Stefan Krauth schrieb:
>> No it connects to remote rsyncd directly. You have to add -rsh=ssh than it
>> connects through ssh.
>>
>
> Sorry, double dash. --rsa=ssh
>
> You can also set env variable RSYNC_CONNECT_PROG. Please check the manpage.
> There is everything explained (some with examples).
>
> Best Regards,
> Stefan
>
Ah... Interesting! I did not know that they had rsync installed at my
site, but actually they have. I always assumed (incorrectly) that my
local rsync automagically uses ssh.

Greets,
Christian

On Thu, 8 Oct 2009 11:15:47 +1100
Adrian Tritschler <Adrian.Tritschler@...> wrote:

> 2009/10/1 Charles Curley <charlescurley@...>:

> >
> > Here is what I use in my make file to use ssh and exclude
> > directories and files that I don't want on the server. This should
> > be all one line. It assumes you are running it from the root of the
> > web site.
> >
> > rsync -va --exclude '*~' --exclude Makefile --exclude "*.txt"
> > --exclude "*.def" --exclude .git --exclude .gitignore --exclude
> > blog/blog.conf --exclude blog/tidy.log --exclude blog/examples
> > --exclude blog/cache --exclude blog/parts --exclude blog/data
> > --exclude blog/templates --exclude blog/cache --delete
> > --delete-excluded -e ssh . <server fqnd>:<path>
>
> With that many excludes I'd suggest putting them in a separate file
> and using rsync's
> "--exclude-from=FILE" switch.

Good idea, and it's been on my list of Things To Do for a while.

--

Charles Curley                  /"\    ASCII Ribbon Campaign
Looking for fine software       \ /    Respect for open standards
and/or writing?                  X     No HTML/RTF in email
http://www.charlescurley.com    / \    No M$ Word docs in email

Key fingerprint = CE5C 6645 A45A 64E4 94C0  809C FFF6 4C48 4ECD DFDB

2009/10/1 Charles Curley <charlescurley@...>:
> On Wed, 30 Sep 2009 19:32:00 +0200
> Christian Herenz <herenz@...> wrote:
>
>> wizumwalt schrieb:
>> >> This works cool via ssh / sftp, cause at my site they dont allow
>> >> ftp acsess!
>> >
>> > I'm not exactly clear on how to do this with ssh. I normally enter
>> > something like "ssh -p 5555 -l me aaa.bbb.ccc.ddd" and then I have
>> > to cd a few directories down. How is this done withing the
>> > blog.conf file?
>> >
>> >
>> havent you read my mail?
>
> I haven't. I don't see it on the list.
>
>>
>> put this in your blog.conf:
>>
>> BLOG_PUBLISH_CMD="rsync -av --delete <blogdir>
>> user@...:./blogdirectory/"
>>
>> <blogdir> is your local directory
>> user is your username
>> bloghost.com is me aaa.bbb.ccc.ddd in your example
>> ./blogdirectory/ is the directory you normaly cd to...
>
> Of course, that won't use ssh.
>
> Here is what I use in my make file to use ssh and exclude directories
> and files that I don't want on the server. This should be all one line.
> It assumes you are running it from the root of the web site.
>
> rsync -va --exclude '*~' --exclude Makefile --exclude "*.txt" --exclude
> "*.def" --exclude .git --exclude .gitignore --exclude blog/blog.conf
> --exclude blog/tidy.log --exclude blog/examples --exclude blog/cache
> --exclude blog/parts --exclude blog/data --exclude blog/templates
> --exclude blog/cache --delete --delete-excluded -e ssh .
> <server fqnd>:<path>

With that many excludes I'd suggest putting them in a separate file
and using rsync's
"--exclude-from=FILE" switch.

Man pages are wonderful things.

--
Adrian

I tried to use Haloscan with nanoblogger as this is the only detailed
instruction available. Unfortunately Haloscan has been bought by JS-Kit so there
is no way to get a Haloscan account anymore.

