DM-8157: #242

lznakano · 2016-11-29T17:59:39Z

1: add image search processor
2: add methods in LssSdssRequestList.js to create the WebRequest which calls the image search processor

To Test the image search processor:

select either Science CCD Exposure or DeepCoadd
Enter target:
9.469,-1.152
Click search, the images should be displayed.
Select each row, an image either highlighted if it is there or load a new one.

Please stop testing this one. The branch was created and tested on 11/23/16 before DM-8308 was merged. There was some merge issue to resolve.

[New}
After the DM-8334 was merged, it worked without doing any modifications. Please check it out and try it again. The previous issue was caused by merging DM-8308 without DM-8334.

ymeiymei · 2016-11-29T18:36:49Z

For this version I can see all DD info and get catalogs but when do image search, I got "Fail to load table. Error: edu.caltech.ipac.firefly.server.query.DataAccessException: DataAccessException:ERROR:No data is found in the search range from:unknown" for both coadd and science ccd tables.

FYI:
The DM-8334 was merged to the dev and was rebased into this branch. Nothing else had changed. I found that the image search at the Enclosed case does not load the images. The rest works OK.

[11/19/16 at 4:10PM]
I ran the debugger to check the ENCOSED search in CCD. With the same URL created in the search processor, if I changed the run, camcol, filterer to the known working result from Yi, it worked. That means the URL is correct. But sever does not return the search result. I don't know why. Maybe Yi has more insight on this.

1: add image search processor 2: add methods in LssSdssRequestList.js to create the WebRequest which calls the image search processor

ymeiymei · 2016-11-30T19:37:37Z

I tested again after Lijun merged DM-8334.

The image processor works fine.

Give a target, select different search types, I got correct image metadata and five images from the five filters with the same patch and tract.

When click one row in the table with different patch or tract, the five images are changed.

The coverage map meets the requirement. (When switch from the image panel back to the coverage panel, I need to click the coverage map image in order to see the layers display. Designed like this?)

Some issues in the catalog polygon search but it's not in this ticket.

Good job!

robyww

JavaScript part looks good. We talked a about the need to use URLFileInfoProcessor instead of subclassing directly from BaseFileInfoProcessor

robyww · 2016-12-02T17:40:35Z

src/firefly/js/metaConvert/LsstSdssRequestList.js

+            builder(plotId+'g', 'g', 'g'),
+            builder(plotId+'r', 'r', 'r'),
+            builder(plotId+'i', 'i', 'i'),
+            builder(plotId+'z', 'z', 'z'),


the plotId should be the same for both types of plots. No reason to have different id's, it will only hold extra memory.

The second parameters is the title DM-8548 is a ticket to improve it but you could just do it here. Make a base title that is scienceCcdExposureId in the ccd case and computed deepCoaddId in the coadd case

robyww · 2016-12-02T17:49:45Z

src/firefly/js/metaConvert/LsstSdssRequestList.js

-        const url= `${baseUrl}?run=${run}&camcol=${camcol}&field=${field}&filter=${filterName}`;
-        return makeUrlPlotRequest(url, title, plotId, rangeValues);
+        const searchId = searchIdBase + Number(bandMap[filterName]);
+        sr.setParam('deepCoaddId', `${searchId}`);


there is a way to pass tract and patch and filter here. I think we decided to do that instead. This affects the search processor as well

…geSearchProcessor and use the tract, patch, filterId and filterName to search the image in DeepCoadd

robyww

if tract:3, filter:0 patch is 0,1 then you create a bad url. I think you are adding an extra 0 on patch

http://lsst-qserv-dax01.ncsa.illinois.edu:5000/image/v0/deepCoadd/ids?tract=3&patch=00,1&fiterId=0&filter=r

robyww · 2016-12-06T20:45:11Z

src/firefly/js/metaConvert/ConverterFactory.js

@@ -57,7 +57,7 @@ export const converters = {
        hasRelatedBands : true,
        canGrid : true,
        maxPlots : 12,
-        makeRequest : makeLsstSdssPlotRequest,
+        makeRequest : makeLsstImagePlotRequest,


You should not change the name here. The right name is makeLsstSdssPlotRequest since we are showing SDSS data using LSST services. I don't think you need to touch this file.

