Re: [chemclipse-dev] Question about AbstractVendorMassSpectrumProxy clas

[Trig Chen <trigchen@xxxxxxxxx>]

The only one method that change isProxy field is a private method and
called more than 20 times in AbstractVendorMassSpectrumProxy class.
Is there an elegant way includes reflection to add the ability to
change the private field isProxy's value?

Best regards,

Trig


2016-10-20 15:51 GMT+08:00 Philip Wenig <philip.wenig@xxxxxxxxxxxxx>:
> Overwriting methods would be better. Copy & Paste causes problems which are
> hard to resolve and code that is hard to maintain. As the method is private,
> please inspect which methods use the private method and overwrite them. It
> also helps to understand how the core works.
>
>
> Best,
> Philip
>
>
>
> Am 20.10.2016 um 09:48 schrieb Trig Chen:
>>
>> Hi Philip,
>>
>> Yes, your consideration is reasonable. Since the isProxy is a private
>> field and has no method to modify it, I did not extend
>> AbstractVendorMassSpectrumProxy class. I just copy it to my package
>> and add a setter for isProxy field. After I have calibrated the m/z
>> conversion factors, I iteratively set all scan's isProxy to true to
>> force them to reload if them are reaccess later. :)
>>
>> Best regards,
>>
>> Trig
>>
>>
>> 2016-10-20 15:30 GMT+08:00 Philip Wenig <philip.wenig@xxxxxxxxxxxxx>:
>>>
>>> Hi Trig,
>>>
>>> this could cause some implications which would decrease the performance
>>> tremendously.
>>> Think of a loop where one looks for a certain m/z values. Each time, the
>>> scan would be loaded and discarded and the loaded again and again. If it
>>> stays in memory, no further I/O operation is needed.
>>>
>>> I don't want to activate additional in the core platform where I'm not
>>> sure
>>> if there are unwanted side-effects. Anyhow, feel free to overwrite the
>>> methods in your own class and add your specific functionality. If that
>>> works
>>> fine, let's see if we can refactor the core.
>>>
>>>
>>> Best,
>>> Philip
>>>
>>>
>>> Am 20.10.2016 um 09:23 schrieb Trig Chen:
>>>
>>> Hi Philip,
>>>
>>> In most cases, such as processing data in OpenChrom, it would be good
>>> loading once data from file through proxy and thereafter access this data
>>> in
>>> memory. In my case, my data is acquired from spectrometer and then
>>> written
>>> to file and sent to GUI for display. That means all scans are gotten
>>> access
>>> and loaded! I prefer to release the scan after it was shown. If user
>>> clicks
>>> the TIC plot and recall the related scan, I can reload it from file
>>> again.
>>> So, it would be good if AbstractVendorMassSpectrumProxy class has the
>>> ability to access data through proxy always, not just once.
>>>
>>>
>>>
>>>
>>>
>>>
>>> Best regards,
>>>
>>> Trig
>>>
>>> 2016-10-19 17:08 GMT+08:00 Philip Wenig <philip.wenig@xxxxxxxxxxxxx>:
>>>>
>>>> Hi Trig,
>>>>
>>>> that's correct.
>>>>
>>>> When analyzing huge files > 1 GB, the performance would be miserable if
>>>> each scan is loaded in advance. Hence, it shall be loaded only when it
>>>> is
>>>> needed. Each vendor supplier best knows on his own ho to extract the
>>>> data
>>>> from a scan, when it's needed (e.g. only he knows the byte position in
>>>> the
>>>> file where the requested scan is located). The member variable is set to
>>>> true after successfully loading the file. I've explicitly not added a
>>>> method
>>>> to set this to false again, cause the scan is loaded after running the
>>>> "checkProxyAndImportOnDemand()" method. So it's not a proxy anymore
>>>> afterwards.
>>>>
>>>>
>>>> Best,
