due to the post process method not being applied while it should
be, as is done when scraping a performer by URL.
* Add tools settings page
* Add tools and move dupe checker
* Make negative number get all
* Show missing phashes
* Add multi-edit button
* Show scene details
* Make config instance-based
* Remove config dependency in paths
* Refactor config init
* Allow startup without database
* Get system status at UI initialise
* Add setup wizard
* Cache and Metadata optional. Database mandatory
* Handle metadata not set during full import/export
* Add links
* Remove config check middleware
* Stash not mandatory
* Panic on missing mandatory config fields
* Redirect setup to main page if setup not required
* Add migration UI
* Remove unused stuff
* Move UI initialisation into App
* Don't create metadata paths on RefreshConfig
* Add folder selector for generated in setup
* Env variable to set and create config file.
Make docker images use a fixed config file.
* Set config file during setup
* Add scraping support for performer tags
* Add performer count to tag cards
* Refactor sqlite test setup
* Add performer tag filtering in gallery and image
* Add bulk update performer
* Add Performers tab to tag page
* Add count filters and sort bys for tags
* Move scene count to icon in performer card #1148
* find correct python executable
For script scrapers using python, both python and python3 are valid depending on the OS and running environment. To save users from having any issues, this change will find the correct executable for them.
Co-authored-by: bnkai <bnkai@users.noreply.github.com>
* Add resolution enum extension
* Add filter builder
* Use filterBuilder for scene query
* Optimise joins
* Add binary operators to scene query
* Use Query for auto-tag
* Expose url for URLReplace in JSON scrapeByURL and scrapeByFragment
* Apply queryURLReplace to xpath scrapers
Co-authored-by: WithoutPants <53250216+WithoutPants@users.noreply.github.com>
* Add regex string filter criterion
* Use query interface for auto tagging
* Use Query interface for filename parser
* Remove query regex interfaces
* Add selective auto tag
* Use page size 0 as no limit
* extend resolutions
- Simplifies logic
- Adds more options including 540p, 1440p, and resolutions common to VR such as 1920p
- Supports vertical/portrait videos and images
* implement new resolution filters
* Fix integer overflow for scene size on 32bit systems
* Cast to double in sqlite to prevent potential overflow
* Add migration to reset scene sizes and scan logic to repopulate if empty
* Add organized boolean to scene model (#729)
* Add organized button to scene page
* Add flag to galleries and images
* Import/export changes
* Make organized flag not null
* Ignore organized scenes for autotag
Co-authored-by: com1234 <com1234@notarealemail.com>
* Added natural sort for scene and images
* Use natural sort for movie names
Co-authored-by: WithoutPants <53250216+WithoutPants@users.noreply.github.com>
* Implement parallel scanning and generation, and combined scanning/preview/sprite generation.
* Added UI component for preview/sprite generation during scan, and configurable number of parallel tasks.
* Add v050 changelog entry
* Fix potential image errors
* Fix issue preventing favoriting of tagged performers
* Add error handling in case of network issues
* Show individual search errors
* Unset scene results if query fails
* Don't abort scene submission if scene id isn't found
* Include studios in movie export
* Generalise cards
* Add selection and export for movies
* Refactor gallery card
* Refactor export dialogs
* Add performer selection and export
* Add selection and export for studios
* Add selection and export of tags
* Include movie scenes and gallery images
* Add gql client generation files
* Update dependencies
* Add stash-box client generation to the makefile
* Move scraped scene object matchers to models
* Add stash-box to scrape with dropdown
* Add scrape scene from fingerprint in UI
Fixed annoyingly noisy transcoding progress log messages.
Fixed minor minor issue with directories than are named "blah.mp4" being detected as files to be scanned.
* Added preview generation fallback feature.
When a preview generation fails (often for wmv/avi files), the new code tries with less stricted (no xerror) and more time consuming options (slow+fast seek).
Fix a minor issue when stash downloads ffmpeg/ffprobe, but doesn't re-detect their paths.
* Allow clearing of tag images
* Allow clearing of studio images
* Allow clearing of performer images
* Allow clearing of movie images
* Add filtering for missing images
* api/urlbuilders/movie: Auto format.
* graphql+pkg+ui: Implement scraping movies by URL.
This patch implements the missing required boilerplate for scraping
movies by URL, using performers and scenes as a reference.
Although this patch contains a big chunck of ground work for enabling
scraping movies by fragment, the feature would require additional
changes to be completely implemented and was not tested.
* graphql+pkg+ui: Scrape movie studio.
Extends and corrects the movie model for the ability to store and
dereference studio IDs with received studio string from the scraper.
This was done with Scenes as a reference. For simplicity the duplication
of having `ScrapedMovieStudio` and `ScrapedSceneStudio` was kept, which
should probably be refactored to be the same type in the model in the
future.
* ui/movies: Add movie scrape dialog.
Adds possibility to update existing movie entries with the URL scraper.
For this the MovieScrapeDialog.tsx was implemented with Performers and
Scenes as a reference. In addition DurationUtils needs to be called one
time for converting seconds from the model to the string that is
displayed in the component. This seemed the least intrusive to me as it
kept a ScrapeResult<string> type compatible with ScrapedInputGroupRow.
