Skip to content

Latest commit

 

History

History
245 lines (186 loc) · 10.6 KB

TODO.md

File metadata and controls

245 lines (186 loc) · 10.6 KB

TODO

LATER

IMDB Interests

  • IMDB provides interests (genres, subgenres, form, style etc.)
  • update-imdb-interests which created data/imdb-interests.json is already created, run with npm run update-imdb-interests

Series: Link IMDB Dialog

  • for episodes:

    • provide all entries for the current series (Series_id_Movies_Owner's IMDB_tconst)
      • new imdb scraper: scrapeSeriesSeasons(tconst) -> https://www.imdb.com/title/tt0206512/episodes/ -> dropdown
    • allow a fallback to standard imdb search

Series

  • Heatmap: allow to customize the colors

  • support for date based series (2023-01-31)

  • BUG: subtitles in .rar files are added to the media list

  • BUG: rescanning a specific series does not run mediainfo on its episodes

    • store.rescanItems

  • Edit Episode: select season, episode number / bonus number

  • if season is unknown or episode number not set: try to find the imdb tconst by comparing names

    • store.findIMDBtconst: series episode find by name

Misc (LATER)

  • download https://easylist.to/easylist/easylist.txt every now and then
  • implement a test which just checks if the graphQL URLs are fine
  • Fix: subdirectory called "extras" is not assigned to the main movie (the files are provided as main movies themselves)
  • lots of video encoders are <not available>
  • BUG: with age rating range (e.g. 6-16+), the dialog's movies and series lists do not expand
    • we have a general issue with age ratings and their ranges
  • BUG: subdirectory called "extras" is not assigned to the main movie (the files are provided as main movies themselves)
  • IMDB Scraper: analyze video URLs in "Videos" section and find a better suited Trailer URL than the one primarily shown in the main page (which is oftentimes an IMDB special)
  • refactor buildINSERTQuery, buildUDPATEQuery to accept only one object as function parameter
  • scanErrors:
    • review how rescan and applyIMDB handle scanErrors (string to object, handle null)
    • artificially test graphql not found errors
    • introduce WARNING vs. ERROR
      • WARNING:
        • imdb implausibility
      • ERROR:
        • (new) mediainfo without result (possibly due to max path length exceeded)
        • IMDB GraphQL errors
    • introduce error types:
      • during filescan
      • during imdb scraping
      • during mediainfo
  • i18n: rescan finished snackbar
  • add ScanErrors functionality in applyMediaInfo (mediainfo may fail on exceeding 259 chars paths)
  • we apparently do not use the timezone (we store in UTC)
  • investigate everything that still works with MediaType (and possibly should also use SpecificMediaType)
  • investigate if we still use Series_id_Movies_Owner to differentiate Series and Episodes -> we have specificMediaType now
  • BUG? when opening person dialog and clicking "filter by this person", the main list "jumps" and THEN reloads with the filter applied (possibly a fix for another issue)
  • #36 Mediainfo always opening
  • rescanHandleDuplicates (also provide snackbar progress for this)
  • Allow user to upload a poster of their liking
  • introduce AND filter for Video Qualities (other filters might also get AND filter added)
  • Rescan: allow to select/unselect certain source paths for the rescan
  • deprecate tbl_Movies.ML_Quality (it isn't used in code anymore)
  • add ffmpeg in order to create screenshots for movies/episodes with missing poster
  • provide "source path selection" in "scan media" dialog
    • allows to select a sub-set of source paths for a media scan
  • include find-imdb-tconst-tests in imdb scraper watchdog
  • fix file/dirbased name when IMDB tconst detection has no result: if the medium is directory-based, then show the directory name and not the file-based name
  • new filter: min. IMDB votes
  • "IMDB score comparison game" (in context of the current list?)
  • curl/wget easylist.txt automatically before build https://easylist.to/easylist/easylist.txt
  • later: properly differentiate V1, V2 and V3 mainPageData handling
  • later: async db funcs (initDb, syncSqlite)
  • add "play random media" functionality to the List Actions menu
  • Trailer Show - trailer dialog is open, when adding to list the scrollbar appears on the side, also the trailer dialog seems to wiggle a bit when the add-to-list dialog closes

Plugin System

Remove Noise from duplicate router push

router.js

// remove the noise from duplicated router pushs (https://stackoverflow.com/a/58439497/5627010)
const originalPush = Router.prototype.push;
Router.prototype.push = function push(location) {
  return originalPush.call(this, location).catch((err) => err);
};

In-List File Management

  • for individual movies as well as for the whole (filtered) list
  • copy/move entries to
    • another sourcepath
    • a user-defined destination

Customizable Quick Info Area

  • have the quick info area customizable?
    • let user decide to hide certain fields
    • let user re-arrange the fields

Find Inspiration from https://github.com/whyboris/Video-Hub-App

Create i18n Editor

  • check out https://crowdin.com/ which is free for opensource projects

  • load message definitions from %workdir%/i18n (e.g. zh.json, fr.json etc.)

