This is a Python script to download multiple different file with same PREFIX and SUFFIX, but different number inside the file (very common in serial files)
For example:
https://www.eyesonanime.org/DDL/ANIME/BlackClover/BlackClover_Ep_01_SUB_ITA.mp4
https://www.eyesonanime.org/DDL/ANIME/BlackClover/BlackClover_Ep_02_SUB_ITA.mp4
...
https://www.eyesonanime.org/DDL/ANIME/BlackClover/BlackClover_Ep_10_SUB_ITA.mp4
... and so on
To install this script and use from terminal every where you are following this guide:
cd ~/Documents/ # or any where you want download
git clone https://github.com/Alfystar/listDownload.git
cd listDownload
source install.sh
source ~/.profile # to add path on current terminal
After reboot, the path will be add on all path in all environment, until the rebooting you have to execute in any new terminal source ~/.profile
.
Possible example to use the command do download list of anime are:
listDownload.py https://www.eyesonanime.org/DDL/ANIME/BlackClover/BlackClover_Ep_ _SUB_ITA.mp4 1 20 -p 100 -d 1 -q
For case with big number of tiny file where the digit aren't always 2:
listDownload.py https://www.egr.msu.edu/~khalil/NonlinearSystems/Sample/Lect_ .pdf 1 41 -p 100 -d 1 -q
More over, if you have to download form different site and with different macro url, it's possible create list in a file, and next use the software to parallelize the download in efficient manner:
Create file named, for example, test.txt
:
https://www.egr.msu.edu/~khalil/NonlinearSystems/Sample/Lect_ .pdf 1 5 -p 100 -d 1 -o test1 # Save in a directory named "test1"
https://www.egr.msu.edu/~khalil/NonlinearSystems/Sample/Lect_ .pdf 10 15 -p 100 -d 1 -o "test 2" # Save in a directory named with a space
https://www.egr.msu.edu/~khalil/NonlinearSystems/Sample/Lect_ .pdf 20 25 -p 100 -d 1 # Save on the default name
and next call the software (remember, test.txt
is the path where the file is saved):
listDownload.py -i test.txt -o testDir
with this output
.
├── testDir
│ ├── Lect_20.pdf
│ ├── Lect_21.pdf
│ ├── Lect_22.pdf
│ ├── Lect_23.pdf
│ ├── Lect_24.pdf
│ └── Lect_25.pdf
├── test1
│ ├── Lect_1.pdf
│ ├── Lect_2.pdf
│ ├── Lect_3.pdf
│ ├── Lect_4.pdf
│ └── Lect_5.pdf
├── test_2
│ ├── Lect_10.pdf
│ ├── Lect_11.pdf
│ ├── Lect_12.pdf
│ ├── Lect_13.pdf
│ ├── Lect_14.pdf
│ └── Lect_15.pdf
└── test.txt
listDownload.py <baseUrl> <endUrl> <startNum> <endNum> [Options]
baseUrl:= First part of the url (until XX number)
endUrl:= Second part of the url (after XX number)
startNum:= First index included
endNum:= Last index (included)
[Options]:
-p --parallelDownload <int> Number of concurrency downdload (default = 5)
-o --outSave <path> Saving directory Path (default = ./listDowndload)
-d --digit <digit> Number of digit (default = 2)
--quiet no show all data
Is also possible setup multiple download in a File:
./parallelDownload.py -i <File List> [Options All]
[Options All]:
-p --parallelDownload <int> Number of concurrency downdload (default = 5)
--quiet no show all data
[File List syntax]:
For EVERY LINE the syntax MUST be is:
<baseUrl> <endUrl> <startNum> <endNum> [Options file]
[Options file] (in case of missing, the global option will be used:
-o --outSave <path> Saving directory Path (default = see global)
-d --digit <digit> Number of digit (default = see global)
- Cross Platform, boot Linux and Windows
- Real parallel download customizable
- Parametric Digit of the number
- Output direcory choosable and recursivly create
- File Setup for multiple download in multiple sub directory
- Debug info
- wget (default)
- xterm (default)
- Invoke-WebRequest(default)
- Multiple terminal for Windows -Gui