pycs-devel Mailing List for Python Community Server (Page 17)

SourceForge Headquarters 1320 Columbia Street Suite 310 San Diego, CA 92101 +1 (858) 422-6466

Hi!

> I've just made a couple of quick changes:

Cool, I hoped you would do that, at least the authorizer.py part ;-)

bye, Georg

I've just made a couple of quick changes:

- swish.py now displays a search box if no query string is entered

- authorizer.py now covers searches (so access to swish.py is governed by
the same rules as comments.py and mailto.py - i.e. the same as access to the
blog) ...

Cheers,
Phil :)

> Just a quick and dirty hack: modules/system/swish.py and make_index.sh
> give
> you the ability to set up per-user search indexes and search them using
> swish++ via a simple web form.

Great!

I'm still struggling on getting ht://Dig in ... it's been quite painful.  I
hate autoconf/automake ;-)

(progress: I've turned ht://dig into a Python module, but it doesn't know
where its dependencies - a whole load of shared libs - are located, so
importing fails unless you set LD_LIBRARY_PATH.)

I'll plug this one in for the meantime - thanks for hacking it up!

Phil

> Now, i should also mention that because i'm doing all this on windows, i
> could not run the makefile, so i've had to manually create all the
> directories and put all the file where they should go (reading a makefile
is
> a bit of a pain! ;). One thing that i noticed is that the weblogUpdate.py
> file is not in the /system directory in the .gz download file - i had to
> manually move the weblogUpdate.py file to the system dir. In the /system
dir
> is only 'update.py'... Not sure if that has anything to do with it, but
this
> always fails when it follows the 'recently updated' link which points to
the
> weblogUpdate.py script.

Aha - here's what you need to do:

- use the version from CVS rather than the .gz file, which is probably
horribly out of date.  As well as having loads of new features, it's
probably heaps more stable ;)

- there _is_ a weblogUpdates.py in the /system directory; it's just that
there's _another_ one in the root.  That's where they're meant to be: the
one in the root is the handler for the weblogUpdates XML-RPC call (that's
the one that puts you on the updates page), and the one in /system is the
updates page itself.

If you check out the archive and copy both of those into the right places,
then restart, you'll probably find that your weblog updates will work again.

Actually, now that I think about it, it used to work on Windows if you just
extracted everything into a directory and ran from there.  You certainly
don't need to copy everything into /var/lib/pycs and stuff.  Read
pycs_paths.py though - it should have some details about where things go on
Windows, which is a bit simpler than on Linux.  (This is one of the things
that is probably different in CVS and in the .gz on SourceForge).

Cheers,
Phil :)

Hi!

Because of the fact that there are now search terms that lead from google
directly to my zeitgeist.py, I thought about disabling this google bomb 
and
added a way to prevent robot indexing of pages in the PyCS. I enabled 
this
in the named three modules, so output of those are not indexed, not 
followed
and not archived in google and other well behaving robots.

The problem is, my zeitgeist.py is a page that is a) heavily linked from 
my
weblog and b) links to many pages outside, so google gives it quite a 
high
rank. Adding to this, there are now popping up search hits that actually
only point to my page. This is fun (search-engine created art, so to 
say), but
actually not really that usefull if those pages clogg up searchspace ;-)

bye, Georg

Hi!

Just a quick and dirty hack: modules/system/swish.py and make_index.sh 
give
you the ability to set up per-user search indexes and search them using
swish++ via a simple web form.

http://muensterland.org/system/swish.py?usernum=1&q=Contax

searches on my main weblog for all Contax related pages. It's really very
crude and simple, just enough to set up searches. Hopefully I escaped the
search string correctly, so it won't create a security risk. I made sure 
that
only a subset of allowed chars is in the query, so it usually shouldn't
make problems.

bye, Georg

Hi!

