[Users] Which files from ~/.claws-mail should be backed up?

[George Anchev]

And which ones are not necessary for a recovery? (e.g. caches or
anything else that is temporary or is anyway retrieved during next mail
or RSS sync)

[Paul]

On Sat, 17 Feb 2018 19:52:50 +0200

Post by George Anchev
And which ones are not necessary for a recovery? (e.g. caches or
anything else that is temporary or is anyway retrieved during next
mail or RSS sync)

[George Anchev]

Thanks Paul!

[George Anchev]

Ok, I have read the whole man page.

However it doesn't describe the sub-folders of ~/.claws-mail but just
the files in the root directory. Perhaps my question should have been
more explicit as I am looking to correctly backup the whole tree
without including things which are temporary/cache.

I assume that all *.bak files are not really necessary too as most of
them seem to be identical copies of their non-*.bak version so it
hardly makes sense to backup the backup.

In tagsdb directory I see tree structures even for accounts which I
have already deleted (is that a bug?) Do I need to backup that
directory?

What about files named:

.claws_cache
.claws_mark
.mh_sequences

Can I safely ignore directories:

imapcache (I don't use "ignore thread" or "mark thread" flags)
mimetmp
newscache
tempfolder

Anything else which can (or must not) be ignored?

--
George

[George Anchev]

Can anyone help please?

My backup program also shows shows entries like:

~/.claws-mail/tagsdb/#imap/<account>/<deleted-folder-name>

Where <deleted-folder-name> are names of IMAP folders which existed
before but were deleted long ago. I don't need to backup (or have)
unnecessary data.

How can I cleanup that mess? I need a clean backup, so that if I need
to restore only correct data is there, not some temp or stale files.

[George Anchev]

Is anyone reading this thread please?

[Bob Williams]

On Wed, 7 Mar 2018 13:43:44 +0200

Post by George Anchev
Is anyone reading this thread please?

Yes, but it's hard to follow. The conversation seems to jump from Subject: to body message without warning.

[George Anchev]

Post by Bob Williams
Yes, but it's hard to follow. The conversation seems to jump from
Subject: to body message without warning.

What do you mean jump? Is it not clear from the mail archive thread? Do
I have to repeat anything? Please explain. I really need to get this
done right.

[Brad Rogers]

On Wed, 7 Mar 2018 14:15:28 +0200
George Anchev <***@anchev.net> wrote:

Hello George,

I'm not one of the developers, so I *don't* speak on their behalf.
Therefore, consider what I say to be the word of a not-so-disinterested
onlooker.

Post by George Anchev
What do you mean jump? Is it not clear from the mail archive thread? Do

Making us do the work you should be doing - using search engines to find
out what you can before asking questions here. Possibly even explaining
what steps you've taken to get to where you are.

Post by George Anchev
I have to repeat anything? Please explain. I really need to get this
done right.

Since 16 Feb, I have received 234 messages from this list, (not counting
messages from bug list, which I put in a different folder), of which 106
have been from you. Very nearly half. Some of those messages had
needless attachments: 30k+ images when a few words of text would have
done. Often you post in reply to yourself because you've forgotten to
add info, or want to correct errors in previous posts, or just want to
ask why you've not had an answer. And you've had the cheek to complain
about the 'spammy' nature of posting bug reports to the list. I'll just
say this: Pot, kettle black.

You've complained that CM isn't gmail. Well, no shit Sherlock. If you
want to use gmail's features, use gmail. Just don't expect CM to conform
to google's world view.

People will have started killfiling you by now, I'm sure.

[George Anchev]

Brad,

Post by Brad Rogers
Making us do the work you should be doing - using search engines to
find out what you can before asking questions here.

I have searched and I have not found, so I asked.

Post by Brad Rogers
Possibly even explaining what steps you've taken to get to where you
are.

I have read the full man page as suggested by Paul but unfortunately
it doesn't answer my question completely. So I have only looked at the
file structure and noticed certain undocumented names which caught my
attention, so I asked specifically about them.

Post by Brad Rogers
Since 16 Feb, I have received...

I am sorry if you have got more messages from me than you expected.
Obviously mailing lists have their cons.

[Bob Williams]

On Wed, 7 Mar 2018 14:15:28 +0200

Post by George Anchev

Post by Bob Williams
Yes, but it's hard to follow. The conversation seems to jump from
Subject: to body message without warning.

What do you mean jump? Is it not clear from the mail archive thread?
Do I have to repeat anything? Please explain. I really need to get
this done right.

Message-ID: <CAKbMyxAw9BKfx8Y1xfYdKtMYdBp-HSoJT=***@mail.gmail.com>
Message-ID: <***@kujata>

[George Anchev]

All I see that Paul replied through modifying the Subject.

[Bob Williams]

On Wed, 7 Mar 2018 20:03:40 +0200

Post by George Anchev
All I see that Paul replied through modifying the Subject.

