3 ms·

NoPriv.py - Easy IMAP email backup to HTML archive

35 points by mdewinter 14y ago

andrewcooke 14y ago
is this a good idea? shouldn't backups be in a standard format (for the medium in question)? what if i want to import emails from this backup into another mail program / mail provider? it seems like a useful static(?) interface to email, but as a backup it's oddly restrictive. maybe it's just confusion over the use of the word "backup". but for me, a tarball of maildirs, while not readable by a human, is closer to what i expect that to mean. perhaps this could be modified to provide both?
- dbaupp 14y ago
 Not only does it generate a non-standard format, the HTML has significant problems: - It would be relatively hard to reconstruct the email with a script, e.g. the headers are formatted as a plain text with for the labels, the data left unmarked in anyway and for line breaks (no ids or classes at all). From http://sparklingnetwork.nl/nopriv/2012/12/26.html http://sparklingnetwork.nl/nopriv/2012/12/26.html: From: "Remy van Elst <relstnl@gmail.com>" To: ""backup@sparklingnetwork.nl" <backup@sparklingnetwork.nl>" - It's invalid markup[1], with the email's HTML contents inserted wholesale, including the <head> and <body> tags. - Not sure if this is a problem with NoPriv.py, or if it has occured earlier in the chain, but something has screwed up character encodings, e.g. in the URLs here: http://sparklingnetwork.nl/nopriv/2012/12/1.html http://sparklingnetwork.nl/nopriv/2012/12/1.html (The page is served with no encoding specified nor is there a <meta http-equiv> so it defaults to UTF-8, and these characters are invalid. However, even in Latin-1 or anything else, they are definitely not '=' which appears to be what they should be.) (The markup is also horrible, with strange indentations and a lot of extraneous whitespace and one-cell tables.) [1]: http://validator.w3.org/check?uri=http%3A%2F%2Fsparklingnetwork.nl%2Fnopriv%2F2012%2F12%2F26.html&doctype=HTML5 http://validator.w3.org/check?uri=http%3A%2F%2Fsparklingnetw...
 - thorum 14y ago
 Simple enough issues to fix, of course, since this project is open source and only 431 lines of Python. https://github.com/RaymiiOrg/NoPriv/blob/master/nopriv.py https://github.com/RaymiiOrg/NoPriv/blob/master/nopriv.py
- mseebach 14y ago
 It's appealing to me. I've got some Maildir (I think) archives sitting around, I've been contemplating feeding them into my GMail, but this seems like a neat and straight-forward way to do it. It's not like I need to interact with the e-mail anymore, it's at least three years out of date, but I've needed something from it a few times and it's fun to take trips down memory lane.
 - jfb 14y ago
 If you have maildirs, just use offlineimap or isync to ingest the mail into GMail.

35 points by mdewinter 14y ago

is this a good idea? shouldn't backups be in a standard format (for the medium in question)? what if i want to import emails from this backup into another mail program / mail provider? it seems like a useful static(?) interface to email, but as a backup it's oddly restrictive. maybe it's just confusion over the use of the word "backup". but for me, a tarball of maildirs, while not readable by a human, is closer to what i expect that to mean. perhaps this could be modified to provide both?

dbaupp 14y ago
Not only does it generate a non-standard format, the HTML has significant problems: - It would be relatively hard to reconstruct the email with a script, e.g. the headers are formatted as a plain text with for the labels, the data left unmarked in anyway and for line breaks (no ids or classes at all). From http://sparklingnetwork.nl/nopriv/2012/12/26.html http://sparklingnetwork.nl/nopriv/2012/12/26.html: From: "Remy van Elst <relstnl@gmail.com>" To: ""backup@sparklingnetwork.nl" <backup@sparklingnetwork.nl>" - It's invalid markup[1], with the email's HTML contents inserted wholesale, including the <head> and <body> tags. - Not sure if this is a problem with NoPriv.py, or if it has occured earlier in the chain, but something has screwed up character encodings, e.g. in the URLs here: http://sparklingnetwork.nl/nopriv/2012/12/1.html http://sparklingnetwork.nl/nopriv/2012/12/1.html (The page is served with no encoding specified nor is there a <meta http-equiv> so it defaults to UTF-8, and these characters are invalid. However, even in Latin-1 or anything else, they are definitely not '=' which appears to be what they should be.) (The markup is also horrible, with strange indentations and a lot of extraneous whitespace and one-cell tables.) [1]: http://validator.w3.org/check?uri=http%3A%2F%2Fsparklingnetwork.nl%2Fnopriv%2F2012%2F12%2F26.html&doctype=HTML5 http://validator.w3.org/check?uri=http%3A%2F%2Fsparklingnetw...
- thorum 14y ago
 Simple enough issues to fix, of course, since this project is open source and only 431 lines of Python. https://github.com/RaymiiOrg/NoPriv/blob/master/nopriv.py https://github.com/RaymiiOrg/NoPriv/blob/master/nopriv.py
mseebach 14y ago
It's appealing to me. I've got some Maildir (I think) archives sitting around, I've been contemplating feeding them into my GMail, but this seems like a neat and straight-forward way to do it. It's not like I need to interact with the e-mail anymore, it's at least three years out of date, but I've needed something from it a few times and it's fun to take trips down memory lane.
- jfb 14y ago
  If you have maildirs, just use offlineimap or isync to ingest the mail into GMail.