Groups | Search | Server Info | Keyboard shortcuts | Login | Register [http] [https] [nntp] [nntps]
Groups > alt.comp.software.thunderbird > #241 > unrolled thread
| Started by | Sticks <Wolverine01@charter.net> |
|---|---|
| First post | 2021-04-25 18:51 -0500 |
| Last post | 2021-04-28 17:34 -0500 |
| Articles | 10 — 6 participants |
Back to article view | Back to alt.comp.software.thunderbird
Another filter question Sticks <Wolverine01@charter.net> - 2021-04-25 18:51 -0500
Re: Another filter question VanguardLH <V@nguard.LH> - 2021-04-25 20:05 -0500
Re: Another filter question Mike Easter <MikeE@ster.invalid> - 2021-04-28 10:40 -0700
Re: Another filter question VanguardLH <V@nguard.LH> - 2021-04-28 15:03 -0500
Re: Another filter question Miguel Tomar Nogueira <mnogueira@mail.telepac.pt> - 2021-05-12 01:09 +0000
Re: Another filter question "J. P. Gilliver (John)" <G6JPG@255soft.uk> - 2021-04-26 03:41 +0100
Re: Another filter question Miguel Tomar Nogueira <mnogueira@mail.telepac.pt> - 2021-04-26 06:19 +0000
Re: Another filter question Andy Burns <usenet@andyburns.uk> - 2021-04-26 08:12 +0100
Re: Another filter question Andy Burns <usenet@andyburns.uk> - 2021-04-26 08:16 +0100
Re: Another filter question Sticks <Wolverine01@charter.net> - 2021-04-28 17:34 -0500
| From | Sticks <Wolverine01@charter.net> |
|---|---|
| Date | 2021-04-25 18:51 -0500 |
| Subject | Another filter question |
| Message-ID | <s64v99$ppv$2@dont-email.me> |
So how do I use the Newsgroups portion of the header to filter on any post that has more than 2 groups in it? I'm guessing if I use a comma (,) I could filter if it has more than one, but for now I'd like to try two. That should eliminate most of the blatant shitheads.
[toc] | [next] | [standalone]
| From | VanguardLH <V@nguard.LH> |
|---|---|
| Date | 2021-04-25 20:05 -0500 |
| Message-ID | <1ddi5dhuq2oke.dlg@v.nguard.lh> |
| In reply to | #241 |
Sticks <Wolverine01@charter.net> wrote: > So how do I use the Newsgroups portion of the header to filter on any > post that has more than 2 groups in it? > > I'm guessing if I use a comma (,) I could filter if it has more than > one, but for now I'd like to try two. That should eliminate most of the > blatant shitheads. Don't use Thunderbird anymore, but I recall it could test using regular expressions. If so, you can probably use a couple that I use: ^Newsgroups: .+,.+,.+,.+, ^FollowUp-To: .+,.+,.+,.+, Besides excessive cross-posting using the Newsgroups header, I also trigger on posters that attempt to get replies excessively cross-posted. For me, I filter on more than 4 cross-posted newsgroups hence why the regex looks for 4 commas with 1, or more, characters between each. The above is testing on the header names, and headers start in column 1 which is why I anchored the string to column 1 (using the ^ special character). Headers must have 1, or more, whitespaces after the colon; however, I don't recall seeing where more than 1 space was used, so I just include the 1 space between header: and headerValue. If I did encounter someone using more than 1 space after the colon, I'd replace the space character with /s+ (one, or more, spaces). I filter on "more than 4" cross-posts because, for example, there are multiple Windows newsgroups to which an inquiry could be relevant. However, more important is I check if the Newsgroups or FollowUp-To headers contain any "garbage" newsgroups, like flaming, proselytizing, kook, warez, soc.*, and other groups that I would never participate, so I would never bother to reply to someone who is highly likely trying to incite reactions from unrelated newsgroups. I don't remember if FollowUp-To is an overview header. Suspect it isn't. I can test on ALL headers because I configure my NNTP client to download not just the overview header (and have to click on an article to then download it all), but to download the entire article (headers and body). That lets me test on everything, because both the overview and non-overview headers get downloaded. I only participate in text only newsgroups, so I'm not concerned about the time to download the new articles (the first retrieve will be all new articles, and take a long time for a busy newsgroup, but subsequent polls for new articles should only list a few). If you do binary newsgroups, you don't want to configure your client to download entire articles. Even the Newsgroups header might be a non-overview header. To check, run "telnet <nntpServer> <port>" and run the "list overview.fmt" command which lists which headers the NNTP server considers overview headers. For me using individual.net, Newsgroups is not an overview header, and another reason I download entire messages, so I can test on any header whether overview or non-overview. If Thunderbird does not support regex (regular expressions) in its filters, you're screwed unless you find an extension that add regex to Tbird, like FiltaQuilla. https://addons.thunderbird.net/en-US/thunderbird/addon/filtaquilla/ No idea if the extension forces a full-message download if you specify a non-overview header. If you don't know regex (typically the PCRE version), then don't bother. You can find lots of online tutorials, snatch what others have used, and learn as you go, but regex is really extensive on what it can test. I've been using it for many years, and still I have to learn something new to get a filter to work (providing the regex parser and interpreter in the program supports all of regex).
