CDR Tickets

Volker, is this issue done and/or tracked elsewhere?

I've noticed a couple of minor things in NVCG PP that I thought I'd mention. I don't know if later refreshes will fix/affect them or not.

1. Table shading. We have requested that the tables not have shading. I believe this has been fixed in NVCG but PP still has the shading.

2. Images. Images look different in PP than in NVCG - better in PP, actually. They are smaller and not skewed (the skewing has been a problem for a long time on the live site). For example, see figures 1, 2, and 3 in the genetics of breast and gynecologic cancers summary (CDR62855). Although I think they look better in PP, this isn't a true picture of what they'll look like on the live site.

Can you give me the URLs you're using to compare?

Images look different in PP than in NVCG - better in PP, actually.

I think the reason for this is that the PP report grabs the image from the CDR and not from Cancer.gov. For new images we have to make sure the PP report still works with a new image that's not yet published to Cancer.gov. Therefore, we're using the original, unskewed image.
Why the images are skewed once they go through gatekeeper in the first place I have never been able to get an answer for.

Sure -

QA Site: http://www-qa.cancer.gov/types/breast/hp/breast-ovarian-genetics-pdq#section/all

Publish Preview: https://cdr-qa.cancer.gov/cgi-bin/cdr/PublishPreview.py?Session=5537FEC1-AF1B68-218-P0Y0MEFU3UBE&DocId=CDR0000062855&Version=cwd#section/all

TIRs filed:

Image distortion: https://tracker.nci.nih.gov/browse/OCEPROJECT-2978
Table row shading: https://tracker.nci.nih.gov/browse/OCEPROJECT-2979

Thanks, Bob!

The table shading problem in PP has been reported as "fixed on Blue." You probably know more about what that means than I do, Robin (I don't have a good picture of the relationship between all of the GateKeeper and cancer.gov color and tiers). the ticket I filed for the image distortion problem (OCEPROJECT-2978) was closed as a duplicate of the very old ticket OCEPROJECT-81, to which a comment was posted yesterday saying that the problem has been tracked down, and the fix will be rolled into Egyptian Mau.

This ticket is marked "On Hold" because (as the description indicates) it's for adjustments we may need to our post-processing Publish Preview code to adjust what we get back from GateKeeper. Other issues mentioned in this ticket (table shading and image distortion) are not really part of this ticket, since those issues have to do with problems on the GateKeeper side, not things we need to adjust in post-processing.

I don't have any idea what "fixed on blue" means, but perhaps Margaret does. At least it sounds like a good sign!

I'm glad to hear the image problem will be fixed, too - I assume Egyptian Mau is the next WCMS release after NVCG (Devon Rex)?

I don't have any idea what "fixed on blue" means,

Fixed on BLUE means the development has been fixed on the WCMS development server. Blue indicates the next release, while Red indicates the current/production release.
For instance, look at the current Cancer.gov site compared to the NVCG site. If a fix needs to be performed on the current site, development would happen on Red. A fix for the new NVCG site is implemented on Blue.
There exist other colors but I think those are only used for software upgrades and we don't want to confuse us too much for right now. :-)

That's a little bit confusing, because we've been doing our CTS work on Red, and I though CTS was happening after NVCG.

That's correct but it's handled like a hot-fix instead of a release and - according to Blair - there won't be any more hot-fix worked on for the current Cancer.gov.

I believe what we need for this ticket is for the users to test Publish Preview on QA and identify any remaining problems in the output that we can fix on our end. I recognize that it can be tricky to distinguish between what we can fix or work around on our end (the CDR), and problems over which we (the CDR developers) have no control. If you know a problem can only be fixed by the WCMS team, please use the usual method for reporting problems with cancer.gov. For everything else, please report the problems here.

We QC'd the summaries in PP on QA on Friday, May 22 and reported several issues to Victoria (Sun) and Amy. I will paste below the issues Victoria identified as having a CDR component.

3. Summaries - Links to Drugs Approved for X Cancer pages bring you to the bottom of that linked page. These links bring you to the top of the page (as expected) on the live site. This will need updates the summary filters to create the URL correctly since the summaries have been moved to new locations for NVCG. Note from Victoria: This will be addressed in the next CDR (Curie) release

5. The following link types do not work: Glossary Refs, Summary Refs, Protocol Refs (we cannot remember which are expected to work and which are not). Clicking on any of these types of links doesn’t do anything. Summary Fragment Refs also do not work but this is a known problem that is tracked in the following issue: OCECDR-3839 Note from Victoria: This is a known CDR issue that will be addressed in the next or future CDR release (either Curie or Darwin)

6. Links to levels of evidence terms (LOE Refs) generate a 404 error. For example, this screenshot is taken from a link to the Level of evidence: 3iDii term in the the Adult Soft Tissue Sarcoma HP summary. Note from Victoria: This is a known CDR issue that will be addressed in the next or future CDR release (either Curie or Darwin)

I'm attaching a picture of the LOE ref error message that is mentioned in the previous comment.

