Configure Feed2toot =================== As a prerequisite to use Feed2toot, you need to authorize a Mastodon app for your account. Just use the script register_feed2toot_app to register the feed2toot app for your account.:: $ ./register_feed2toot_app This app generates Mastodon app credentials needed by Feed2toot. feed2toot_clientcred.txt and feed2toot_usercred.txt will be written in the current dir /home/chaica/progra/python/feed2toot. One connection is initiated to create the app. Your password is *not* stored. Mastodon instance url (defaults to https://mastodon.social): Mastodon login:chaica@ohmytux.com Mastodon password: The feed2toot app was added to your preferences=>authorized apps page As described above, two files were created. You'll need them in the feed2toot configuration. In order to configure Feed2toot, you need to create a feed2toot.ini file (or any name you prefer, finishing with the extension .ini) with the following parameters:: [mastodon] ; Here you need the two files created by register_feed2toot_app user_credentials=/etc/feed2toot/credentials/feed2toot_usercred.txt client_credentials=/etc/feed2toot/credentials/feed2toot_clientcred.txt [cache] cachefile=/var/lib/feed2toot/feed2toot.db cache_limit=10000 [rss] uri: https://www.journalduhacker.net/rss uri_list: /etc/feed2toot//rsslist.txt tweet: {title} {link} title_pattern: Open Source title_pattern_case_sensitive: true no_uri_pattern_no_global_pattern=true [hashtaglist] several_words_hashtags_list: /etc/feed2toot/hashtags.txt For the [mastodon] section: - user_credentials: a file with the user credentials, generated by the command register_feed2toot_app - client_credentials: a file with the client credentials, generated by the command register_feed2toot_app For the [cache] section: - cachefile: the path to the cache file storing ids of already tweeted links. Absolute path is mandatory. This file should always use the .db extension. - cache_limit: length of the cache queue. defaults to 100. For the [rss] section: - uri: the url of the rss feed to parse - uri_list: a path to a file with several adresses of rss feeds, one by line. Absolute path is mandatory. - tweet: format of the tweet you want to post. It should use existing entries of the RSS fields like {title} or {link}. Launch it with this field empty to display all available entries. - {one field of the rss feed}_pattern: takes a string representing a pattern to match for a specified field of each rss entry of the rss feed, like title_pattern or summary_pattern. - {one field of the rss feed}_pattern_case_sensitive: either the pattern matching for the specified field should be case sensitive or not. Default to true if not specified. - no_uri_pattern_no_global_pattern: don't apply global pattern (see above) when no pattern-by-uri is defined in the uri_list. Allows to get all entries of a rss in the uri_list because no pattern is defined so we match them all. Defaults to false, meaning the global patterns will be tried on every rss in the uri_list NOT HAVING specific patterns and so ONLY entries from the specific uri in the uri_list matching the global patterns will be considered. For the [hashtaglist] section: - several_words_hashtags_list: a path to the file containing hashtags in two or more words. Absolute path is mandatory. By default Feed2toot adds a # before every words of a hashtag. List of rss feeds ================= Simple list of rss feeds ------------------------ With the parameter **uri_list**, you can define a list of uri to use. Starting from 0.10, Feed2toot is now able to match specific patterns for each of the rss feeds from this list. Consider the following rss section of the configuration file:: [rss] uri_list=/home/john/feed2toot/rsslist.txt tweet={title} {link} Now let's have a look at the =/home/john/feed2toot/rsslist.txt file:: https://www.journalduhacker.net/rss https://carlchenet.com/feed Each line of this file is a url to a rss feed. Pretty simple. Match specific patterns of rss feeds in the uri_list files ---------------------------------------------------------- You can use specific pattern matching for uri in the uri_list file to filter some of the rss entries of a rss feed. Lets modify the previous file:: https://www.journalduhacker.net/rss|title|hacker,psql https://carlchenet.com/feed|title|gitlab Each line of this file starts with an uri, followed by a pipe (|), followed by the name of the available section to parse (see below), again followed by a pipe (|), followed by patterns, each pattern being separated from the other one by a semi-colon (,). In the example file above wee get every rss entries from the feed available at https://www.journalduhacker.net/rss where a substring in the title section of this entry matches either "hacker" or "psql". Specific patterns are not case sensitive. For the second line, we match every rss entries from the feed available at https://carlchenet.com/feed where a substring in the title section of this entry matches "gitlab". Consider every entries of a rss feed from a uri in the uri_list file -------------------------------------------------------------------- It is possible to get all entries from a rss feed available in the uri_list file. You need an option to deactivate the global pattern matching for uri in the uri_list NOT having specific patterns:: [rss] ... no_uri_pattern_no_global_pattern=true In you rsslist.txt, just don't give anything else than the needed feed url to get all the entries:: https://www.journalduhacker.net/rss|title|hacker,psql https://carlchenet.com/feed|title|gitlab https://blog.linuxjobs.fr/feed.php?rss The last line of the file above only has the url of a rss feed. All entries from this feed will be tweeted. How to display available sections of the rss feed ================================================= Feed2toot offers the **--rss-sections** command line option to display the available section of the rss feed and exits:: $ feed2toot --rss-sections -c feed2toot.ini The following sections are available in this RSS feed: ['title', 'comments', 'authors', 'link', 'author', 'summary', 'links', 'tags', id', 'author_detail', 'published'].