>>>> Philip
>>>>
>>>>
>>>>
>>>> Am 19.10.2016 um 10:57 schrieb Trig Chen:
>>>>
>>>> Hi all,
>>>>
>>>> The class
>>>> org.eclipse.chemclipse.msd.model.core.AbstractVendorMassSpectrumProxy
>>>> has
>>>> one private boolean type field "isProxy" with true initial value. This
>>>> value
>>>> is changed to false as method checkProxyAndImportOnDemand() called.
>>>> checkProxyAndImportOnDemand method does the import ions job and then set
>>>> isProxy to false. Then isProxy stays false always and will never change
>>>> to
>>>> true again. That means I can't import ions again.
>>>>
>>>> private void checkProxyAndImportOnDemand() {
>>>>
>>>> if(isProxy) {
>>>> isProxy = false;
>>>> importIons();
>>>> }
>>>> }
>>>>
>>>> In my case, I firstly import ions and find the m/z values incorrect, so
>>>> I
>>>> make a m/z calibration on current mass spectrum. Here comes the problem,
>>>> I
>>>> can't reimport ions for current mass spectrum or any other spectra
>>>> because
>>>> isProxy is false. There is not a way to set isProxy to true in
>>>> AbstractVendorMassSpectrumProxy class.
>>>>
>>>>
>>>> Best regards,
>>>>
>>>> Trig
>>>>
>>>>
>>>> _______________________________________________
>>>> chemclipse-dev mailing list
>>>> chemclipse-dev@xxxxxxxxxxx
>>>> To change your delivery options, retrieve your password, or unsubscribe
>>>> from this list, visit
>>>> https://dev.eclipse.org/mailman/listinfo/chemclipse-dev
>>>>
>>>> --
>>>> ~~~~~~~~~~~~~~~~~~~~~~~~
>>>> OpenChrom - the open source alternative for chromatography / mass
>>>> spectrometry
>>>> Dr. Philip Wenig » Founder » philip.wenig@xxxxxxxxxxxxx »
>>>> http://www.openchrom.net
>>>> ~~~~~~~~~~~~~~~~~~~~~~~~
>>>>
>>>> _______________________________________________ chemclipse-dev mailing
>>>> list chemclipse-dev@xxxxxxxxxxx To change your delivery options,
>>>> retrieve
>>>> your password, or unsubscribe from this list, visit
>>>> https://dev.eclipse.org/mailman/listinfo/chemclipse-dev
>>>
>>> _______________________________________________
>>> chemclipse-dev mailing list
>>> chemclipse-dev@xxxxxxxxxxx
>>> To change your delivery options, retrieve your password, or unsubscribe
>>> from
>>> this list, visit
>>> https://dev.eclipse.org/mailman/listinfo/chemclipse-dev
>>>
>>> --
>>> ~~~~~~~~~~~~~~~~~~~~~~~~
>>> OpenChrom - the open source alternative for chromatography / mass
>>> spectrometry
>>> Dr. Philip Wenig » Founder » philip.wenig@xxxxxxxxxxxxx »
>>> http://www.openchrom.net
>>> ~~~~~~~~~~~~~~~~~~~~~~~~
>>>
>>>
>>> _______________________________________________
>>> chemclipse-dev mailing list
>>> chemclipse-dev@xxxxxxxxxxx
>>> To change your delivery options, retrieve your password, or unsubscribe
>>> from
>>> this list, visit
>>> https://dev.eclipse.org/mailman/listinfo/chemclipse-dev
>>>
>> _______________________________________________
>> chemclipse-dev mailing list
>> chemclipse-dev@xxxxxxxxxxx
>> To change your delivery options, retrieve your password, or unsubscribe
>> from this list, visit
>> https://dev.eclipse.org/mailman/listinfo/chemclipse-dev
>
>
> --
> ~~~~~~~~~~~~~~~~~~~~~~~~
> OpenChrom - the open source alternative for chromatography / mass
> spectrometry
> Dr. Philip Wenig » Founder » philip.wenig@xxxxxxxxxxxxx »
> http://www.openchrom.net
> ~~~~~~~~~~~~~~~~~~~~~~~~
>
> _______________________________________________
> chemclipse-dev mailing list
> chemclipse-dev@xxxxxxxxxxx
> To change your delivery options, retrieve your password, or unsubscribe from
> this list, visit
> https://dev.eclipse.org/mailman/listinfo/chemclipse-dev