Precisely. I expect the conversation to take place in the message body.

BTW, I think Paul might have done that to point out that your message, to which he was replying, posed one question in the Subject and another in the message body.

It sometimes means repeating the same question in both Subject and body, but that does make it easier for your audience.

HAND

Bob

[George Anchev]

Post by Bob Williams
BTW, I think Paul might have done that to point out that your
message, to which he was replying, posed one question in the Subject
and another in the message body.

In this particular case the text in the body rather clarifies the
purpose of the question. It doesn't ask about anything different.

[Mike Yetto]

Post by George Anchev

Post by Bob Williams
BTW, I think Paul might have done that to point out that your
message, to which he was replying, posed one question in the Subject
and another in the message body.

In this particular case the text in the body rather clarifies the
purpose of the question. It doesn't ask about anything different.
_______________________________________________
Users mailing list
http://lists.claws-mail.org/cgi-bin/mailman/listinfo/users

The Subject: should be a short descriptive statement about the message. The body should be able to stand on its own. In this way there is nothing lost if or when the subject is changed or the thread drifts.

This is an simple basic rule of email, mailing lists, and Usenet. The discusion belongs completely in the body.

Oh, trim and don't top post.

[George Anchev]

Post by Mike Yetto
The Subject: should be a short descriptive statement about the
message. The body should be able to stand on its own. In this way
there is nothing lost if or when the subject is changed or the thread
drifts.

Is that not the case? Or is there anything unclear in the question that
I have to ask in a different way? Or are you explaining in general?

Post by Mike Yetto
This is an simple basic rule of email, mailing lists, and Usenet. The
discusion belongs completely in the body.

I am sure Paul knows much better than me how these things work.
FWIW: This is the first time I am using a mailing list. Never touched
Usenet.

[wwp]

Hello George,

Post by George Anchev

Post by Mike Yetto
The Subject: should be a short descriptive statement about the
message. The body should be able to stand on its own. In this way
there is nothing lost if or when the subject is changed or the thread
drifts.

Is that not the case? Or is there anything unclear in the question that
I have to ask in a different way? Or are you explaining in general?

Post by Mike Yetto
This is an simple basic rule of email, mailing lists, and Usenet. The
discusion belongs completely in the body.

I am sure Paul knows much better than me how these things work.
FWIW: This is the first time I am using a mailing list. Never touched
Usenet.

Mike did summarize very well the very few common rules in mailing lists
(1), without mentioning the whole netiquette (2) or local rules (like
avoid big attachments, don't post twice the same question, avoid
political or off-topic subjects): don't communicate using the subject
only (it's just a short version of the whole question asked in body -
the discussion resides in the body), don't top-post, use prefixed
quotation (see the > chars above), trim long stuff, don't steal topics
by asking your own question into a thread that has nothing to do with
it (post a new email instead of replying to an existing one, in that
case), generally don't change subject when replying, be polite, watch
your tone, etc.

There are also norms and standardization efforts about email and other
communications ways (some RFC's you can consult online (3)) but they
are quite technical. In addition to that, since it's very hard to guess
one's tone and intentions by reading him/her (unless you know him/her
well), the use of smileys can save us flame wars, sometimes. The notion
of trolling and flooding also matters. All that usually helps in
mailing list where individualities, cultures and personal characters try
to communicate.

But, if you're new to mailing lists, (and nobody here can presume about
your use to email software) nobody should invoke common sense since it's
all about (a very little bit of common sense but a lot of) education
and learning, so be forgiven and feel free to ask what are the rules in
case of any doubt. Commonly, if you get to reply it's either because
you did something wrong (nobody wants to reply because of that), nobody
reads or nobody knows.

(1) https://en.wikipedia.org/wiki/Posting_style
(2) https://en.wikipedia.org/wiki/Netiquette
(3) https://www.ietf.org/rfc/rfc1855.txt


Regards,

[George Anchev]

...

Thanks for the added explanations. I think the initial question and the
additional ones added to it do not conflict any of the things you
explained. If you are criticizing anyone for changing the subject -
well, it wasn't me who did that. :)

I am taking a general note for not "pinging" for an answer. Still it is
fair to note that this is due to incomplete documentation and lack of
any response for 17 days. It is logical and natural that if there is no
other info source providing the answer one needs to ask.

[wwp]

Hi George,

Post by George Anchev

...

Thanks for the added explanations. I think the initial question and the
additional ones added to it do not conflict any of the things you
explained. If you are criticizing anyone for changing the subject -
well, it wasn't me who did that. :)

My intention was to inform, not to criticize. :-)

Post by George Anchev
I am taking a general note for not "pinging" for an answer. Still it is
fair to note that this is due to incomplete documentation and lack of
any response for 17 days. It is logical and natural that if there is no
other info source providing the answer one needs to ask.

Agree on all that!


Regards,

[Liz]

On Wed, 7 Mar 2018 14:15:28 +0200

Post by George Anchev

Post by Bob Williams
Yes, but it's hard to follow. The conversation seems to jump from
Subject: to body message without warning.

