"Scan From String" and > symbol

[pyatachyok]

I want to scan time from string like "10>20>30" (12 hrs, 20 mins and 30 secs). I cannot build proper format string. For "10_20_30" format "%<%H_%M_%S>t" would work but it is impossible to include ">" symbol instead of "_". I tried to screen it with backslash ("%<%H\>%M\>%S>t") but it doesn't work neither.

Another funny stuff with scanning time values. If I have string "10_20_30" and try to scan with "%<%H____%M_%S>t" or "%<%H_sdfdf__%M_%S>t" format string, everything works while I would rather expect "LabVIEW:  Scan failed. The input string does not contain data in the expected format." error.

The reason I try to scan such freaky strings is that I want to write a routine to scan an arbitrary data from string; I wanted to split the format strings into several one-scan parts scanning each part into the appropriate type Variant with "Scan From String" routine. To do this I need to figure out proper rules determining format specifier boundaries. I can develop my own rules conforming with my applications (surely the examples above are extremely freaky) but I thought its better to mimic "Scan From String" behaviour.

Are there any strict official format specifier rules?

PS: I work with LabVIEW 2011.

[Darin.K]

Try escaping with the % character (I can't test myself right now):

%<%H%>%M%>%S>t

[pyatachyok]

Funny enough, it works. And the same strange behaviour (as in case of underscores or other symbols) exists: even if I have "%<%H%>abc%>%M%>%S>t" format string, it scans "10>20>30" as 10 hrs, 20 mins, 30 secs, "forgetting" excessive symbols in delimiter (so in case of "%<%H%>abc%>%M%>%S>t" format string it

• will read "10>20>30", "10>ab>20>30", "10>>20>30" or "10>abc>20>30" correctly,
• will read "10>>>20>30" as 10 hrs, 0 mins, 20 secs leaving current position at last ">",
• and will not read "10>abb>20>30" or "10>ac20>30").

Very strange behaviour. It would be good to find official rules or just develop my own for my own routine as the last example really seems quite weird.

[ben64]

---

pyatachyok a écrit :

Funny enough, it works. And the same strange behaviour (as in case of underscores or other symbols) exists: even if I have "%<%H%>abc%>%M%>%S>t" format string, it scans "10>20>30" as 10 hrs, 20 mins, 30 secs, "forgetting" excessive symbols in delimiter (so in case of "%<%H%>abc%>%M%>%S>t" format string it

• will read "10>20>30", "10>ab>20>30", "10>>20>30" or "10>abc>20>30" correctly,
• will read "10>>>20>30" as 10 hrs, 0 mins, 20 secs leaving current position at last ">",
• and will not read "10>abb>20>30" or "10>ac20>30").

Very strange behaviour. It would be good to find official rules or just develop my own for my own routine as the last example really seems quite weird.

---

That make sense but it is not obvious.

First it searches a match for %H

Then it will try to match >abc, Partial possible matches are >, >a, >ab and the complete match >abc. Since these are not time format they will not modify the result ( any other combination will throw an error: >b, >abb, ...)

It will then search for the next match : > and then for a match to %M

It will then search for the next match : > and then for a match to %S.

Look at the case 10>>>20>30

10 matches %H

the first > matches partially %>abc (the scan position is now here: 10>|>>20>30)

it now look to match the %> before %M  (the scan position is now here: 10>>|>20>30)

It now try to match %M but the next character is > so there is no match (return 0) (the scan position is still here: 10>>|>20>30)

It now try to match the %> before %S  (the scan position is still here: 10>>>|20>30)

It now try to match %S and find 20  (the scan position is still here: 10>>>20|>30)

There is no more things to match in the format string so >30 is returned in the remaining string output of the Scan From String function.

Ben64