On Fri, Feb 28, 2003 at 10:18:01AM -0500, digital hyakugei wrote:
> I've got pyDS running. I (sort of) have it talking to pycs.net

Hey, that is cool, as I never ever tried it with W2K :-)

> (http://www.pycs.net/users/0000074/) tho phil said that it wasn't really
> talking to pycs correctly - could be - the pyDS docs leave a bit to be

Hmm. Looks fine to me, except that several files are missing (images and
stylesheet). This might be due to the fact that PyDS isn't really
Window-aware and needs a $HOME directory for the user and want's to
create a .PyDS directory there and populate it. So all this might go bad
on Windows, as there isn't such a thing like users home directories and
hidden directories in the Unix sense.

Did you set a HOME environment variable to point to some place where
your current user has write permissions?

Oh, and you created a user at muensterland.org, too. But I assume that
was because some experiments, right? Doesn't look too populated, that
directory ;-)

> desired ;). In the upstream pref i think i used the url:
> http://www.pycs.net/RPC2... Thats about it.

Yep, you usually don't need to do more. Everything else comes from the
community server.

> <error>
> Traceback (most recent call last):
>   File "pycs_module_handler.py", line 155, in continue_request
>     mod = cached_module( realPath )
>   File "pycs_module_handler.py", line 78, in __init__
>     self.Recompile()
>   File "pycs_module_handler.py", line 86, in Recompile
>     self.module = compile(code,self.path,'exec')
>   File "<string>", line 116
> 
> SyntaxError: invalid syntax
> </error>

What is in line 116 of weblogUpdates.py on your machine? Maybe there is
just a broken version in the tar.gz? I never installed it that way, as I
always use CVS :-)

> I also had another question about pyCS - is there any chance that it will
> speak the blogger API? does this request even make sense? I've looked into

No, PyCS can't speak it, as it isn't a weblog application but a
community server. The blogger API should go into PyDS, as that is the
desktop part of the bunch. But I currently don't work on it.

> Radio and know what it does, but i'm a bit unclear about RCS/pyCS's function
> and if it/they require a client like Radio/pyDS...

RCS/pyCS are just the server that pushes out your static content, you
generated with Radio or PyDS or bzero. A community server has several
aspects and functions:

- user registration (mostly done automatic) for new users
- uploading of changed files
- pushing static files to the world
- comment functionality
- referer, searches or other reports

So it is a bit like an FTP server with some CGI functionality, if you
want to see it like that.

bye, Georg

Hello all.

I sent an email to Phillip Pearson who was kind enough to get back to me and
point me here - so, let me give you the low-down on what i've managed to do
so far.

I've got a win2k box. Running python 2.2

I've got pyDS running. I (sort of) have it talking to pycs.net
(http://www.pycs.net/users/0000074/) tho phil said that it wasn't really
talking to pycs correctly - could be - the pyDS docs leave a bit to be
desired ;). In the upstream pref i think i used the url:
http://www.pycs.net/RPC2... Thats about it.

I got pyCS going locally (after a fair bit of futzing with PIL). However!!!
I get the following error when i try to access the 'recently updated' link
off the default main page:

<error>
Traceback (most recent call last):
  File "pycs_module_handler.py", line 155, in continue_request
    mod = cached_module( realPath )
  File "pycs_module_handler.py", line 78, in __init__
    self.Recompile()
  File "pycs_module_handler.py", line 86, in Recompile
    self.module = compile(code,self.path,'exec')
  File "<string>", line 116

SyntaxError: invalid syntax
</error>

Now, i should also mention that because i'm doing all this on windows, i
could not run the makefile, so i've had to manually create all the
directories and put all the file where they should go (reading a makefile is
a bit of a pain! ;). One thing that i noticed is that the weblogUpdate.py
file is not in the /system directory in the .gz download file - i had to
manually move the weblogUpdate.py file to the system dir. In the /system dir
is only 'update.py'... Not sure if that has anything to do with it, but this
always fails when it follows the 'recently updated' link which points to the
weblogUpdate.py script.

I also had another question about pyCS - is there any chance that it will
speak the blogger API? does this request even make sense? I've looked into
Radio and know what it does, but i'm a bit unclear about RCS/pyCS's function
and if it/they require a client like Radio/pyDS...

Thanks very much,
jos

Hi!

