IBM Security i2

Expand all | Collapse all

SDK

Jump to Best Answer
  • 1.  SDK

    Posted Mon November 30, 2020 09:56 AM
    Afternoon everyone, I was wondering if someone could assist me in an add on? I have seen it done before on previous versions of I2, but I have not been able to utilize the function on version 9. The add on is the ability to search (ctrl+f) multiples items. Natively a user can only search one item at a time. Having the ability to search multiples at once has been an extreme timesaver for our users. I don't know if one has to access the sdk to create the tool. Looking forward to your advice or contributions. Thank you in advance.


    Emmi

    ------------------------------
    a emmi
    ------------------------------


  • 2.  RE: SDK

    Posted Tue December 01, 2020 03:08 PM

    Hi Emmi,

    I suppose you are talking about Analyst's notebook.
    When you press ctrl+F you will get a window with setting. On the right-hand side, under "Action" you can choose between "select all: (Default) and "Select First".

    When you choose "Select First"  you can step through the selection with the F3-button.

    Does that answer your problem?



    ------------------------------
    Jan-Willem Lapoutre
    Senior software developer
    DataExpert b.v.
    Veenendaal
    ------------------------------



  • 3.  RE: SDK

    Posted Tue December 01, 2020 11:10 AM
    Hello.

    Do you mean search in a diagram whit multiple entities?

    Try to do a condicional formating, for example, if the entities have the name you are searching they get a red frame.

    If the search is in ibase instead of analyst, in ibase os possible to do that search whit out any add on.

    Do you have the add on?

    Best regards

    ------------------------------
    Paulo Ribeiro
    ------------------------------



  • 4.  RE: SDK

    Posted Tue December 08, 2020 04:34 PM
    Thank you for responding, but not actually. What I am looking for is the ability to use the ctrl+f function and search multiple entitles at once, not just one at a time, within I2. I have seen the plugin created before on the instances of I2, but I do not have the plug in myself.

    ------------------------------
    a emmi
    ------------------------------



  • 5.  RE: SDK

    Posted Wed December 09, 2020 02:34 AM
    Hi Emmi,

    I am afraid that I still don't understand what you want to do.




    When "Select Frist" and "Select All" isn't what you want to do, what or where else do you want to search?


    ------------------------------
    Jan-Willem Lapoutre
    Senior software developer
    DataExpert b.v.
    Veenendaal
    ------------------------------



  • 6.  RE: SDK
    Best Answer

    Posted Wed December 09, 2020 03:53 AM
    Hello Emmi & Jan-Willem

    Forgive me for being presumptuous, but I read Emmi's request as wanting the ability to run a bulk search of multiple terms.  This would be very similar to the In List function within iBase, and I agree it would probably be useful.

    To give just one hypothetical example, say there was a car number plate reader near to the scene of a crime, and we were able to extract 100 car registration numbers that passed the camera 5 minutes beforehand.

    If the data we wanted was in an iBase build, the search term would be e.g. 
    Car Registration Plate | In List | [the 100 car plates in a list or saved as a text file]
    ...a search that takes seconds

    Whereas if we had an ongoing i2 ANB chart listing all our local criminals and the cars they use, we would have to search those 100 registration nnumbers separately, and that can take hours.

    As we move towards ever larger datasets, bulk search tools will become more of a pressing need. 


    Ant

    ------------------------------
    Anthony Patamia
    ------------------------------



  • 7.  RE: SDK

    Posted Wed December 09, 2020 03:53 AM

    Hi

    i can't help re SDK however I am curious if there is another option - what particular feature of find text matters to you in this example? I ask because if I am looking for several things at a time - eg 3 people - say Vidal Picard, Francine Smith and David Thompson I would type all 6 words into the search bar, and not use find text at all - and it is highly likely the 3 people I want will be at the top of the results. 

    Cheers Mark



    ------------------------------
    Mark Fleet
    ------------------------------



  • 8.  RE: SDK

    Posted Wed December 09, 2020 04:11 AM
    Hello Anthony,

    Yes that would make sense. Didn't get that from Emmi posts. I can immagin somone worte a plugin for that. I would be interrested my self.