From JS-Kit I got this code:

Echo:
<div class="js-kit-comments" permalink=""></div><script
src="http://js-kit.com/for/blog.domain.com/comments.js"></script>

Navigator:
<div class="js-kit-top" style="width: 300px" title="Easy Site
Navigator"></div><script
src="http://js-kit.com/for/blog.domain.com/top.js"></script>

That looks to me completely different than the examples on
http://nanoblogger.sourceforge.net/articles/nanoblogger/install_haloscan/

Does anyone know how to modify the the "install_haloscan" instructions so they
fit the JS-Kit Javascript? Otherwise I'll give up on integrating comments with
nanoblogger.

Thorsten

Hello,

can someone give me a (preferrably) detailed step-by-step instruction how to
make comments work with nanoblogger? I checked out  NanoBlogger Comments,
CGIComment, blogkomm.

NanoBlogger Comments requires a dedicated SQL server so this is not an option. 
CGIComment hasn't been maintained, so blogkomm would probably the best choice.

S-Kit and Haloscan.com are paid services and as the blog is on an intranet-site
I would rather not use those.

I tried integrating blogkomm according to
http://www.fiatjustitia.net/unblahg/index.php?itemid=19 and
http://acn192.csie.ncu.edu.tw/~clark/nblog/archives/2007-11.php

but I was not successful.

* Stefan Krauth schrieb:
> No it connects to remote rsyncd directly. You have to add -rsh=ssh than it
> connects through ssh.

Sorry, double dash. --rsa=ssh

You can also set env variable RSYNC_CONNECT_PROG. Please check the manpage.
There is everything explained (some with examples).

Best Regards,
Stefan

* Christian Herenz schrieb:
> Charles Curley schrieb:
> >
> >> BLOG_PUBLISH_CMD="rsync -av --delete <blogdir>
> >> user@...:./blogdirectory/"
> >>
> >> <blogdir> is your local directory
> >> user is your username
> >> bloghost.com is me aaa.bbb.ccc.ddd in your example
> >> ./blogdirectory/ is the directory you normaly cd to...
> >
> > Of course, that won't use ssh.
> >
> /It does! I have only ssh-access to my site.
>
No it connects to remote rsyncd directly. You have to add -rsh=ssh than it
connects through ssh.

Best Regards,
Stefan

Charles Curley schrieb:
>
>> BLOG_PUBLISH_CMD="rsync -av --delete <blogdir>
>> user@...:./blogdirectory/"
>>
>> <blogdir> is your local directory
>> user is your username
>> bloghost.com is me aaa.bbb.ccc.ddd in your example
>> ./blogdirectory/ is the directory you normaly cd to...
>>
>
> Of course, that won't use ssh.
>
>
>
/It does! I have only ssh-access to my site.

Greetz,
Christian
/

Charles Curley schrieb:
> Emacs lets you specify the mode(s) to use for a given file by putting a
> bit of magic text at the top of the file. For html-helper-mode, add
> this:
>
> ; -*- html-helper -*-
>
> to the top of templates/entry.metadata and templates/file.metadata.
> That will make Emacs automatically use html-helper-mode in entries and
> other files when you visit them. Of course, you'll need
> html-helper-mode. Or some other suitable mode.
>
> Are there other files where that trick would be useful? A similar trick
> for vi?
>
>
a) I dont use html-mode, but markdown-mode for writing my entries.
b) I added the following lines to my .emacs

     ;; open new nanoblogger entries automatically in markdown-mode
     (setq auto-mode-alist
        (cons '("\\nb_edit-newentry-.*\\.txt" . markdown-mode)
auto-mode-alist))

c) similar trick for vi? :q!

:-=)

On Wed, 30 Sep 2009 19:32:00 +0200
Christian Herenz <herenz@...> wrote:

> wizumwalt schrieb:
> >> This works cool via ssh / sftp, cause at my site they dont allow
> >> ftp acsess!
> >
> > I'm not exactly clear on how to do this with ssh. I normally enter
> > something like "ssh -p 5555 -l me aaa.bbb.ccc.ddd" and then I have
> > to cd a few directories down. How is this done withing the
> > blog.conf file?
> >
> >
> havent you read my mail?

I haven't. I don't see it on the list.

>
> put this in your blog.conf:
>
> BLOG_PUBLISH_CMD="rsync -av --delete <blogdir>
> user@...:./blogdirectory/"
>
> <blogdir> is your local directory
> user is your username
> bloghost.com is me aaa.bbb.ccc.ddd in your example
> ./blogdirectory/ is the directory you normaly cd to...

