parsing single expression trouble [Solved x3]

[fonpaolo]

Thanks fonpaolo - incidentally the "." at the end could be misconstrued as part of the RegExp example
...

Usually, the .* is used as "skip everything after", maybe you can add $ at the end.

[Mor3bane]

Usually, the .* is used as "skip everything after", maybe you can add $ at the end.

Cool! Thanks for explaining that

[fonpaolo]

You're welcome.
...and happy New Year!

[Mor3bane]

So I have gotten 3 suggestions I felt I could use in my case.

I will list all three:

1: "(?siU)<ut>(.*)</ut>.*<dnam>(.*), Australia</dnam>.*<hi>(.*)</hi>.*<low>(.*)</low>"

2: "(?siU)<ut>(.*)</ut>.*<dnam>(.*), (.*), (.*)</dnam>.*<hi>(.*)</hi>.*<low>(.*)</low>"

3: "(?siU)<ut>(.*)</ut>.*<dnam>(.*)\W+\s+(.*)\W+\s+\w+</dnam>.*<hi>(.*)</hi>.*<low>(.*)</low>"

#1 is specific to the output of my location - so not portable for sharing really.
#2 allows me to reference each string segment with a StringIndex
#3 breaks down the string leaving only the first segment for output, however this RegExp can also reference the second segment with an incremental StringIndex. This version leaves me with fewer StringIndexes to worry about (I think ).

Hope I got that right

[balala]

Code: Select all

(?siU)<dnam>(.*)\W+\s+(.*)\W+\s+\w+<\/dnam>

The slash (/) is not a reserved character in regular expressions. The reserved characters are: . ^, $, *, +, ?, (, ), [, {, \ and finally | (https://forum.rainmeter.net/viewtopic.php?p=124711#p124711). So, you don't have to escape it with a backslash in <\/dnam>. You can use the simple </dnam>. It'll work equally well.

[balala]

For every (.*) there will increment another StringIndex

Yes?

Well, seems to work that way in my first test.

You're right. For example in my RegExp=(?siU)<dnam>(.*), (.*), (.*)</dnam>, the measure with StringIndex=1 will relate to the first captured string, before the first comma (in this case Mount Isa), StringIndex=2 to the second captured string (QNS), while finally StringIndex=3 to the last captured string (Australia).

[Mor3bane]

You're right. For example in my RegExp=(?siU)<dnam>(.*), (.*), (.*)</dnam>, the measure with StringIndex=1 will relate to the first captured string, before the first comma (in this case Mount Isa), StringIndex=2 to the second captured string (QNS), while finally StringIndex=3 to the last captured string (Australia).

Thanks for confirming balala,
I keep heading back to the RegExp section of the manual, and realised with the HTML code for a page one can request sepcific parts and not have to parse the whole page as it would seem from some examples I have seen in other skin suites.
That made things much simpler, then breaking down selective syntax became simpler also.
As always, I appreciate the input from the more experienced Rainmeter hobbyists.

Happy New Year!

[balala]

one can request sepcific parts and not have to parse the whole page

Right. Parsing the whole page very rarely is a good idea. Usually parsing the needed parts is a better one.
Happy New Year to you too.