Commit: 31a39df64d03048012f5bc414179e3c24c59b590
Parent: 1e6bc397c8b57d695e5b2bdfdb3e560759e051da
Author: Ren Tatsumoto
Date: Sun, 20 Dec 2020 13:33:21 +0300
trim long filenames individually depending on their timestamps
Diffstat:
| M | subs2srs.lua | | | 118 | +++++++++++++++++++++++++++++++++---------------------------------------------- |
1 file changed, 49 insertions(+), 69 deletions(-)
diff --git a/subs2srs.lua b/subs2srs.lua
@@ -252,60 +252,6 @@ local function subprocess(args, completion_fn)
return command_native(command_table, completion_fn)
end
-local anki_compatible_length
-do
- -- Anki forcibly mutilates all filenames longer than 119 bytes when you run `Tools->Check Media...`.
- local allowed_bytes = 119
- local timestamp_bytes = #'_99h99m99s999ms-99h99m99s999ms.webp'
- local limit_bytes = allowed_bytes - timestamp_bytes
-
- anki_compatible_length = function(str)
- if #str <= limit_bytes then
- return str
- end
-
- local bytes_per_char = contains_non_latin_letters(str) and #'車' or #'z'
- local limit_chars = math.floor(limit_bytes / bytes_per_char)
-
- if limit_chars == limit_bytes then
- return str:sub(1, limit_bytes)
- end
-
- local ret = subprocess {
- 'awk',
- '-v', string.format('str=%s', str),
- '-v', string.format('limit=%d', limit_chars),
- 'BEGIN{print substr(str, 1, limit); exit}'
- }
-
- if ret.status == 0 then
- ret.stdout = remove_newlines(ret.stdout)
- ret.stdout = remove_leading_trailing_spaces(ret.stdout)
- return ret.stdout
- else
- return 'subs2srs_' .. os.time()
- end
- end
-end
-
-local function construct_media_filenames(timings)
- local filename = mp.get_property("filename") -- filename without path
-
- filename = remove_extension(filename)
- filename = remove_text_in_brackets(filename)
- filename = remove_special_characters(filename)
- filename = anki_compatible_length(filename)
-
- filename = string.format(
- '%s_%s-%s',
- filename,
- human_readable_time(timings['start']),
- human_readable_time(timings['end'])
- )
-
- return filename .. config.snapshot_extension, filename .. config.audio_extension
-end
-
local function construct_note_fields(sub_text, snapshot_filename, audio_filename)
return {
[config.sentence_field] = sub_text,
@@ -318,10 +264,6 @@ local function minutes_ago(m)
return (os.time() - 60 * m) * 1000
end
-local function get_time_pos()
- return string.format('%.3f', mp.get_property_number("time-pos", 0))
-end
-
local function join_media_fields(new_data, stored_data)
for _, field in pairs { config.audio_field, config.image_field } do
new_data[field] = table.get(stored_data, field, "") .. table.get(new_data, field, "")
@@ -468,31 +410,67 @@ end
local filename_factory = (function()
local filename
+ local anki_compatible_length = (function()
+ -- Anki forcibly mutilates all filenames longer than 119 bytes when you run `Tools->Check Media...`.
+ local allowed_bytes = 119
+ local timestamp_bytes = #'_99h99m99s999ms-99h99m99s999ms.webp'
+
+ return function(str, timestamp)
+ -- if timestamp provided, recalculate limit_bytes
+ local limit_bytes = allowed_bytes - (timestamp and #timestamp or timestamp_bytes)
+
+ if #str <= limit_bytes then
+ return str
+ end
+
+ local bytes_per_char = contains_non_latin_letters(str) and #'車' or #'z'
+ local limit_chars = math.floor(limit_bytes / bytes_per_char)
+
+ if limit_chars == limit_bytes then
+ return str:sub(1, limit_bytes)
+ end
+
+ local ret = subprocess {
+ 'awk',
+ '-v', string.format('str=%s', str),
+ '-v', string.format('limit=%d', limit_chars),
+ 'BEGIN{print substr(str, 1, limit); exit}'
+ }
+
+ if ret.status == 0 then
+ ret.stdout = remove_newlines(ret.stdout)
+ ret.stdout = remove_leading_trailing_spaces(ret.stdout)
+ return ret.stdout
+ else
+ return 'subs2srs_' .. os.time()
+ end
+ end
+ end)()
+
local make_media_filename = function()
filename = mp.get_property("filename") -- filename without path
filename = remove_extension(filename)
filename = remove_text_in_brackets(filename)
filename = remove_special_characters(filename)
- filename = anki_compatible_length(filename)
end
local make_audio_filename = function(speech_start, speech_end)
- return string.format(
- '%s_%s-%s%s',
- filename,
+ local filename_timestamp = string.format(
+ '_%s-%s%s',
human_readable_time(speech_start),
human_readable_time(speech_end),
config.audio_extension
)
+ return anki_compatible_length(filename, filename_timestamp) .. filename_timestamp
end
local make_snapshot_filename = function(timestamp)
- return string.format(
- '%s_%s%s',
- filename,
+ local filename_timestamp = string.format(
+ '_%s%s',
human_readable_time(timestamp),
config.snapshot_extension
)
+ return anki_compatible_length(filename, filename_timestamp) .. filename_timestamp
end
mp.register_event("file-loaded", make_media_filename)
@@ -517,10 +495,11 @@ local function export_to_anki(gui)
sub['text'] = string.format([[<span id="mpv%s">mpvacious wasn't able to grab subtitles</span>]], os.time())
end
- local snapshot_filename = filename_factory.make_snapshot_filename(get_time_pos())
+ local snapshot_timestamp = mp.get_property_number("time-pos", 0)
+ local snapshot_filename = filename_factory.make_snapshot_filename(snapshot_timestamp)
local audio_filename = filename_factory.make_audio_filename(sub['start'], sub['end'])
- encoder.create_snapshot(get_time_pos(), snapshot_filename)
+ encoder.create_snapshot(snapshot_timestamp, snapshot_filename)
encoder.create_audio(sub['start'], sub['end'], audio_filename)
local note_fields = construct_note_fields(sub['text'], snapshot_filename, audio_filename)
@@ -542,11 +521,12 @@ local function update_last_note(overwrite)
return
end
- local snapshot_filename = filename_factory.make_snapshot_filename(get_time_pos())
+ local snapshot_timestamp = mp.get_property_number("time-pos", 0)
+ local snapshot_filename = filename_factory.make_snapshot_filename(snapshot_timestamp)
local audio_filename = filename_factory.make_audio_filename(sub['start'], sub['end'])
local create_media = function()
- encoder.create_snapshot(get_time_pos(), snapshot_filename)
+ encoder.create_snapshot(snapshot_timestamp, snapshot_filename)
encoder.create_audio(sub['start'], sub['end'], audio_filename)
end