When trying to list all SSO groups one belongs to, the good old URL /yourgroups/display now leads to an Internal Server Error.

An auto-completion from CADI is desirable for CMS workflow. An available API should be studied together with CMS.

Check with Ana if

• the metadata integrated in root can extend the LHCb data model #45
• parts from the LHCb data model can be auto-filled because information can be integrated in root n-tuples

and put ORCID in metadata for all research objects

The recjson configuration for storing submitted analyses should be finalised.

Either hide fields if they contain no data or set default to None.

In 65cf772, @jirikuncar committed @tiborsimko's amendments to the home page. It missed experiments logos, though. They should be added.

ssh -f [email protected] -L 3000:data-demo.cern.ch:80 -N

When a user expands a deposition box, we may perhaps want to change icon from downward-pointing to upward-pointing, to better indicate expand/collapse status; or we may want to choose a more neutral static one.

After JSON model is finalised in #9, the Deposit should be configured to call Upload as the last step, so that the data would be uploaded.

Only the SSO login method should be shown in the page header and on the login page, just like we do for LABS.

Commit demo collection setup.

As usrers are adding data files in "Final N-tuples" box, the "Data Files" item is well multiplied, but the "Description" item is not. So we end up with N-to-1 ration. What we need is a description field for each new data file added, so that every data file can be described; N-to-N ratio.

(Ditto for harvest-or-link decision radio, there has to be N of them, for each data file.)

Test the submission form based on the new json data model and get feedback for further improvement

Some of the submitted files will be stored on Ceph, some on AFS. A filesystem abstraction should be introduced when it is ready in upstream invenio.

Collections should be SSO restricted to respective collaboration member groups.

In the "Physics" information box, after "Trigger Selection", a new
facility to enter information about "Final state particles" is wanted.

basically this is a matrix of information looking like:

| particle | number | PT cut | eta cut |
|----------+--------+--------+---------|
| electron | 1      | 20 GeV | 0.4     |
| muon     | ...    | ...    | ...     |
| gamma    | ...    | ...    | ...     |
| b-jet    | ...    | ...    | ...     |
| tau-jet  | ...    | ...    | ...     |
| MET      | ...    | ...    | ...     |

The UI could work like this:

Final state particles:
----------------------
[--electron--] [number?] [PT cut?] [eta cut?]  [+ ADD MORE]

The user could choose particle from the drop-down selection menu
(electron, muon, etc) and enter PT cut and eta cut information
manually in input text areas.

The user could then click on "[+ ADD MORE]" to add more lines to fill
above matrix.

The tasks:

(1) Create "Final state particles" box either as a big static matrix
of the above kind, or better yet as a progressively deploying matrix
lines as people will be clicking on "[+ ADD MORE]".

(2) Interesting things happen if people choose more than one particle
(say 2 electrons) in the above "number?" box. A new selection box
should then pop up below saying something like:

combination?  (*) same sign ( ) opposite sign ( ) invariant mass [--free text--]

(3) Similarly to the above selection matrix, one can have have "Vetos"
matrix that will "unselect" events in similar way. (e.g. vetoing this and this
particle with that and that eta cut)

The author/title information for analysis notes such as LHCb-ANA-2012-049 should be auto-completed from CDS. Need user token and IP address restriction.

create json data model for first submission form prototype

@espacial will suggest a logo

Test the submission form based on the new json data model and get feedback for further improvement

Observed with @katilp: after successful authentication, going to CMS deposit works:

• https://analysis-preservation.cern.ch/deposit/cms
• https://analysis-preservation.cern.ch/deposit/questions2

but going to the deposit index does not:

• https://analysis-preservation.cern.ch/deposit/

The page does not show available permitted deposit workflows based on logged-in user.

(P.S. It works for admins.)

For auto-completion for datasets (an import of their metadata) from DAS (https://cmsweb.cern.ch/das/) the API needs to be studied

• based on feedback from Paris
• with real data

Complete CERN OAuth registration.

Set up demo CMS collection and corresponding restriction.

Primary CMS datasets that users will enter in deposit forms look like
this:

/Electron/Run-2010B-Apr21ReReco-v1/AOD

or like this:

/ZeroBias/Run-2010B-Apr21ReReco-v1/AOD

If we represent this by /X/Run-YYYY* form, then for certain
combinations of X and YYYY the "Trigger Selection" box can be
auto-completed with suitable trigger values that people could simply
select from.

Here are the rules on how the trigger mappings look like for various
years:

• https://fwyzard.web.cern.ch/fwyzard/hlt/2010/dataset
• https://fwyzard.web.cern.ch/fwyzard/hlt/2011/dataset
• https://fwyzard.web.cern.ch/fwyzard/hlt/2012/dataset
• https://fwyzard.web.cern.ch/fwyzard/hlt/2013/dataset

For example, if a user submitted
/Electron/Run-2010B-Apr21ReReco-v1/AOD primary data set, then
X=Electron and Y=2010 and looking up the trigger values from the
first URL, one can get X->TRIGGER value mappings, leading to the
following trigger selection drop-down menu:

HLT_DoubleEle15_SW_L1R
HLT_DoubleEle17_SW_L1R
HLT_DoubleEle8_SW_HT70U_L1R
HLT_Ele10_MET45
HLT_Ele10_SW_EleId_HT70U_L1R
HLT_Ele10_SW_HT100U_L1R
HLT_Ele10_SW_HT70U_L1R
HLT_Ele12_SW_TightEleIdIsol_L1R
HLT_Ele12_SW_TightEleIdIsol_NoDEtaInEE_L1R
HLT_Ele12_SW_TighterEleIdIsol_L1R
HLT_Ele17_SW_CaloEleId_L1R
HLT_Ele17_SW_EleId_L1R
HLT_Ele17_SW_LooseEleId_L1R
HLT_Ele17_SW_TightCaloEleId_Ele8HE_L1R
HLT_Ele17_SW_TightCaloEleId_SC8HE_L1R
HLT_Ele17_SW_TightEleIdIsol_L1R
HLT_Ele17_SW_TightEleId_L1R
HLT_Ele17_SW_TighterEleIdIsol_L1R
HLT_Ele17_SW_TighterEleId_L1R
HLT_Ele22_SW_CaloEleId_L1R
HLT_Ele22_SW_TighterCaloIdIsol_L1R
HLT_Ele22_SW_TighterEleId_L1R
HLT_Ele27_SW_TightCaloEleIdTrack_L1R
HLT_Ele32_SW_TightCaloEleIdTrack_L1R
HLT_Ele32_SW_TighterEleId_L1R
HLT_Ele40_SW_L1R
HLT_IsoEle12_PFTau15

The tasks:

(1) We need to take X---YYYY---TRIGGER mappings from the above URLs
and create auto-completion facility that would analyse entered primary dataset
value and generate trigger selection drop-down box with appropriate values.

(2) Note that one can add more than one dataset, so more than one
trigger selection drop-down should be auto-completed in this way.

(3) Note that to be on the safe side, the deposit UI could offer users
a free text box to fill in case trigger selection auto-completion did
not work correctly. (Say X->Trigger mapping was not complete.)

Set up demo LHCb collection and restrictions.

• API needed from ALICE in order to fetch data from LEGO System.

Upgrade data.cern.ch overlay to respect mid-summer JS/CSS budle changes in latest invenio/pu.

Box "User Code", option "How to reproduce", value "Other (please upload)" opens well a file upload possibility.

SImilar thing is probably needed for box "User Code", option "Code Type", value "Other (please specify)" that should open a text input box or a text area, but it does not seem to do anything now. So that users are forced to use common "Code Comment" text area.

Since various steps of the same analysis may be performed by different persons, it would be profitable to allow different group of people to fill the same deposit.

(1) E.g. sharing could happen via "Save" facility before the record is uploaded, which is harder WRT locking and correct diff-propagation.

(2) E.g. sharing could happen simply via post-upload editing of the record, which is harder WRT DOI assignements etc.

The collaboration-wide visibility of temporarily co-edited & co-submitted material does not seem to be an issue.

Create schema to translate json to json-ld for export based on the ontology work done with DASPOS

see mail from @katilp from April 30th 2015 that contains the questionnaire.

Needs json model of the questionnaire to be searchable.

Test the submission form based on the new json data model and get feedback for further improvement

• doodle
• demo prep
• prepare notes for update on "different tracks"

Currently all three collaboration depositions are listed under the same generic "Data Analysis" type. This makes one extra user click. As we won't have many deposition types listed there, it would be more ergonomic to start with various collaboration deposition types directly, without single common heading.

(BTW note the name should be "ALICE", not "Alice".)

The site comes with default fields some of which are not applicable (reference) which should be taken out, while some of existing fields should be created based on data model (analysisname and friends).

When accessing a wrong deposition such as /deposit/foobarblah, the system replies with an Internal Server Error. An error handler is missing in the overlay, probably.

A nicer handling of missing doctype situation would be desirable, e.g. replying the user with "The deposition foobarblash is not available". The same may be needed for upstream Invenio.

Map the ALICE metadata in json coming from the LEGO system to the CAP data model and extend/adjust this accordingly for the ALICE submission forms

For ALICE and CMS, the 'physics information' section doesnt display some fields correctly if there is only one value: ESD Primary Data Set, MC Data Set, Keywords, Primary Data Set & Trigger selection

Make radio & select fields display the same text as what is selected on the form

• check dependencies
• check if we need specific testing conditions with users
• reach out to CMS to plan test (participants)

Done by Anxhela, feedback by @pherterich and myself.

From LHCb data samples input path like:

sim://LHCb/Collision12/Beam4000GeV-VeloClosed-MagDo wn/RealData/Reco14/Stripping20/90000000 ( Full stream )/BHADR

as well as event samples input path:

sim://MC/MC11a/Beam3500GeV-2011-MagDown-Nu2-50ns-EmNoCuts/Sim05a/Trig0x40760037Flagged/Reco12a/Stripping17NoPrescalingFlagged/42

we should auto-complete several fields in concerned boxes on the form, such as:

• Reco14
• Stripping version 20

We have the go-ahead.

(is this going to be the repository for future development, or will it go into the open data project or a new project?)

Tailor the general json model to LHCb needs

Page footer templates should be amended to use Deposit instead of Submit. (Ditto in invenio generic "pu" branch.)

Replace the current ATLAS logo with the new one similar to CODP request cernopendata/opendata.cern.ch#789

Analysis name comes first, and is mandatory, and has consistent capitalisation.