I was inspired by http://jimfl.tensegrity.net/zeitgeist/ to not only 
hack up something similar, but also shamelessly steal the color codes 
and concept ;-)

If you want to see how this looks with the Python Community Server, have 
a look at:

http://muensterland.org/system/zeitgeist.py?usernum=0000001

I think it's quite cool, although utter stupid and silly. :-)

It's currently not linked from any of the pages, so you have to 
construct your own URL (should be simple, just plug in your user id and 
off you go. Oh, and my version only works if your browser is capable of 
using the style="..." attribute on tags and supports CSS (ok, only 
font-size with px sizes and color is used, so most browsers should work).

bye, Georg

Hi!

Just noticed that there was a stupid bug in the fullfeeds, where the 
filter with full=1 or full=2 didn't return the comments of the last 14 
days, but only comments _older_ than 14 days. There _is_ a difference 
between < and >=, after all ;-)

So if somebody uses the fullfeeds and not the comment monitor, it would 
be better to upgrade.

Ob-PyDS-plug: it's now actually running quite nicely, even though there 
were several updates. I have found something to render plaintext into 
HTML (Docutils), so now postings don't need to be in HTML source any 
more. Only thing really missing from Radio is a news aggregator, but I 
hacked up the AmphetaDesk index.html and now use that for aggregating 
and post from there to PyDS. If somebody actually is using PyDS and 
misses a news aggregator, drop me a line and I send him/her the 
index.html

bye, Georg

Hi!

I made a stupid error in the password handling in PyDS: I sent the
password directly, where Radio sends the md5 Hexdigest, instead. So PyDS
users couldn't login via login.py, as the password was stored in clear. I
fixed PyDS, but an upgrade requires action by the Community Server Admin,
as he must set the password for the user new. Phil had written a
changepasswd.py for that purpose and I just checked in a pycsadm.py way to
do it (pycsadm.py doesn't require the community server to be stopped, as
it uses xmlrpc). A description is in the newest entry at
http://pyds.muensterland.org/

Ok, now I know why I call it an Alpha version ;-)

bye, Georg

Hi!

> No, it's the rendering stuff. Ok, the first post will cache all 
> templates and most posting indices, so the second post with still 
> running pyds should be faster. But rendering still takes quite some 
> time. That's why I plan to put rendering into a background task, like 
> upstreaming already is.

Done. And several other new stuff. Just re-download and upgrade. Oh, and 
don't forget to activate upstreaming, as it is disabled by default in 
the new version. With the new background rendering, it is _fast_ :-)

bye, Georg

Hi!

> Hey, this works beautifully.  I love it!

Cool :-)

> The posting is a little slow (is it doing all the upstreaming before
> returning?)

No, it's the rendering stuff. Ok, the first post will cache all 
templates and most posting indices, so the second post with still 
running pyds should be faster. But rendering still takes quite some 
time. That's why I plan to put rendering into a background task, like 
upstreaming already is.

> and it doesn't seem to be pinging the local community
> server,

Yep, that's on my todo. Actually my todo is my instant radio outline, so 
you can see it (you can use the rendered version under the "Pick my 
brain" link on http://hugo.muensterland.org/).

> Officially I'm not allowed to like it *too* much, because I'm meant to
> be working with UserLand on community servers, and I also have my own
> competing blogging tool (because I have no business sense whatsoever),

LOL

> Anyway, the useful thing I found out was how to embed Python and use
> wxWindows to make little Windows app shells for Python programs.

Yep, that's something I want to do for OS X, too. Some small controller 
App like the one from AmphetaDesk, just to monitor the running 
background process. But it's very low on my priority list currently, 
first I would like to get all todo items done.

>   So, if I have time, I'll do exactly that for this and make it behave 
> like
> Radio, with a system tray icon and probably a console log tailer that
> shows in a window.  If I have time ...

Would be really cool. Actually I didn't do any tests with windows up to 
now, so I don't know wether it works with windows. I only have Linux and 
OS X available.

bye, Georg (and now _really_ off to bed)

Hi!

> I will check with the validator. The rss feed wasn't actually much 
> tested, I just hacked some template for that from my Radio generated 
> feeds.

Ok, rss feeds now validate. Just download the 0.1 package again :-)

bye, Georg

Hi!