    ------------------------------
    Jan-Willem Lapoutre
    Senior software developer
    DataExpert b.v.
    Veenendaal
    ------------------------------



  • 9.  RE: SDK

    Posted Tue December 15, 2020 04:58 PM
    Exactly Anthony. I deal with bulk entitles often and having that ability either native or via an add on/plugin would be extremely helpful to my colleagues and I. Having this capability in ANB standard would be great for all users. We currently do not have IBASE in our version.

    ------------------------------
    a emmi
    ------------------------------



  • 10.  RE: SDK

    Posted Fri September 03, 2021 01:45 PM
    Edited by Karst van den Bosch Mon September 06, 2021 08:30 AM

    Hi Anthony, Emmi, 

    I'm not sure if this is still an active issue, but I'm new to the forums and was scrolling through some posts. 

    Would the complex search -> regular expression function not solve this issue, using a pipe character to separate the items?


    It isn't the neatest solution, but would allow you to search for multiple items at once. Though not sure if there is a limit to the search string length, I did a quick test in ANB and was able to find 466 entities without any issues (search string was 15394 characters long). 

    To quickly compile the list from new line separated values to pipe separated I used a find & replace in a text editor (e.g. Notepad++/Sublime).


    I hope this helps.



    ------------------------------
    Karst van den Bosch
    Consultant @ DataExpert
    ------------------------------



  • 11.  RE: SDK

    Posted Wed December 09, 2020 04:58 AM
    Let us hope that I have interpreted Emmi correctly. 

    I have the benefit that this forum is in English which is my first language, whereas I suspect for Emmi and yourself it is at best your 2nd language.  And whilst I do not only speak English, my other language skills are nowhere near strong enough for me to consider writing about data science nor software development - so my admiration to both of your for your presence here.

    Back to i2 ANB, and whilst a plug-in would be nice, it would be better if it was native to the original app.  Visual Search allows a low-level construction of Boolean search terms (an iBase-lite); perhaps that could be enhanced to include multiple terms.


    Ant

    ------------------------------
    Anthony Patamia
    ------------------------------



  • 12.  RE: SDK

    Posted Mon September 06, 2021 06:51 AM
    Edited by Anthony Patamia Mon September 06, 2021 07:46 AM

    Hello Karst

    Thanks for that.  I've been using a form of ANB or its predecessors for over 20 years now and I never knew that trick.  It's essentially the In List operator from iBase but within the ANB Find Text - exactly what I suggested Emmi needs.  No need for the software to be updated then.

    I just did my own experiment - nowhere near as large as yours - and it worked fine.  I'll just mention for anybody reviewing this in the future that for the text editor part of the process I used Find and Replace within Word (accessed via Ctrl + H or from the Home tab).  I pasted as plain text a list from Excel and then I replaced the paragraph marks = ^p with the pipe character

    [Edit: I tried to include a screenshot of Find & Replace - it failed three times, sorry]. 

    Incidentally I've only just noticed Mark's reply below about the Search bar.  I re-ran the same search using both a pipe character as a delimiter and a space character as the delimiter, and they both worked.  I think it is down to user preference as to deploying Find Text vs Search - both have pros and cons. 

    The only test that failed was I tried to copy a list from Excel into search and it wouldn't accept that (whereas in iBase using In List that is a viable way to set up the search parameters).
    Thanks again for that bit of advice - it will genuinely change my usage of ANB. 


    Ant 


    ------------------------------
    Anthony Patamia
    ------------------------------



  • 13.  RE: SDK

    Posted Mon September 06, 2021 08:20 AM
    Edited by Karst van den Bosch Mon September 06, 2021 08:30 AM
    Hi Anthony, 

    I'm glad I could help! 

    Microsoft Word
    Your Find & Replace trick in MS Word is very helpful as well, though it is important to remove the last pipe character so the string doesn't end with a pipe character (if you don't remove it, it results in an invalid error "invalid search string").

