(Xpath) How do I get a specific numeric value?

Question

How do I get a specific numeric value?

example image

how can I get only the digit "6" ?

example "ตอนที่ 6" i need only output for "6" or " 6 "

i'm try //li[@class='active']/text() for simple get text, but i not sure how should i do next

Have a look at substring-after. I.e. see this answer: https://stackoverflow.com/a/334251/3710053 — Siebe Jongebloed, Jul 09 '22 at 09:39
i'm try //li[@class='active']/text()[substring-after(. , 'ตอนที่')] still output "ตอนที่ 6" :( — parames3010, Jul 09 '22 at 09:46
everything inside square brackets is called the predicate. They are meant as filter , not to give results. See the answer of @zx485, how to solve your question — Siebe Jongebloed, Jul 09 '22 at 09:54
By posting the XML as an image, you've ensured that no-one can test their proposed answer to be sure that it works. This may be why you're getting answers that don't work for you. — Michael Kay, Jul 09 '22 at 17:12
See how to get only the digits in any given string - below :) — Dimitre Novatchev, Jul 10 '22 at 17:35

zx485 · Answer 1 · 2022-07-09T10:38:28.067

1

You can try this XPath-1.0 expression:

normalize-space(substring-before(substring-after(//li[@class='active']/text(),'ตอนที่ '),'&quot;'))

You can also use XPath-2.0's fn:replace function to use a RegEx to filter the string:

normalize-space(replace(//li[@class='active']/text(),'.*&#x0A;\s*&quot;.*(\d+)\s*&quot;.*','$1'))

This RegEx may have to be adjusted. It consists of several parts:

.*
 matches the whole "::before" line
\s*".* matches the string " ตอนที่
(\d+) selects one or more numbers later copied to the output with $1
\s*".* matches the rest of the line " = 50

edited Jul 09 '22 at 10:38

answered Jul 09 '22 at 09:51

zx485

28,498
28
50
59

Text is not selected Or is there something I missed? https://prnt.sc/83vzhFIVGn3J – parames3010 Jul 09 '22 at 10:00
I added the `/text()` to the end. I tested it, but there are sometimes tiny differences between the parsers. Maybe this does work better. – zx485 Jul 09 '22 at 10:12
If this doesn't work either, you can try adding a `[1]` near the end the XPath-expression: `//li[@class='active'][1]/text()`. – zx485 Jul 09 '22 at 12:43
it still doesn't work https://prnt.sc/3myMQCgHLac3 – parames3010 Jul 09 '22 at 13:03
Can you provide the XML? Maybe the value is created dynamically. – zx485 Jul 09 '22 at 14:18

Dimitre Novatchev · Answer 2 · 2022-07-10T17:41:13.483

The way to do this, without even knowing what non-digit characters could be there, is using the double-translate method:

    translate(//li[@class='active']/text(),
              translate(//li[@class='active']/text(), '0123456789', ''),
              '')

Explanation:

The inner translate(//li[@class='active']/text(), '0123456789', '') gets all non-digit characters in the first argument.
The outer translate() replaces all those non-digit characters, found in the inner translate() with the empty string.
What remains are only the digit characters

XSLT - based verification:

Given this simplified XML document:

<t>ตอนที่ 6</t>

The following transformation evaluates the correct (solution) XPath expression and outputs the result of this evaluation:

<xsl:stylesheet version="1.0" xmlns:xsl="http://www.w3.org/1999/XSL/Transform">
 <xsl:output omit-xml-declaration="yes" indent="yes"/>

  <xsl:template match="/*">
    <xsl:value-of select="translate(., translate(., '0123456789', ''), '')"/>
  </xsl:template>
</xsl:stylesheet>

The wanted result is produced:

(Xpath) How do I get a specific numeric value?

2 Answers2