I'm assuming it's not happening on all posts right?

My guess is that for whatever reason, it couldn't find the date in the article header. Before the nntppostingdate column was introduced the script used to put the date straight into the dateline column. This dateline value was checked and set to the current date if null. This extra error checking seems to have been forgotten this time.

However, it doesn't explain why there are some articles coming through that the script can't parse. I've not actually seen it happen on my setup, so my guess is it's specific to a newserver and not an article. I did see it now and again when I was using Newsfeeds.com as a provider.

Let me know if you find anything out. :)

I've just had another thought about this. The only other thing different between instalations that would affect the date is the version of the Date::Parse module. I use the str2time function from this module to parse the date and convert it into unix time.

There is a similar version number incompatibility with the DBI module and the 'mysql_insertid' function. In the older version, the function was returning null. I was pulling my hair out for weeks over this one as only captnroger had the problem. Then just last week, eric figured it out when he encountered the problem.

I think... I found it.

The problem was not posts coming from usenet, it's on the local end when I replied to or started a new thread. What I did was add what I think is missing nntpdateline updates:

newthread.php
newreply.php.
Unless I'm missing something I think that's it? At least it seems to work right so far. I'm still waiting on usenet propagation to see what they look like externally.

You're right. I was just about to say it's in the instructions as a necessary code change, but I just went and checked and it's not there. That's exactly what I have on my setup though, so I guess I forgot to put it in the instructions. ooops :o

I intend to put up another little release tomorrow to include gilbys fix so I'll fix the instructions at the same time.

I've also got another little option that allows you to reduce the font size of quotes. I think it makes the posts easier to read. I've put it on dbforums.com already so you can see what I mean.

The quote option sounds great. :)

One other tiny problem you may want to fix is in newnews.pl
When doing a history pull and all articles on a run are not available last msg: counter does not increment and has to be updated by hand.

Also, it would be nice if when there are no messages in a group that the program would skip checking orphans and jump right to the next group. With a lot of groups I think that little diddy would speed things a bit.

Don't want to load you down again Paul, I just thought you might want to know about these being that you are updating again so soon.

Cheers!
-Larry

I am getting this problem




[QUOTE]

update your DBI module. The version you have is not returning the threadid.

I think someone else had this problem and I couldn'd understand what they meant. However, I switched news serrvers this weekend and re-pulled all news and I saw the problem with my own eyes. I'll get on it.

I hope to get all these fixes done today and a new release out. If not today, definately tomorrow.

While I was validating the output HTML I found that signatures from usenet messages can have invalid characters like (&, and high ASCII chars). htmlentities() action needed?

Also, I would like to see a nl2br in usenet signatures.

Oh-oh I think I hooked a big one!

I knew something wasn't quite right with... function update_last_active_thread because I have one thread title that displays B& amp;W Album. Looking into it further I think I see what is happening.

In function update_last_active_thread you are using
addslashes(htmlspecialchars($lastactivethread))
Unfortunately htmlspecialchars is not needed because vB has already preformed special character conversion and stores it converted in the DB.

This caused me to look further and I found that newnews.pl saves the thread title to the DB without processing for special characters. That would be OK if vB did it that way too.

Paul, tell me I'm wrong... please! If I'm not wrong this means a total purge of all usenet posts and new history pull. This is not that big of a deal for me because I don't have many posting users. Although the news feed I use is terribly slow. :(

God I love this stuff! :D

Cheers!
-Larry

The download link disappeared. In fact, so did the site. Could be down for maintenance but I'd love to get hold of this hack and implement on my system.

The site could be a bit dodgy over the next few days as I've just updated the DNS and moved it to a new server. Keep trying and you should get it... if I've set up all the redirects correctly :(

Hmmm, I think Fast forward's site (not dbforum, BritishExpats) is undergoing DNS propagation.. The old site redirects to an IP and the IP redirects to the domain name. Hope things are o.k. Fastforward, this may take a few hours to clear-up.