> Cool.  Also seems that pyds-browse is missing; here are the last 3
> lines from the 'python setup.py' output:

Fixed.

> Also I can't see any .tmpl files in there (when I tried installing
> manually).

Fixed.

Both were just missing from the MANIFEST, too. Just refetch the archive, 
now it should be complete (at least I can't think of files still 
missing, but if they are, you surely will find them ;-) )

bye, Georg

Hi!

> The <guid> element in the generated RSS doesn't have a full URL (it's
> just "/weblog/foo" rather than "http://bar/weblog/foo").

Hmm. I just looked into the newly generated rss.xml and it does have 
server name and all.

> Also it looks like <pubDate> is missing the "GMT" or timezone
> specifier at the end.

That might be possible, as I didn't know it was missing :-)

I will check with the validator. The rss feed wasn't actually much 
tested, I just hacked some template for that from my Radio generated 
feeds.

bye, Georg

Another bug report:

The <guid> element in the generated RSS doesn't have a full URL (it's
just "/weblog/foo" rather than "http://bar/weblog/foo").

Also it looks like <pubDate> is missing the "GMT" or timezone
specifier at the end.

Validator results:

http://feeds.archive.org/validator/check?url=http%3A%2F%2Fmuensterland.org%2Fusers%2F0000006%2Frss.xml

Cheers,
Phil :)

> >Looks like setup.py is missing from the PyDS archive.  Georg?
> 
> Fixed. Was a typo in the MANIFEST. Just download the package again, same 
> name. I am not yet ready for version number changes ;-)

Cool.  Also seems that pyds-browse is missing; here are the last 3
lines from the 'python setup.py' output:

copying and adjusting pyds-start -> build/scripts-2.2
copying and adjusting pyds-stop -> build/scripts-2.2
error: file 'pyds-browse' does not exist

Also I can't see any .tmpl files in there (when I tried installing
manually).

Cheers,
Phil :)

Hi!

> Looks like setup.py is missing from the PyDS archive.  Georg?

Fixed. Was a typo in the MANIFEST. Just download the package again, same 
name. I am not yet ready for version number changes ;-)

bye, Georg

Hi,

Looks like setup.py is missing from the PyDS archive.  Georg?

Cheers,
Phil :)

Hi!

> doesn't complain either.  Now that you've changed ISO- to iso-, IE
> isn't complaining either though, so that seems to have been what it
> objected to.

Nah, it was the <?xml?> vs. <xml> :-)

> I see.  Regardless of that, it might be a good idea to put in the
> alias, otherwise you'll confuse Google and get people linking to the
> wrong place all the time.

Oh, the / does actually work. It's just that when you click at 
"Startseite" you go to /weblog/ - but you can call the main index, and 
can index it. But I think I will change it, nonetheless. Looks nicer and 
cleaner.

> Heh - pretty quick!

It was my weekend work :-)

I now have the weblog, have stories, have macros (actually those are 
python functions) and have shortcuts. And have muchos infrastructure 
stuff up and running.

> package for PyCS - it would be cool if we can do that for both of
> them, then it will be trivial to set them both up together.

Sure. I did a standard setup.py installer, so it should be easy 
debianized. And one of the things I did this weekend was cleaning up the 
installer, so it is mostly FHS compliant now.

bye, Georg

> >- the rss not well-formed xml: try viewing
> >http://muensterland.org/users/0000006/weblog/rss.xml in internet
> >explorer (or running it through expat's 'xmlwf' checker).  Looks like
> >a character-encoding problem.  Should "ISO-8859-1" be "iso-8859-1" in
> >the <?xml?> header?
> 
> No, ISO-8859-1 is the same as iso-8859-1, actually the former is the 
> "official" name, although all tools I know about support both names. 
> Hmm. Chimera (Mozilla based) doesn't complain.

Yeah - upon further checking, Python's xmllib (presumably expat)
doesn't complain either.  Now that you've changed ISO- to iso-, IE
isn't complaining either though, so that seems to have been what it
objected to.

