vimagick · November 4, 2025 18:34 · vimagick · Oct 20, 2022
diff --git a/imdb-top-1000.tsv b/imdb-top-1000.tsv
diff --git a/imdb.sh b/imdb.sh
 #!/bin/bash
 #
 # IMDB Search
 #
 # - https://gist.github.com/vimagick/fbc551ecece8639dc0bea18ac7113450
 # - https://developer.imdb.com/documentation/api-documentation/sample-queries/
 # - https://github.com/Gujal00/Kodi-Official/blob/master/plugin.video.imdb.trailers/resources/lib/imdb_trailers.py
 # - https://github.com/theMK2k/Media-Hoarder/blob/master/src/data/imdb-graphql-urls.json
 # - https://www.imdb.com/search/title/
 # - https://imdbapi.dev/docs/graphql/schema/title
 #

 title=${1:?title is empty}
 year=${2}
 count=${3:-5}

 if [[ $year =~ ^([0-9]{4})(:([0-9]{4}))?$ ]]; then
  year_start=${BASH_REMATCH[1]}
  if [[ -z ${BASH_REMATCH[3]} ]]; then
    year_end=${BASH_REMATCH[1]}
  else
    year_end=${BASH_REMATCH[3]}
  fi
 else
  year_start=1900
  year_end=2099
 fi

 if [[ $title =~ ^tt[0-9]{7,8}$ ]]; then
  jq -sR '{query:.}' <<- _GQL_ |
 	query {
 	  title(id: "$title") {
 	    id
 	    titleType { text }
 	    titleText { text }
 	    originalTitleText { text }
 	    releaseYear { year }
 	    primaryImage { url }
 	    certificate { rating }
 	    ratingsSummary {
 	      aggregateRating
 	      voteCount
 	    }
 	    runtime { seconds }
 	    titleGenres {
 	      genres {
 	        genre { text }
 	      }
 	    }
 	    latestTrailer { id }
 	  }
 	}
 _GQL_
    curl -s https://caching.graphql.imdb.com/ --json @- |
      jq -c '.data.title | {
              id,
              type:.titleType.text,
              title:.titleText.text,
              year:.releaseYear.year,
              runtime:(.runtime.seconds//0|strftime("%H:%M:%S")),
              rating:.ratingsSummary.aggregateRating,
              genres:((.titleGenres.genres//[])|map(.genre.text)|join(", ")),
          }' |
        qsv jsonl |
          csvlook -I
  exit $?
 fi

 jq -sR '{query:.}' <<- _GQL_ |
 	query {
 	  advancedTitleSearch(
 	    first: $count
 	    constraints: {
 	      titleTextConstraint: { searchTerm: "$title" }
 	      titleTypeConstraint: { anyTitleTypeIds: ["movie", "tvSeries", "tvEpisode", "tvMiniSeries", "tvMovie"] }
 	      releaseDateConstraint: { releaseDateRange: { start: "${year_start}-01-01", end: "${year_end}-12-31" } }
 	      originCountryConstraint: { anyCountries: ["US", "GB", "CA", "AU", "DE", "FR", "IT", "NZ", "JP", "HK"] }
 	    }
 	  ) {
 	    edges {
 	      node {
 	        title {
 	          id
 	          titleType { text }
 	          titleText { text }
 	          originalTitleText { text }
 	          releaseYear { year }
 	          primaryImage { url }
 	          certificate { rating }
 	          ratingsSummary {
 	            aggregateRating
 	            voteCount
 	          }
 	          runtime { seconds }
 	          titleGenres {
 	            genres {
 	              genre { text }
 	            }
 	          }
 	          latestTrailer { id }
 	        }
 	      }
 	    }
 	  }
 	}
 _GQL_
  curl -s https://caching.graphql.imdb.com/ --json @- | # tee /dev/stderr |
    jq -c '.data.advancedTitleSearch.edges[].node.title | {
              id,
              type:.titleType?.text,
              title:.titleText.text,
              year:.releaseYear.year,
              runtime:(.runtime.seconds//0|strftime("%H:%M:%S")),
              rating:.ratingsSummary.aggregateRating,
              genres:((.titleGenres.genres//[])|map(.genre.text)|join(", ")),
          }' |
      qsv jsonl |
        csvlook -I

