The Arcive of Official vBulletin Modifications Site.It is not a VB3 engine, just a parsed copy! |
|
|
vB Global Translator - Multiply your indexed pages & put search traffic on autopilot Details »» | |||||||||||||||||||||||||||
vB Global Translator - Multiply your indexed pages & put search traffic on autopilot
Developer Last Online: Dec 2013
**Text Removed**
Show Your Support
|
Comments |
#252
|
||||
|
||||
Ok, but i ran your duplication update a few days ago, i have to run it again for this update too?
|
#253
|
||||
|
||||
YES! As I wrote - before my update it is possible that data are duplicated, when some pages are translated in same time. spiders usually works with friends So since your last removal you have again duplication. It will not happen again after this update (ok it will but only for table wt_cache - it's mysql limitation, but tables with biggest amount of data will not have duplicated data again).
|
#254
|
||||
|
||||
Understood, i'll try again in a bit and then update the main install. Thanks.
|
#255
|
||||
|
||||
Does anyone use vBseo Relevant Replacement?
RR also be translated? Unfortunately, not on my forum :/ |
#256
|
||||
|
||||
Can someone help me understand this. The goal of this MOD is to get a flood of international traffic. This traffic is unlikely to stick around because of the language issue so whats the point of wanting all this traffic.
|
#257
|
||||
|
||||
Quote:
For me I use it for ad revenue, branding, future referrals from word of mouth to UK traffic, it's free exposure. Your mileage may vary. |
#258
|
||||
|
||||
Quote:
|
#259
|
||||
|
||||
We'll I tried again with your new update.
Had to get my host to run this bit as it times out in the browser. 4. Delete data duplication like described here (note this is very important - otherwise unique indexes will not be set): They say the query is stuck on delete from cleaner... So i am going to stick with the current update for now. |
#260
|
||||
|
||||
Quote:
wt_cache cannot have unique index for whole oryginaltext and tl, because oryginaltext is too big for that, and making it's unique only for part of it has some disadvantages. I made tests and mysql allows me to set unique index olny for first 323 letters from oryginaltest and whole tl. So if I do that, then data duplication will be a history in this mod - that's the good part. Worst part is that if we will have 2 different translations with same first 323 letters, then only first one will be cached, and second one will be translated by google every time when page will be generated. I also realize that it is rare to have 2 different posts which starts from such long same text (323 letters), but it is possible - especially when users like to quote others posts. I've just made a query and I have such cases in my cache. So we can have actual solution which allows data duplication, but we know how to delete duplicated data, and once translated text will never be translated again. Or change it to solution where duplication never happens, but we have small chance that some texts will be translated over and over again. Usually when someone quotes other long post, and removes some part at the end – in such case we have 1 long translation of whole original text and one long translation of shorter quoted text. Now both are cashed - if we set unique index on wt_cache then one of those will be translated each time when page is generated. So - If users want this solution I can give the solution. Who wants it? But , till wt_cache allows for data duplication - I advice to let cleaner and saver tables stay (you can delete all data from there). Empty tables doesn't bother anyone, and can be helpful for cleaning wt_cache from time to time. In my installation I will keep those at least till I crawl all translation sites with my spider. And probably let them stay after that too. |
#261
|
||||
|
||||
Quote:
Note that this step if very time consuming and probably you will need to connect to DB with other client that WWW. Remember - always you can use the easy way New solution is better and faster. For those who makes fresh install it is same easy as it is right now. For update - 2 ways easy: one and good one. If you cannot handle the good one, and still want have your cache - so ok, you can stay with old version. But I think that for users it would be good if you test new solution in fresh install (if have difficulties on old one) - see it works fine, and make it official release - for all new users, who I think would like to have faster DB without data duplication. |
Thread Tools | |
|
|
X vBulletin 3.8.12 by vBS Debug Information | |
---|---|
|
|
More Information | |
Template Usage:
Phrase Groups Available:
|
Included Files:
Hooks Called:
|