Hi Tomas,
I get all this (it sticks with what I had suspected). And I agree about
keeping development side simple.
About your last suggestion, well, I'm not sure. It is not such painful
for the user to select files correctly for the very first step the
SetupDialog rather than from selection tree. The point is: if some files
were already generated using a previous batch, they'll all fit the next
steps name pattern, then already existing files will be processed again
while running the batch a second time (with parameters change(s)...). In
this use case, if I get it correct, the user has to change, let's say,
quite all the suffixes and/or the pattern fields at every step, unless
he has to think a bit (not straight forward).
May be I missed something. Please correct me if I'm wrong.
May be, we could provide, to ease the process of filling again the
fields, to add a combobox to the "Add name pattern" allowing to choose
among already typed patterns (during the MZmine session) such that we
can select one of the patterns defined in a previous step for the
currently being edited one... and/or add a button that dispatches
downward one or all the patterns from one step to all the steps
following it (which would cause a possible mess while changing the steps
order, but this would be the user's responsibility to use this button or
not).
Well, just thinking. I know what I wrote above is not an answer...
Actually, I don't really see a better way than the one you already
implemented and mentioned.
Cheers
Gauthier
On 12/17/2014 09:37 AM, Tomas Pluskal wrote:
Hi Gauthier,
I really didn't like the previous design of batch mode, for several
reasons:
1) Some modules are simply not supposed to take the previous module's
output as their input. A typical example is the alignment module,
which may require a combined output of several previous steps.
2) In the previous design it was absolutely impossible to process 2
groups of files simultaneously with slightly different methods or
parameters. For example, in our lab we often acquire data in positive
and negative ionization modes, and I want to be able to process all
this data using a single batch.
3) The requirement that modules must report newly generated objects is
a burden for module developers. Based on previous experience, I want
to make module development as simple as possible.
However, you have a good point that the user should be able to run the
same batch on different files in the GUI mode. I propose the following
solution: the RawDataFileParameter and PeakListParameter can have a
special checkbox "Use currently selected files", which, if selected,
will run the methods on the files that are selected in the tree (just
like in the previous implementation). In the subsequent batch steps,
the input data files or peak lists can be defined using wildcards
(e.g. "*filtered") - that is already possible now.
What do you think?
Tomas
On Dec 17, 2014, at 16:04, Gauthier Boaglio
<gauthier.boag...@gmail.com <mailto:gauthier.boag...@gmail.com>> wrote:
Hi Tomas,
I understand that this is important to have it work this way for
better compatibility in Headless Mode, but wouldn't it be more
practical to have the choice of using pattern or using the result of
the previous step (as it used to be), at least in GUI mode? That was
very handy that way, because we were able to run several times a
batch in the same session/project on different input files (string
pattern - regex - doesn't allow to make the distinction between 2
different runs of a given batch). A simple suggestion: couldn't we
restore that possibility and in some way (for example, adding some
sort of predefined tags to the
RawDataFilesComponent/PeakListsComponent), such that we can, for
example say: "${previous.step}" as a valid/recognized
"pattern"/entry. Or, another way could be to give the possibility to
prepend a user define prefix to the resulting file/list of the result
of the first step (which would be used recursiveley throug the steps,
such that we don't have to update every
RawDataFilesParameter/PeakListsParameter, for every single step, each
time we run the batch one more time in the same project). As a simple
user, I would prefer to have "${previous.step}" recognized as a valid
pattern entry. As a developer, I understand that this can make the
code a bit more complicated/less flexible (since we would have to
restore the ability of a Task to remember what objects it actually
generated).
What do you say?
Best
Gaut
On 12/16/2014 01:40 AM, Tomas Pluskal wrote:
- I updated the RawDataFileParameter and PeakListParameter
(selection of inputs of individual modules) to be real, visible
parameters with their own selection components. In these components,
users can set wildcards (*) to apply methods to certain groups of
data files or peak lists. This also changes the philosophy of the
batch mode. In previous versions, the batch mode steps were always
applied to the output of the previous step. Now, the user can setup
what each batch mode step is actually applied to. This change is
still a work in progress, as some modules will have to be updated to
reflect the new philosophy (e.g., the spectra visualizer now
complains when multiple raw data files are selected - but these are
cosmetic issues).
--
Gauthier BOAGLIO
CEFE - UMR 5175
1919 route de Mende
F-34293 Montpellier cedex 5
Tel: +33/0 4 67 61 32 15
Fax: +33/0 4 67 61 33 36
email: gauthier.boag...@cefe.cnrs.fr
<mailto:gauthier.boag...@cefe.cnrs.fr>
www:
http://www.cefe.cnrs.fr/en/evolutionary-ecology-and-epidemiology/gauthier-boaglio
http://www.evolepid.org/people.php?name=boaglio
===============================================
Tomas Pluskal
G0 Cell Unit, Okinawa Institute of Science and Technology Graduate
University
1919-1 Tancha, Onna-son, Okinawa 904-0495, Japan
WWW: https://groups.oist.jp/g0
TEL: +81-98-966-8684
Fax: +81-98-966-2890
------------------------------------------------------------------------------
Download BIRT iHub F-Type - The Free Enterprise-Grade BIRT Server
from Actuate! Instantly Supercharge Your Business Reports and Dashboards
with Interactivity, Sharing, Native Excel Exports, App Integration & more
Get technology previously reserved for billion-dollar corporations, FREE
http://pubads.g.doubleclick.net/gampad/clk?id=164703151&iu=/4140/ostg.clktrk
_______________________________________________
Mzmine-devel mailing list
Mzmine-devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/mzmine-devel
--
Gauthier BOAGLIO
CEFE - UMR 5175
1919 route de Mende
F-34293 Montpellier cedex 5
Tel: +33/0 4 67 61 32 15
Fax: +33/0 4 67 61 33 36
email: gauthier.boag...@cefe.cnrs.fr
www:
http://www.cefe.cnrs.fr/en/evolutionary-ecology-and-epidemiology/gauthier-boaglio
http://www.evolepid.org/people.php?name=boaglio
------------------------------------------------------------------------------
Download BIRT iHub F-Type - The Free Enterprise-Grade BIRT Server
from Actuate! Instantly Supercharge Your Business Reports and Dashboards
with Interactivity, Sharing, Native Excel Exports, App Integration & more
Get technology previously reserved for billion-dollar corporations, FREE
http://pubads.g.doubleclick.net/gampad/clk?id=164703151&iu=/4140/ostg.clktrk
_______________________________________________
Mzmine-devel mailing list
Mzmine-devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/mzmine-devel