I've been there before. :)

I just talked with fastforward tonight and he is working on an update that should be ready tomorrow or the next day.

You can disregard my posts above about finding a big error, I wasn't really in sharp mode when I posted those. There will be no need to do a new history pull with the updated version. The posts that are affected by the above problem will cycle out of the database quickly enough to not be a bother.

slinky, your gonna love this hack. :D

Cheers!
-Larry

Shezz, I think the three of us should play the LOTTO today. Check out the post time on the last three posts. :)

Yes, I figured the DNS thing at the same time fastward was posting about it!

O.k. what numbers are we picking? :)

I think some ISPs will get circular redirects (infinite loops) for a while :(

Heheh, can't wait. I set up a usenet passthrough into a graphical BBS back in 1992 and haven't done so since. (I'm still ROFL from the lotto comment.)

It seems that now I'm being directed to the right site as expats.com is up (nice work). Problem is that it seems that there is no download directory and with bad addresses I get redirected to the home page.

slinky, I feel your anticipation. Here's a gift to help tide you over until Paul finishes the update. This will give you a leg up by giving you time to get all the perl modules talking to newnews.pl. You don't have to worry about doing things twice, the new version should drop in with just a few minor changes. Just don't come crying to me if I'm wrong. ;)

This is version 2.0 for vB 2.0 final.
http://www.makeitsimple.com/download.php

Cheers!
-Larry

PS: I'm playing 3,6,9 tomorrow. :)

Ooooooh.... many thanks for helping me with my condition of instant vb hack gratification! :) I accept your disclaimer and let the hacker beware! Good luck with the lottery. :D

This release fixes a few bugs and adds one new option.

1. Made use of HTML::Entities to strip usenet post subjects of HTML characters prior to inserting into database. This is consistent with the way vB does it.
2. Fixed last message number updating when no posts are found during a batch run.
3. Added option to size usenet quote text.
4. Another fix to the remove bbcode function.

Upgrading

1. Replace newnews.pl
2. Install the HTML::Entities Perl module from CPAN if necessary. This is part of the HTML-Parser bundle and is most likely already installed.
3. Check the optional code change number 4 and change if necessary.
4. Run the SQL statements in UPGRADE.txt. This will erase your settings so make a note of your 'Useragent' setting so that you can reset it to the same value after the upgrade. You spam and replacement filters will NOT be touched.

Download version 20010528

Wow... just looking at the cp impressed me! But nuts.... I'm getting a lot of errors trying to install the modules. :( Tried to find some of the through CPAN.org and was having a tough time.... any ideas?



cpan> install DBI
Running make for T/TI/TIMB/DBI-1.15.tar.gz
Unwrapped into directory yes/build/TIMB000

CPAN.pm: Going to build T/TI/TIMB/DBI-1.15.tar.gz

Couldn't chdir yes/build/TIMB000: No such file or directory at (eval 26) line 42
19
======================================

cpan>install Bundle::MySQL
Can't install Bundle::MySQL, don't have an associated bundle file. :-(
at (eval 26) line 1806



===============================


cpan> install Text::Autoformat
cpan> install Date::Parse
cpan> install Email::Find
cpan> install URI::Find
cpan> install MIME::WordDecoder
cpan> install HTML::Entities

yields these errors:


Running make for D/DC/DCONWAY/Text-Autoformat-1.04.tar.gz
yes/sources/authors/id/D/DC/DCONWAY/Text-Autoformat-1.04.tar.gz: No such file or
directory
Checksum mismatch for distribution file. Please investigate.

Distribution id = D/DC/DCONWAY/Text-Autoformat-1.04.tar.gz
CALLED_FOR Text::Autoformat
CONTAINSMODS Text::Autoformat
CPAN_USERID DCONWAY (Damian Conway <damian@conway.org>)
MD5_STATUS
localfile yes/sources/authors/id/D/DC/DCONWAY/Text-Autoformat-1.04.tar.gz

I'd recommend removing
yes/sources/authors/id/D/DC/DCONWAY/Text-Autoformat-1.04.tar.gz. It seems
to
be a bogus file. Maybe you have configured your `urllist' with a
bad URL. Please check this array with `o conf urllist', and
retry.