What do you mean jump? Is it not clear from the mail archive thread?
Do I have to repeat anything? Please explain. I really need to get
this done right.

I don't read the subject. I belong to many lists where the subject line
is not changed when it should be.

However, I read the text.
So yes, people are reading this thread.
Some like me may be tired of reading your questions. I learn by
experimentation. I know each of us learns in a different way, but we do
appreciate when someone tries to find a solution for themselves.

Liz

[George Anchev]

Post by Liz
Some like me may be tired of reading your questions. I learn by
experimentation.

Can you please share how you find through experimentation the
undocumented purpose of each file, how it is used by the program and
whether it is necessary to include it in backup?

[Brian Morrison]

On Thu, 8 Mar 2018 12:49:41 +0200

Post by George Anchev

Post by Liz
Some like me may be tired of reading your questions. I learn by
experimentation.

Can you please share how you find through experimentation the
undocumented purpose of each file, how it is used by the program and
whether it is necessary to include it in backup?

The simple answer is that the source code is there to be read but I
realise that it takes time to learn how to do that and how to find the
information you need.

When it comes to backup, just backup all of .claws-mail, you can't go
wrong that way. The extra space it takes is irrelevant in today's
computers and you won't be risking anything. Claws is a work in
progress, there are always bugs and things that aren't clear.

I've been using Claws and its antecedents for well over a decade and I
don't know it all, but I have stuck with Claws because there are no
better alternatives. Constructive criticism and questions are fine,
demanding answers is not.

[George Anchev]

Post by Brian Morrison
The simple answer is that the source code is there to be read but I
realise that it takes time to learn how to do that and how to find the
information you need.

I am glad you realize that because that would rather be an escape from
answering :)

Post by Brian Morrison
When it comes to backup, just backup all of .claws-mail, you can't go
wrong that way. The extra space it takes is irrelevant in today's
computers and you won't be risking anything. Claws is a work in
progress, there are always bugs and things that aren't clear.

It is not just a question of space but rather of order and information
collision. For example I notice the following structure exists:

~/.claws-mail/RSSyl/.../<some feed>/1
~/.claws-mail/RSSyl/.../<some feed>/2
...
~/.claws-mail/RSSyl/.../<some feed>/N

The messages seems to always be 1-N. When I delete (inside CM) certain
RSS messages which I have read and then refresh the feed - new messages
may arrive. So consider a scenario like this:

1. Your refresh an RSS feed
2. For simplicity lets suppose 3 messages arrive only (1, 2, 3)
3. You backup everything
4. On the next day your read messages 1, 2, 3 and delete them
5. Refresh RSS feed
6. This time new 5 messages arrive (1, 2, ... 5)
7. Something happens and you have to restore from backup
8. During the restore messages 1, 2, 3 get overridden with older files

The question is: how will CM handle such logical/sequence conflict?
Obviously it is even more complicated with a non-sequential mix.

Another example:

~/.claws-mail/tmp/claws-crashed

contains only a string "foo". There are also other tmp files in the
same directory. There are also other files which look as temp data but
it is not clear how CM uses them. Also as mentioned in an earlier
message CM seems not to clean up file structure for deleted IMAP
folders.

So considering all that: Why would anyone want to back up
unnecessary or volatile data which can only create conflict upon
restore? That contradicts the very essence of having a backup.

As you see - the answer is not as simple as "backup everything". After
all one of the reasons for which one chooses to use a powerful mail
client as CM is to have communication organized and handled better, not
merely piled up and randomly colliding with itself.

So I still think it is relevant to have more detailed clarifications
re. the file organization.

Post by Brian Morrison
Constructive criticism and questions are fine, demanding answers is
not.

A demand for clarity should not be automatically condemned as a caprice.
It may be an actual and reasonable necessity, just like I explained. In
software projects documentation is often improved due to such demands.
So it should not be easily refuted or marked as anti-netiquette. The
questions are quite valid and on topic.

[Jeremy Nicoll]

Post by George Anchev
It is not just a question of space but rather of order and information
~/.claws-mail/RSSyl/.../<some feed>/1
~/.claws-mail/RSSyl/.../<some feed>/2
...
~/.claws-mail/RSSyl/.../<some feed>/N
The messages seems to always be 1-N. When I delete (inside CM) certain
RSS messages which I have read and then refresh the feed - new messages
1. Your refresh an RSS feed
2. For simplicity lets suppose 3 messages arrive only (1, 2, 3)
3. You backup everything
4. On the next day your read messages 1, 2, 3 and delete them
5. Refresh RSS feed
6. This time new 5 messages arrive (1, 2, ... 5)
7. Something happens and you have to restore from backup
8. During the restore messages 1, 2, 3 get overridden with older files
The question is: how will CM handle such logical/sequence conflict?

