I love knitr & rmarkdown, but I often find myself in situations where I have a lengthy report that takes some nontrivial amount of time to run. After it's generated, I notice my inevitable typos in text. However, re-knitting everything to just fix a couple typos (just in text, not code) takes a long time and seems avoidable. I was about to start taking a hack at developing my own solution to this, but I'm thinking it's the kind of thing that could already have a mature solution which would likely be more robust than the one I'd build.
I'm wondering if there is solution out there within knitr or third party that would allow me to edit just the text of my reports without rerunning code, generating plots and outputs etc. I know, I can simply edit the generated html text, but then those changes must be replicated in the R/Rmd code that generated it, or they get out of sync. I'm envisioning a function like this:
argument 1: the R/Rmd script with text edits (no code changes)... perhaps a warning is generated when code chunks change
argument 2: the html output file from the last time the R script in argument was knitted without the text edits.
return: the html report (argument 2) updated with the comments in the R/Rmd script (argument 1).
I use the cache option sometimes for large datasets. I toggle eval and echo on and off when developing if I'm just working on the text of my report. However, I'm looking for a function that would take care of all this for me, so one doesn't have to mess with the code and chunk options to make small edits to text.
Here's an interim solution that lets you retain the speed of making changes directly to the rendered text, but you have to do a little work after you're done making changes.
Assuming the following files:
input.knitr is the original Knitr file with text and code integrated.
output.html is the resulting HTML code that has been rendered by Knitr.
Consider making direct text edits to output.html and then running something like Meld visual merge tool:
meld output.html input.knitr
Then manually select the edits in output.html that are new and should be fixed in the original source input.knitr. Tools such as Meld do a pretty good job of aligning the texts so that the chunks and knitted output will appear as large "changes" that, in practice, you would ignore. You would focus on the small changes in the non-chunk sections.
Related
To preface this, I think my question is related to this, but it's not exactly the same: How to source R Markdown file like `source('myfile.r')`?
Basically, I perform most of my data cleaning and analysis in Rmarkdown files because the visual separation between chunks of code and my own comments on what should be done for the analysis/cleaning is very helpful to me. It also helps that within Rstudio if you run a table, df, then it'll display an interactive snippet of it in the document. This is all very helpful in complicated cleaning/analyses. So in other words, I'd like to develop in one R-Markdown file and write in another R-Markdown file. Splitting/writing the code into a source.R file is not ideal, unless there was a very automated and reproducible way to do it.
The issue is that for reports, I'd sometimes like to take specific objects that were generated from these lengthy data-cleaning and analyses files in Rmarkdown. For example, let's say that during my data-cleaning in Rmarkdown-file-1, there was a particular table that was giving me trouble problematic.df and that I'd like to call in my report or possibly perform further manipulations in my report (Rmarkdown-file-2).
So ultimately I think this is the question:
How can I call any arbitrary object generated at any arbitrary point of one Rmarkdown file in another Rmarkdown file?
Obviously, the above would be the ideal, but it sounds unreasonable, so perhaps this is a better question/request:
How can I call any arbitrary objects generated by the end of one Rmarkdown file in another Rmarkdown file?
Upon further reflection, my question might already be answered in the post I linked, but it's been a while since that question was posted and perhaps there are new solutions or perspectives on this issue.
So I recently helped write a code for my lab which takes our processed data and makes a merged data frame of it. For purpose of keeping the lab updated, we keep our data tables updated on a secure wiki and thus I need an HTML made so I can basically upload the dataframe onto the wiki easily. It's worked before - all I did was basically copy what was already written and working and edited it to work for a different time point in our data collection. I have no errors given back to me and the data looks how I want it to look. As far as I know this script should be written logically and working well and so far it does except for one issue: R will make a file for the HTML, but there is no HTML written in the text document.
I have HTML's written from the other data time points which are written the exact same as this one, so I don't think it is a script construction thing.
Any ideas as to why this could be happening? I just need to know where to triage.
The package used for HTML is R2HTML, included in my packages list up at the top of the script. For HTML(, file=paste()), you will need to use your own directory to see if the HTML is written as a text file.
If I am not wrong , You are trying to get the dataframe in html format .