Fetching with LWP:
ftp://ftp.perl.org/pub/CPAN/authors/...NWAY/CHECKSUMS
yes/sources/authors/id/D/DC/DCONWAY/Text-Autoformat-1.04.tar.gz: No such file or
directory
Checksum mismatch for distribution file. Please investigate.

Distribution id = D/DC/DCONWAY/Text-Autoformat-1.04.tar.gz
CALLED_FOR Text::Autoformat
CONTAINSMODS Text::Autoformat
CPAN_USERID DCONWAY (Damian Conway <damian@conway.org>)
MD5_STATUS
localfile yes/sources/authors/id/D/DC/DCONWAY/Text-Autoformat-1.04.tar.gz

I'd recommend removing
yes/sources/authors/id/D/DC/DCONWAY/Text-Autoformat-1.04.tar.gz. It seems
to
be a bogus file. Maybe you have configured your `urllist' with a
bad URL. Please check this array with `o conf urllist', and
retry.

sh: yes/sources/authors/id/D/DC/DCONWAY/Text-Autoformat-1.04.tar.gz: No such fil
e or directory
Could not open >yes/build/DCONWAY000/Makefile.PL at (eval 26) line 4219

=============================

I've never seen errors like that slinky :(
What version of Perl do you have and what platform?
You might have better luck installing them using the traditional method.

Not sure which version of perl but I went to CPAN to find the modules to install and not all of them are there. Here are the ones that are either not installed or that I couldn't find.

Bundle::mysql could be one of several different bundles.
Date::Parse isn't listed.
MIME::Decoder isn't listed
HTML::Entities isn't listed.

Here's the links:

Bundle::mysql (part of the Msql-Mysql-modules bundle)
Date::Parse (part of the TimeDate bundle)
MIME::Decoder (part of the MIME-tools bundle
HTML::Entities (part of the HTML-Parser bundle)

There's a good chance that these particular four are already installed on your system. They are very common.

You're a saint... :) I booted it up and realize that MIME::WordDecoder isn't installed. After searching all over the place I found out that it is part of the MIME-tools package, FYI... I'm going to try to install them now... :)

I see you're using the vBPortal Slinky. It looks good! I must have a closer look at that when I find time. That's a nice domain name too! No doubt you have plenty of offers for it :)

Yes - the Portal is coming along very nicely. Testing for wajones and I've added in a few twists of my own! :) Now all I need to do is hire a designer just to design the overall site look. The domain name is a long story but it worked out really well for me... got it over 6 years ago, then was involved with the former mayor of New York, now I'm back in control... long story but great ride. :D

Almost there... some of the problems I've encountered could be that I'm using a virtual host and I've had to install the modules into another directory /usr/thelaw/perl/ which I created. The only error I'm getting now upon install and make are:

Warning: I could not locate your pod2man program. Please make sure, your pod2man program is in your PATH before you execute 'make'

Getting closer....

MMmm,
an scripts "that" check wich modules your need for this hack..!

no...yes...to much work :rolleyes:

carrillo

Please Can Help Me ??

I found this error when i try use newnews.pl to import newsgroup into my forum but it so many error like this .. :D

Please Who can help me fix it ??? Thank you Very Much.. :D

I've seen a few other posts with similar errors (not related to this hack). I believe the advice given was to empty the word and searchindex tables and reindex. you could also try just deleting that particular record 'where title = '¡»'.

If you don't want to either of the above, you can simply turn off usenet post indexing for one batch run using the usenet control panel.

and Thank you very much.. :D I will do this ..

and 2. Question.