~oseipokuw: I thought I had posted this request earlier, but I think JIRA swallowed it silently after timing out my session: could you share any notes you have for the CIAT users on what is known to work and what is known not to work for Publish Preview?

Thanks!

could you share any notes you have for the CIAT users on what is known to work and what is known not to work for Publish Preview?
Thanks!

I have attached a file that was last updated last year (Bastion Host and XMetal 4.5 era) so there may be one or two things that may be incorrect. I will review the list next week and note any features that don't work or work differently in the new environment.

Perfect! Thanks, William.

Yes, I agree with Bob! This is a great set of information and we should add it to the CDR documentation so that it's accessible to everyone. We should create a document for Testing Reports.

I have created a page in the CDR online documentation system to capture what should and shouldn't work in QC reports and publish preview:

https://cdr.cancer.gov/cgi-bin/cdr/Filter.py?DocId=CDR0000773122&Filter=name:Documentation+Help+Screens+Filter

This covers all of the issues in Robin's 2015-06-02 comment above except "Links approved for drugs ...." I can't find the link to that page in the example I'm looking at:

https://cdr-qa.cancer.gov/cgi-bin/cdr/PublishPreview.py?Session=guest&DocId=CDR0000062864&Version=cwd

Can you tell me where I should be looking?

I think those links are just added to the patient summaries. Here's an example:

https://cdr-qa.cancer.gov/cgi-bin/cdr/PublishPreview.py?Session=guest&DocId=CDR0000258013&Version=cwd#section/all

Search for "Drugs Approved For" and you'll see a few links to those pages. However, they seem to be going to the right places now (this is on QA).

As we discussed in the status meeting, Bob will see if he's able to remove the LOE pop-up error message.

Is there a summary that takes less time to process than the Adult Soft Tissue Sarcoma HP summary and still has the LOE problem? I've been waiting for about a half hour for my first run just so I can see what the markup I need to modify looks like, and it's still spinning. At this rate Curie won't be on QA until next week. :-)

Any of the HP treatment summaries should have LOEs. How about Bladder Cancer Treatment (CDR62908)?

CDR62902 is one of the smaller HP summaries. It's my go to for fast.

Sorry - apparently "any of the HP treatment summaries" wasn't accurate. I don't see any LOEs in 62902, so I'd stick with 62908. It's pretty short.

Thanks to both of you. Something funky is going on. The browser is sitting there forever while I wait. Then as soon as I give up and cancel the request the summary is displayed immediately.

It doesn't look as if these links have any helpful classes I can search on in the HTML markup. Should I disable all links which have an onclick attribute starting with "javascript:popWindow(...)"?

I think that would work. There shouldn't be any pop-up windows in the PP reports.

I have disabled all popup windows in Publish Preview on DEV. I believe we have identified (in the new online documentation page) the other issues as features which are not expected to work in PP. Marking this is resolved and ready for user testing on DEV.

• R13196 /branches/Curie/Inetpub/wwwroot/cgi-bin/cdr/PublishPreview.py

I have confirmed that we are no longer receiving the LOE pop-up box with an error message. Thanks. Verified on DEV.

We should talk about reference links though - I think those should work since they are links pointing to elsewhere in the same summary, much like a summary fragment ref. We can table this for Darwin since we're in crunch time but I don't want to lose the thought.

Reference links didn't make it, so please open a new ticket for Darwin. Thanks.

Turns out Citation Links ARE working on QA - but were not on DEV. A bonus! However, clicking "back" after clicking on a citation link brings us to the top of the summary rather than where we were when we clicked the link. The same thing happens with Summary Fragment Refs so I'll post a similar comment in that issue.

Isn't it correct that this is the same behavior as on PROD? If I remember correctly this is an issue with the JavaScript and is not directly related to this ticket.

Yes, this happens on the live site, too. We reported it during the initial NVCG testing in the context of summary fragment refs. I assume this same problem with citation links is related to that.

SummaryRefs ARE working in PP on QA, so we should move that from the list of things that cannot be verified to things that can be verified in PP. We might want to add a note that they take a very long time though, since they essentially generate Publish Preview of the related summary. They aren't something we necessarily want to be testing in PP since they take so long, but it's good to know that we can.

The right-nav is not displaying in Publish Preview on QA. I'm not sure if this is a Javascript problem, too, but it is not a problem on PROD.

I'm not sure where best to report this but the Genetics of Breast and Gynecologic Cancers summary (CDR62855) is not coming up in PP. I'm getting a "page cannot be displayed" error (twice) and Val got it once for the same summary, too.

The right-nav is not displaying in Publish Preview on QA.

