Re: Re: Random?

with generate-id() , I have the same "id" at witch call,
The use of generate-id() is here :
...
<sentence>
  <xsl:attribute name="stamp">
      <xsl:value-of select="generate-id()"/>
  </xsl:attribute>
  <xsl:attribute name="state">nonDesamb</xsl:attribute>
  <xsl:value-of select="concat($arg1/sentence[last()], $arg2)"/>
</sentence>
...

the result is:
...
  <sentence stamp="id108131" state="nonDesamb">111 111 111</sentence>
  <sentence stamp="id108131" state="nonDesamb"> 222 222</sentence>
  <sentence stamp="id108131" state="nonDesamb"> 333 333 333</sentence>
  <sentence stamp="id108131" state="nonDesamb"> 444 444 444</sentence>
  <sentence stamp="id108131" state="nonDesamb">555 555 555</sentence>
  <sentence stamp="id108131" state="nonDesamb"> 666 666 666</sentence>
  <sentence stamp="id108131" state="nonDesamb"> 777 777 777</sentence>
  <sentence stamp="id108131" state="nonDesamb"> 888 888 888, 999 999</sentence>
...

I have 8 call to generate-id() (8 sentences) witch are equal.
you think I am doing an error?
I will try the last solution gave me Dimitre. 

Thanks for your help.
 

Quoting Ramkumar Menon <ramkumar.menon@xxxxxxxxx>:

> Not sure if I got your question right - but yes, generate-id() is a
> valid  XSLT 2.0 Function. if you pass in a node to it, it generates an
> id unique for that node.
> so it you pass generate-id(<a sentence node>), it wd generate a unique
> id for each "sentence" node.
> 
> On 5/2/05, Ali Choumane <ali.choumane@xxxxxxx> wrote:
> > 
> > If we can use a function like generate-id()...  that solve my problem.
> because
> > that we are trying to do with Dimitre has as goal to generate unique
> values
> > for the value of stamp,... but manualy...
> > a function as generate-id() exist ??
> > 
> > 
> > Quoting Ramkumar Menon <ramkumar.menon@xxxxxxxxx>:
> > 
> > > I had a question on this. Can't we use a function like generate-id()
> > > on each "sentence" node in the target to generate unique values for
> > > the value of stamp ?
> > > Just curious....
> > >
> > > rgds,
> > > Menon
> > >
> > > On 5/2/05, Dimitre Novatchev <dnovatchev@xxxxxxxxx> wrote:
> > > > Hi Ali,
> > > >
> > > > On 5/2/05, Ali Choumane <ali.choumane@xxxxxxx> wrote:
> > > > > Hi Dimitre,
> > > > > first, thanks for your help,
> > > > > second,
> > > > > I am using the seconds part of the current time -- as the value of
> the
> > > pSeed
> > > > > parameter (by using the library  of exslt).
> > > > >
> > > > > But the problem is : I am calling the function randomSequence
> several
> > > times
> > > > > per second ,so I have the same value of the parameter pSeed to the
> > > > > randomSequence.
> > > > >
> > > > > I want to use the third part of the current time (milliseconds) ???
> its
> > > > > possible?? I can't find anything!
> > > >
> > > > It seems that you are generating a lot of sequences of random numbers
> > > > -- with what length?
> > > >
> > > > I would suggest two ways around the problem:
> > > >
> > > >   1. Generate only one (but sufficiently long) sequence of randoms,
> > > > then use its subsequences as needed.
> > > >
> > > >   2. In case you'd need N sequences of random numbers, then generate
> > > > a sequence of N random numbers and use these as the seeds to generate
> > > > the corresponding N sequences.
> > > >
> > > > Cheers,
> > > > Dimitre Novatchev.
> > > >
> > > > >
> > > > >
> > > > > Quoting Dimitre Novatchev <dnovatchev@xxxxxxxxx>:
> > > > >
> > > > > > Hi Ali,
> > > > > >
> > > > > > On 4/29/05, Ali Choumane <ali.choumane@xxxxxxx> wrote:
> > > > > > > Hi Dimitre,
> > > > > > >
> > > > > > > I am using the template "randNext" in random.xsl :
> > > > > > >
> > > > > > > ...
> > > > > > > <sentence>
> > > > > > > <xsl:attribute name="stamp">
> > > > > > >     <xsl:call-template name="randNext">
> > > > > > >         <xsl:with-param name="arg1" select="1"/>
> > > > > > >     </xsl:call-template>
> > > > > > >  </xsl:attribute>
> > > > > > >  <xsl:attribute name="state">nonDesamb</xsl:attribute>
> > > > > > >  <xsl:value-of select="concat($arg1/sentence[last()], $arg2)"/>
> > > > > > > </sentence>
> > > > > > > ...
> > > > > > >
> > > > > > > and this is a part of result of transformation:
> > > > > > > ...
> > > > > > >  <sentence stamp="39022" state="nonDesamb">111 111
> 111</sentence>
> > > > > > >  <sentence stamp="39022" state="nonDesamb"> 222 222</sentence>
> > > > > > >  <sentence stamp="39022" state="nonDesamb"> 333 333
> 333</sentence>
> > > > > > >  <sentence stamp="39022" state="nonDesamb"> 444 444
> 444</sentence>
> > > > > > >  <sentence/>
> > > > > > >  <sentence stamp="39022" state="nonDesamb">555 555
> 555</sentence>
> > > > > > >  <sentence stamp="39022" state="nonDesamb"> 666 666
> 666</sentence>
> > > > > > >  <sentence stamp="39022" state="nonDesamb"> 777 777
> 777</sentence>
> > > > > > >  <sentence stamp="39022" state="nonDesamb"> 888 888 888, 999
> > > > > > 999999</sentence>
> > > > > > > ...
> > > > > > >
> > > > > > > As the result show, "stamp" it's not a key now. I think because
> > > "arg1" is
> > > > > > > fixed.
> > > > > > >
> > > > > > > and I am using the template "randomSequence" in random.xsl, but
> it
> > > can't
> > > > > > > resolve the problem.
> > > > > >
> > > > > > randomSequence produces a sequence of (the pLength parameter) N
> > > > > > numbers. If the seed (the pSeed parameter) is the same on two
> > > > > > invocations of randomSequence, the two produced sequences are, of
> > > > > > course, the same.
> > > > > >
> > > > > > > I think , if I put a variable value of "arg1" , can resolve the
> > > problem?
> > > > > >
> > > > > > In order to achieve truly "random" results, the transformation has
> to
> > > > > > use some non-fixed number, such as the seconds part of the current
> > > > > > time -- and this can be passed as a parameter to the transformation
> by
> > > > > > the external code that invokes it.
> > > > > >
> > > > > > Then this number can be used as a seed -- to be passed as the value
> of
> > > > > > the pSeed parameter.
> > > > > >
> > > > > > > But how, when we can't change the value of variables in XSLT?
> > > > > >
> > > > > > We can't and actually we don't need to.
> > > > > >
> > > > > > Cheers,
> > > > > > Dimitre novatchev.
> > > > > >
> > > > >
> > > > > Ali CHOUMANE
> > > > > M2R III ,Univ Joseph fourrier, UFR IMA
> > > > > WAM/INRIA & GETA/CLIPS
> > > > > portable.      06 71 55 80 65
> > > > > Bureau(INRIA). 04 76 61 54 13
> > > > > Email. ali.choumane@xxxxxxx
> > > > >
> > > > > -------------------------------------------------
> > > > > envoyi via Webmail/IMAG !
> > > >
> > > >
> > >
> > >
> > > --
> > > Shift to the left, shift to the right!
> > > Pop up, push down, byte, byte, byte!
> > >
> > > -Ramkumar Menon
> > >  A typical Macroprocessor
> > >
> > >
> > 
> > Ali CHOUMANE
> > M2R III ,Univ Joseph fourrier, UFR IMA
> > WAM/INRIA & GETA/CLIPS
> > portable.      06 71 55 80 65
> > Bureau(INRIA). 04 76 61 54 13
> > Email. ali.choumane@xxxxxxx
> > 
> > -------------------------------------------------
> > envoyi via Webmail/IMAG !
> > 
> > 
> 
> 
> -- 
> Shift to the left, shift to the right!
> Pop up, push down, byte, byte, byte!
> 
> -Ramkumar Menon
>  A typical Macroprocessor
> 
> 


Ali CHOUMANE
M2R III ,Univ Joseph fourrier, UFR IMA
WAM/INRIA & GETA/CLIPS
portable.      06 71 55 80 65
Bureau(INRIA). 04 76 61 54 13 
Email. ali.choumane@xxxxxxx

-------------------------------------------------
envoyi via Webmail/IMAG !