There's no conflict. If you replace (by doing a restore) one complete set
of Claws files by another complete set, you will have restored Claws to
the state it was in when that backup was originally made.

What you must not do is pick and choose which files within a backup
you restore. It must be the whole lot.

(That's not Claws-specific. Many applications' backups require the
whole set of logically consistent files to be backed-up or restored in
one complete set.)

[George Anchev]

Post by Jeremy Nicoll
What you must not do is pick and choose which files within a backup
you restore. It must be the whole lot.

Yes, generally that is the way it should be done. However it is still
not clear how this will work for IMAP messages for example. In
particular: IMAP messages are also numbered files locally. I don't know
how it is on the server (if they have other numbers etc). So if you
delete IMAP message X locally which results in deleting it on the
server and then you restore from backup in which that same message X is
not deleted - will that re-upload the message to the server or what?

The other examples which I mentioned still remain too.

[Andreas Fink]

On Thu, 8 Mar 2018 18:41:15 +0200

Post by George Anchev

Post by Jeremy Nicoll
What you must not do is pick and choose which files within a backup
you restore. It must be the whole lot.

Yes, generally that is the way it should be done. However it is still
not clear how this will work for IMAP messages for example. In
particular: IMAP messages are also numbered files locally. I don't
know how it is on the server (if they have other numbers etc). So if
you delete IMAP message X locally which results in deleting it on the
server and then you restore from backup in which that same message X
is not deleted - will that re-upload the message to the server or
what?
The other examples which I mentioned still remain too.
_______________________________________________
Users mailing list
http://lists.claws-mail.org/cgi-bin/mailman/listinfo/users

How is your described scenario in any way different than
logging into your mail account via web frontend, delete the mails
there, and open afterwards claws-mail. Claws-mail would have in its
cache the email from a previous run and realize that it is not there on
the server anymore, so it should be deleted also locally.

A full backup of the whole folder is the most consistent thing, that
you can do, and expect it to behave in exactly the same way, as if you
haven't opened claws-mail from the backup date until today.

Cheers
Andreas

[George Anchev]

Claws-mail would have in its cache the email from a previous run and
realize that it is not there on the server anymore, so it should be
deleted also locally.

If it works this way - then that is fine. But it seems tagsdb keeps
deleted stuff. Check bug 3982.

[Andreas Fink]

On Thu, 8 Mar 2018 19:01:35 +0200

Post by George Anchev

Claws-mail would have in its cache the email from a previous run and
realize that it is not there on the server anymore, so it should be
deleted also locally.

If it works this way - then that is fine. But it seems tagsdb keeps
deleted stuff. Check bug 3982.
_______________________________________________
Users mailing list
http://lists.claws-mail.org/cgi-bin/mailman/listinfo/users

And does the existence of these files/folders stop claws-mail from
functioning, or is it just some garbage lying around on your harddisk,
and claws-mail shows the correct things to you as user.
I agree this is a bug and could be fixed, but it is unrelated to your
issue of a backup. You want a safe way for backup, and I suggested to
backup the whole folder, because this is the sanest thing you could do.

Claws-mail is not the first program that would leave some cruft on the
harddisk, and it is not the last to do so. As long as it doesn't choke
on it, I guess it's acceptable to be fixed rather later than sooner ;)

Cheers
Andreas

[George Anchev]

You are right Andreas and I have already filed a bug report about it as
you may have noticed.

--
George

[Colin Leroy]

On 08 March 2018 at 18h41, George Anchev wrote:

Hi,

So if you delete IMAP message X locally which results in deleting it
on the server

Deleting an IMAP message locally deletes it on the server first, then
the local copy. If you just rm the local copy, the mail will still be
on the server and will get re-cached to disk should you open it. With a
new number.

[George Anchev]

Thanks Colin.

Yes, I think understand the process of deletion. My
confusion is/was rather about what happens if a message deleted on the
server gets restored from backup locally, i.e if it will be re-uploaded
to the server thus creating a duplicate (one copy in inbox/folder and
another in Trash).

Similarly it is not completely clear if restoring from backup a local
cache file would prevent the proper synchronization. That depends on
the program logic. For example the following logic would prevent the
sync (pseudo code):

if (local_cache_file_exists) {
// Do not sync
} else {
// Fetch new items
// Update tags for existing items
// ...
}

--
George

[Andrej Kacian]

On Sat, 10 Mar 2018 19:35:48 +0200

Post by George Anchev
Yes, I think understand the process of deletion. My
confusion is/was rather about what happens if a message deleted on the
server gets restored from backup locally, i.e if it will be re-uploaded
to the server thus creating a duplicate (one copy in inbox/folder and
another in Trash).

No, this will not happen. The local copy is only used as a ephemeral
"mirror" of the message with the same ID on server, for purposes of
displaying the message. If there is no message with that ID on server,
Claws Mail will not even try to access the local copy.