robyww · 2016-12-06T20:58:44Z

src/firefly/js/metaConvert/LsstSdssRequestList.js

+            builder('lsst-g',  titleBase+'-g', 'g'),
+            builder('lsst-r',  titleBase+'-r', 'r'),
+            builder('lsst-i',  titleBase+'-i', 'i'),
+            builder('lsst-z',  titleBase+'-z', 'z'),


lsst-sdss is the right plot id prefix since it is sdss data using lsst

robyww · 2016-12-06T20:59:14Z

src/firefly/js/metaConvert/LsstSdssRequestList.js

        ];
        if (retval.standard[filterId]) retval.highlightPlotId= retval.standard[filterId].getPlotId();
    }

    if (threeColorOps) {
-        retval.threeColor= threeColorOps.map( (b) => b && builder('lsst-sdss-threeC', 'SDSS 3 Color', b) );
+        retval.threeColor= threeColorOps.map( (b) => b && builder('lsst-threeC', 'SDSS 3 Color', b) );


same here lsst-sdss-threeC

My mistake, I thought Sdss means science CCD. I will change all related text back as they were.

Made all the requested changes. Please review again!

robyww · 2016-12-06T21:00:02Z

src/firefly/js/metaConvert/LsstSdssRequestList.js

@@ -76,39 +93,41 @@ function makeCoaddReqBuilder(table, rowIdx) {
 * @param threeColorOps
 * @return {{}}
 */
-export function makeLsstSdssPlotRequest(table, row, includeSingle, includeStandard, threeColorOps) {
+export function makeLsstImagePlotRequest(table, row, includeSingle, includeStandard, threeColorOps) {


change name back to makeLsstSdssPlotRequest

robyww · 2016-12-06T21:01:48Z

src/firefly/js/metaConvert/LsstSdssRequestList.js

    r.setTitleOptions(TitleOptions.NONE);
    r.setTitle(title);
    r.setPlotId(plotId);
    r.setMultiImageIdx(0);
-    r.setPreferenceColorKey('lsst-sdss-color-pref');
+    r.setPreferenceColorKey('lsst-coadd-color-pref');


start key with lsst-sdss

… back to 'lsst-sdss'

robyww · 2016-12-06T23:54:57Z

OK, looks good.

DM-8157:

650983f

1: add image search processor 2: add methods in LssSdssRequestList.js to create the WebRequest which calls the image search processor

lznakano force-pushed the DM-8157-ImgSearchProcessor branch from ec3799d to 650983f Compare November 29, 2016 22:16

robyww requested changes Dec 2, 2016

View reviewed changes

DM-8157: change the extended class to URLFileInfoProcessor in LSSTIma…

4fc334d

…geSearchProcessor and use the tract, patch, filterId and filterName to search the image in DeepCoadd

robyww reviewed Dec 6, 2016

View reviewed changes

Lijun Zhang added 2 commits December 6, 2016 14:15

DM-8157: fixed the padding bug in patch field and changed the 'lsst-'…

39ae8fa

… back to 'lsst-sdss'

DM-8157:removed fieldId in the URL since it is no needed

bbd566a

robyww approved these changes Dec 6, 2016

View reviewed changes

lznakano merged commit 0a1df68 into dev Dec 7, 2016

lznakano deleted the DM-8157-ImgSearchProcessor branch December 7, 2016 00:33

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

Uh oh!

DM-8157: #242

DM-8157: #242

Uh oh!

lznakano commented Nov 29, 2016 •

edited

Loading

Uh oh!

ymeiymei commented Nov 29, 2016 •

edited by lznakano

Loading

Uh oh!

ymeiymei commented Nov 30, 2016

Uh oh!

robyww left a comment

Uh oh!

robyww Dec 2, 2016

Uh oh!

robyww Dec 2, 2016

Uh oh!

robyww Dec 2, 2016

Uh oh!

robyww left a comment

Uh oh!

robyww Dec 6, 2016

Uh oh!

robyww Dec 6, 2016

Uh oh!

robyww Dec 6, 2016

Uh oh!

lznakano Dec 6, 2016

Uh oh!

lznakano Dec 6, 2016

Uh oh!

robyww Dec 6, 2016

Uh oh!

robyww Dec 6, 2016

Uh oh!

robyww commented Dec 6, 2016

Uh oh!

Uh oh!

DM-8157: #242

DM-8157: #242

Uh oh!

Conversation

lznakano commented Nov 29, 2016 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

ymeiymei commented Nov 29, 2016 • edited by lznakano Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

ymeiymei commented Nov 30, 2016

Uh oh!

robyww left a comment

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

robyww left a comment

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

robyww commented Dec 6, 2016

Uh oh!

Uh oh!

lznakano commented Nov 29, 2016 •

edited

Loading

ymeiymei commented Nov 29, 2016 •

edited by lznakano

Loading