Of course, that won't use ssh.

Here is what I use in my make file to use ssh and exclude directories
and files that I don't want on the server. This should be all one line.
It assumes you are running it from the root of the web site.

rsync -va --exclude '*~' --exclude Makefile --exclude "*.txt" --exclude
"*.def" --exclude .git --exclude .gitignore --exclude blog/blog.conf
--exclude blog/tidy.log --exclude blog/examples --exclude blog/cache
--exclude blog/parts --exclude blog/data --exclude blog/templates
--exclude blog/cache --delete --delete-excluded -e ssh .
<server fqnd>:<path>

--

Charles Curley                  /"\    ASCII Ribbon Campaign
Looking for fine software       \ /    Respect for open standards
and/or writing?                  X     No HTML/RTF in email
http://www.charlescurley.com    / \    No M$ Word docs in email

Key fingerprint = CE5C 6645 A45A 64E4 94C0  809C FFF6 4C48 4ECD DFDB

wizumwalt schrieb:
>> This works cool via ssh / sftp, cause at my site they dont allow ftp acsess!
>>
>
> I'm not exactly clear on how to do this with ssh. I normally enter something
like "ssh -p 5555 -l me aaa.bbb.ccc.ddd" and then I have to cd a few directories
down. How is this done withing the blog.conf file?
>
>
havent you read my mail?

put this in your blog.conf:

BLOG_PUBLISH_CMD="rsync -av --delete <blogdir>
user@...:./blogdirectory/"

<blogdir> is your local directory
user is your username
bloghost.com is me aaa.bbb.ccc.ddd in your example
./blogdirectory/ is the directory you normaly cd to...

What is the problem with carefully reading others mails?

Greets,
Christian

On Tue, 25 Aug 2009 20:57:38 -0700 (PDT)
"Kevin W." <un1xt3r@...> wrote:

> Try the following as a guideline.
>
> non-essentials: cache/ parts/ tidy.log
> essentials: data/ plugins/ templates/ drafts/ blog.conf
> website: archives/ articles/ images/ styles/ index.html rss.xml etc.

Thanks. I finally got a chance to do some experimenting. Here's what I
came up with.

My blog is in the repo, in a directory called (with great imagination)
"blog". git does not track directories per se, so I added files, like
so:

blog/articles/*.txt
blog/data/*.txt
blog/data/*.db
blog/images/*
blog/styles/*
blog/templates/*.htm
blog/templates/*.metadata

You don't need the subdirectories under blog/articles as nb will create
them for you at run time.

I was perhaps more careful than I needed to be to avoid adding backup
files (*~).

The first time you run "nb update all" in a new clone, nb will remind
you to "mkdir cache parts". git doesn't track empty directories, and I
thought that making them once was less obnoxious than creating the
directories and then a dummy empty file in each of them.

You should set git to ignore the following:

# nanoblogger goodies
atom.xml
rss.xml
tidy.log
blog/cache

Which I did in .gitignore, not in .git/info/exclude. Depending on how
your repo is set up you may have to propagate those manually.

>
> >
> >From: Charles Curley <charlescurley@...>
> >To: nanoblogger@yahoogroups.com
> >Sent: Tuesday, August 25, 2009 9:32:58 PM
> >Subject: [nanoblogger] NB and Revision Control Systems?
> >
> >
> >I'd like to commit the source of my NB blog to an RCS, in this case
> >>git. Has anyone figured out which files to add to the repository and
> >>which to exclude?
> >
> >>Thank you.
> >
> >>--
> >
> >>Charles Curley                  /"\    ASCII Ribbon Campaign
> >>Looking for fine software       \ /    Respect for open standards
> >>and/or writing?                  X     No HTML/RTF in email
> >http://www.charlescurley.com / \    No M$ Word docs in email
> >
> >>Key fingerprint = CE5C 6645 A45A 64E4 94C0  809C FFF6 4C48 4ECD DFDB
> >
> > > >
>
>
>


--

Charles Curley                  /"\    ASCII Ribbon Campaign
Looking for fine software       \ /    Respect for open standards
and/or writing?                  X     No HTML/RTF in email
http://www.charlescurley.com    / \    No M$ Word docs in email

Key fingerprint = CE5C 6645 A45A 64E4 94C0  809C FFF6 4C48 4ECD DFDB

On Mon, 07 Sep 2009 23:21:46 -0000
"wizumwalt" <wizumwalt@...> wrote:

> > This works cool via ssh / sftp, cause at my site they dont allow
> > ftp acsess!
>
> I'm not exactly clear on how to do this with ssh. I normally enter
> something like "ssh -p 5555 -l me aaa.bbb.ccc.ddd" and then I have to
> cd a few directories down. How is this done withing the blog.conf
> file?

