Sikuli sometimes not activated, causing some visual automation to fail

[siowyisheng]

This fails if it is the whole script, although it is basic usage of visual automation.

if present('test.png')
    echo 5

If I just activate sikuli separately using a click on coordinates, then it works.

click (1,1)
if present('test.png')
    echo 5

To fix.

[kensoh]

Oh.. Yes, there isn't a mechanism to detect this. The assumption is if a user uses exist() and present() for images, there is most likely other steps in the workflow that does visual automation. Thus those steps would already trigger sikuli. Very unlikely that a workflow uses exist and present image without any other sikuli VA steps. Though not impossible.

[adegard]

Hi guys, Click text using ocr seems very attractive new feature!

Is there still needed Sikuli, for that? There ins't instructions on the new V6 webpage Could you explain Sikuli installation?

thanks

[siowyisheng]

Sikuli is packaged together with tagui in the zip file. No need for a separate installation.

On Sun, Mar 22, 2020, 19:33 adegard notifications@github.com wrote:

Hi guys, Click text using ocr seems very attractive new feature!

Is there still needed Sikuli, for that? There ins't instructions on the new V6 webpage Could you explain Sikuli installation?

thanks

— You are receiving this because you authored the thread. Reply to this email directly, view it on GitHub https://github.com/kelaberetiv/TagUI/issues/756#issuecomment-602185220, or unsubscribe https://github.com/notifications/unsubscribe-auth/AGB7JBEZGFHCN447WMNMIDDRIXZQHANCNFSM4LGGXOPA .

[adegard]

OK Yi, thank you, however can you better explain the usage of 'using OCR ' feature. I was trying in live, but it is not responding if I indicate some text to be clicked in webpage...

Il dom 22 mar 2020, 12:38 Yi Sheng notifications@github.com ha scritto:

Sikuli is packaged together with tagui in the zip file. No need for a separate installation.

On Sun, Mar 22, 2020, 19:33 adegard notifications@github.com wrote:

Hi guys, Click text using ocr seems very attractive new feature!

Is there still needed Sikuli, for that? There ins't instructions on the new V6 webpage Could you explain Sikuli installation?

thanks

— You are receiving this because you authored the thread. Reply to this email directly, view it on GitHub <https://github.com/kelaberetiv/TagUI/issues/756#issuecomment-602185220 , or unsubscribe < https://github.com/notifications/unsubscribe-auth/AGB7JBEZGFHCN447WMNMIDDRIXZQHANCNFSM4LGGXOPA

.

— You are receiving this because you commented. Reply to this email directly, view it on GitHub https://github.com/kelaberetiv/TagUI/issues/756#issuecomment-602185788, or unsubscribe https://github.com/notifications/unsubscribe-auth/AFTO2EP6X4YW5RMQGXIFKLDRIX2CTANCNFSM4LGGXOPA .

[siowyisheng]

Sure. I'll also update the docs. Thanks for pointing out the shortcoming.

click some text using ocr

This looks for 'some text' on the user's screen using ocr and moves the mouse to the middle of the found matched text (from top left of the screen to the bottom right) and clicks it. Sometimes it fails because ocr is not 100% reliable.

It's most useful with click, rclick and dclick steps.

When using it in live mode, make sure your command prompt is below your target text, or it will likely click the matched text within your command prompt itself.

On Sun, Mar 22, 2020, 20:30 adegard notifications@github.com wrote:

OK Yi, thank you, however can you better explain the usage of 'using OCR ' feature. I was trying in live, but it is not responding if I indicate some text to be clicked in webpage...

Il dom 22 mar 2020, 12:38 Yi Sheng notifications@github.com ha scritto:

Sikuli is packaged together with tagui in the zip file. No need for a separate installation.

On Sun, Mar 22, 2020, 19:33 adegard notifications@github.com wrote:

Hi guys, Click text using ocr seems very attractive new feature!

Is there still needed Sikuli, for that? There ins't instructions on the new V6 webpage Could you explain Sikuli installation?

thanks

— You are receiving this because you authored the thread. Reply to this email directly, view it on GitHub < https://github.com/kelaberetiv/TagUI/issues/756#issuecomment-602185220 , or unsubscribe <

https://github.com/notifications/unsubscribe-auth/AGB7JBEZGFHCN447WMNMIDDRIXZQHANCNFSM4LGGXOPA

.

— You are receiving this because you commented. Reply to this email directly, view it on GitHub <https://github.com/kelaberetiv/TagUI/issues/756#issuecomment-602185788 , or unsubscribe < https://github.com/notifications/unsubscribe-auth/AFTO2EP6X4YW5RMQGXIFKLDRIX2CTANCNFSM4LGGXOPA

.

— You are receiving this because you authored the thread. Reply to this email directly, view it on GitHub https://github.com/kelaberetiv/TagUI/issues/756#issuecomment-602192531, or unsubscribe https://github.com/notifications/unsubscribe-auth/AGB7JBF56P2VDJBXPTSF6KLRIYAGVANCNFSM4LGGXOPA .

[adegard]

ok, perfect, I found the way, now. thank. It tooks some time (around 2 min) and pointed to the words when they were at beginning. So, my understand is that it takes a picture and take time to analyse it. Then I need to don't move meanwhile he is searching.

[kensoh]

Hi @siowyisheng, I created https://github.com/kelaberetiv/TagUI/pull/774 which fixes this issue using regex string matches.