> >- 'Startseite' links to /weblog/index.html rather than just /
> 
> Yep, that's just because the "Homepage" is only a copy of one of the 
> installed tools homepages. Every tool installes into it's own namespace. 
> Those generate corresponding subdirectories. The homepage /index.html is 
> just the direct copy of the /<tool>/index.html of the installed tool 
> with level 0, that happens to be the weblog tool.

I see.  Regardless of that, it might be a good idea to put in the
alias, otherwise you'll confuse Google and get people linking to the
wrong place all the time.

(That reminds me... I have to put in 301 redirects from pycs.net/* to
www.pycs.net/* because I've got the same problem there).

> >Will download and check out soon!
> 
> Would be great. Some more input  on it is highly appreciated. Current 
> version already has severall new tools added.

Heh - pretty quick!

Will try and get it going on Linux.  I'm thinking about doing a Debian
package for PyCS - it would be cool if we can do that for both of
them, then it will be trivial to set them both up together.

(... and yes, I'll be doing the same for bzero, but being non-free and
all it doesn't fit in to Debian nearly as well).

Cheers,
Phil :)

Hi!

>> - the rss not well-formed xml: try viewing
>> http://muensterland.org/users/0000006/weblog/rss.xml in internet
>> explorer (or running it through expat's 'xmlwf' checker).  Looks like
>> a character-encoding problem.  Should "ISO-8859-1" be "iso-8859-1" in
>> the <?xml?> header?

Just checked again this time with rxp and found the problem. It actually 
_should_ be a <?xml?> header, not a <xml> one :-)

Dang, and I even made it wellformed by adding a corresponding </xml> 
line to the end :-)

bye, Georg

Hi!

> - the numbers on the calendar link to urls like
> http://muensterland.org/weblog/;2003-01-10, which don't seem to exist.

Hmm. That's not the case, actually not _now_ :-)

I hacked the whole day, and made heavy reworks on the calendar code, so 
it might be that you checked a version with broken l inks.

> - likewise for the # marks below entries, which link to
> http://muensterland.org/users/0000006/http://muensterland.org/users/0000006/
> weblog/2003/01/12.html#P9

Yep, that's actually there. I have to fix it. Looks like some reworks in 
the getStaticUrl code broke that. It did work once, but I haven't 
noticed that a bug crept in there.

> - the rss not well-formed xml: try viewing
> http://muensterland.org/users/0000006/weblog/rss.xml in internet
> explorer (or running it through expat's 'xmlwf' checker).  Looks like
> a character-encoding problem.  Should "ISO-8859-1" be "iso-8859-1" in
> the <?xml?> header?

No, ISO-8859-1 is the same as iso-8859-1, actually the former is the 
"official" name, although all tools I know about support both names. 
Hmm. Chimera (Mozilla based) doesn't complain.

> - 'Startseite' links to /weblog/index.html rather than just /

Yep, that's just because the "Homepage" is only a copy of one of the 
installed tools homepages. Every tool installes into it's own namespace. 
Those generate corresponding subdirectories. The homepage /index.html is 
just the direct copy of the /<tool>/index.html of the installed tool 
with level 0, that happens to be the weblog tool.

> I just noticed that pyds.muensterland.org doesn't have the broken
> links; maybe I'm complaining about an out of date version here :)

Actually pyds.muensterland.org is a very outdated version now, as I 
hacked last time on that one on Friday :-)

> Will download and check out soon!

Would be great. Some more input  on it is highly appreciated. Current 
version already has severall new tools added.

bye, Georg

Heh - very nice.

Some bug reports for you:

- the numbers on the calendar link to urls like
http://muensterland.org/weblog/;2003-01-10, which don't seem to exist.

- likewise for the # marks below entries, which link to
http://muensterland.org/users/0000006/http://muensterland.org/users/0000006/weblog/2003/01/12.html#P9
etc

- the rss not well-formed xml: try viewing
http://muensterland.org/users/0000006/weblog/rss.xml in internet
explorer (or running it through expat's 'xmlwf' checker).  Looks like
a character-encoding problem.  Should "ISO-8859-1" be "iso-8859-1" in
the <?xml?> header?

Note: to convert iso-8859-1 to utf-8, use
string.decode('iso-8859-1').encode('utf-8').