In this case you need to use xtable package in R
Just the below code in bottom of the script
## install the xtable package before importing it
library("xtable")
print(xtable(ChildSRPtotsFU_wiki), type="html", file="check_stack_overflow.html")
I am a new user to R Markdown and was wondering if there was a way to "incrementally" compile an html markdown page as I am writing the code. Say I add 20 lines of code to an existing markdown file today. Is there a way to have the program "remember" the past compilations so that only the 20 new lines are added to the html file preserving the past rendering. I have a lot of memory heavy steps in my code (loading unloading files), and I find that when I add new bits of code I am having to compile everything from the beginning.
I tried looking into the "cache" option but it does not seem to be working.
I am assuming that all the variables I will be needing are present in my environment. In other words, i want to incrementally build an html markdown file without having to compile everything the moment I add an extra line to an existing document. Thanks for your help!
You should look into using the excellent editR package. It does exactly what you're looking for:
https://github.com/swarm-lab/editR
I have a .Rnw document in which I include childs. The childs produce tables via the 'latex' command of the Hmisc library in R.
When I make changes in the child documents, these changes do not anymore change the pdf document. My first guess was to use the chunk option 'eval=TRUE', but this does not change anything. Then, I saw, that the tables are actually saved to a .tex file with same name as the .Rnw document. I deleted this file and after compilation with knitr I got an error:
Error: Latexmk: Could not find file documentname.tex.
I assume, this is not the way to do it. Now I am out of ideas what to do. I appreciate some help on my problem.
Best
Simon
Allright, when trying to construct a simple example, I actually found out, that neither the packages I included nor the nesting of child documents interfere with the compilation via knitr. The reason was a simple error in the low-level .Rnw document, where a Hmisc latex table had a label, that missed a closing speech mark.
This causes then the output pdf not to change - I assume, that in this case the already constructed .tex file is included instead of letting knitr recompile the .Rnw documents and this hasn't changed since the last compilation?
What I wonder about is the different format of the landscape ctable in the document. Using a simple knitr document just with \documentclass{article} produces well placed tables. In my document using a template for the JFE, I get a table that extends over the whole page and even in footnotesize it is far away from the great appearance in the simple document. There is only a margin of less than half a cm on the right and the left. Page size is the same: both US letter... Can I probably control that via knitr or only via resizebox?
In conjunction with trying to find a solution for myself in regards to this question, I find myself plunged into trying to write valid Rd markup. What I want is to add a section named Raw Function Code and put the code of the function under it. I've achieved limited success in this regard by writing a script to modify the Rd files to include
\section{Raw Function Code}{\code{
# some piece of R script will eventally provide this part of the text
}}
However, even if I manually properly spaced text into the .Rd file (using either spaces or tabs), the initial white space of each line seems to get stripped away leaving an undesirable looking function. I've noticed that if I provide a starting character before the white space the white space is retained. However, I did not want to provide a starting character because I'd like people to be able to copy and paste directly from the produced PDF.
I have reviewed parseRd and I know there are three types of text LaTeX-
like, R-like, and verbatim. I have tried to put my function code in \code and in \verb and neither seemed to yield the desired results. What can I do to hold onto my initial white space?
The \section macro contains LaTeX type of text, but as you want to write code, you could use \synopsis macro, i.e.
\synopsis
# some piece of R script will eventally provide this part of the text
}
There is one problem with this though; you cannot give name to this section, it is automatically named as another usage section. Same thing could be achieved by using \examples macro, but now the name of the section is Examples, which is probably even more dubious (not to mention that you probably already have Examples section).
It isn't possible without modifying the usage or examples sections of your Rd code. See Hemmo's answer for a usable workaround. It produces text in the verbatim mode which is sub-optimal, but far better than nothing.
(This answer is set community Wiki in case this state of affairs changes. This result is current as of R-2.15.1)
If you want a super hacky way to do it, you can use \Sexpr to make zero width characters and add spaces between them:
#' first line \cr
#'\Sexpr{"\u200B"} \Sexpr{"\u200B"} \Sexpr{"\u200B"} \Sexpr{"\u200B"} indented line
A warning however - your package will build fine, but R CMD CHECK will throw a fit.