5 Random SE Auto Selection Button :)

seagul

I like to select 5 random Google SE (+2 yahoo). Would be nice to have auto feature select 5 random different Gs from last time. Time saver for sure.

Or go one step further and choose 5 random at a time interval say every week.

Ozz

create it by your own ... somehow

1. create a file called user_se.dat with a text editor

2. copy/paste those lines 5 times into the file

3. rename the line [google 1/RND: Country] ... [google 5/RND: Country] and save the file

4. move the file to you %appdata%/GSA Search Engine Ranker folder

5. restart SER and select the new created SEs in your project options

6. done

if you like to do it for Yahoo as well just study the se.dat file you find in your installation folder of SER and i'm sure you'll find a workaround for this too.

thats why i love SER so much. you are able to mod more or less everything and just need to be creative

seagul

Ya I will try that thanks Ozz

LeeG

Ozz, your a true star

That share, is way beyond genius

If you was a female, twenty pint, five bagger, I would be offering to inseminate you for that share

If it pulls a random google on each time its accessed, at an educated guess, you would probably only need it ticked once

Ozz

if you like to do it a step further than just keep the "bigger" well indexed Google in the list as some smaller Google engines doesn't give that much results if i'm not wrong (if its not a common footprint like "powered by wordpress").

so just keep Google like US, UK, CN, RU, DE, FR, ZA, CA, JP, KR, ES, BR, AR, IN, IT, MX and so on until you have like 20-30 of them. there is no need to use the exotic ones imo.

for testing i suggest to use the "search online for urls" tool.

LeeG

Is that every google from ser in that file Ozz?

Im on a ser maintenance day, checking 1.2 million links to see if they are live

Not much I can do until that lots run through

Its a blinding share to say the very least

Ozz

thats one of my SE mods i already did some days ago. as there are only ~60 google engines in that list its more or less pre-filtered.

LeeG

Nice to see you still aint given up testing and tweaking

Right chuffed with you sharing that

Kreist

This is sweet. So, is it random each time a project does a new search? SER is started up? Thanks Ozz

Ozz

yes, and with 5 of them (differently named) you are searching at 5/60 google SE at the same time. thinking about this mod i believe it may be better to remove the "link:" line from it. thats only used for "competitor search" and you can do that with the normal Google INT search engine.

LeeG

I will admit to being lazy prior to asking this Ozz

Are the two google internationals and the other similar one included or are they just the country specific ones?

 

I might even offer to answer AlexR's questions for you for an hour after you shared this

Ozz

US and UK are included, INT not. if you like to use them exclusive  than search for ".com/" within the lines and delete both. you get a clue about the origin of the SE when you scan the "Country=" operator and the domain in that file in case you want to use some other SEs exclusively.

i'm not 100% sure though what you meant with "the two google internationals". 

the mod itself just uses the lines for search queries of the other Google SEs in spintax format and were just copy/pasted to this file.

seagul

Awesome info here! Will try tomorrow.

dogGoogles

@Ozz . . . I must be missing something totally simple, but I don't see the new search engines when I restart SER.  I've placed the user_se.dat file into my  %appdata%/GSA Search Engine Ranker folder, but I don't seen them in SER.  Are they supposed to just appear with all the other search engines on the Project->Options page?

Hunar

They Will Show up as whatever you named them when you replaced the [google 1/RND: Country]

I named mine Google Random 1, 2nd copy I named that Google Random 2 etc. etc. I scrolled down in the SER engine list and there it was.

Ozz

@dogGoogles: if you've done everything like i told you then they will show up.

to make sure the file is named and placed correctly you can add a dummy SE first when you click the "Add" button nearby the SE selection field. just fill the field with some nonsense, save it and go to your appdata folder. then you open the user_se.dat file that was created by SER with a text editor and modify the file with the code from pastebin. save that file, restart SER and scroll down to where the google SEs are liste where you should find them according to the name you have given them like "google 1/ RND: Country" for example.

dogGoogles

Thanks @Ozz @Hunar Apparently SER was running under my Administrator user, and not the other user on my VPS.  Didn't realize that before.  The trick with adding a dummy SE showed me where the file was.

LeeG

Im sure this is the day Ozz reinvented the wheel =D> ^:)^

Ozz

i suppose its working well then

i use that mod for some time now but i didn't thought of using more than one instance of it until @seagul asked for it

Meatplow

What I did was give them a unique name, like "Betsy1" etc... That way if you have a lot of projects, you can select by mask, without having to scroll to check them. You could also, name them ***Custom1, so they show up first and you can tick them without scrolling

dogGoogles

It's a pretty slick fix.  I was thinking about a feature like this the other day, when I was trying to update a bunch of projects, but couldn't use the mass update feature since I didn't want them all to have the same SE's.  Thanks @seagul for actually asking the question as well as @ozz for the ultra cool mod.  Add to the fact that my LPM seems to be at an all time high of 80+ and I'm just thrilled

LeeG

More va va voom is easy.

Just clear the target url cache after adding all the new random engine system

 

Ozz, did you try just running one instance of this

At an educated guess, you would not need more than one anyway, since every engine call would be random

 

Ozz ^:)^ ^:)^ ^:)^ ^:)^

Startrip

It's working. I'll do the same for yahoo now to mix that in too (less banned proxies maybe )

Ozz

<Ozz, did you try just running one instance of this

yes, i did that all the time until yesterday. atm i'm using 3 instances so it searches for 3/60 with the same footprints every time. i'm not 100% sure though how it scrape the pages if there are more than 100 results. in the end it shouldn't make a difference if you are using 1 instance with 10 second query time or 3 instances with 30 seconds (if i'm not wrong).

@Sven: whenever we get more than 100 search results than it caches the search query and just change the page number, right? i hope SER does it this way and doesn't change the url for the SE as well like "google US / page 1" and then uses "google DE / page 2" and so on.

Startrip

Works like a charm with yahoo too. Not much less speed than G I must say...

dogGoogles

@LeeG  why does clearing the target cache add more va va voom?  Doesn't that just mean we're clearing out targets that SER has already done the work to identify, but hasn't posted to yet?

LeeG

Try it and see.

All you do is clear the present list of targets

So not too much at stake in all honesty

If you have been using an engine that's been giving limited or poor results, the new Ozz method should bring in more results and rely less upon the global sites lists

Sven

@Ozz no, it creates the URL and just changes the page number

rodol

@sven just implement this option on SER =D

tommy99

+1

Ozz

if i've understood Sven correctly regarding my last question than it make no sense to add this at current state.

i see a problem as the random SE domain isn't cached. so it could happen that you find 1000 results with "Google US" and SER scrape the results of page one. now it moves a page forward to get the results of page two and uses because of the random url spintax "Google SE" this time. "Google SE" has only 120 results overall so SER will stop to scrape for more results at page two because it thinks that it got all results. all the other remaining 7 pages of the original search in "Google US" will be skipped.

but again, i'm not 100% sure if i've understood Sven correctly. it could also be possible that SER just uses a random page for each search and each engine. than it wouldn't matter and there were no differences to the searches like you do with "normal" search engines (and doesn't start from page one each time when doing its initial scrape).