Look at scp. Even better, look at rsync, particularly the -e option.

--

Charles Curley                  /"\    ASCII Ribbon Campaign
Looking for fine software       \ /    Respect for open standards
and/or writing?                  X     No HTML/RTF in email
http://www.charlescurley.com    / \    No M$ Word docs in email

Key fingerprint = CE5C 6645 A45A 64E4 94C0  809C FFF6 4C48 4ECD DFDB

> This works cool via ssh / sftp, cause at my site they dont allow ftp acsess!

I'm not exactly clear on how to do this with ssh. I normally enter something
like "ssh -p 5555 -l me aaa.bbb.ccc.ddd" and then I have to cd a few directories
down. How is this done withing the blog.conf file?

Hey all,

I really like working w/ NB from the command line, but I'm probably using it the
wrong way. Not so much as for blogging, rather, trying to use it more like a
CMS. So I have the following questions if I may ...

Q1. Can an entry be placed like a "sticky", so that it will always display at
the top (or in
some order) of the main page?

Q2. Any way to file an entry under a category/tag and not have it display on the
main page?

Any help much appreciated.

On Sun, 06 Sep 2009 07:06:44 -0000
"wizumwalt" <wizumwalt@...> wrote:

> One more dumb question if I may ...
>
> I don't see how links are added ... but inside the index.html file
> there is a comment aobut adding links to the div class "side". Are
> links just manually added to the index.html file?

nb rebuilds index.html every time you run it to change an article or
entry (and probably other times as well). Take a look at
templates/main_links.htm.

--

Charles Curley                  /"\    ASCII Ribbon Campaign
Looking for fine software       \ /    Respect for open standards
and/or writing?                  X     No HTML/RTF in email
http://www.charlescurley.com    / \    No M$ Word docs in email

Key fingerprint = CE5C 6645 A45A 64E4 94C0  809C FFF6 4C48 4ECD DFDB

One more dumb question if I may ...

I don't see how links are added ... but inside the index.html file there is a
comment aobut adding links to the div class "side". Are links just manually
added to the index.html file?

On Fri, 4 Sep 2009 11:19:58 -0600
Charles Curley <charlescurley@...> wrote:

> I have some articles in my blog (http://www.charlescurley.com/blog).
> Since they don't get into the RSS feed, I also made a short entry
> describing and linking to them. The source looks like so:
>
>   <li><a href="articles/free_software_and_freedom/index.html">Free
>   Software and Freedom</a>: the connection between free software and
>   freedom.</li>
>
> The relative links are fine for the main page. But in the archived
> version of the entry, the relative links are broken. Is there a way to
> automate fixing those links to work correctly?
>

Ask and ye shall receive: use %base_url%, like so:

   <li><a
   href="%base_url%articles/free_software_and_freedom/index.html">Free
   Software and Freedom</a>: the connection between free software and
   freedom.</li>

Thanks for reading, anyway.

--

Charles Curley                  /"\    ASCII Ribbon Campaign
Looking for fine software       \ /    Respect for open standards
and/or writing?                  X     No HTML/RTF in email
http://www.charlescurley.com    / \    No M$ Word docs in email

Key fingerprint = CE5C 6645 A45A 64E4 94C0  809C FFF6 4C48 4ECD DFDB

I have some articles in my blog (http://www.charlescurley.com/blog).
Since they don't get into the RSS feed, I also made a short entry
describing and linking to them. The source looks like so:

   <li><a href="articles/free_software_and_freedom/index.html">Free
   Software and Freedom</a>: the connection between free software and
   freedom.</li>

The relative links are fine for the main page. But in the archived
version of the entry, the relative links are broken. Is there a way to
automate fixing those links to work correctly?

