The interpretation of that string looks to be done by gnu.jel (in lcsim).
It claims to use the same rules for arithmetic as in standard java language,
which should be doing what you want, and in the case of this math is also the
same as in C:
a - -b = a + b
a - - b = a + b
a - +b = a - b
etc.
But of course good to check it just by diffing lcdd's like Sho suggested.
-Nathan
On Aug 30, 2016, at 2:32 PM, Sho Uemura <[log in to unmask]> wrote:
> I guess the best thing to do would be to check? Make two versions of the compact.xml (one with "0.0 - -0.000625" and one with "0.000625"), convert both to LCDD, do a diff?
>
> But you could also change Pelle's code in BuildMillepedeCompact so it doesn't use minus signs. The minus signs are coming from this code:
>
> if (p.getType() == MilleParameter.Type.ROTATION.getType() && !flipRotationCorrection) {
> newValue = oldValue + " - " + String.format("%.6f",correction);
> System.out.println("NOFLIP");
> } else {
> newValue = oldValue + " + " + String.format("%.6f",correction);
> System.out.println("FLIP");
> }
>
> and you could instead do String.format("%.6f",-1.0*correction). Should be safe.
>
> On Tue, 30 Aug 2016, Alessandra Filippi wrote:
>
>> Hi Norman,
>> I'm not quite sure as I haven't taken a closer look yet. This is the output of the procedure buildCompact.py which is supposed to read the output
>> of millepede and add (namely, subtract) the offsets in the compact.xml of the starting geometry, to produce a different aligned geometry.
>> I guess the relevant code is the BuildMillepedeCompact.java class, and the issue may be that these numbers have to be treated as text strings as it can happen that the offset must be subtracted from some parameters over which a correction by means of a formula is already applied in the compact.xml (as the constraints that were suggested by Sho).
>> If you tell me that the "- -" sequence does not make sense, as I was supposing, I think I can spot the problem by myself and apply the necessary patch (whenever the offset it negative). In the meanwhile, I'm anyway correcting my new xml files by hand.
>> Thanks, cheers
>> Alessandra
>>
>>
>>
>>
>>
>>
>>
>> On Tue, 30 Aug 2016, Graf, Norman A. wrote:
>>
>>> Hello Alessandra,
>>> That is, indeed, odd behavior. What code is producing this xml file?
>>> Norman
>>> -----Original Message-----
>>> From: [log in to unmask] [mailto:[log in to unmask]] On Behalf Of Alessandra Filippi
>>> Sent: Tuesday, August 30, 2016 5:49 AM
>>> To: hps-software <[log in to unmask]>
>>> Subject: xml behavior
>>> Hi all,
>>> I have a stupid question about xml, that somebody with some more knowledge of the language can answer very quickly I think. In the compact.xml file that I'm producing (automatically) upon alignment I found some lines like, for instance:
>>> <millepede_constant name="12304" value="0.0 - -0.000625" />
>>> I'm wondering how xml treats the two double - signs. I cannot see any complain anywhere but I actually do not know how this string is converted into a number.
>>> Is this an acceptable syntax, to get as a result for value = 0.000625 ?
>>> (I would anyhow be tempted for the moment to replace every "- -" string with a "+", by hand).
>>> Thanks, cheers
>>> Alessandra
>>> ########################################################################
>>> Use REPLY-ALL to reply to list
>>> To unsubscribe from the HPS-SOFTWARE list, click the following link:
>>> https://listserv.slac.stanford.edu/cgi-bin/wa?SUBED1=HPS-SOFTWARE&A=1
>>> ########################################################################
>>> Use REPLY-ALL to reply to list
>>> To unsubscribe from the HPS-SOFTWARE list, click the following link:
>>> https://listserv.slac.stanford.edu/cgi-bin/wa?SUBED1=HPS-SOFTWARE&A=1
>>
>> ########################################################################
>> Use REPLY-ALL to reply to list
>>
>> To unsubscribe from the HPS-SOFTWARE list, click the following link:
>> https://listserv.slac.stanford.edu/cgi-bin/wa?SUBED1=HPS-SOFTWARE&A=1
>>
>
> ########################################################################
> Use REPLY-ALL to reply to list
>
> To unsubscribe from the HPS-SOFTWARE list, click the following link:
> https://listserv.slac.stanford.edu/cgi-bin/wa?SUBED1=HPS-SOFTWARE&A=1
########################################################################
Use REPLY-ALL to reply to list
To unsubscribe from the HPS-SOFTWARE list, click the following link:
https://listserv.slac.stanford.edu/cgi-bin/wa?SUBED1=HPS-SOFTWARE&A=1