 # vim: set noet:
diff --git a/payload.json b/payload.json
 {
  "extensions": {
    "persistedQuery": {
      "sha256Hash": "",
      "version": 1
    }
  },
  "operationName": "AdvancedTitleSearch",
  "variables": {
    "first": 100,
    "rankedTitleListConstraint": {
      "allRankedTitleLists": [
        {
          "rankRange": {
            "max": 1000
          },
          "rankedTitleListType": "TOP_RATED_MOVIES"
        }
      ]
    },
    "sortBy": "USER_RATING",
    "sortOrder": "DESC"
  }
 }
diff --git a/update.sh b/update.sh
 #!/bin/bash
 #
 # IMDB Top 1000
 #

 set -euo pipefail

 # point to this gist itself
 GIST_ID=fbc551ecece8639dc0bea18ac7113450

 # network captured from <https://www.imdb.com/search/title/?title=godfather> (update when changed)
 QUERY_ID=be358d7b41add9fd174461f4c8c673dfee5e2a88744e2d5dc037362a96e2b4e4

 gh gist view $GIST_ID -r -f payload.json | jq --arg query_id $QUERY_ID '.extensions.persistedQuery.sha256Hash=$query_id' > payload.json

 echo -e 'id\ttitle\tyear\trating\tvotes\tgenres' > imdb-top-1000.tsv

 for((i=0;i<10;i+=1)); do
  echo "Downloading ... [$((i*100+1))..$((i*100+100))]"
  if ((i==0)); then
    cursor=''
  else
    cursor=$(jq -r '.data.advancedTitleSearch.pageInfo.endCursor' imdb-$((i-1)).json)
  fi
  jq --arg cursor "$cursor" 'if $cursor=="" then . else .variables.after=$cursor end' payload.json | curl -s -H 'Content-Type:application/json' https://caching.graphql.imdb.com/ -d @- > imdb-$i.json
  jq -r '.data.advancedTitleSearch.edges[].node.title|[.id,.titleText.text,.releaseYear.year,.ratingsSummary.aggregateRating,.ratingsSummary.voteCount,([.titleGenres.genres[].genre.text]|join(","))]|@tsv' imdb-$i.json >> imdb-top-1000.tsv
 done

 echo "Sorting ..."
 qsv sort -d '\t' -s rating,votes,year -N -R -o imdb-top-1000.tsv imdb-top-1000.tsv

 echo "Uploading ..."
 gh gist edit --desc "IMDB Top 1000 ($(date +%F))" $GIST_ID imdb-top-1000.tsv -f imdb-top-1000.tsv
	#!/bin/bash
	#
	# IMDB Search
	#
	# - https://gist.github.com/vimagick/fbc551ecece8639dc0bea18ac7113450
	# - https://developer.imdb.com/documentation/api-documentation/sample-queries/
	# - https://github.com/Gujal00/Kodi-Official/blob/master/plugin.video.imdb.trailers/resources/lib/imdb_trailers.py
	# - https://github.com/theMK2k/Media-Hoarder/blob/master/src/data/imdb-graphql-urls.json
	# - https://www.imdb.com/search/title/
	# - https://imdbapi.dev/docs/graphql/schema/title
	#

	title=${1:?title is empty}
	year=${2}
	count=${3:-5}

	if [[ $year =~ ^([0-9]{4})(:([0-9]{4}))?$ ]]; then
	year_start=${BASH_REMATCH[1]}
	if [[ -z ${BASH_REMATCH[3]} ]]; then
	year_end=${BASH_REMATCH[1]}
	else
	year_end=${BASH_REMATCH[3]}
	fi
	else
	year_start=1900
	year_end=2099
	fi

	if [[ $title =~ ^tt[0-9]{7,8}$ ]]; then
	jq -sR '{query:.}' <<- _GQL_ \|
	query {
	title(id: "$title") {
	id
	titleType { text }
	titleText { text }
	originalTitleText { text }
	releaseYear { year }
	primaryImage { url }
	certificate { rating }
	ratingsSummary {
	aggregateRating
	voteCount
	}
	runtime { seconds }
	titleGenres {
	genres {
	genre { text }
	}
	}
	latestTrailer { id }
	}
	}
	_GQL_
	curl -s https://caching.graphql.imdb.com/ --json @- \|
	jq -c '.data.title \| {
	id,
	type:.titleType.text,
	title:.titleText.text,
	year:.releaseYear.year,
	runtime:(.runtime.seconds//0\|strftime("%H:%M:%S")),
	rating:.ratingsSummary.aggregateRating,
	genres:((.titleGenres.genres//[])\|map(.genre.text)\|join(", ")),
	}' \|
	qsv jsonl \|
	csvlook -I
	exit $?
	fi