  • provide selectable languages from actually loaded messages

  • include *.json export in Media Hoarder

  • investigate external i18n sources - so that tranlastors can access their creation

  • dropdown of available (to be translated) languages

  • read available en.json ($fieldName, $expectedText)

  • provide $expectedText as Input, let user type the translation in the language of choice

  • establish a way of file exchange

  • implement i18n/where.json as copy of en.json. The value describes the location in the UI where the string is shown.

Raspberry Pi (armhf) Build

JSDoc everywhere

/**
 * Description of the function
 * @param {string} myParam
 */

Memory Leak (multiple reloads of medialist)

  • more intelligent loading?
  • better garbage collection?
  • -> we have a memory leak using eventBus.$on -> implement eventBus.$off on beforeDestroy() lifecycle hook
  • -> we still have the memory leak: don't fetch all data at once (re-fetch for each page)
  • -> check if memory leak is only in dev-mode (yes, it still persists even in prod-mode)

Use Alternative Search Method

  • IMDB Detection uses Suggestion API which doesn't support non-latin names
  • IMDB Link Dialog uses Advanced Title Search API which doesn't support non-latin names
  • maybe we should use https://www.imdb.com/find?q= (which is IMDB search incl. ENTER)?
  • -> replace it in LinkIMDBDialog
  • -> also use it for IMDB detection by filename
  • IMPORTANT: Find API only yields results if the movie title is complete
  • => we have to use find API only as fallback for the advancedTitleSearch
  • => maybe we should implement an "ultimate" search which combines results of all three searches?

Filters Customization

  • UI: implement App - Filters as array with Sort field
  • add Settings - Filters tab
  • user may reorder and show/hide filters
  • also use this in the filtersList creation

I18N - Basic

  • support all languages supported by DeepL or AWS (API): en, de, fr, es, it, nl, pl, ja, pt-PT, pt-BR, ru, zh

I18N - Advanced

  • Create AWS-based Service for automatic translation
    • Integrate with DeepL API or AWS
    • Integrate Payment API (sorry, DeepL/AWS are great but also want some cash)

Mediainfo Languages

  • we get languages like "German" from Mediainfo and map them to e.g. "De" using languages.js and store.js' ensureLanguageMapping
  • however, we can't be sure that we know all possible Mediainfo provided languages
  • how do we cope with that???
    • local logging?
    • webservice?

Progress - Movies sorted by Name

  • page: 56 (God bless America)

Other (later)

  • investigate Linux "stores" (see: https://www.electron.build/configuration/linux.html)

  • investigate scroll-snap https://markodenic.com/css-tips/

  • have individual lists of "my lists" clickable, show dialog (analog to genres, people, companies etc.) and thus "filter by this list"

  • create a MediaInfo watchdog

    • autodownload latest mediainfo
    • check for expected fields (error on fail)
    • check for new fields (warning on fail)

  • Test mediainfo and VLC in Linux/MacOS (we now use "" in the exec)

    • Win: OK
    • Linux: OK
    • MacOS: ??

  • MediaInfo supports "--Output=JSON", better use this instead of the default XML Output

  • .iso handling?

  • IMDB Awards (Oscars etc.) as data and filter criteria

  • apply shared.*AppliedContains to the media item once after completely fetching media

    • huge refactoring as movie data must then contain the information if it is affected by the filter

  • items which have a scan error skew the next scan's time-per-item

  • allow imdbid in relink dialog

  • refactor

    • use find-imdb-tconst.js and remove functions from store
    • find proper place for imdb-scraper.js, find-imdb-tconst.js
    • identify other functions in store.js which don't really belong there
    • provide util.promisified stuff in helpers.js and use it exclusively from there

  • handle helpers.isPortable via env-var (get rid of set-portable.js)

  • add a new rescan method: "rescan entries with errors"

  • layout the app with css grid (see: https://layout.bradwoods.io/customize)

  • treat rescan of particular titles as a queue (don't disable all other titles when a rescan runs)

    • same queue as a "Scan Media" process

  • introduce #tags instead of lists (keep the datastructures in .db though)

  • correctly implement mk-scrollcontainer class (e.g. Medialist)

  • OK? - fix moment's missing local time (see MediaList.lastAccessDisplayText)

Youtube Support

QA

  • check how fetchMedia/MediaList works if only filescan has been performed (no MI/IMDB data)

Unsure: Implement Backend as express-like server

  • this way we can have front- and backend as independent apps
  • major re-write neccessary