Talk:Statistics of the COVID-19 pandemic in Brazil

Cases by Capital

An editor under the ip 88.136.208.70 ^{how do i link those?} added the above graph. But i am unable to find the source of the data he used to parse it. I would gladly update my scripts to update it, but without that information i am unable to, and have therefore decided to remove it from the article until the original author can help me with it, or update it himself. --Hagnat (talk) 22:07, 24 May 2020 (UTC)

Group by Week

@Ânes-pur-sàng, MatheusGamezi, Wvictor07, and Albertoleoncio: as we near the 3th month since the first case, the graphs have too much data to show. I am thinking about grouping them by week, this way things becomes more manageable and less cluthered. If someone wants to see fine grained growth, they can see the timeline tables, which additionally shows data by state. What you guys think ? --Hagnat (talk) 12:09, 25 May 2020 (UTC)

Seems sensible to me, this is not going away for at least 2-3 months so much more data needs to be shown. Ânes-pur-sàng talk 13;57 25 May 2020 (UTC)

Sounds good. --Wvictor07 (talk) 13:12, 25 May 2020 (UTC)

Agree. Albertoleoncio (talk) 14:51, 25 May 2020 (UTC)

After parsing the graphs using weekly data, they now have too little data. I have settled on showing the graphs every 3 days. --Hagnat (talk) 18:27, 25 May 2020 (UTC)

@CJK09: as you can see, i had a brief discussion with the main contributors of this page, and there was no opposition to group data, since the timeline tables show the same information and more. --Hagnat (talk) 22:10, 25 May 2020 (UTC)

@Hagnat: Is it a viable option to generate the graphs outside of Wikipedia and use .svg files to display the data? I think python certainly has a way to group the data into a bar chart like we want it to. The down side is that one would need to use another programming language (and a different set of tools) for that. --Wvictor07 (talk) 22:18, 25 May 2020 (UTC)

@Wvictor07: The guys at the portuguese wikipedia are using svg files for it, but it does not look like anything here on the english wp. It becomes even harder for people to collaborate, as it requires people to know how to modify svg files and/or have the tools to generate them. Right now, with the current version on the english wp anyone can update the graphs without no need for special tooling, as it only requires to input the values in the correct places. --Hagnat (talk) 12:02, 26 May 2020 (UTC)

Yeah, you are right - it really becomes harder for other people to colaborate. -- Wvictor07 (talk) 12:05, 26 May 2020 (UTC)

Ministry of Health data source

I am going to be honest here. I am getting tired of the incompetent job that the Ministry of Health is going with the shared data. Not only the current format is clunky, running on a closed source file, but it also contains a LOT of invalid data that needs to be sanitized -- such as duplicate rows, and rows that point to no valid region/state/city. I just wasted another two hours fixing their incompetence. AGAIN. It kind of frustrates me from continuing on updating this. Maybe i should start using the data from the guys at BrasilIO. Let them sanitize the data for me. /rant --Hagnat (talk) 14:58, 1 June 2020 (UTC)

@Hagnat: Has the information/data format been improved (or at least gone back to "normal") since the Ministry of Health took the data down last week? I didn't have the time to go over the data since then. --Wvictor07 (talk) 19:20, 13 September 2020 (UTC)

@Wvictor07: after the outage in June -- caused by the President ordering the Ministry of Health to stop releasing the COVID data -- i kind of stopped updating these templates/pages. Fortunately, there was a lot of public and political outcry and the data has once again returned to be published. The script i created seems to be still working, and you may have noticed that Brunoff has been updating it daily since i stopped. --Hagnat (talk) 14:59, 28 September 2020 (UTC)

Semi-protected edit request on 20 June 2020

4-day update. updated: all graphs except for one external svg (logscale graph). not updated: external tables and the external svg. Brunoff (talk) 03:52, 20 June 2020 (UTC)

Main article: COVID-19 pandemic in Brazil

 Not done: it's not clear what changes you want to be made. Please mention the specific changes in a "change X to Y" format and provide a reliable source if appropriate. — Yours, Berrely • ^Talk∕_Contribs 07:55, 20 June 2020 (UTC)

Backlogs and negative values displaying in statistical graphs

Hi @Brunoff: please do not delete the yAxisMax and xAxisMin functions as these help prevent data backlogs and negative values from disrupting the continuity of the graph (please, see the Statistics of the COVID-19 pandemic in Argentina page to have an appreciation on how these metrics prevent data outliers from disrupting the graphs). Also, keep in mind logarithms of negative values produce errors therefore you will see some disruptions in the logarithmic graphs if you decide to include negative values. Thank you. Joplin201017 (talk)

@Joplin201017: Thanks for pointing out. I think I have done it right now. I apologize for this and for the future times it will eventually happen again on future when new modifications arrive in the page. I regenerate the whole page and I manually check before applying changes but this one was small enough that I did not realize. The only way to avoid this would be to automate the difference detection but for now I don't have time for this. Brunoff (talk) 00:53, 29 September 2021 (UTC)

External links

• https://covid.saude.gov.br/ – Ministry of Health Statistics Panel, updated daily
• https://brasil.io/dataset/covid19/caso_full/ - Brasil.io, an aggregator of the data published by the Brazilian Ministry of Health

PEIS warning

The following warning appears on the page when you preview it:

Warning: Post-expand include size is too large. Some templates will not be included.

In charts and tables, I suggest switching their source data from daily numbers to something more practical like weekly or monthly numbers. If any time-sensitive data isn't going to be updated, then delete it. Jroberson108 (talk) 16:09, 25 February 2022 (UTC)

@Brunoff: you need to stop updating this page with daily stats. Wikipedia is not a database, this amount of stats is excessive and has broken the page. My browser crashed when I tried to open it when it appeared in Lint error reports due to use of <center>...</center>, which I had replaced in the artcile a couple of weeks ago and readded by you. Please update whatever script you are using to use <div class="center"></div> instead of center tags.

Also per above, the references and bottom half of the page does not render properly due to excessive tables and graphs. I will be working to condense the article to use weekly number. ಮಲ್ನಾಡಾಚ್ ಕೊಂಕ್ಣೊ (talk) 13:33, 5 November 2022 (UTC)

It looks like the biggest reason for the page exceeding template expand limit is the graph charts. It is useless to add daily data of the last two and half years to graphs since it doesn't make much difference to page rendering and gretly increases parsing time. Even weekly is not necessary, I have condensed it monthly which gives adequate scale of how the numbers progressed. ಮಲ್ನಾಡಾಚ್ ಕೊಂಕ್ಣೊ (talk) 15:14, 5 November 2022 (UTC)

You are right. I will stop updating daily data for now and I will continue updating once I adapt to the new format. Thanks for your suggestions. Brunoff (talk) 01:51, 8 November 2022 (UTC)