[toc] | [prev] | [next] | [standalone]
| From | Mike Easter <MikeE@ster.invalid> |
|---|---|
| Date | 2021-04-28 10:40 -0700 |
| Message-ID | <ietl0dFqjh8U1@mid.individual.net> |
| In reply to | #243 |
VanguardLH wrote: > Don't use Thunderbird anymore, but I recall it could test using regular > expressions. Definitely not; no regex, no wildcards, no scoring. That's a big Tb weakness compared to 'real' news agents. -- Mike Easter
[toc] | [prev] | [next] | [standalone]
| From | VanguardLH <V@nguard.LH> |
|---|---|
| Date | 2021-04-28 15:03 -0500 |
| Message-ID | <7l1yz21pymaw.dlg@v.nguard.lh> |
| In reply to | #365 |
Mike Easter <MikeE@ster.invalid> wrote: > VanguardLH wrote: > >> Don't use Thunderbird anymore, but I recall it could test using regular >> expressions. > > Definitely not; no regex, no wildcards, no scoring. That's a big Tb > weakness compared to 'real' news agents. Then, I guess, a TBird user is stuck testing if the FiltaQuilla extension still works. Its extension page says it has been updated to work with Tbird 78. But this still depends on whether or not Tbird retrieves both overview and non-overview headers to allow testing on non-overview headers. Since the client adds the Newsgroups header, often it is not an overview header. The user would need to configure Tbird to download ALL of every message (headers+body) to have access to the non-overview headers. Even if Tbird supported XPAT, few NNTP servers do since that incurs a lot of overhead of doing searches at the server that a more capable client should be doing. Downloading ALL of a message grants access to ALL headers (overview and non-overview). If you're into porn (seems the major use of binary newsgroups), you wouldn't be using Tbird, anyway, and you certainly would not want Tbird downloading complete articles. However, if you only visit text-only newsgroups then downloading complete articles is not a big deal, especially after the initial retrieve on subscribing to a new newsgroups and thereafter the downloads are incremental (only new articles are retrieved). If after configuring Tbird to retrieve complete articles, that doesn't mean Tbird grants access to all headers within an article. The FiltaQuilla extension is useless if Tbird doesn't let you test on both overview and non-overview headers.
[toc] | [prev] | [next] | [standalone]
| From | Miguel Tomar Nogueira <mnogueira@mail.telepac.pt> |
|---|---|
| Date | 2021-05-12 01:09 +0000 |
| Message-ID | <s7f9so$nhp$1@neodome.net> |
| In reply to | #365 |
Mike Easter <MikeE@ster.invalid> writes: > VanguardLH wrote: >> Don't use Thunderbird anymore, but I recall it could test using regular >> expressions. > > Definitely not; no regex, no wildcards, no scoring. That's a big Tb > weakness compared to 'real' news agents. Theoretically, you can run local server (Hamster for Windows, for example) and do regex matching using internal scripts of the local server software or external scripts operating directly on article spool.