And style notes:

- 'Startseite' links to /weblog/index.html rather than just /

I just noticed that pyds.muensterland.org doesn't have the broken
links; maybe I'm complaining about an out of date version here :)

Will download and check out soon!

Cheers,
Phil

On Fri, Jan 10, 2003 at 11:30:55PM +0100, Georg Bauer wrote:
> Hi!
> 
> Just thought I first tell you here, as it might be something of interest 
> for you. What happened? I hacked the last days quite heavy. Didn't do 
> much PyCS work, though, but instead PyDS. Yep, PyDS :-)
> 
> As a supplement to the Python Community Server, now there is a Python 
> Desktop Server. It's actually mostly an implementation of an application 
> much like Radio Userland, but in Python. And for Posix systems. It's 
> still alpha, but it is useable, as you can see at
> 
> http://muensterland.org/users/0000006/
> 
> The page is german, an english one is at http://pyds.muensterland.org/, 
> but that was done earlier today and so uses an older version of PyDS and 
> has maybe some broken links. At that site you can download the current 
> version. Usually it's a daily snapshot, but I just made it current.
> 
> What is implemented:
> 
> - weblog editing, categories, rendering, upstreaming (in the background)
> - preferences and preferences-hooks for tools
> - dynamic userdefined macros
> - online documentation of tool API
> - templating using Cheetah templates
> - theoretically themeable, although currently done by hand
> - dynamically installed tools
> - eventlog
> - full translations for english and german
> - almost but not fully without documentation :-)
> 
> I currently still use Radio for now, but that's just for now - I don't 
> have stories implemented yet, and I would need them to transfer all my 
> stuff to PyDS. And the News Aggregator isn't implemented yet, but there 
> one can use AmphetaDesk (and make a little patch to the template of 
> AmphetaDesk to post to PyDS).
> 
> Would be fun if some of you using Posix systems (some BSD flavor, Linux 
> or Mac OS X - actually I am developing both on Mac OS X and Linux) would 
> try it out. Installation instructions are given in the Archive.
> 
> When there is enough to discuss I can set up a mailing list for it, so 
> we don't flood Phils pycs-devel, but I thought you to be a good targeted 
> audience for first experiments :-)
> 
> Comments?
> 
> bye, Georg
> 
> 
> 
> -------------------------------------------------------
> This SF.NET email is sponsored by:
> SourceForge Enterprise Edition + IBM + LinuxWorld = Something 2 See!
> http://www.vasoftware.com
> _______________________________________________
> PyCS-devel mailing list
> PyC...@li...
> https://lists.sourceforge.net/lists/listinfo/pycs-devel

2002	Jan	Feb	Mar	Apr (1)	May	Jun	Jul	Aug (2)	Sep (3)	Oct (1)	Nov (70)	Dec (41)
2003	Jan (20)	Feb (9)	Mar (36)	Apr (11)	May (3)	Jun (6)	Jul (3)	Aug (13)	Sep (2)	Oct (32)	Nov (4)	Dec (7)
2004	Jan (14)	Feb (16)	Mar (3)	Apr (12)	May (1)	Jun (4)	Jul (13)	Aug (1)	Sep (2)	Oct (1)	Nov (2)	Dec (3)
2005	Jan (7)	Feb	Mar	Apr (4)	May	Jun (2)	Jul	Aug (5)	Sep	Oct	Nov (2)	Dec (1)
2006	Jan	Feb	Mar	Apr	May (1)	Jun	Jul	Aug (1)	Sep (2)	Oct (7)	Nov (18)	Dec (22)
2007	Jan (10)	Feb (11)	Mar (1)	Apr (6)	May (5)	Jun (5)	Jul (14)	Aug (28)	Sep (4)	Oct (6)	Nov (9)	Dec (8)
2008	Jan (10)	Feb (19)	Mar (38)	Apr (17)	May (13)	Jun (7)	Jul (36)	Aug (15)	Sep (2)	Oct	Nov	Dec

pycs-devel Mailing List for Python Community Server (Page 17)

pycs-devel — Python Community Server development/chat