iterative reporting - docx from dataframe

[zoe-xanthellae]

Thanks for the super helpful content here, I've been looking for something like this to help me iteratively produce reports! I have a few questions about tweaking the functionality of this code:

• Currently you have information for formatting html/pdf output, is there a way to iteratively create outputs to docx? The ultimate format of my documents is docx using a template docx as specified in the YAML header in the rmd file

• From my understanding, the example data table, mtcars, has unique car names for the loop to pull from, then grabs data from that cells in other columns in that row for the report. In my dataset (test, see below), I would like to have the loop function over each "doc", printing all the rows in "markd" in order to create a marked down word document (should produce two, blah 1 and blah 2). For reference, this is the chunk settings/code I was previously using in the rmd file to produce one "doc" from this dataframe:

# results='asis', echo=FALSE, warning=FALSE, message=FALSE, evaL=FALSE
doc1 <- test %>% 
filter(., doc == "blah 1")
cat(doc1$markd, sep="\n", fill = 2)

Rendering docx files

If I'm understanding your code correctly, this is what the render code would like in the Rscript file (though I'd want it to print docx files):

for (doc in unique(rownames(test))){
  rmarkdown::render('./Docs/renderReports.Rmd',  # file 2
                    output_file =  paste(doc, ".docx", sep=''), 
                    output_dir = './Docs')

Example data.frame file here:
test.txt

I'm somewhat of an R novice, so let me know if you need any more information!

[majerus]

Great questions. I’m currently on a hiking trip and will take a look at this next week. If you don’t hear from me in a week please follow up.

On Thu, Mar 22, 2018 at 5:26 PM Zoë ***@***.***> wrote: Thanks for the super helpful content here, I've been looking for something like this to help me iteratively produce reports! I have a few questions about tweaking the functionality of this code: - Currently you have information for formatting html/pdf output, is there a way to iteratively create outputs to docx? The ultimate format of my documents is docx using a template docx as specified in the YAML header in the rmd file - From my understanding, the example data table, mtcars, has unique car names for the loop to pull from, then grabs data from that cells in other columns in that row for the report. In my dataset (test, see below), I would like to have the loop function over each "doc", printing all the rows in "markd" in order to create a marked down word document (should produce two, blah 1 and blah 2). For reference, this is the chunk settings/code I was previously using in the rmd file to produce one "doc" from this dataframe: # results='asis', echo=FALSE, warning=FALSE, message=FALSE, evaL=FALSE doc1 <- test %>% filter(., doc == "blah 1") cat(doc1$markd, sep="\n", fill = 2) Rendering docx files If I'm understanding your code correctly, this is what the render code would like in the Rscript file (though I'd want it to print docx files): for (doc in unique(rownames(test))){ rmarkdown::render('./Docs/renderReports.Rmd', # file 2 output_file = paste(doc, ".docx", sep=''), output_dir = './Docs') Example data.frame file here: test.txt <https://github.com/majerus/r_code_tips/files/1839272/test.txt> I'm somewhat of an R novice, so let me know if you need any more information! — You are receiving this because you are subscribed to this thread. Reply to this email directly, view it on GitHub <#1>, or mute the thread <https://github.com/notifications/unsubscribe-auth/AIlSLeWjrogGJYNo_Gn0XgMu4KTDI4Zaks5thBcTgaJpZM4S30jU> .

-- Rich Majerus richmajerus.com

[zoe-xanthellae]

@majerus Hi Rich, hope you had a great hiking trip! Just following up on your email, any thoughts on my above issue?

[majerus]

@zoe-xanthellae, thanks for the follow-up!

I just took a quick shot at putting together an example that I think gets at what you are asking. Check it out here and let me know if this is what you are looking for:

https://github.com/majerus/r_code_tips/tree/master/iterative_reporting/r-tmp

[zoe-xanthellae]

@majerus One step closer! This definitely creates docx reports, very exciting. It also prints the "i", or the unique doc name there which is handy. However, none of the Names catenated out for whatever reason...but when I put your "Names: (cat..." code inside a chunk and gave it the chunk options I mentioned earlier, it knit them out!

BUT ,this knits out the entire "names" column, rather than just the part of that column that belongs to the "i" doc. In other words, the doc 1 report has all names printed, when it ought to only be printing up until (and including) row 5. Thoughts for how to specify?

[zoe-xanthellae]

@majerus Update: I was able to fix it to print only the data having to do with the unique "doc" while doing iterative reports, just needed to add the code to create a filtered dataframe within the loop on the .R file! See code below (GitHub won't let me upload files here/to the branch):

Loop.R:

# Create sample data
tmp <- tibble(
  doc = c(rep("doc1", 5), 
          rep("doc2", 5)),
  name = randomNames(10)
)

for (i in unique(tmp$doc)){
  tmp1 <- tmp %>% # Creates a temporary dataframe with just names belonging to that "i"
    dplyr::filter(., doc == i)
  rmarkdown::render(input = paste0(getwd(), "/RmdRender/Loop.rmd"),
                    output_file = paste0("test_report_", i, "_", Sys.Date(), ".docx", sep=''),
                    output_dir = paste0(getwd(), "/RmDRender/reports/")
  )
  
}

Loop.Rmd:
Doc: r i

cat(tmp1$name, sep="\n", fill = 2)

Thanks for all your help! This is going to save heaps of time.

[majerus]

Glad to here this is working for you. Nice work! Let me know if you run into any other issues.

…

On Fri, Apr 6, 2018 at 3:27 PM, Zoë ***@***.***> wrote: @majerus <https://github.com/majerus> Update: I was able to fix it to print only the data having to do with the unique "doc" while doing iterative reports, just needed to add the code to create a filtered dataframe within the loop on the .R file! See code below (GitHub won't let me upload files here/to the branch): *Loop.R:* # Create sample data tmp <- tibble( doc = c(rep("doc1", 5), rep("doc2", 5)), name = randomNames(10) ) for (i in unique(tmp$doc)){ tmp1 <- tmp %>% # Creates a temporary dataframe with just names belonging to that "i" dplyr::filter(., doc == i) rmarkdown::render(input = paste0(getwd(), "/RmdRender/Loop.rmd"), output_file = paste0("test_report_", i, "_", Sys.Date(), ".docx", sep=''), output_dir = paste0(getwd(), "/RmDRender/reports/") ) } *Loop.Rmd:* Doc: r i cat(tmp1$name, sep="\n", fill = 2) Thanks for all your help! This is going to save *heaps* of time. — You are receiving this because you were mentioned. Reply to this email directly, view it on GitHub <#1 (comment)>, or mute the thread <https://github.com/notifications/unsubscribe-auth/AIlSLU_dSzPXyzOYzwOs8hDCyJYA_xl_ks5tl8GogaJpZM4S30jU> .

-- Rich Majerus richmajerus.com

[GutoBarros]

Just to say a big THANKS! for your solution. I've been struggling with a huge bookdown project and your approach taking the loop outside the .rmd file made it possible. Thank you very much!

[majerus]

Awesome!

[zoe-xanthellae]

@GutoBarros You're welcome! This ended up saving so much time, and opened up more possibilities for other projects, so I'm excited to hear it's useful for you too!