Re: [xsl] I desire this function: substring-before(string, regex charset)

["G. Ken Holman g.ken.holman@xxxxxxxxx" <xsl-list-service@xxxxxxxxxxxxxxxxxxxxxx>]

Wouldn't the non-greedy '?' version of regex in XPath help you here?

replace('THEN THE CURTAIN FELL','(^.*?)[AEIOU].*$','$1')

... returns "TH".

I hope this helps. But I haven't tested edge cases.

. . . . . . Ken

At 13/04/2025 12:43 +0000, Roger L Costello costello@xxxxxxxxx wrote:
> Hi Folks,
>
> The XPath substring-before function returns "that part of the given 
> input string that occurs before the first occurrence of the string 
> given in $arg2." [definition from SAXON web page]
>
> substring-before($arg1 as xs:string?, $arg2 as xs:string?) --> xs:string
>
> It's a shame that the value of $arg2 can't be a regex character set, e.g.,
>
> substring-before("THEN THE CURTAIN FELL", '[AEIOU]')
>
> returns TH.
>
> Even better, it would be nice if there was a third argument which 
> specified that you also want the character that was matched from the 
> character set:
>
> substring-before("THEN THE CURTAIN FELL", '[AEIOU]', 'plus matched 
> charset character')
>
> returns THE.
>
> I believe such a function would be useful.
>
> SNOBOL has such a function.
>
> Let's see how such functionality could be used. I have this text:
>
> THEN THE CURTAIN FELL
>
> Fetch the string preceding the first vowel, plus the vowel:
>
> THE
>
> However, instead of fetching the string plus vowel, modify the text 
> by nullifying the string plus vowel:
>
> N THE CURTAIN FELL
>
> Repeat on the new, shortened text.
>
> Here is the text as it is repeatedly shortened:
>
> THEN THE CURTAIN FELL
> N THE CURTAIN FELL
>  CURTAIN FELL
> RTAIN FELL
> IN FELL
> N FELL
> LL
>
> General Problem Statement: There is a text string. There is a 
> character set. Strip off the string prior to the first occurrence of 
> a character from the character set, plus the character. Repeat until 
> the end of text is reached.
>
> Below I show how to implement this in SNOBOL and then in XSLT. My 
> XSLT solution is large and complex. Is there a simpler, shorter solution?
>
> First, the SNOBOL solution:
>
> Assign the variable TEXT a string:
>
> TEXT = "THEN THE CURTAIN FELL"
>
> BREAK is a built-in SNOBOL function. It has one argument, which is a 
> character set. BREAK returns a pattern that matches a string up to 
> but not including the character from the character set. E.g.,
>
> BREAK("AEIOU")
>
> returns a pattern that matches characters up to but not including a 
> vowel. This pattern:
>
> BREAK("AEIOU") LEN(1)
>
> matches characters up to a vowel, plus the vowel.
>
> Note: LEN(N) means, match any N-length character string. It is 
> SNOBOL's version of the regex .{N}
>
> The following statement applies the pattern to TEXT, replacing the 
> string plus vowel with null:
>
> TEXT BREAK("AEIOU") LEN(1) =
>
> To incrementally strip away the string, put the statement inside a loop:
>
> LOOP    TEXT BREAK("AEIOU") LEN(1) 
> =                            :F(END)
>                 OUTPUT =  TEXT                                  :(LOOP)
>
> Here is the output from running the SNOBOL program:
>
> THEN THE CURTAIN FELL
> N THE CURTAIN FELL
>  CURTAIN FELL
> RTAIN FELL
> IN FELL
> N FELL
> LL
>
> Nice.
>
> Below is my XSLT solution. It uses the replace idea that Liam 
> provided a few weeks back, which is neat. Whereas the SNOBOL 
> solution takes only 2 lines of code, the XSLT solution requires many 
> lines of code. Is there a simpler, shorter solution?
>
> Lesson Learned: when designing a new language, it might be useful 
> for the language to provide something like the SNOBOL BREAK function.
>
> <xsl:stylesheet xmlns:xsl="http://www.w3.org/1999/XSL/Transform";
>                 xmlns:xs="http://www.w3.org/2001/XMLSchema";
>                 xmlns:f="function"
>                 exclude-result-prefixes="#all"
>                 version="3.0">
>
>     <xsl:function name="f:remove-up-to-vowel" as="xs:string*">
>         <xsl:param name="TEXT" as="xs:string"/>
>         <xsl:choose>
>             <!-- end of string? -->
>             <xsl:when test="$TEXT eq ''"/>
>             <xsl:otherwise>
>                 <xsl:variable name="substring-after-vowel"
>                            select="replace($TEXT, 
> '^[^AEIOU]*[AEIOU](.*)$', '$1')"
>                            as="xs:string*"/>
>                 <xsl:sequence select="$substring-after-vowel"/>
>                 <xsl:choose>
>                     <xsl:when 
> test="not(matches($substring-after-vowel,'[AEIOU]'))"/>
>                     <xsl:otherwise>
>                         <xsl:sequence 
> select="f:remove-up-to-vowel($substring-after-vowel)"/>
>                     </xsl:otherwise>
>                 </xsl:choose>
>             </xsl:otherwise>
>         </xsl:choose>
>     </xsl:function>
>
>     <xsl:template match="/*">
>         <xsl:variable name="result"
>                    select="f:remove-up-to-vowel('THEN THE CURTAIN FELL')"
>                   as="xs:string*"/>
>         <xsl:for-each select="$result">
>             <xsl:message>
>                 <xsl:value-of select="."/>
>             </xsl:message>
>         </xsl:for-each>
>     </xsl:template>
>
> </xsl:stylesheet>


--
Contact info, blog, articles, etc. http://www.CraneSoftwrights.com/s/ |
Check our site for free XML, XSLT, XSL-FO and UBL developer resources |
Streaming hands-on XSLT/XPath 2 training class @US$125 (5 hours free) |
Essays (UBL, XML, etc.) http://www.linkedin.com/today/author/gkholman |