Post by George Anchev
Similarly it is not completely clear if restoring from backup a local
cache file would prevent the proper synchronization. That depends on
the program logic. For example the following logic would prevent the
if (local_cache_file_exists) {
// Do not sync
} else {
// Fetch new items
// Update tags for existing items
// ...
}

In some cases, this can cause issues. That is what "Discard folder
cache" button in folder properties dialog is there for - it will delete
everything cached locally for that particular folder.

Regards,

[George Anchev]

Post by Andrej Kacian
In some cases, this can cause issues. That is what "Discard folder
cache" button in folder properties dialog is there for - it will
delete everything cached locally for that particular folder.

Thanks for explaining Andrej. Can you also recommend a workaround for
cleaning up of

~/.claws-mail/tagsdb/#imap/<account>/<deleted-folder-name>

until bug 3982 is fixed?

--
George

[Brian Morrison]

On Thu, 8 Mar 2018 15:07:32 +0200

Post by George Anchev

Post by Brian Morrison
The simple answer is that the source code is there to be read but I
realise that it takes time to learn how to do that and how to find
the information you need.

I am glad you realize that because that would rather be an escape from
answering :)

Well you need to realise that nothing comes for nothing, if you want
knowledge you need to invest time and effort.

Post by George Anchev

Post by Brian Morrison
When it comes to backup, just backup all of .claws-mail, you can't
go wrong that way. The extra space it takes is irrelevant in today's
computers and you won't be risking anything. Claws is a work in
progress, there are always bugs and things that aren't clear.

It is not just a question of space but rather of order and information
~/.claws-mail/RSSyl/.../<some feed>/1
~/.claws-mail/RSSyl/.../<some feed>/2
...
~/.claws-mail/RSSyl/.../<some feed>/N
The messages seems to always be 1-N. When I delete (inside CM) certain
RSS messages which I have read and then refresh the feed - new
1. Your refresh an RSS feed
2. For simplicity lets suppose 3 messages arrive only (1, 2, 3)
3. You backup everything
4. On the next day your read messages 1, 2, 3 and delete them
5. Refresh RSS feed
6. This time new 5 messages arrive (1, 2, ... 5)
7. Something happens and you have to restore from backup
8. During the restore messages 1, 2, 3 get overridden with older files
The question is: how will CM handle such logical/sequence conflict?

It won't. Everything will be fucked up. You are asking for a database,
Claws doesn't have one. RSS feeds are viewed as volatile and
non-protected.

Post by George Anchev
Obviously it is even more complicated with a non-sequential mix.
~/.claws-mail/tmp/claws-crashed
contains only a string "foo". There are also other tmp files in the
same directory. There are also other files which look as temp data but
it is not clear how CM uses them. Also as mentioned in an earlier
message CM seems not to clean up file structure for deleted IMAP
folders.

Temp folders are never intended to be permanent, if it says "temp" or
"tmp" then ignore that folder. If you restore from a backup then do
that in a new directory and sanitise it before you copy it over your
real .claws-mail, always have copies of the current file tree in case
you make a mistake.

Post by George Anchev
So considering all that: Why would anyone want to back up
unnecessary or volatile data which can only create conflict upon
restore? That contradicts the very essence of having a backup.
As you see - the answer is not as simple as "backup everything". After
all one of the reasons for which one chooses to use a powerful mail
client as CM is to have communication organized and handled better,
not merely piled up and randomly colliding with itself.
So I still think it is relevant to have more detailed clarifications
re. the file organization.

Free Software is not expected to be perfect, when there is something
wrong that bugs you then it is reasonable to expect you to fix it
either by submitting code or by offering improved documentation. If it
breaks, you get to keep all the pieces.

Post by George Anchev

Post by Brian Morrison
Constructive criticism and questions are fine, demanding answers is
not.

A demand for clarity should not be automatically condemned as a
caprice. It may be an actual and reasonable necessity, just like I
explained. In software projects documentation is often improved due
to such demands. So it should not be easily refuted or marked as
anti-netiquette. The questions are quite valid and on topic.

I've tried to be helpful and have replied to you several times with
assistance and suggestions. I don't have to, no one pays me to be here
or read anything anyone writes. I don't want to be rude but as you can
see some people have already been provoked into telling you you're
being too demanding. I think you need to consider what you can
reasonably expect, especially because you are clearly very demanding of
your email experience and in terms of security and privacy. None of
those things are easy or come for nothing.

HTH

[Michael]

As a sidenote (no need to follow-up):

When writing to developers (like, mailing lists or bug reports or personal) i try to point out my problem and also tell how i expect it to work, instead (which always is part of a good bug report) and if i have that time, i try to find out how it could work 'internally' and what part might be difficult to implement.

Sometimes this leads me to realize that i am very demanding (implying a large refactoring or redesign of the whole thing) and then i change the problem to some core element that could-be-achieved-rather-easily even if it does not fully solve my full needs.
This often results in re-wording the subject to something more precise, too.

It's a little precompiling of the problem, trying to see it from the developers point of view.