--

Charles Curley                  /"\    ASCII Ribbon Campaign
Looking for fine software       \ /    Respect for open standards
and/or writing?                  X     No HTML/RTF in email
http://www.charlescurley.com    / \    No M$ Word docs in email

Key fingerprint = CE5C 6645 A45A 64E4 94C0  809C FFF6 4C48 4ECD DFDB

I'd like to commit the source of my NB blog to an RCS, in this case
git. Has anyone figured out which files to add to the repository and
which to exclude?

Thank you.

--

Charles Curley                  /"\    ASCII Ribbon Campaign
Looking for fine software       \ /    Respect for open standards
and/or writing?                  X     No HTML/RTF in email
http://www.charlescurley.com    / \    No M$ Word docs in email

Key fingerprint = CE5C 6645 A45A 64E4 94C0  809C FFF6 4C48 4ECD DFDB

Emacs lets you specify the mode(s) to use for a given file by putting a
bit of magic text at the top of the file. For html-helper-mode, add
this:

; -*- html-helper -*-

to the top of templates/entry.metadata and templates/file.metadata.
That will make Emacs automatically use html-helper-mode in entries and
other files when you visit them. Of course, you'll need
html-helper-mode. Or some other suitable mode.

Are there other files where that trick would be useful? A similar trick
for vi?

--

Charles Curley                  /"\    ASCII Ribbon Campaign
Looking for fine software       \ /    Respect for open standards
and/or writing?                  X     No HTML/RTF in email
http://www.charlescurley.com    / \    No M$ Word docs in email

Key fingerprint = CE5C 6645 A45A 64E4 94C0  809C FFF6 4C48 4ECD DFDB

I set up a union script that ran after I created an entry locally that uploaded
the new stuff to my remote server.



1. remote NB sites?
     Posted by: "wizumwalt" wizumwalt@... wizumwalt
     Date: Sat Aug 1, 2009 9:06 pm ((PDT))

Just learned about NB tonight, and I really like it so I've installed it on my
web host providers server. But I'm curious to know how you guys do posts w/ NB.
Currently, I ssh to my web host and execute my ./nb from there. But I'm sure
there's a better way, like to do it locally, then update my remote server site?

How do you guys do this?

wizumwalt schrieb:
> Just learned about NB tonight, and I really like it so I've installed it on my
web host providers server. But I'm curious to know how you guys do posts w/ NB.
Currently, I ssh to my web host and execute my ./nb from there. But I'm sure
there's a better way, like to do it locally, then update my remote server site?
>
> How do you guys do this?
>
>
I do it locally and then publish it via rsync to my webspace. For this I
have following line in my blog.conf:
BLOG_PUBLISH_CMD="rsync -av --delete <blogdir>
user@...:./blogdirectory/"

This works cool via ssh / sftp, cause at my site they dont allow ftp acsess!

Using a trick, documented in [1], you dont have to enter your password
when you publish your blog (use with caution!).

Greetings,
Christian

[1]: http://www.suseblog.com/ssh-without-a-password

> I tested, on my computer with an ext3 linux file system, how
> much files (empty) I could populate a directory.

      Interestin,  Thanks.

> According the tiny community of French speaking users of NB,
> upgrading to a newer version is beneficial for the time spent
> with the generation of web pages.
>
> From my side I prefer to upgrade by creating a new blog aside.

      I have considered that.  But I would like not to mess
      up anyone's links or subscriptions to my my blog, so
      I would want to use the current names for the new blog.
      Perhaps I could move the current blog to a dir named
      'archive' or some such, run a script that changes all
      of the internal links to use the new path, and then
      put a link to the archive on the new blog using a newer
      NB in a directory with the original name.

---- Eugene

Hi Eugene,

>       Unix starts to behave
>       strangely when a directory gets close to
>       1000 entries.

I tested, on my computer with an ext3 linux file system, how much files (empty)
I could populate a directory.

