Forward RSS feeds to your email address, community maintained

Hi!

The current location ~/.local/share/ is a weird place to keep state with regard to to the file hierarchy standard. It's the .local equivalent of /usr/share which is intend for "Architecture-independent" (see /usr/share) "read-only" (see /usr) data. Maybe ~/.local/var/lib/ — "State information. Persistent data modified by programs as they run" (see /var/lib)— is a better place for the current use semantic?

Best, Sebastian

Some sites refuses fetching rss, when user-agent is not known to them. I prepared simple change which allow define "http-user-agent-option" and pass it to parser (diff is against debian version but I think this is no issue)

Hi!

I learned today that rss2email==3.12.2 with feedparser<6 (with is 5.2.1) does not work with Python 3.9:

$ r2e add name1 url2 account3
Traceback (most recent call last):
  File "/usr/lib/python-exec/python3.9/r2e", line 3, in <module>
    import rss2email.main
  File "/usr/lib/python3.9/site-packages/rss2email/main.py", line 34, in <module>
    from . import command as _command
  File "/usr/lib/python3.9/site-packages/rss2email/command.py", line 37, in <module>
    from . import error as _error
  File "/usr/lib/python3.9/site-packages/rss2email/error.py", line 30, in <module>
    import feedparser as _feedparser
  File "/usr/lib/python3.9/site-packages/feedparser.py", line 93, in <module>
    _base64decode = getattr(base64, 'decodebytes', base64.decodestring)
AttributeError: module 'base64' has no attribute 'decodestring'

Hence it would rock to have a rss2email release that works with more recent feedparser>=6. Are there any plans to make a release off master some time soon?

Thanks and best

Sebastian

PS:

$ python3.8 -c 'import base64; print(hasattr(base64, "decodestring"))'
True

$ python3.9 -c 'import base64; print(hasattr(base64, "decodestring"))'
False

This PR corrects how the locking works, which caused issues in e.g. https://github.com/rss2email/rss2email/issues/83. The problem is outlined in a comment I wrote there but basically, the desired behaviour is that if there are multiple instances of r2e about, they should take it in turns to run and avoid all running at the same time, possibly mangling the data file or the config file.

I implement this by acquiring an exclusive lock on the data file right at the start and releasing it right at the end. As far as I can tell, this eliminates the error @ezbik was experiencing, but it's always possible I broke something else subtle.

Also, there is a branch that only happens if we're not on Solaris...I certainly hope no-one is.

The previous format (multipart/digest), was incompatible with many popular e-mail readers (such as gmail), which made it essentially unusable.

The new format is multipart/mixed, with each part consisting of a single non-digest message (of type text/plain or text/html). One possible incorrectness is that the parts still retain message headers (subject, date, etc.) that are not really used by normal text attachments.

This is an incompatible change. It looks different, and may break existing digest-post-proc functions.

This runs notably faster even with the small number of tests we currently have and can make debugging easier in certain situations.

However I'd suggest we continue using subprocesses in CI for better isolation. This will ensure that we don't accidentally introduce inter-test state dependence which in my experience is a common cause of false negative situations where tests don't actually test what they're supposed to.

% r2e list  | grep netch
23: [*] netch (https://dottedmag:<redacted password>@netch.dreamwidth.org/data/atom?auth=digest -> [email protected])
% r2e run --no-send
processing error: expected bytes-like object, not str: netch (https://dottedmag:<redacted password>@netch.dreamwidth.org/data/atom?auth=digest -> [email protected])
% r2e --version
rss2email 3.9
%

It used to work in an ancient Python 2 version of rss2email.

Hi, i am rebuilding my pi with rss2email running bullseye. I installed exim4 and tested to send mail to my hosting proviers with authentication. Sending mail by the mail command did result in a delivered mail in my mailbox.

Running r2e run seem not to generate mails or are not been distributed by exim4 when i look in the config there is an executable file in /usr/sbin/sendmail (it seems a exim ) root@nms:/home/pi# /usr/sbin/sendmail Exim is a Mail Transfer Agent. It is normally called by Mail User Agents, not directly from a shell command line. Options and/or arguments control what it does when called. For a list of options, see the Exim documentation.

The question is: does this emulate traditional sendmail and should there be an email in the exim maillog when r2e runns...??

And, is there a trick i can use to create testmails with r2e and have some debug entrys to check?

Regards

Eelco

Not all e-mail clients (ahem, mail user agents) display the multipart digest format in a usable way, for example k-9 on android and gmail just show the parts as downloadable attachments.

It would be nice to have an optional digest format as a single (html or plaintext) body. It can basically look like the concatenation of the digest parts with some separator. Alternatively (the cheap and nasty version), we can just make the message multipart/mixed wrapped around text/html, instead of multipart/digest around message/rfc822 around text/html.

