To block the Jason 20 Gb makes sense, but to processs one video is okay.

@zhuyifei1999 The tools.wmflabs.org is served in https:// only, so http:// is a redirect. Could you adjust the URLs?

Oh, by the way, if you tweak the output, as far as I'm concerned,, the dream output would be something like:

**URLs**

* https://tools.wmflabs.org/video2commons/static/ssu/Shree_420,_Hindi_film.webm

//Description files are available too: append `.txt` to the URLs.//

**Checksums**

| **File** | **MD5** |
| Shree_420,_Hindi_film.webm | 5bc858d4ca990881e99374ba9c776f97 |

That would give this:

URLs

• https://tools.wmflabs.org/video2commons/static/ssu/Shree_420,_Hindi_film.webm

Description files are available too: append .txt to the URLs.

Checksums

This is clearly helpful to have a table to check MD5 and to be able to grab a list of the URLs. Lost in URLs, that's not legible.

To select a part of the command is annoying (I use an alias to invoke curl with the right parameters, not wget for example, to get files for example).

When you have a big list of files I need to use a tool to convert the output to ^wget (http://*.?\.webm).*$ to $1.txt then $1.

The note about .txt is useful if someone else handles the task, so they know where to find them. To avoid to waste space, no need for any duplicate URL or to any baroque syntaxe only usable in some tools like {.txt}.

With that we have all the essential information clear and ready to use by copy/paste to compose any command sequences we want.