[Solved] Regex pattern need help

[guner7]

Hello,

I need some help to parse the Green highlighted value with from below text:

RESISTOR  THICK FILM 4.64K ±1% 1/4W ±100PPM/°C 1206 SMT
RESISTOR  THICK FILM 3.83K ±1% 1/4W ±100PPM/°C 1206 SMT
RESISTOR CARBON FILM 22K ±10% 1/2W AXIAL THT
RESISTOR  WIREWOUND  22 ±5% 3W ±30PPM/°C AXIAL THT
RESISTOR  METAL OXIDE 4.7K ±5% 2 W ±300PPM/°C AXIAL THT
RESISTOR  THICK FILM 0   1/8W  0805 SMT

I am using positive look behind.:

(?<=FILM|WOUND|OXIDE).+

Can only pull this off:

4.64K ±1% 1/4W ±100PPM/°C 1206 SMT
3.83K ±1% 1/4W ±100PPM/°C 1206 SMT
22K ±10% 1/2W AXIAL THT
 22 ±5% 3W ±30PPM/°C AXIAL THT
4.7K ±5% 2 W ±300PPM/°C AXIAL THT
0   1/8W  0805 SMT

I'm trying the \b word boundary to no avail at this point. Appreciate if anyone would guide me on this?

[FrancescoDiMuro]

@guner7

You don't need to use lookbehind (even if you could), and since you are trying to extract digit(s) only, plus the character K (which could be replaced as \w* if you have any other engineering unit), you could use a pattern like this one:

'(?:FILM|WIREWOUND|OXIDE)\s*(\d+\.?\d*K?)'

[guner7]

9 minutes ago, FrancescoDiMuro said:

'(?:FILM|WIREWOUND|OXIDE)\s*(\d+\.?\d*K?)'

That works like charm. Thank you.

Why \s* able to tell regex not to capture?

[FrancescoDiMuro]

@guner7
You're welcome.

1 hour ago, guner7 said:

Why \s* able to tell regex not to capture?

Everything that is in a pair of round parenthesis is captured, as long as there's no captouring groups, atomic groups, or lookaround.
Since the \s* is outside the round parenthesis, it is not captured

[iamtheky]

think you can also pull it off focusing on the K (or lack thereof)

#include<array.au3>

$s = "RESISTOR  THICK FILM 4.64K ±1% 1/4W ±100PPM/°C 1206 SMT" & @LF & _
"RESISTOR  THICK FILM 3.83K ±1% 1/4W ±100PPM/°C 1206 SMT" & @LF & _
"RESISTOR CARBON FILM 22K ±10% 1/2W AXIAL THT" & @LF & _
"RESISTOR  WIREWOUND  22 ±5% 3W ±30PPM/°C AXIAL THT" & @LF & _
"RESISTOR  METAL OXIDE 4.7K ±5% 2 W ±300PPM/°C AXIAL THT" & @LF & _
"RESISTOR  THICK FILM 0   1/8W  0805 SM"

$a = stringregexp(stringstripws($s , 4) , "(?m)(\d+\.*\d*?K*)(?:\s+).*" , 3)
_ArrayDisplay($a)

 

[mikell]

And assuming that there are no digits in the item names, it could even be done like this

$a = stringregexp($s, "(?m)^\D+(\S+)" , 3)

... my 2 cents

[jchd]

Don't forget resistors of high and low values: MΩ, GΩ (if working with very high voltage), mΩ and even µΩ (e.g. current sense resistors). Also, kilo is k, not K but that is possible to cope with this.

It's probable that a BOM can refer to "CAPACITOR MKP X2", hence digits in designations may occur, depending on the nomenclature used.

Provided fractional values have a leading zero (e.g. 0.27 = 270mΩ):

#include <array.au3>

Local $sRes =   "RESISTOR  THICK FILM 4.64K ±1% 1/4W ±100PPM/°C 1206 SMT" & @LF & _
                "RESISTOR  METAL STIRP 4.7m ±0.5% 3 W ±100PPM/°C AXIAL THT" & @LF & _
                "RESISTOR  THICK FILM 3.83K ±1% 1/4W ±100PPM/°C 1206 SMT" & @LF & _
                "RESISTOR CARBON FILM 22K ±10% 1/2W AXIAL THT" & @LF & _
                "RESISTOR  WIREWOUND  22 ±5% 3W ±30PPM/°C AXIAL THT" & @LF & _
                "RESISTOR  METAL OXIDE 4.7K ±5% 2 W ±300PPM/°C AXIAL THT" & @LF & _
                "RESISTOR  METAL OXIDE 1.13G ±1% 1/2 W ±50PPM/°C AXIAL THT" & @LF & _
                "RESISTOR  THICK FILM 0   1/8W  0805 SM"

$a = StringRegExp($sRes, "(?m)^[\w\s]+\s(\d+\.?\d*[GMKmµ]?)\s", 3)
_ArrayDisplay($a)