[toc] | [prev] | [next] | [standalone]
| From | "J. P. Gilliver (John)" <G6JPG@255soft.uk> |
|---|---|
| Date | 2021-04-26 03:41 +0100 |
| Message-ID | <QQqE$cMJhihgFwwH@255soft.uk> |
| In reply to | #241 |
On Sun, 25 Apr 2021 at 18:51:06, Sticks <Wolverine01@charter.net> wrote (my responses usually follow points raised): >So how do I use the Newsgroups portion of the header to filter on any >post that has more than 2 groups in it? > >I'm guessing if I use a comma (,) I could filter if it has more than >one, but for now I'd like to try two. That should eliminate most of >the blatant shitheads. I don't know TB's syntax, but I _think_ it would be difficult to filter on _only_ two; as you say, detecting a comma (in either To: or Followup-To:) will detect two _or more_. (Which, as you say, will eliminate most.) If you only want to detect ones aiming at three _or more_, then searching for two commas (with wildtext between - that would be ,.*, in Turnpike - a post in this thread suggests it might be ,.+, in TB) should work; specifying _exactly_ three (i. e. two and only two commas) would I think be difficult. -- J. P. Gilliver. UMRA: 1960/<1985 MB++G()AL-IS-Ch++(p)Ar@T+H+Sh0!:`)DNAf War doesn't determine who's right. War determines who's left.
[toc] | [prev] | [next] | [standalone]
| From | Miguel Tomar Nogueira <mnogueira@mail.telepac.pt> |
|---|---|
| Date | 2021-04-26 06:19 +0000 |
| Message-ID | <s65m1p$6aj$2@neodome.net> |
| In reply to | #241 |
Sticks <Wolverine01@charter.net> writes: > So how do I use the Newsgroups portion of the header to filter on any > post that has more than 2 groups in it? > > I'm guessing if I use a comma (,) I could filter if it has more than > one, but for now I'd like to try two. That should eliminate most of > the blatant shitheads. Are you already downloading all the headers? If not, you might want to filter on Xref. Xref is usually in overview, while Newsgroups is not.
[toc] | [prev] | [next] | [standalone]
| From | Andy Burns <usenet@andyburns.uk> |
|---|---|
| Date | 2021-04-26 08:12 +0100 |
| Message-ID | <ien7f1Fiv1cU1@mid.individual.net> |
| In reply to | #241 |
Sticks wrote: > So how do I use the Newsgroups portion of the header to filter on any > post that has more than 2 groups in it? > > I'm guessing if I use a comma (,) I could filter if it has more than > one, but for now I'd like to try two. That should eliminate most of the > blatant shitheads. Not sure you can ... the "contains a comma" method does work for two or more groups, but there isn't a built-in regex search, and I'm not sure if any of the regex addons still work?
[toc] | [prev] | [next] | [standalone]
| From | Andy Burns <usenet@andyburns.uk> |
|---|---|
| Date | 2021-04-26 08:16 +0100 |
| Message-ID | <ien7mvFj01fU1@mid.individual.net> |
| In reply to | #247 |
Andy Burns wrote: > I'm not sure if any of the regex addons still work? FiltaQuilla 3.1 claims to work again with TB78 <https://addons.thunderbird.net/en-US/thunderbird/addon/filtaquilla/versions/>
[toc] | [prev] | [next] | [standalone]
| From | Sticks <Wolverine01@charter.net> |
|---|---|
| Date | 2021-04-28 17:34 -0500 |
| Message-ID | <s6cnt4$tln$1@dont-email.me> |
| In reply to | #241 |
On 4/25/2021 6:51 PM, Sticks wrote: > So how do I use the Newsgroups portion of the header to filter on any > post that has more than 2 groups in it? > > I'm guessing if I use a comma (,) I could filter if it has more than > one, but for now I'd like to try two. That should eliminate most of the > blatant shitheads. Thanks to everyone for the suggestions. I decided to just use a single comma and only allow what is posted to the group I am reading. I can live with that.
[toc] | [prev] | [standalone]
Back to top | Article view | alt.comp.software.thunderbird
csiph-web