Here is this Bash script:

###############################################################
#!/bin/bash
# create huge amount of empty files to populate a directory
limit=10000 #10 000 files to be ceated. Adjust the amount.
i=1
while [ "$i" -le $limit ]
         do
                 echo -n -e "creating empty file foo$i\r"
                 : > foo$i
                 let "i+=1"
         done
echo
echo "you could now do [rm foo*] to clean the directory."
echo "if the command [rm] can't do it in one pass, do [rmfoo1*]\
then [rm foo2*] and so on."
exit 0
##################################################################

I reached 1 million of files inside the same directory without problem. Excepted
with the command "rm" to remove so much files! So, Kevin or somebody else, sould
create a stress test (with an amount of a daily entry over 50 years...) to have
an idea of the limits of NB. And why not a benchmark to test the speed of each
version!

According the tiny community of French speaking users of NB, upgrading to a
newer version is beneficial for the time spent with the generation of web pages.

From my side I prefer to upgrade by creating a new blog aside. You can have
several versions of NB on the same system if you put them in separate
directories; then you have to start these blog engines locally in their own
directory using "./nb" in lieu of "nb". You can use links to use the same data
or template directory: this is usefull if you must have several versions of a
website according the langage or the style sheets design. Have a look over these
sites (merged with a wiki engine standing on an other server), using modified
templates to include a menu bar and modified style sheets to experiment new
flavors:

http://www.oceamer.com/~nanoblogger/index.html
http://oceamer.com/~nanoblogger/clean/
http://oceamer.com/~nanoblogger/spring/
http://oceamer.com/~nanoblogger/steelblue/
http://oceamer.com/~nanoblogger/rusted/

My web site hasn't htaccess file, so you can browse my modified templates.

Again, these sites with deep modifications to provide a single column blog with
menu bar and be text browser compatible.

http://oceamer.com/~nanoblogger/console/
http://oceamer.com/~nanoblogger/gothick/
http://oceamer.com/~nanoblogger/cons-steelblue/

--Denis

> There's been a lot of optimization since version 2.8. The script
> relies more on Bash internals, which translates to less time wasted
> outside the script's process. Unfortunately, there's also been some
> changes that will make upgrading from the 2.x series more
> difficult.  ...

       Thanks for the feedback, Kevin.  I'd just as
       soon stay with Nanoblogger, so maybe I'll set
       aside some time and try to upgrade.  I have
       saved a couple of old forum messages from
       people who had done so; maybe they will help.

> By the way, I could not help but be impressed with the sincerity
> and quality of your writing when I checked your blog. Keep it up
> and good luck with your latest endeavor.

       Thanks for the kind words.

---- Eugene

> I have a blog
> http://www.cs. uni.edu/~ wallingf/ blog/
> powered by Nanoblogger 2.8. I never got
> around to upgrading to a 3.x version because
> 2.8 more than met my needs. I now have
> 918 entries. In the last couple of weeks,
> I have noticed that posting a new entry
> takes a much longer time, both generating
> top-level elements like the feed and
> generating the pages themselves.
>
> I figure this may have something to do
> with the blog having so many files in a
> single directory. Unix starts to behave
> strangely when a directory gets close to
> 1000 entries.
>
> Have any of you had such an experience?
> Will upgrading to a 3.x version help, or
> does it also keep all the entry files in
> a single dir? Or should I look for a
> different blog tool? Thanks for any
> pointers.

I have a blog
           http://www.cs.uni.edu/~wallingf/blog/
       powered by Nanoblogger 2.8.  I never got
       around to upgrading to a 3.x version because
       2.8 more than met my needs.  I now have
       918 entries.  In the last couple of weeks,
       I have noticed that posting a new entry
       takes a much longer time, both generating
       top-level elements like the feed and
       generating the pages themselves.

       I figure this may have something to do
       with the blog having so many files in a
       single directory.  Unix starts to behave
       strangely when a directory gets close to
       1000 entries.

       Have any of you had such an experience?
       Will upgrading to a 3.x version help, or
       does it also keep all the entry files in
       a single dir?  Or should I look for a
       different blog tool?  Thanks for any
       pointers.

---- Eugene

Is there any chance to get tags names with non-latin characters. I'm trying to
add tags in russian, but only gets blank ('') tags names. The same thing with a
article name - russian characters are replaced by "_" - may be there is a way to
transliterate them, since I can leave with tags in english only, but the
articles such as "_________mutt_archlinux____" looks ugly?