All of us were students. Everyone has an episode in their lives when they didn’t know what they actually knew and decided to take part in something, start working in that field or just learning more and more. And within the time we learned that we didn’t know anything. Zero. I hope that Google Summer of Code students are different in that case from other regular students. I hope they know what they do and every week they gather knowledge that helps them in fulfilling requirements for their project goals. Today, we’ll focus on problems which they have to face and if they found solutions for issues.
Bartłomiej from Google Summer of Code
Let’s start with Bartłomiej who works in “The REST of Administration” project.
He had to face problems in the first week. They consisted of difficulties in installing the environment in which he would work. Fortunately, these weren’t complicated mistakes and he was able to deal with them quickly.
Unfortunately, the next week did not spare him and this time he encountered great difficulties in documentation. As he mentions, “not every endpoint is documented well, and in some cases, required features were already implemented but not yet documented :)”. Unfortunately, it is a common problem in organizations, especially in open-source, that documentation is neglected and often made haphazardly. unfinished or containing only a scrap of information.
There were also problems with tasks that seemed to be relatively trivial. Unfortunately, the characteristics of OpenMRS often make it impossible to copy code from one part of it to another. Often the code has been written by different people and in different years, so it can differ significantly from one another. Bartholomew did not predict it, which unfortunately caused further turmoil and complications. Within 4 weeks, he managed to complete 6 tasks and send 3 tasks to code-review. Unfortunately, the task https://issues.openmrs.org/browse/RESTWS-842 and the associated blocker https://issues.openmrs.org/browse/TRUNK-6010 still remain unresolved.
Medhavi pracuje przy projekcie “Support for Extended Operation in FHIR.”. She also has to face up troubles during the community bonding period. She wrote “I realised that my initial implementation was not performing exactly what the operation was required to perform, so I changed my implementation and faced another problem- Hibernate’s Criteria API did not support some of the required functions. I briefed my mentor about these changes and issues. We discussed a workaround to the lack of window functions(dense ranking) in Criteria API. Another issue was that the OperationParam does not support chaining functions. My mentor advised me to drop chaining for the first pass and, come back to it later. I created the relevant tickets and epic to get ready for the coding period.”
Fortunately, later on she hasn’t occur any big issues which blocked her for a while. There was a little one, like not creating a unit tests so she has to do it in seperate task.
It looks like she has everything under control and are doing a great job!. We keep fingers crossed for her
Saurabh is working on Android Client Project this year. During that 4-week period he has to solve one proble (which is quite fresh)
“We had some Issues with our HTTP debugging library due to some NPE Issues turns out it’s a known bug here. It will get resolved by updating the dependencies and the Issue has been made ready. Kudos to Shivam for work on this Issue !! Moreover, another Issue was in the PatientPhotoActivity due to the ImageView throwing up NPE but only sometimes, so I added a sanity check there to avoid this Issue. One last bug that I solved this week was with the providerDashBoard activity where the updates to the providers were not getting reflected in the recycler view, so I made sure that the recycler view gets”
Heshan’s work is different from others because he not only tries to solve the problems but also prepare a video/ documentation where he describe the solution and how to achieve that. His first issue was how to export and import period indicator reports from instances. Of course he found a solution for that, what’smore he crated a video in which everything is described (https://youtu.be/fytXa-tk-1A). The next weeks weren’t easier as he wrote “At first there were some issues that I had to face because I kept getting an error after error. But somehow with the help of my mentor, Kaweesi Jospeh, and hours of debugging I finally got it to export a DHIS mapping with the corresponding Period Indicator Report and import them back to a different instance.”.
And more problems: “for some reason, it doesn’t map the period indicator report with the other metadata and when I try to run the report I get this error.
org.openmrs.module.reporting.evaluation.EvaluationException: Failed to evaluate Error evaluating because: Cannot find parameter 'endDate' in org.openmrs.module.reporting.dataset.definition.CohortIndicatorDataSetDefinition null “
Vineet shares a great post about beginnings as a GSOC student, what features you need to have to be a good student and contributor, and how to unstoppable develop depending only on your own skills. He is a part of Micro Frontend Squad, and working on Clinical Data Management.
Piumal was selected for project called ‘Improving Functionalities of DHIS Connector Module’. It’s under the ‘Module Development’ Category. During his work on DHIS2 API, he faced a few problems which stopped them for a week until he found a solution for them. To not waste a time and get a higher chance to get a help he created tasks which you can see here: https://issues.openmrs.org/projects/DCM/issues/DCM-47?filter=allopenissues . “Next week I hope to start working on other issues parallelly, instead of working only on the location mapping feature. At the moment I have already created a set of issues in JIRA, so I hope I’ll be able to send 2–3 pull requests in the upcoming week.” There is also good news that things are moving forward, “You may remember the Internal Server error that occurred In OpenMRS API last week. And I had created a thread to find the problem. The good news is, after some discussions, the openmrs community found the reason for the problem. So now I can find a way to fix it.” Good to hear that 🙂
That’s almost all. As you see our students have to face a lot of problems, bugs and issues and for sure they are frustrated because of that reason. We haven’t listed here all GSOC contributors as the rest of them haven’t shared their issues, or they weren’t described accurately enough. Don’t worry, later or sooner everybody will have to face to code monsters.
Keep in touch with us and GSOC students! More informations and newses u can find here.