	jq -sR '{query:.}' <<- _GQL_ \|
	query {
	advancedTitleSearch(
	first: $count
	constraints: {
	titleTextConstraint: { searchTerm: "$title" }
	titleTypeConstraint: { anyTitleTypeIds: ["movie", "tvSeries", "tvEpisode", "tvMiniSeries", "tvMovie"] }
	releaseDateConstraint: { releaseDateRange: { start: "${year_start}-01-01", end: "${year_end}-12-31" } }
	originCountryConstraint: { anyCountries: ["US", "GB", "CA", "AU", "DE", "FR", "IT", "NZ", "JP", "HK"] }
	}
	) {
	edges {
	node {
	title {
	id
	titleType { text }
	titleText { text }
	originalTitleText { text }
	releaseYear { year }
	primaryImage { url }
	certificate { rating }
	ratingsSummary {
	aggregateRating
	voteCount
	}
	runtime { seconds }
	titleGenres {
	genres {
	genre { text }
	}
	}
	latestTrailer { id }
	}
	}
	}
	}
	}
	_GQL_
	curl -s https://caching.graphql.imdb.com/ --json @- \| # tee /dev/stderr \|
	jq -c '.data.advancedTitleSearch.edges[].node.title \| {
	id,
	type:.titleType?.text,
	title:.titleText.text,
	year:.releaseYear.year,
	runtime:(.runtime.seconds//0\|strftime("%H:%M:%S")),
	rating:.ratingsSummary.aggregateRating,
	genres:((.titleGenres.genres//[])\|map(.genre.text)\|join(", ")),
	}' \|
	qsv jsonl \|
	csvlook -I

	# vim: set noet:
	{
	"extensions": {
	"persistedQuery": {
	"sha256Hash": "",
	"version": 1
	}
	},
	"operationName": "AdvancedTitleSearch",
	"variables": {
	"first": 100,
	"rankedTitleListConstraint": {
	"allRankedTitleLists": [
	{
	"rankRange": {
	"max": 1000
	},
	"rankedTitleListType": "TOP_RATED_MOVIES"
	}
	]
	},
	"sortBy": "USER_RATING",
	"sortOrder": "DESC"
	}
	}
	#!/bin/bash
	#
	# IMDB Top 1000
	#

	set -euo pipefail

	# point to this gist itself
	GIST_ID=fbc551ecece8639dc0bea18ac7113450

	# network captured from <https://www.imdb.com/search/title/?title=godfather> (update when changed)
	QUERY_ID=be358d7b41add9fd174461f4c8c673dfee5e2a88744e2d5dc037362a96e2b4e4

	gh gist view $GIST_ID -r -f payload.json \| jq --arg query_id $QUERY_ID '.extensions.persistedQuery.sha256Hash=$query_id' > payload.json

	echo -e 'id\ttitle\tyear\trating\tvotes\tgenres' > imdb-top-1000.tsv

	for((i=0;i<10;i+=1)); do
	echo "Downloading ... [$((i100+1))..$((i100+100))]"
	if ((i==0)); then
	cursor=''
	else
	cursor=$(jq -r '.data.advancedTitleSearch.pageInfo.endCursor' imdb-$((i-1)).json)
	fi
	jq --arg cursor "$cursor" 'if $cursor=="" then . else .variables.after=$cursor end' payload.json \| curl -s -H 'Content-Type:application/json' https://caching.graphql.imdb.com/ -d @- > imdb-$i.json
	jq -r '.data.advancedTitleSearch.edges[].node.title\|[.id,.titleText.text,.releaseYear.year,.ratingsSummary.aggregateRating,.ratingsSummary.voteCount,([.titleGenres.genres[].genre.text]\|join(","))]\|@tsv' imdb-$i.json >> imdb-top-1000.tsv
	done

	echo "Sorting ..."
	qsv sort -d '\t' -s rating,votes,year -N -R -o imdb-top-1000.tsv imdb-top-1000.tsv

	echo "Uploading ..."
	gh gist edit --desc "IMDB Top 1000 ($(date +%F))" $GIST_ID imdb-top-1000.tsv -f imdb-top-1000.tsv