    For reference, the input (MS Word):

    result:


    Microsoft Excel
    I had to do some additional research and testing for getting this to work in Excel, and found that there are several ways you can achieve the same result. By far the easiest I was able to find is using the TextJoin function in Excel. TextJoin allows you to join rows and/or columns into a single string. Below example is of a simple list, but you can join a complete spreadsheet of rows and columns if you'd like to.

    Cell A1 to A5 (A1:A5) is the list. Cell C2 shows the result. Cell C3 shows the formula I used in cell C2.


    ------------------------------
    Karst van den Bosch
    Consultant @ DataExpert
    ------------------------------



  • 14.  RE: SDK

    Posted Mon September 06, 2021 08:43 AM

    Good point Karst

    I know about not keeping the final '|' from iBase In List searching - and I make it very clear when I deliver iBase training to remove it (it's normally critical when searching a large iBase build). 

    I did remove the last pipe before running the search in ANB, but didn't make that clear to others reading this thread - so thanks for highlighting it.  It is very important to do this.


    That tip about TEXTJOIN is clever but unfortunately not useful for my organisation just yet.  We're still on Excel 2016, and whilst Excel 2016 & 2019 & O365 are purportedly all the same "version", Excel 2016 lacks both TEXTJOIN and CONCAT (the newer version of CONCATENATE).  At some point in the coming year we're supposed to be moving away from Office 2016 to O365 by which time I may well have forgotten that tip.  But it is great advice that hopefully someone else will make good use of.

    I am however jealous of your ability to incorporate pictures into a reply.  I tried PNG and JPEG and I tried dragging the pic in and navigating via Windows Explorer - the pic shows up in the preview but then 'breaks' after I push 'upload'. 



    ------------------------------
    Anthony Patamia
    ------------------------------



  • 15.  RE: SDK

    Posted Mon September 06, 2021 09:54 AM
    Hi Anthony, 

    I don't use the "Upload file" function, but directly paste it into the chat box, while using the copy function from a screenshotting tool (e.g. Snipping Tool, a default Windows screenshot tool). 

    Regarding not having TEXTJOIN available, that's a shame... Though there are some other options. 

    Option 1
    Copy the list from Excel, and paste as plain text into Word.
    Now you can use your previous method to utilize Find & Replace in Word to replace the paragraph marks with a pipe character.


    Option 2
    This one is a bit less straight forward, but should still work in Excel 2016, utilizing the TRANSPOSE function.

    Step 1:
    Select the row of cells you want to use, and copy them (cut does not work)

    Step 2:
    Paste the cells using the Transpose option. After pasting it, make sure to remove the original data.

    Your data should now look like this:

    This format already works for the Search function in ANB. 

    Step 3:
    Here you have 2 options again.
    Option 1 is to save the file as a normal CSV, and continue to replace the comma's with a pipe character. 
    Option 2 is to first edit your Windows settings (control panel -> Region -> additional settings) to use a pipe character as a list separator (see https://www.howtogeek.com/howto/21456/export-or-save-excel-files-with-pipe-or-other-delimiters-instead-of-commas/ for the details). After this change, when you save your Excel file as a CSV, it will automatically use a pipe character as a delimiter instead of a comma. This option will save you quite some time in the long run if you regularly use this.




    ------------------------------
    Karst van den Bosch
    Consultant @ DataExpert
    ------------------------------



  • 16.  RE: SDK

    Posted Mon September 06, 2021 10:16 AM
    Edited by Anthony Patamia Mon September 06, 2021 10:17 AM
    Thanks.  Option 1 is how I tend to get data that cannot be cleansed easily in Excel into Word to use Find & Replace (but being very old skool I frequently deploy the Ctrl+Alt+V method).

    The transpose function in Excel is one I have made use of regularly, but never thought to use it this way.  And I've just tried then copying the transposed terms straight into ANB.  It didn't work with the Find Text box you suggested earlier, but was successful with Mark Fleet's suggestion of the Search function.  So it looks like we have a few options here. 

    And yes, I've now managed to paste in a screenshot - the one I wanted to post earlier.  Thanks. 


    ------------------------------
    Anthony Patamia
    ------------------------------