[BUG] VLC displays captions, ccextractorwin not finding captions #471

New Issue

claunia · 2026-01-29T16:44:44Z

claunia commented

2026-01-29 16:44:44 +00:00

Originally created by @DHLgh on GitHub (Jan 1, 2019).

CCExtractor version (using the --version parameter preferably) : 0.87

[ X] I have used the latest available version of CCExtractor to verify this issue exists.
[X ] I absolutely love CCExtractor, but have not contributed previously.

Necessary information

Is this a regression (did it work before)? [X ] NO | [ ] YES - please specify the last known working version
What platform did you use? [ X] Windows - [ ] Linux - [ ] Mac
What were the used arguments? -ttxt -2

0.77 reports that CC is type 1 (field 2 closed captions).

**Video links (replace text below with your links) **

FS2_sample.zip

Originally created by @DHLgh on GitHub (Jan 1, 2019). CCExtractor version (using the --version parameter preferably) : **0.87** - [ X] I have used the latest available version of CCExtractor to verify this issue exists. - [X ] I absolutely love CCExtractor, but have not contributed previously. **Necessary information** - Is this a regression (did it work before)? [X ] NO | [ ] YES - *please specify the last known working version* - What platform did you use? [ X] Windows - [ ] Linux - [ ] Mac - What were the used arguments? -ttxt -2 0.77 reports that CC is type 1 (field 2 closed captions). **Video links (replace text below with your links) ** [FS2_sample.zip](https://github.com/CCExtractor/ccextractor/files/2719624/FS2_sample.zip)

claunia added the needs-confirmation-of-being-broken difficulty: medium labels 2026-01-29 16:44:44 +00:00

claunia closed this issue

2026-01-29 16:44:45 +00:00

claunia commented

2026-01-29 16:44:45 +00:00

@navimakarov commented on GitHub (Jan 1, 2019):

I'm on it

@navimakarov commented on GitHub (Jan 1, 2019): I'm on it

claunia commented

2026-01-29 16:44:46 +00:00

@saurabh18213 commented on GitHub (Feb 1, 2019):

I am also unable to extract captions in Ubuntu 18.04.
The command which I used from ccextractor/linux directory.
./ccextractor FS2_sample.ts -2 -out=ttxt

@saurabh18213 commented on GitHub (Feb 1, 2019): I am also unable to extract captions in Ubuntu 18.04. The command which I used from ccextractor/linux directory. `./ccextractor FS2_sample.ts -2 -out=ttxt`

claunia commented

2026-01-29 16:44:46 +00:00

@thelastpolaris commented on GitHub (Apr 25, 2019):

@saurabh18213 I also tried it with -2 flag but for some reason, it doesn't work, though CCExtractors seems to be extracting the second CC type.

Total frames time: 00:00:45:945 (1377 frames at 29,97fps) CC type 0: 0 (NTSC line 21 field 1 closed captions) CC type 1: 1360 (NTSC line 21 field 2 closed captions) CC type 2: 0 (DTVCC Channel Packet Data) CC type 3: 0 (DTVCC Channel Packet Start)

I'm currently looking into this issue.

@thelastpolaris commented on GitHub (Apr 25, 2019): @saurabh18213 I also tried it with -2 flag but for some reason, it doesn't work, though CCExtractors seems to be extracting the second CC type. `Total frames time: 00:00:45:945 (1377 frames at 29,97fps) CC type 0: 0 (NTSC line 21 field 1 closed captions) CC type 1: 1360 (NTSC line 21 field 2 closed captions) CC type 2: 0 (DTVCC Channel Packet Data) CC type 3: 0 (DTVCC Channel Packet Start)` I'm currently looking into this issue.

claunia commented

2026-01-29 16:44:46 +00:00

@thelastpolaris commented on GitHub (Apr 28, 2019):

So for some reason CCExtractor treats Program name, which is given by XDS, as subtitles track. If you convert the given sample to .mp4 or even .mkv CCExtractor extracts subtitles without any problem. I will dive deeper into write_cc_buffer function to solve this problem.