* Refactor xpath scraper code
* Make post-process a list
* Add map post-process action
* Add fixed xpath values
* Refactor scrapers into cache
* Refactor into mapped config
* Trim test html
* Improve layout and add buttons
* Move functionality into ListFilter
* Make modal style dark
* Convert scene options into edit scenes dialog
* Add delete scenes dialog
* Clear selected ids on delete
* Refetch after update/delete
* Use DeleteScenesDialog in Scene page
* Show scene check boxes in small screens
* Change default multi-set mode to set
* Add reload scraper option to performer details
* Add scraper reload to scene edit page
* Show scene scraper menu when no queryable scrapers
* Add 0.3 changelog
* Change scrape matching (studio, movies, tag, performers) to case insensitive
* * fix collate order
* * make filename parser findbyname calls case insensitive
* * add unit testing for Tags GetFindbyName/s
* Add lint/format checks to build
* Make travis get full repo to get tags
* Run packr2 once in cross-compile
* Fix quotes in package.json
* Fix linting issues
* Formatting
* Fix vet issue
* Fix go lint issues
* Show start of each platform compilation
* Add validate target
* Set gitattributes for go fmt and mod vendor
* Fix tag name
* Add fmt-ui target
* Add movie migration
* Update server and UI code for type changes
* Add studio to movies
* Movie blobs to end
* Document movie duration
* Add filtering on movie studio
* Ignore generated-graphql.tsx in 2.5
* Make movie name mandatory
* Port #395 fix to v2.5
* Differentiate front/back image browse buttons
* Move URL, Synopsis to separate rows
* Fix unknown query params crashing UI
* Make mutating metadata ops mutation
* Implement scene generate screenshot
* Remove fetch policy on metadata mutations
* Port UI changes to v2.5
* Set generated image in database
* Add sub-scraper functionality
* Add scraping of performer image
* Add scene cover image scraping
* Port UI changes to v2.5
* Fix v2.5 dialog suggest color
* Don't convert eol of UI to support pretty
* Tag develop builds as latest. Version in makefile
* Rename latest to latest_develop
* Update GetLatestVersion
* Separate master and develop releases
* Add release date back into development version
* Remove unused code
* Two minute limit per file to upload to transfer.sh
* Add check version functionality
* add backend support
* add ui support
* minor fixes
* cosmetic fixes
* workaround query refetch not working after network error
* revert changes to Makefile after testing is complete
* switch to Releases Github API endpoint, add latest Release URL to UI
* latest version is only shown in UI when version is available and data is ready
* resolve conflict , squash rebase
* Show localhost in console output instead of 0.0.0.0
* Updated message to reflect both navigation and listening address
Co-authored-by: InfiniteTF <infinitekittens@protonmail.com>
* Add release make target
* Use Box.Find now that Box.Bytes is depreciated
Pretty much directly mocked off of the post-depreciation implementation
of Box.Bytes in packr. In theory we should totally be checking the
returned error but I'm lazy.
* Add stash scraper type
* Add graphql client to vendor
* Embed stash credentials in URL
* Fill URL from scraped scene
* Nil IDs returned from remote stash
* Nil check
* Added exclude file from scan feature
* Abort exclusion instead of panicking when pattern isn't valid
* Added UI configuration for exclude patterns
* * cosmetic fixes
* changed behavior of exclude function to continue and ignore invalide regex patterns
* added some more tests (windows networks and continue after regex error)
* Move duration and resolution to overlay
* Improve display of portrait videos
* Condense filter controls
* Add performer images to scene tags
* Add studio overlay to scene cards
* Fade out scene overlays on hover
* CSS grid tweaks
* Align overlay to bottom of video preview
* Fix opacity value
* Fix performer thumbnails
* Show studio overlay on mouseover
* Correct display colour for display mode buttons
* Add scene zoom slider
* Add show studio as text option
* Move select all/none to more button
* Add auto tagging for performers, studios and tags
* Fix scene title logging
* Fix golang 1.11 compile error
* Improve regex pattern. Only log when added
* Remove paging from parse query
* Add integration test
* Fix prefix detection on paths
* Add unit and integration test to make file
* Add gitattributes for go.mod/go.sum files
* Always run integration tests
* Removed redundant call. Clarified targets
* Make regex matching case-insensitive
* Port filename parser code to backend
* Add performers to scene filename parser UI
* Finish porting parser to backend
* Add performer, studio and tag parsing
* Hide fields not being parsed
* Don't query for empty performer/studio/tag
* Use exact matches
* Fix panic
* Fix arrays changed false positive. Fix layout
* Initial UI prototype
* Add backend support to update multiple scenes
* Fix title editing issues
* Add query regex support. UI improvements
* Rewrite parser. Add fields button and page size
* Add helper text for escaping {} characters
* Validate date
* Only set values if different from original
* Only update scenes that have something changed
* Add built in parser input recipes
* Make pattern matching case-insensistive
* Add websocket connection
* Add logs to the log page
* Make debug color more readable
* Remove TODO from front page
* Put all log entries in latest first order
* Add filtering of log entries by level
* Limit log entries and throttle updates
* Fix logger not throttling broadcasts
* Remove now unnecessary UI-side log throttling
* Filter incoming logs by log level
* Make log view more terminal-like
make delete logging info instead of debug
update the logged text for clarity when deleting db data
changed from "delete" to "cleaning" to clarify the actual type of action taken
* Added flags to customize the host and port
* Start up just one server rather than a server for HTTP and HTTPS. HTTPS server will only start if a cert and key are found
* Allow configuration of generated file path and database path. Closes#33
* Added checkboxes to choose what gets generated for the generate task. Closes#32