(I admit that i still often suck with spontaneous tatter, for many human reasons, don't nail me down on that)

[George Anchev]

Post by Brian Morrison
Well you need to realise that nothing comes for nothing, if you want
knowledge you need to invest time and effort.

I already explained that I have read the man page as suggested. If you
suggest that one should become an electronics engineer if the TV manual
is incomplete that is hardly the best suggestion :)

Post by Brian Morrison
It won't. Everything will be fucked up. You are asking for a database,
Claws doesn't have one. RSS feeds are viewed as volatile and
non-protected.

Which contradicts the whole suggestion to backup everything.

Post by Brian Morrison
Temp folders are never intended to be permanent, if it says "temp" or
"tmp" then ignore that folder. If you restore from a backup then do
that in a new directory and sanitise it before you copy it over your
real .claws-mail, always have copies of the current file tree in case
you make a mistake.

Sure, this sanitizing is a common sense. But although file names can
imply they cannot reveal the complete role of the file or its time span
in the functionality of a program. Also as I mentioned there are
leftovers which are not named "temp".

Post by Brian Morrison
I've tried to be helpful and have replied to you several times with
assistance and suggestions...

Thanks - for all the times. Unfortunately the question still remains
and it seems that only a developer who knows the intricacies of the
program can shed light on the subject. There is hardly any need to
accuse the questioner for being too interested just because others
also don't have the info he is asking for. Surely there are people
who have it, so my question is to whoever this may apply. Peace.

[Steve Litt]

Post by George Anchev

Post by Brian Morrison
Well you need to realise that nothing comes for nothing, if you want
knowledge you need to invest time and effort.

I already explained that I have read the man page as suggested. If you
suggest that one should become an electronics engineer if the TV
manual is incomplete that is hardly the best suggestion :)

The best suggestion is to do what I just did: Filter all email from
George Anchev to /dev/null. Life's too short to read 5 emails a day
from a guy criticizing software given them for free, stridently asking
for help, and then arguing with help suggestions instead of trying them.

SteveT

[George Anchev]

Life's too short to read 5 emails a day from a guy criticizing
software given them for free, stridently asking for help, and then
arguing with help suggestions instead of trying them.

Actually if you really read those emails you would see that I am not
the one who criticizes.

[Michael A. Yetto]

On Thu, 8 Mar 2018 18:50:17 +0200

Unfortunately the question still remains and it seems that only a
developer who knows the intricacies of theprogram can shed light on
the subject.

Offer to pay that developer's customary consulting fee and you might
a more complete answer.

A frequent reply to a storm of questions such as your is "Do your own
homework." Fortunately, you have been treated much better than the
privileged students who typically try this.

Mike Yetto

[George Anchev]

Post by Michael A. Yetto
Offer to pay that developer's customary consulting fee and you might
a more complete answer.

Fair enough. How much do I have to pay and to who in particular so that
I know which files do (not) need backup?

[Michael A. Yetto]

On Thu, 8 Mar 2018 20:58:19 +0200

Post by George Anchev

Post by Michael A. Yetto
Offer to pay that developer's customary consulting fee and you might
a more complete answer.

Fair enough. How much do I have to pay and to who in particular so that
I know which files do (not) need backup?

Publish an RFC.

Mike Yetto

[Dave Howorth]

On Thu, 8 Mar 2018 20:58:19 +0200

Post by George Anchev

Post by Michael A. Yetto
Offer to pay that developer's customary consulting fee and you might
a more complete answer.

Fair enough. How much do I have to pay and to who in particular so
that I know which files do (not) need backup?

Full marks to Andrej for his recent intervention, and I too have
noticed that your questions are well-intentioned although your way of
asking sometimes appears thoughtless.

If we look at your question above,and with my perspective as a
developer although NOT a Claws developer, I would suggest you first need
to ask yourself some further questions and answer them.

Firstly, is it a serious question?
Secondly what level of accuracy and completeness would you want to pay
for?
And thirdly, what warranty would you expect?

It seems quite likely to me that nobody has ever thought about the
notion of an exact backup. It seems much more likely that developers'
concerns have centred on things like not crashing if there is an
inconsistency between the claws file collection and the various remote
message stores for any reason at all, of which an inconsistent backup
would be just one possible case. If they had thought about the notion,
I'd have expected them to step forward by now, especially in response
to your present question.

So answering your question is going to involve somebody spelunking the
code, trying to determine the result of various situations. And that
won't be easy, given that the code (probably) wasn't written with that
as an identified requirement. It will likely involve a lot of code
reading and a bunch of experiments, such as making a whole load more
tests for claws. And even after they've done that, they won't be sure
they've caught all the possible cases, and they won't want to be
committed forevermore to fixing and documenting whatever new use case
you come up with. They certainly won't want to be liable for the
possibility of any consequential damages.