Yes, that's correct. There is a problem with one of the JavaScripts because Bryan was testing something. He told me what I needed to do in order to fix this but I haven't had the time to do so.
I bet the TOC isn't coming up either, right?

You are correct 🙂

I'm not sure where best to report this but the Genetics of Breast and Gynecologic Cancers summary (CDR62855) is not coming up in PP. I'm getting a "page cannot be displayed" error (twice) and Val got it once for the same summary, too.

Can you give me the exact URL you're using?

Thanks.

Sure, here it is:

https://cdr-qa.cancer.gov/cgi-bin/cdr/PublishPreview.py?Session=5588478B-960693-218-VJI4Q4FGM6RM&DocId=CDR0000062855&Version=cwd#section/all

I am unable to reproduce the problem. It takes a while, but I get back the PP page. Can you try again tomorrow? Perhaps you could use this URL (with debug logging turned on) so I can see what's going on if it fails for you:

https://cdr-qa.cancer.gov/cgi-bin/cdr/PublishPreview.py?Session=guest&DocId=CDR0000062855&Version=cwd&DebugLog=1#section/all

Trying this now - thanks.

In the meantime, I also wanted to report that summary frag refs to a specific section of a different summary are working and the back button will take you back where you began. Like summary refs, these links take a long time to come up since they pull up PP of the related summary, but this is another thing we should adjust on the list of things we expect to work.

The summary didn't come up in Firefox. (attaching error)
Trying IE (which I meant to try in the first place, since that's our default for PP) now.

I'm getting the "page cannot be displayed" error in IE.

I just ran the Genetics of Breast/Gyn and Genetics of Colorectal Cancer summaries in PP on QA and both came up successfully, so that's good. Did you find anything helpful in the debugging?

That's good news. The debug logs confirmed that the server was doing what it should be doing. If you hadn't succeeded (and I continued to be unable to reproduce the problem on my computers) I was going to have you open a help desk ticket.

Sounds good.

I believe the only remaining issue related to Publish Preview is the right-nav and TOC not displaying. Should I be waiting to retest that, Volker?

Also just a reminder that we'll need to update the documentation to reflect the fact that SummaryRefs and SummaryFragmentRefs (External) are both able to be verified in PP.

I believe the only remaining issue related to Publish Preview is the right-nav and TOC not displaying. Should I be waiting to retest that, Volker?

Since we found out yesterday your JavaScript is working and the problem is related to this one summary (CDR0000517309) that's creating a JS error I'll double-check with Bryan. He was working on the JS on QA. Bryan had recommended to update a JS file in Percussion. I'll try to do this today and then we can check PP for this document again.

The problem is actually more widespread on QA - all summaries I'm running are missing the right-nav and TOC on QA. The summary we looked at yesterday (517309) was on PROD. I'm checking with Victoria to see if she's seeing the right-nav and TOC on QA on her computer so we can determine if it's something on my computer. Thanks.

Here's Victoria's report:

Hi Robin,

On QA, I do NOT get the right-nav and TOC; I get the whole opened document.

On PROD, I get the right-nav and TOC.

Victoria.

The JS file has been updated on QA. Please try again.

The summary we looked at yesterday (517309) was on PROD.

I'm sorry, I completely missed that point. I was under the impression that JS was working for you on QA just not for that summary we looked at. Hopefully, the section navigation and TOC are now working on QA for everyone. At least it started working for me after the update of the JS file.

So, this is odd, but I'm now seeing the right-nav and TOC for every summary that I am trying (initially I didn't see it for one summary [Childhood ALL HP] but after rerunning a second time it came up), but Victoria is not seeing it in any of the summaries she is running (even if she reruns them a second time). We tested using some of the same summaries and verified that we are each on QA. I had her run the Javascript test you had given me and she is getting the same results as I am on the red/blue test, in case that is helpful.

Also just a reminder that we'll need to update the documentation to reflect the fact that SummaryRefs and SummaryFragmentRefs (External) are both able to be verified in PP.

Done:

https://cdr.cancer.gov/cgi-bin/cdr/Filter.py?DocId=CDR0000773122&Filter=name:Documentation+Help+Screens+Filter

Thanks, Bob! I think we can remove the second bullet about SummaryFragmentRefs since the first bullet now says "internal and external".

SummaryFragmentRef (both internal and external)
SummaryFragmentRef (external)

Thanks, I meant to do that when I was editing the document. The superfluous bullet is gone.

This did turn out to be a caching problem. Victoria tried again holding down the shift key as you suggested and the right-nav and TOC are there. Thanks!

Verified on QA.

Please verify the publish preview report on PROD and close this issue.

I've verified this on PROD. I had the same problem reported above where the TOC and right section-nav did not display, but refreshing the page brought those back. As we learned on QA, another option is to shift+refresh if we see this again.

William, are you comfortable closing this issue? Thanks.

Yes, it is okay to close this issue.