@thelastpolaris commented on GitHub (Apr 28, 2019): So for some reason CCExtractor treats Program name, which is given by XDS, as subtitles track. If you convert the given sample to .mp4 or even .mkv CCExtractor extracts subtitles without any problem. I will dive deeper into write_cc_buffer function to solve this problem.

claunia commented

2026-01-29 16:44:46 +00:00

@cfsmp3 commented on GitHub (Jan 25, 2020):

@DHLgh Is this still happening in current master?

@cfsmp3 commented on GitHub (Jan 25, 2020): @DHLgh Is this still happening in current master?

claunia commented

2026-01-29 16:44:47 +00:00

@cfsmp3 commented on GitHub (Dec 14, 2025):

Investigation Results

I've thoroughly analyzed the sample file FS2_sample.ts and found that this file does not contain actual CC3/CC4 caption text.

What the file contains

CC Type	Count	Description
CC type 0	0	No field 1 captions
CC type 1	1360	Field 2 bytes present

However, examining the field 2 data reveals:

All 1360 bytes are XDS (Extended Data Services) metadata, not caption text
Only one caption control code found: 14 2C (EDM - EraseDisplayedMemory) at the very end
No actual CC3/CC4 caption characters anywhere in the file

XDS metadata found in the file

Program name: "UFC UNLEASHED-Jon Jones vs Ly"
Network: "FS2"
Program type: Sports (25)
Content advisory information
Program timing information

Why VLC might "show captions"

VLC can display XDS metadata (program names, network info) in its OSD or caption area, which might appear as "captions" but is actually just program metadata - not spoken dialogue captions.

FFmpeg confirmation

FFmpeg also cannot extract captions from this file, reporting:

Data ignored due to columns exceeding screen width

(repeated 714 times)

Conclusion

CCExtractor is working correctly - there simply are no CC3/CC4 closed captions to extract from this file. The field 2 data consists entirely of XDS metadata plus one screen-clear command.

Closing as "not a bug" - the sample file doesn't contain the caption content that was expected.

@cfsmp3 commented on GitHub (Dec 14, 2025): ## Investigation Results I've thoroughly analyzed the sample file `FS2_sample.ts` and found that **this file does not contain actual CC3/CC4 caption text**. ### What the file contains | CC Type | Count | Description | |---------|-------|-------------| | CC type 0 | 0 | No field 1 captions | | CC type 1 | 1360 | Field 2 bytes present | However, examining the field 2 data reveals: - **All 1360 bytes are XDS (Extended Data Services) metadata**, not caption text - Only **one caption control code** found: `14 2C` (EDM - EraseDisplayedMemory) at the very end - **No actual CC3/CC4 caption characters** anywhere in the file ### XDS metadata found in the file - Program name: "UFC UNLEASHED-Jon Jones vs Ly" - Network: "FS2" - Program type: Sports (25) - Content advisory information - Program timing information ### Why VLC might "show captions" VLC can display XDS metadata (program names, network info) in its OSD or caption area, which might appear as "captions" but is actually just program metadata - not spoken dialogue captions. ### FFmpeg confirmation FFmpeg also cannot extract captions from this file, reporting: ``` Data ignored due to columns exceeding screen width ``` (repeated 714 times) ### Conclusion CCExtractor is working correctly - there simply are no CC3/CC4 closed captions to extract from this file. The field 2 data consists entirely of XDS metadata plus one screen-clear command. Closing as "not a bug" - the sample file doesn't contain the caption content that was expected.

claunia commented

2026-01-29 16:44:47 +00:00

@cfsmp3 commented on GitHub (Dec 14, 2025):

Closing as the sample file does not contain actual closed captions - only XDS metadata.

@cfsmp3 commented on GitHub (Dec 14, 2025): Closing as the sample file does not contain actual closed captions - only XDS metadata.

claunia referenced this issue

2026-01-29 17:00:52 +00:00

[PR #471] [CLOSED] GCI task: Added error for blank output #1288

Sign in to join this conversation.