我的Twitter画师插画自动下载方案

前言

由于推特上的画师挺多的，偶尔看到好看的图片保存下来也比较麻烦，就借助了一些工具可以对某个推特用户进行监控，定时自动下载。

环境要求

Python 3.4+

Linux

开始配置工具

安装gallery-dl

仓库地址：https://github.com/mikf/gallery-dl

安装：

pip install gallery-dl

更新命令：

pip install --upgrade gallery-dl

gallery-dl用法

M:\cmder
λ gallery-dl --help
usage: gallery-dl [OPTION]... URL...

General Options:
  -h, --help                Print this help message and exit
  --version                 Print program version and exit
  -d, --dest DEST           Destination directory
  -i, --input-file FILE     Download URLs found in FILE ('-' for stdin). More than one --input-file can be specified
  --cookies FILE            File to load additional cookies from
  --proxy URL               Use the specified proxy
  --clear-cache [MODULE]    Delete all cached login sessions, cookies, etc.

Output Options:
  -q, --quiet               Activate quiet mode
  -v, --verbose             Print various debugging information
  -g, --get-urls            Print URLs instead of downloading
  -G, --resolve-urls        Print URLs instead of downloading; resolve intermediary URLs
  -j, --dump-json           Print JSON information
  -s, --simulate            Simulate data extraction; do not download anything
  -E, --extractor-info      Print extractor defaults and settings
  -K, --list-keywords       Print a list of available keywords and example values for the given URLs
  --list-modules            Print a list of available extractor modules
  --list-extractors         Print a list of extractor classes with description, (sub)category and example URL
  --write-log FILE          Write logging output to FILE
  --write-unsupported FILE  Write URLs, which get emitted by other extractors but cannot be handled, to FILE
  --write-pages             Write downloaded intermediary pages to files in the current directory to debug problems

Downloader Options:
  -r, --limit-rate RATE     Maximum download rate (e.g. 500k or 2.5M)
  -R, --retries N           Maximum number of retries for failed HTTP requests or -1 for infinite retries (default: 4)
  -A, --abort N             Abort extractor run after N consecutive file downloads have been skipped, e.g. if files with the same filename already exist
  --http-timeout SECONDS    Timeout for HTTP connections (default: 30.0)
  --sleep SECONDS           Number of seconds to sleep before each download
  --filesize-min SIZE       Do not download files smaller than SIZE (e.g. 500k or 2.5M)
  --filesize-max SIZE       Do not download files larger than SIZE (e.g. 500k or 2.5M)
  --no-part                 Do not use .part files
  --no-skip                 Do not skip downloads; overwrite existing files
  --no-mtime                Do not set file modification times according to Last-Modified HTTP response headers
  --no-download             Do not download any files
  --no-check-certificate    Disable HTTPS certificate validation

Configuration Options:
  -c, --config FILE         Additional configuration files
  -o, --option OPT          Additional '<key>=<value>' option values
  --ignore-config           Do not read the default configuration files

Authentication Options:
  -u, --username USER       Username to login with
  -p, --password PASS       Password belonging to the given username
  --netrc                   Enable .netrc authentication data

Selection Options:
  --download-archive FILE   Record all downloaded files in the archive file and skip downloading any file already in it.
  --range RANGE             Index-range(s) specifying which images to download. For example '5-10' or '1,3-5,10-'
  --chapter-range RANGE     Like '--range', but applies to manga-chapters and other delegated URLs
  --filter EXPR             Python expression controlling which images to download. Files for which the expression evaluates to False are ignored. Available keys are
                            the filename-specific ones listed by '-K'. Example: --filter "image_width >= 1000 and rating in ('s', 'q')"
  --chapter-filter EXPR     Like '--filter', but applies to manga-chapters and other delegated URLs

Post-processing Options:
  --zip                     Store downloaded files in a ZIP archive
  --ugoira-conv             Convert Pixiv Ugoira to WebM (requires FFmpeg)
  --ugoira-conv-lossless    Convert Pixiv Ugoira to WebM in VP9 lossless mode
  --write-metadata          Write metadata to separate JSON files
  --write-tags              Write image tags to separate text files
  --mtime-from-date         Set file modification times according to 'date' metadata
  --exec CMD                Execute CMD for each downloaded file. Example: --exec 'convert {} {}.png && rm {}'
  --exec-after CMD          Execute CMD after all files were downloaded successfully. Example: --exec-after 'cd {} && convert * ../doc.pdf'

开始使用

由于需要下载多个画师的插画，这里需要使用–input-file参数。

首先创建一个gallery-dl目录作为下载路径：

mkdir gallery-dl

然后编写twitter_dl.sh代码如下

#!/bin/bash
dir=/root/gallery-dl
cd $dir && /root/.pyenv/versions/3.8.6/bin/gallery-dl --input-file list.txt

将twitter_dl.sh放在gallery-dl目录下，赋予权限：

chmod +x gallery-dl

然后在当前目录创建list.txt文件：

touch list.txt

自己用vim或者其它任意编辑器将需要下载的画师的推特账户链接粘贴到list.txt里面，一行一个；

list.txt文件示例：

https://twitter.com/BillGates

定时执行

最后定时任务加入这个脚本即可，定时任务的使用不再详细赘述，谷歌即可；

/bin/bash /root/gallery-dl/twitter_dl.sh

前言