You can dump a JSON with --dump-single-json (or simply -J) or use the --no-download parameter.
writes info.json in working dir
yt-dlp --write-comments --no-download “$url”
dump complete JSON with comments
yt-dlp --write-comments --dump-single-json -o “ f i l e " " file" " file""url”
yt-dlp --write-comments --dump-single-json “ u r l " > " url" > " url">"file”
Pretty print with JQ (JSON processor, package needs to be installed)
yt-dlp --write-comments --dump-single-json “ u r l " ∣ j q y t − d l p − − w r i t e − c o m m e n t s − − d u m p − s i n g l e − j s o n " url" | jq yt-dlp --write-comments --dump-single-json " url"∣jqyt−dlp−−write−comments−−dump−single−json"url” | jq > “$file”
Then, use JQ to query the comments.
jq ‘.comments’ “KaTeX parse error: Expected 'EOF', got '#' at position 28: …ents_only.json #̲ file input jq …json” > comments_only.json # var input
or
some_command | jq ‘.comments’
.comments is the JQ selector for the JSON property ‘comments’ here.
The JQ approach becomes somewhat more complicated if you want to integrate those comments that you isolated with JQ into a another JSON file.
Here is an example where I’m doing this with entries to boost you forward in case you chose to go this route.
old_json= ( j q ′ . ′ " (jq '.' " (jq′.′"json_file")
entries= ( j q − s ′ . [ 0 ] . e n t r i e s + = . [ 1 ] . e n t r i e s ∣ . [ 0 ] . e n t r i e s ∣ s o r t b y ( . d a t e ) ∣ r e v e r s e ′ < < < " (jq -s '.[0].entries += .[1].entries | .[0].entries | sort_by(.date) | reverse' <<<" (jq−s′.[0].entries+=.[1].entries∣.[0].entries∣sortby(.date)∣reverse′<<<"old_json" <<<" j s o n " ) j s o n = json") json= json")json=(jq -s ‘.[0].entries = .[1] | .[0]’ <<<“ o l d j s o n " < < < " old_json" <<<" oldjson"<<<"entries”)
Share
Improve this answer
