[will-moore]

Fixes #234 and also adds support for exporting ROI Tags.

To test:

• Run Batch_ROI_Export script with default parameters (e.g. just run it on a single Image - quicker):
  • Only a single channel is exported (instead of multiple channels)
  • We don't export Points Coords by default
• Run with Export ROI Tags checkbox checked
  • ROI Tags column should be included with Tag text values separated by commas

[will-moore]

Test at test/integration/test_export_scripts.py::TestExportScripts::test_batch_roi_export[True] FAILED [ 59%]
is failing above because the csv.writer writes csv files with an extra row compared with the manual csv writing we had before.

        row_count = 1 + size_c + polygon_planes
>       assert len(csv_text.split("\n")) == row_count
E       assert 6 == 5

[will-moore]

@pwalczysko Good images for testing are in the dataset at https://merge-ci.openmicroscopy.org/web/webclient/?show=dataset-569
Just choose a single image to run the script.

[pwalczysko]

• Only a single channel is exported (instead of multiple channels)

Yes, wfm. But, now, I would like to export all channels. Pretty confused about how to do it. Which checkbox to check ?

Edit: I know the answer now, and this is from long past memory. One has to enter 1,2,3 into the Channels textfield... Which worked. But I would still say an obvious RFE is "Export all channels" (wording does not matter much) checkbox

[will-moore]

Updated script parameters to give a new Pixel Intensity Stats option. The default is None - in which case we don't include stats. This is much faster. e.g testing with a Dataset containing 7 images and 500 ROIs, took 0.5 seconds with None whereas choosing Selected Channels and just channel 1 took 17 secs (as printed in logging to stdout).

[pwalczysko]

aha, so you returned the Channel 1,2,3 filled in by default (i.e. by default all channels get exported) which fixes my comment #235 (comment) and compensated for the lost time by not doing the stats by default if I get it right @will-moore ? Thank you

[pwalczysko]

bug (possibly in omero-biofilefinder) ? but reporting here:

Going through the workflow of adding Tags in iviewer to ROIs I added newly 3 tags:

• Dead
• Nice Cel
• Ben's Tag

I have run this script on them then and created on the Dataset a new CSV. The CSV when inspected in Excel lgtm, the Ben's tag row does not seem extraordinary in comparison with the Dead rows. But the Ben's tag ultimately does not load in BFF, see screenshot below.

another case with different server (outreach) crashes on date (Uploaded)

[will-moore]

aha, so you returned the Channel 1,2,3 filled in by default (i.e. by default all channels get exported) which fixes my comment #235 (comment) and compensated for the lost time by not doing the stats by default if I get it right @will-moore ? Thank you

No. By default Pixel Intensity Stats: None -> no pixel intensity stats are exported.
If you choose Pixel Intensity Stats: All Channels -> then pixel intensity stats exported for all channels
if you choose Pixel Intensity Stats: Selected Channels (enter below) then the 1,2,3 Channels will get stats exported, or different Channels if you edit it.

[will-moore]

Don't know what's going on in BFF error there

[pwalczysko]

Don't know what's going on in BFF error there

It is the apostrophy:
Proof: I have removed the apostrophy (Ben's -> Bens), saved the CSV again, uploaded it to the Dataset as ...noapo.csv and started BFF. See below, it works

[pwalczysko]

I am approving this as it works fine. The issue with BFF not taking apostrophy, and, more importantly claiming some valid dates to be invalid, how do we solve it ? Maybe opening an issue at BFF ? @will-moore

[will-moore]

@pwalczysko I tested latest BFF and it handles apostrophes just fine. So then I decided to try and upgrade the BFF we ship in omero-biofilefinder at ome/omero-biofilefinder#18 but having some issues (see PR).

[pwalczysko]

@pwalczysko I tested latest BFF and it handles apostrophes just fine. So then I decided to try and upgrade the BFF we ship in omero-biofilefinder at ome/omero-biofilefinder#18 but having some issues (see PR).

Thank you. If in deep weeds, please do not hesitate to revert. imho, this is a very annoying bug, but NOT a blocker.