You are not logged in. Please login or register.
PunBB Forums → Posts by MattF
Just on a slight side track with regards to this, are the db tables search_cache, search_matches and search_words only used for single byte character sets, i.e: not UTF-8? If so, and everything is running and converted to using UTF-8, code wise, is it safe to drop those tables and any reference to them throughout the code?
Cheers.
Cheers. 
It's just been one of those rare things where I knew it must have some use, but couldn't quite figure what. 
So am I correct in surmising that it's main use, (and apologies if I'm being a bit on the thick side here), is that someone can link to information on another site and have it verified as belonging to them, and if so, it's imported or linked via whatever means the other site or service uses?
Again, my apologies if I'm being a bit dense here.
That only mentions the benefits of hcards, as such. And those I can see a use for if they're ever requested or needed. MicroID, however, seems to be useful but not actually have any use, if that makes sense?
Edit: Or is it of use in the context of verifying that the origin of the other micro-formats are legitimate?
Just a quick question. Been browsing the dev site and seen some of the features of 1.3 listed. One of them is support for MicroID. Hence the question. What actually is the point of MicroID? I've looked over the docs on their site too, and I honestly can't conceive any actual use for it. Am I just missing the point again?
They all have their plus points and downsides.
Possible solutions to avoid II (bot registering)
II.1 - IP control (blacklisted IP or domain) - not suffisant
Good: Effective, plus external lists are updated frequently.
Bad: Possibility of false positives. Possible time delays introduced. Also, as some MX admins will attest to, could bite you in the arse if one of the sites decides to kill it's service. Some have been known to return a positive response to every request in such scenarios.
II.2 - User name and address analysis (mode difficult than for post)
What classifies as legitimate and not legitimate?
II.3 - Captcha, cf I.3
Good: Fairly effective.
Bad: Accessibilty. Plus, some captcha's are horrendous to use even when you have perfect vision.
II.4 - moderation (registration queing), cf I.4
Effective but resource, (human), intensive.
II.5 - Automatic deletion if no post during N days after registering)
Only viable if you allow guests free access to anything other than posting. If you require people to register to view certain areas, there's a good chance they'll only be registering to view content, and may never post.
Why not just use a session with matching image/answer in arrays for a random selection? Btw, it should be local to the server, not remote if possible. You'll run the possibility of time delays and no image if the external site is down, or for a number of other possible reasons.
Cheers. I already have it gzipping the html files and such, but for some reason, (and I'm sure it was due to some IBM techinfo), got it into my head that the CSS and JS should be left as is. Thought it seemed a bit silly not compressing them though.
Been trying to get rid of any excess of late, (or make any fancy features all on/off selectable at a minimum in the profile, as with smilies and such), so started wondering about those two sets of code again.
Thanks ever so much for the advice gents. Greatly appreciated.
Matt
I'm not bumping, honest. Been searching the web again regarding this, and I believe the answer is it's safe to do so nowadays? T'would appear old Netscape 4 was the main problem with doing this in the past? I've not been able to find any references to anything later having problems, but would still rather have a second opinion before I go ahead and attempt it.
Cheers again,
Matt
It definitely makes life easier. Functions were the main reason for changing things here. I've been running full E_ALL on errors for the last week or two, (just tracking down any slight typo's or such I may have missed), and that method has definitely saved a lot of function checking for required lang files, especially with the redirect function.
Can it cause any problems letting the httpd server compress CSS and JS files, or should they as a matter of course be compressed?
Cheers.
What exactly are you wanting to do?
Is there any easy way to incorporate a counter that tallies unique page loads/views, rather than just overall page loads/views, i.e: for the same user, within a certain time period, the page view counter would only be incremented once even if they refreshed or navigated back to the same page several times?
Cheers,
Matt
That script above is only part of an overall setup. As it is above, it wouldn't work within PunBB as is.
code wrong
not correct
I mean this in a nice way, but if you're going to give such useless responses regarding specific questions, then expect useless replies.
And what does not working mean?
Look over on PunRes. Those mods are available there.
At the end of the day, without making any parser unnecessarily long and complicated, you'll never catch every variant. It's an uphill battle which you'll never win. If users are purposefully bypassing known censor words, (in moderation, might I add, as what some filters [catch/term as offensive] is pathetic), then you should penalise the user, not try to anticipate every future move.
Same as in design: your work is finished when there's nothing left to remove; not when there's nothing more to add.
Love that. So true.
Just curious what you think about it, is it appropriate or is the barrier to big?
Can't see any problem with it personally, other than perhaps accessibility related, (but with all the stuff required in that regards, it's a losing battle anyhow). Might be worthwhile altering the alt= text on that image to just ship, however, rather than sailing ship.
It's similar to the V.I.P code mod on other forums software which you tend to see in use on quite a few sites. Quite like the concept myself.
I quite like that style.
Cheers. I can stop obsessing over that one then. Thanks again.
Well, I've just managed to isolate one occasion that triggered it. A client machine pulling the RSS feed. Is it safe to surmise that this is just due to some people still having the old layout cookies, due to not having logged in since I altered the cookie code?
Cheers.
I thought I might have misunderstood it slightly. I only gave it a cursory browse. (There appears to be a good several hours reading there).
Unless I'm mistaking what that info means, it's basically a case of you're dropping the PunBB search tables, and letting MySQL create it's own and then using MySQL's builtin fulltext search command(s)?
All other db types in 1.3 will just do searches as in 1.2.*?
PunBB Forums → Posts by MattF
Powered by PunBB, supported by Informer Technologies, Inc.