What do y'all think?

Since I want this feature, I'll be happy to write the code, obvs.

Hi, I'm facing an issue when I try to run r2e:

r2e -V run could not connect to mail server smtp.gmail.com cause: server_hostname cannot be an empty string or start with a leading dot. check your config file to confirm that smtp-server and other mail server settings are configured properly

Could you please help me to fix it? Here is my rss2email.cfg:

[DEFAULT] from = [email protected] use-8bit = False force-from = False use-publisher-email = False name-format = {feed-title}: {author} to = [email protected] proxy = feed-timeout = 60 active = True digest = False date-header = True date-header-order = modified, issued, created, expired bonus-header = trust-guid = True trust-link = False encodings = US-ASCII, ISO-8859-1, UTF-8, BIG5, ISO-2022-JP post-process = digest-post-process = html-mail = True use-css = False css = h1 { font: 18pt Georgia, "Times New Roman"; } body { font: 12pt Arial; } a:link { font: 12pt Arial; font-weight: bold; color: #0000cc; } blockquote { font-family: monospace; } .header { background: #e0ecff; border-bottom: solid 4px #c3d9ff; padding: 5px; margin-top: 0px; color: red; } .header a { font-size: 20px; text-decoration: none; } .footer { background: #c3d9ff; border-top: solid 4px #c3d9ff; padding: 5px; margin-bottom: 0px; } border: solid 4px #c3d9ff; } margin-left: 5px; margin-right: 5px; } unicode-snob = False links-after-each-paragraph = False body-width = 0 email-protocol = smtp sendmail = /usr/sbin/sendmail smtp-auth = True smtp-username = [email protected] smtp-password = password smtp-server = smtp.gmail.com:587 smtp-ssl = True smtp-ssl-protocol = SSLv3 imap-auth = False imap-username = username imap-password = password imap-server = imap.yourisp.net imap-port = 143 imap-ssl = False imap-mailbox = INBOX verbose = warning

Thanks in advance for your kind help.

When emails are being sent in text format instead of HTML, links are converted to markdown syntax and then the body of the text is line-wrapped. The line-wrapping doesn't treat links in any special way, which means that links with hyphens in them can get wrapped at the hyphen, thus breaking the link.

Fixes #229 by not attempting to enumerate all non-error statuses explicitly.

While researching for this fix, I noticed that HTTP status 308 (Moved Permanently) is another form of permanent redirect and corrected the code to include that case. Note that currently feedparser does not handle HTTP status 308 as a redirect, but this will change when (hopefully) https://github.com/kurtmckee/feedparser/pull/326 gets merged and released there.

In addition, while I was in the area, I added handling of status 410 (Gone) in similar vein to the handling of permanent redirects, by marking the feed as inactive (as suggested in the feedparser documentation).

When using rss2email on https://www.nature.com/nphys.rss I'm getting HTTP 303 errors. Feedparser seems to handle getting the feed content OK, but feed.py raises an error. I'm wondering if this line should be changed to also exclude 303 status.

Happy to create a PR if needed.

In current master, there is a hardcoded path during program initialisation in main.py

https://github.com/rss2email/rss2email/blob/4ac56248e81c82affec75e4cef499437e8614208/rss2email/main.py#L186

(Reproducible with a fresh install on Termux.)

/tmp should be replaced with tempfile.gettempdir(), because on Android the /tmp is usually not available, result a initialisation failure on hardcoded path.

Example RSS

<rss version="2.0">
<channel>
<title>My RSS Today</title>
<link>http://www.example.com/</link>
<image>
<url>http://www.example.com/img/myepisodes_logo.jpg</url>
<title>www.example.com</title>
<link>http://www.example.com/</link>
</image>
<description>
This is my RSS example!
</description>
<language>en-us</language>
<item>
<guid isPermaLink="false">123456</guid>
<title>
[ rss2email ][ 123 - some more text ][ More text ][ 02-Sep-2022 ]
</title>
<link>
https://www.example.com/weirdsite
</link>
<description>
This is a description for my example
</description>
</item>
<item>
<guid isPermaLink="false">123456</guid>
<title>
[ thisisnotrss2email ][ 5689 - blabla ][blabla ][ 02-Sep-2022 ]
</title>
<link>
https://www.example.com/nicesite
</link>
<description>
More description!
</description>
</item>
</channel>
</rss>

I would like to extract "rss2email" and "thisisnotrss2email" from the item title and put it the name-format line. Is it possible to capture/split this and put this in the name?

If this is not possible, would it be possible to put the whole item title (not feed title) as the name?