Set against that we notice that nobody, neither a developer nor any
other user, has apparently ever had a problem by just taking a full
backup and restoring from it. Sure they may have had some local
breakage of some message stream or lost a few messages, but things like
that can usually be recovered out of band - by going to a web archive,
by asking senders to resend etc. So the community seems to have
historically judged that there's little value in answering your
question.

So you'd be asking a developer to work on something essentially of use
just to you, in preference to work that would benefit everybody. That's
not an attractive proposition. Further, there's the risk of scope
growth and ongoing support issues (depending on your answers to my
supplementary questions). So anybody standing up to offer to answer
your question is going to want to be well paid for their time on what's
likely to prove a tedious and thankless task. Are you up for spending
thousands of dollars to answer your question?

An alternative possibility might actually be for you to answer the
question yourself. If you were prepared to write tests and/or learn to
read the code in order to answer your question, and then write it up
for the documentation, I'm sure the developers would be more than happy
to help you learn how to do it. It seems to me that you have the
necessary attitude and abilities. It's not rocket science.

[d***@gmail.com]

On Thu, 8 Mar 2018 18:50:17 +0200

Post by George Anchev
Thanks - for all the times. Unfortunately the question still remains
and it seems that only a developer who knows the intricacies of the
program can shed light on the subject. There is hardly any need to
accuse the questioner for being too interested just because others
also don't have the info he is asking for. Surely there are people
who have it, so my question is to whoever this may apply. Peace.

If you can not get the answers you are looking for which leaves you
uncomfortable with what Claws provides, then it probably isn't for you.

I hope you can find another free client that suits all your needs and a
developer that is willing to be your personal consultant. I
heard the Gmail web interface is peachy. Cheers!

[Andrej Kacian]

Ok, can we all please stop dogpiling on George now? Sure, he ruffled a
few feathers with some aspects of how he communicates on this list (I
too am grinding my teeth everytime he asks a question in subject and
only says "Subject" in the message body :) ), but I believe he is in
earnest and not acting with any malice.

Plus, like any other new user with a fresh perspective, he often asks
very pertinent (if sometimes uncomfortable, at least for the developers)
questions, and notices bugs that developers or long-time users do not
even notice, or take for granted. That alone is worth it, in my view.

George, lot of the stuff you mention is something that could use some
work, be it documentation that we know is often lacking, or incorrect
behavior in the app itself. Claws Mail is a rather big program with lot
of layers of code, both newer and older, and there are only a few active
developers left, with very limited time, and big TODO lists each.

By all means, keep asking questions, but be aware that sometimes the
answers won't be exactly what you need or want. There are hidden
bugs that need uncovering, but there are also features which have been
the way they are for many years for a good reason.

Anyway, I wrote a bit more than I initially set out to. Basically, we
all discharged some of the tension, but let's dial it down now,
please. :)

Regards,

[Paul]

On Thu, 8 Mar 2018 20:45:45 +0100

Post by Andrej Kacian
By all means, keep asking questions,

but ask them in the message body and not in the subject, yes? :)

with regards

Paul

[George Anchev]

Thank you guys. Surely I am not ready to spend thousands of dollars for
anything like that. I don't feel it is an utterly complicated question
and surely a developer who knows how the program works and how its
files are organized should be able to answer instantly, without testing any
complicated scenarios etc. E.g. what are the "cache" and "tmp"
files/dirs used for, if they need to be backed up etc.

Post by Paul
but ask them in the message body and not in the subject, yes? :)

Can you please explain what would be the proper form to ask this current
question?

I think the subject is the proper place to put the question shortly
because it is the first thing which one sees and based on that one
can decide whether to read the body or not. Then the body can contain
details. In some cases I put in the body just "subject" to avoid
repetition. It is more time efficient, wouldn't you agree? Or would you
insist that I repeat the subject in the body too?

[Michael A. Yetto]

On Fri, 9 Mar 2018 00:55:24 +0200

Post by George Anchev
Thank you guys. Surely I am not ready to spend thousands of dollars for
anything like that. I don't feel it is an utterly complicated question
and surely a developer who knows how the program works and how its
files are organized should be able to answer instantly, without
testing any complicated scenarios etc. E.g. what are the "cache" and
"tmp" files/dirs used for, if they need to be backed up etc.

Post by Paul
but ask them in the message body and not in the subject, yes? :)

Can you please explain what would be the proper form to ask this
current question?
I think the subject is the proper place to put the question shortly
because it is the first thing which one sees and based on that one
can decide whether to read the body or not. Then the body can contain
details. In some cases I put in the body just "subject" to avoid
repetition. It is more time efficient, wouldn't you agree? Or would you
insist that I repeat the subject in the body too?

The Subject: is just for an eye-catcher or headline if you prefer. The
body of the message should be complete. That is, it should not depend
on a changeable Subject:, it should have enough information to explain
the problem without being to long to read, it should define what was
expected of the program, and it should enumerate all that occurred
instead.