Sometime I try use newnews.pl to get some newgroups posts
but I have try many newsgroups , but why some group I just can not import any news into forum, But I am sure this newsgroup have many posts ??

would you can access newgroup "tw.bbc.comp.linux" or "tw.bbs.comp.hardware" ?? this two chinese newsgroups have many posts everyday . But I just can not import any things ??

Well finally got this installed again and wow, looking brilliant
http://www.ascifi.com/forums2/forumd...hp?forumid=980

couple of things:

i) the merge into new thead for lost posts didn't work and gave this error;

i have a feeling this might be because i have an old version of mysql but amnot sure

2. Doing "new threads" instead gives this error:

but insert into thread worked like a charm.

3. News server

which one? I need:

i) cheapish, less than $15 per month
ii) no speed restriction, restriction done on monthly bandwidth
iii) does NOT post a footer. This is why i am leaving newsfeeds
iv) decent selection of newsgoups, ie need to be able to check first
v) and preferably uses a 3rd party credit card processor. I don't want to be easily liable for any spam muck ups that someone might attempt to do and get charged $150 an hour "clean up". They wouldn't be able to do this if they use paysystems/instabill which is one of the reasons i like newsfeeds.

Any suggestions?

oh yeah, are you planning on finishing of a "vb2.0 final template set for newsgroups"? would be very cool ;)

Chris,

The merge into lost threads and all other functions within that option need updating. I made a lot of table structure changes without changing this function. It was low on my priority list as I don't use it :) I willl get it fixed though.

As for a news feed, I switched to onlynews.com from newsfeeds. They are very fast and have no speed restriction, but the retention is pretty crap (4-5 days) and the number groups is quite low. I'm looking for another one myself. Let me know if you find one.



mobilebbs,

There are several reasons why posts may not get imported.
1) You haven't corrrectly mapped the local forum to the usenet group.
2) The article numbers have somehow got out of synch. (try resetting the numbers to zero for those groups)
3) you are pulling historical posts and have autopurge turned on and set too low. This will cause old posts to be immediately deleted after being imported.
4) The articles are all getting rejected for spam (unlikely)

. Thanks .. Let me try .. Sorry My English is very bad . But Really
Thank for your replys.. :D

For some reason I get a little messed up html code with quoted replies in version 20010528.

Here's what i mean:
<br>
User <name@email.dom> wrote in message<br>
<a href="http://news:somerefnum@somenewserv.dom" target="_blank">news:somerefnum@somnewserv.dom</a>...<br>
<font size="1">&nbsp;&nbsp;<i><font color="blue<br>">>

Which obviously produces a nasty color far from blue.

I don't quite know how to locate the bug...

There was a change in this section of code to allow for the size setting. However, I don't seem to have this problem on my setup. What other usenet display settings do you have set ie. indents, color etc. I have all set to yes on mine.

All set to yes, wrap 90, quote 1, West Europan-Latin1

finding a news server is a nightmare, an incredibly disorganized and unprofessional looking sector incredibly annoying.

Anyway, realised that spamkiller.newsfeeds does not in fact add any signature so i am going to use that. Paul, why did you switch from newsfeeds? is there something i should know about?

i guess that for us retention does not matter as we are mirroring everything anyway apart from the initial pull of posts which will be less? is that correct?

finally, sorry to nag, are you going to do your template in v2.0 style. There were a lot of updates to the templates i think ):

I left them out of principle after they accused me of spamming due to one post that did not have the body in the message, but did have the signature with my url! They didn't actually make the complaint obviously, but they still followed it up.

Quote:

i guess that for us retention does not matter as we are mirroring everything anyway apart from the initial pull of posts which will be less? is that correct?

True, but I need a newsfeed for home aswell, with binaries and longer retention.

Quote:

finally, sorry to nag, are you going to do your template in v2.0 style. There were a lot of updates to the templates i think ):

I'm afraid I won't be redoing them until I have finished redoing my britishexpats site. So it could be a month away.