Missing data in extracted data table after user adds more references

[MJWainwright]

Hi @damian-garrido A user reached-out to me with an urgent bug that they have found. Their review is due next Friday so it is very time sensitive. As a first priority we should try to fix the issue for the user so that they can move on with their GRADE-CERQual assessments so they can meet their deadline. Currently, it's a "loss of data" situation as the extracted data is nowhere else in iSoQ. I hope you have a way to see it in the background and person restore it. Secondly, we should try to understand and fix the bug to prevent it happening again.

Here's what happened

• The user created an iSoQ project with review findings and completed all the GRADE-CERQual assessments over 9 months ago.
• They then updated the review and added 8 references to their review by importing a second RIS file they exported from Endnote
• They assigned 2 of the 8 new studies to the first subtheme/review finding (information trustworthiness).
• When they then went into the worksheet for that review finding (first sub-theme) to re-assess GRADE-CERQual to account for the fact that new references were added, they found that all the data (quotes) for the original studies were no longer visible.
• Note that the person who reached out to me was not the creator of the original project, but I think that they were the person to import the additional 8 references. If this information is relevant let me know and I can confirm this with them.

I will send you a separate email with the user's account so you can log-in.

I looked quickly through the project and the only abnormality apart from the missing data that I could see was that there was an additional blank first row in the Methodological assessments table that shouldn't be there. See below

[damian-garrido]

hi @MJWainwright i made a hotfix for this issue, could you check on the test server please, at least for the first one, the second one i couldn't reproduce so i need investigate about this.

[MJWainwright]

@javierapena it would be great if you could also test this fix and let us know how it goes. Thanks!

[javierapena]

OK! I'll take a look and let you know.

On Fri, Mar 22, 2024 at 6:11 AM MJWainwright @.***> wrote:

@javierapena https://github.com/javierapena it would be great if you could also test this fix and let us know how it goes. Thanks!

— Reply to this email directly, view it on GitHub https://github.com/epistemonikos/isoq/issues/241#issuecomment-2014662932, or unsubscribe https://github.com/notifications/unsubscribe-auth/BCTKOWLI2KNEZWNWX5PFQBLYZPYUZAVCNFSM6AAAAABELXQH46VHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMZDAMJUGY3DEOJTGI . You are receiving this because you were mentioned.Message ID: @.***>

--

[image: Epistemonikos Foundation] https://www.epistemonikos.cl/ Javiera Peña Metodóloga @.***

[javierapena]

Hi!

I couldn't replicate the issue. I've followed the same steps you described (created project > complete GRADECERQual assessments > upload a new RIS file [with different references but also with some duplicates to test if this could influence the error] > assigned references to findings > re-assess GRADECERQual > all the extracted data were still available in .

I've also tried eliminating a previous reference to 'replace it' with a new study, and all the extracted data were visible for the updated included references.

For the second one (additional row in methodological lim.), I couldn't replicate the issue.

The only strange thind I noticed was the "Author , year" distribution in the first column when I downloaded the csv file as we discussed in the last meeting. I don't know if this is a different issue.

Let me know if I can help with additional testing for this issue. Javiera.

[MJWainwright]

@javierapena Thanks for checking! I found an issue, see below. Can you see if you also find this problem, on BOTH the test server and the production server and report back here? (The issue you encountered is a separate problem and I'm going to open a ticket for that.) @damian-garrido I've just done a test and I've encountered an issue that wasn't there before. When I import the new load of references the findings that I did have look like they have no references assigned which isn't the case. See images below. The process:

• I imported refs
• I imported tables and added inc. exl. criteria
• i added three findings and assigned them references
• i completed the CERQual assessments for the three findings
• I imported a second set of references
• when I went to add references to finding #1 I saw that in the column references it said "There are 0 references."
• When I add two more references, then the number of references appears as it should, but if I refresh the screen or navigate to another page and come back to the iSoQ page it changes back to "There are 0 references".
• The extracted data for the original studies is there however, so that seems fixed. So perhaps you can try to replicate and decide whether the issue of "there are 0 references" is a separate Bug or related to the hot fix you made for the disappearing extracted data bug.

[damian-garrido]

@MJWainwright i'm going to check, maybe is related! i'll be back with an answer.

[damian-garrido]

@MJWainwright i follow the steps that you mention but i didn't have the failure that you mention, did you this on the test or production server? i made on the test server.

[MJWainwright]

@damian-garrido I did it on the test server. @javierapena can you try?

[javierapena]

Hi, I couldn't replicate it either :/ but we will meet with Damian to explore it.

[MJWainwright]

@damian-garrido I've gone through testing this again a-z and did not reproduce the problem I had last time (of the number of references being incorrect until I refreshed). Perhaps this was just a server communications/internet speed issue I had at the time.

In any case, can you please publish this fix to issue #241 to the production server and I'll check it there. Thanks!

[MJWainwright]

@damian-garrido Is this ready for testing on the publication server?

[MJWainwright]

@damian-garrido It doesn't look like you published the fix to the production server. Can you please confirm?

[damian-garrido]

@MJWainwright this update should be deployed on the production server but maybe i never let you know. im going to check!

[MJWainwright]

@damian-garrido . Checked, but this was never closed. Closing now