Any message that consists in its entirety of the word "Subject" is very
likely to be ignored, possibly even by someone who is familiar with the
problem and a solution.

Mike Yetto

[d***@gmail.com]

To take the OP's needs and desires in a related direction that are not
addressed by Claws, perhaps googling "database-driven mail user-agent"
would lead one to something more promising. Investigating the first
item may be quite informative.

[Bob Williams]

On Fri, 9 Mar 2018 00:55:24 +0200

Post by George Anchev
I think the subject is the proper place to put the question shortly
because it is the first thing which one sees and based on that one
can decide whether to read the body or not. Then the body can contain
details.

Correct

Post by George Anchev
In some cases I put in the body just "subject" to avoid
repetition. It is more time efficient, wouldn't you agree? Or would
you insist that I repeat the subject in the body too?

It may be more efficient for you, but not for your readers.

Good luck

Bob

[George Anchev]

Post by Bob Williams
It may be more efficient for you, but not for your readers.

Are you saying that people would rather prefer to read the same thing
twice? To avoid any further confusion - can you please give an example
how you would formulate the current question?

Perhaps it would be good to have short guidelines about "How to post in
mailing CM's lists" if the messages must be strictly structured up to a
certain protocol.

[Paul]

On Fri, 9 Mar 2018 14:11:28 +0200

Post by George Anchev
Are you saying that people would rather prefer to read the same
thing twice? To avoid any further confusion - can you please give
an example how you would formulate the current question?

E.g.
Subject: backing-up
Body: Which files from ~/.claws-mail should be backed up? (+ thanks
in advance, etc, etc)

Post by George Anchev
Perhaps it would be good to have short guidelines about "How to
post in mailing CM's lists" if the messages must be strictly
structured up to a certain protocol.

There's nothing particularly special about this list when compared to
most lists out there.

with regards

Paul

[George Anchev]

Post by Paul
Subject: backing-up
Body: Which files from ~/.claws-mail should be backed up? (+ thanks
in advance, etc, etc)

Thanks.

Post by Paul
There's nothing particularly special about this list when compared to
most lists out there.

In such case: Is there a particular establishment for lists as a whole?
I have seen all kinds of craziness on mailing list archives.

Sorry to ask so much about that but I don't want any other future
outrage regarding my questions so I am trying to learn not to be a
pain. Personally I communicate by email with all kinds of people and so
far nobody has complained, on the contrary. Logically it makes one
think that this (or any other) mailing lists may be special in a way.

[wwp]

Hello George,

Post by George Anchev

Post by Bob Williams
It may be more efficient for you, but not for your readers.

Are you saying that people would rather prefer to read the same thing
twice? To avoid any further confusion - can you please give an example
how you would formulate the current question?
Perhaps it would be good to have short guidelines about "How to post in
mailing CM's lists" if the messages must be strictly structured up to a
certain protocol.

I started writing this email upper case, but remembered that this was
not so important, no need to lose my temper. You've been posted
netiquette and common rules in previous emails, I guess you didn't
spend a minute reading all that nor trying to understand what everyone
here is telling you about the subject. If you're too lazy to make the
effort to put your question in the body and a summary of the request in
subject, it's either that you don't want to follow common rules, or you
want to do differently. Then do it, but you'll be ignored, if not
banned. Starting from now, I will blacklist you locally, had enough of
your dump crap, I'm losing my time reading or answering you.


Regards,

[Ricardo Mones]

On Fri, Mar 09, 2018 at 02:11:28PM +0200, George Anchev wrote:
[
]

Post by George Anchev
Perhaps it would be good to have short guidelines about "How to post in
mailing CM's lists" if the messages must be strictly structured up to a
certain protocol.

As Paul said there's nothing special on this list, but since it seems
you like to read things you can start reading RFC-1855's § 3.0 titled
“One-to-Many Communication (Mailing Lists, NetNews)”¹. Reading the rest
of the document won't harm either.

regards,

¹ https://www.rfc-editor.org/info/rfc1855

[George Anchev]

Thanks Ricardo.

I have read all the info right after these links were previously
shared. Although that RFC is not a strict standard I apologize once
again if any of my posts were not shaped properly or sounded too
demanding. It was never my intention to offend anyone.

--
George

[Michael A. Yetto]

On Fri, 9 Mar 2018 14:11:28 +0200

Post by George Anchev
Are you saying that people would rather prefer to read the same thing
twice? To avoid any further confusion - can you please give an example
how you would formulate the current question?

This has been around for a while and pertains to mailing lists as well
as Usenet or any othe help forum.

<http://www.catb.org/~esr/faqs/smart-questions.html>
Now you know.

Mike "and knowing is half the battle" Yetto

[Olaf Hering]

Am Sat, 17 Feb 2018 19:52:50 +0200

Post by George Anchev
And which ones are not necessary for a recovery? (e.g. caches

.claws-mail/imapcache/ is required if "ignore thread" or "mark thread" flags are used.

Olaf