topic Re: Comparison of 2 strings in LabVIEW

Comparaison between 2 strings

fxblabview — Mon, 07 May 2018 15:33:34 GMT

Good afternoon,

I have a question about the comparition between 2 strings.

My VI write in a txt file a message to indicate the state of the test. If this state is unchange I don't want to write on this file that's why I test the message and the message write in the last line of the txt file to be sure they are not similar and write.

But the comparator tools I use don't give the expected solution.

Someone could explain me how compaire these 2 strings?

Thanks

Re: Comparaison between 2 strings

altenbach — Mon, 07 May 2018 15:37:17 GMT

@fxblabview wrote:

But the comparator tools I use don't give the expected solution.

Attach code instead of pictures
Look for differences in hidden characters (e.g. CR, vs LF, vs CRLF).
If the result is not as expected, you need to tell us what you expect!

Re: Comparaison between 2 strings

altenbach — Mon, 07 May 2018 16:08:46 GMT

To detect successive duplicate messages, use a feedback node and compare current and previous.

Re: Comparaison between 2 strings

billko — Mon, 07 May 2018 17:20:10 GMT

@fxblabview wrote:

Good afternoon,

I have a question about the comparition between 2 strings.

My VI write in a txt file a message to indicate the state of the test. If this state is unchange I don't want to write on this file that's why I test the message and the message write in the last line of the txt file to be sure they are not similar and write.

But the comparator tools I use don't give the expected solution.

Someone could explain me how compaire these 2 strings?

Thanks

You send the strings with CR as your EOL but your read is set to convert EOL. In Windows, EOL is CR + LF, so the changes the EOL and the strings no longer match. It's only a guess (but a good one) since you didn't supply real code, and the probes and stuff don't show the backslash codes.

Re: Comparaison between 2 strings

tyk007 — Mon, 07 May 2018 23:50:50 GMT

Since you only really care about the "readable" string content perhaps you could use the Trim Whitespace node on both before comparing them?

Re: Comparaison between 2 strings

fxblabview — Tue, 08 May 2018 06:16:54 GMT

Right the txt file replace my CR by an EOL. Thats why I should put a supplementary character on the reading to have complete sentence.

Thanks for your help.

Re: Comparaison between 2 strings

fxblabview — Tue, 08 May 2018 06:26:50 GMT

It's a good solution to not be bored by these invisible types.

Re: Comparison of 2 strings

GerdW — Tue, 08 May 2018 06:33:42 GMT

Hi fx,

It's a good solution

The better solution would be to store the previous message in a feedback node for comparison: ne need to read the file before writing to it! (As has been suggested before!)

Simple like this:

Re: Comparaison between 2 strings

fxblabview — Tue, 08 May 2018 06:33:43 GMT

Sorry Knight to my approximations about my first post on the forum.

My expected response was that the comparator tools give a false response when the 2 sentences tested were similar... and a true when they were different.

Re: Comparaison between 2 strings

GerdW — Tue, 08 May 2018 06:36:19 GMT

Hi fx,

My expected response was that the comparator tools give a false response when the 2 sentences tested were similar... and a true when they were different.

The "unequal?" comparison gives TRUE when the inputs are unequal (aka "not the same").

When you are talking about this "unequal?" comparison then your expectation was right. 😄

Re: Comparaison between 2 strings

billko — Tue, 08 May 2018 06:39:33 GMT

@fxblabview wrote:

Right the txt file replace my CR by an EOL. Thats why I should put a supplementary character on the reading to have complete sentence.

Thanks for your help.

Actually, you can configure the read to NOT translate the EOL.

Re: Comparison of 2 strings

fxblabview — Tue, 08 May 2018 07:42:39 GMT

Hi GerdW,

I tried your suggestion, with the retroactive node (in structures menu) with a direction change.

But the boulean value is never at false, it looks like my retroactiv node don't stock the precedent value.

Thanks fo your help

Re: Comparison of 2 strings

Yamaeda — Tue, 08 May 2018 07:50:55 GMT

Probe the wires and check, it definatly stores the previous value.

Re: Comparison of 2 strings

fxblabview — Tue, 08 May 2018 07:51:33 GMT

GerdW,

It works!!

great simplification of the code...

Thanks

Re: Comparison of 2 strings

altenbach — Tue, 08 May 2018 14:51:18 GMT

@fxblabview wrote:
I tried your suggestion, with the retroactive node (in structures menu) with a direction change.

But the boulean value is never at false, it looks like my retroactiv node don't stock the precedent value.

Please keep the terminology straight to avoid confusion, It's a "feedback node" and that's what I was suggesting already in message #3 above. The "direction change" has no functional meaning and just solves a cosmetic issue giving fewer backwards wires and wire bends. Both "directions" give the same result.

One thing you need to be careful about when testing with a "globally initialized on first call" FN is the fact that it will not work as you might expect if the code is not in a loop and you re-rerun the VI several times. Even if you keep the value the same, each new run will compare with an empty string instead. Make sure you understand all initialization options fully.