[knitr] Pandoc: Convert Markdown to other formats via Pandoc | /knitr/demo/pandoc/

It is not possible to produce an Rnw file from md. You can only get a tex file. To do that, you need to specify the output option, e.g.

format: latex
o: foo.tex

Originally posted on 2013-06-20 00:19:00

fixed; thanks! please also see: http://yihui.name/en/2013/06/fix-typo-in-documentation/ :)

Originally posted on 2013-10-19 21:27:49

Sorry, I do not see how your version is better than Pandoc, and I do not see the limit of the number of characters in the Pandoc online demo, either. Your statement about that limit is probably false.

BTW, I feel it is a little bit unfair for you to claim Pandoc is less capable than your version, when you developed your version based on Pandoc without releasing the source code and a detailed list of comparison.

Originally posted on 2013-11-25 04:41:32

Guest *Alec Kinnear* @ 2013-11-26 00:35:29 originally posted:

To see how our version is better than Pandoc, please try some conversions back and forth between Textile and Markdown for instance. As we moved from Basecamp (Textile) to TeamworkPM (Markdown) we needed that pair to work perfectly.

What part of "Please limit input to 1000 characters or less." is false? Graphic is from: http://johnmacfarlane.net/pandoc/try/

We are providing a service for free. We are not interested in replacing Pandoc for local installations or undertaking its maintenance.

Not quite sure why we should face slander and reproach for offering our internal tools for free to the community.

I hope this answers your questions.

I just checked the HTML source, and the textarea does have a limit of 3000 characters. I have also attached a screenshot of 3000 chars. The error message from the online demo is false, and it is actually not your problem. When I read your page, tested more than 1000 characters, and it worked, so I thought there was something wrong. Sorry about that. For Pandoc as a personal project, I think it makes sense to limit the number of characters for the online demo. On one hand, it is just a demo to give you a feel instead of a service; on the other, the author may have limited resource and he has no obligation to put together more of his own resources for other people to use. For me, I just download it and use it locally. I do not want to upload my real data to a "cloud" that I have never heard of.

To clarify, I did not mean slander and reproach. If there was a misunderstanding, I apologize. But I do think you need to provide more details to support your claim that your modified version is better, and Pandoc is "partially broken" (broken in what sense?). It gives the reader an impression that your version is absolutely superior without realizing that Pandoc has also done a huge amount of work. Isn't it a good thing to see what you want are included in Pandoc directly? Have you contacted the author? Did he refuse to listen? If not, I'd say it is a better idea to collaborate than to go on your own path. If you believe something is broken, filing bug reports should be easier than starting hacking at it. Fixing bugs will benefit all people.

I do not mean you should make your version open source. That is totally up to you. I'll just be excited if people can collaborate to make the tools even better.

Originally posted on 2013-11-26 18:03:51

Guest *Alec Kinnear* @ 2013-11-29 03:33:29 originally posted:

Hi Yihui,

Interesting that it actually breaks at 3000 character and not the announced 1000. Still far too small.

We are providing a 100% free and unlimited Pandoc conversion service which is permanently online and maintained (and yes tools like this break if they are not maintained). Still not sure how we became the bad guys here.

We are not interested in maintaining or taking over the Pandoc project but are helping as we can, by making sure that a working and tweaked copy is online all the time for people all over the world to use. It's not a cloud service, requiring an account or identification but just a script which runs in real time delivering converted text (not even documents).

Personally I think trying to maintain Pandoc on different computers is a huge pain in the butt which is why I had our team create this, so that everyone in Foliovision and the rest of the wide world could run conversions without dealing with the hassles of installing and maintaining Pandoc.

I understand the good intention, and I appreciate your generosity. All I wanted to say is, you may talk to the Pandoc author(s) to make both parties in a win-win situation, if you believe Pandoc is broken in some sense. You do not have to write code or maintain the whole package, but your feedback will be highly appreciated if it pushes Pandoc to a better direction. You do not lose anything if your suggestions are turned down, but wouldn't it be great if the things that you wanted just end up in the official Pandoc distribution some day? In short, I suggest "talk before fragmentation".

To clarify again, I did not say you were the bad guys. I'm not the judge in any sense. I was just wishing people could combine forces to make one thing better. You certainly have your freedom choosing not to do so, and tweaking Pandoc without letting other people know what you really did.

Originally posted on 2013-11-29 05:58:46

Rnw是LaTeX和R的混合体，建议你放弃LaTeX转Word的想法。LaTeX的天然输出是PDF，建议放弃HTML的想法。

https://github.com/yihui/r-ninja/blob/master/11-auto-report.md

——原帖发布于 2013-11-30 02:57:58

I think Pandoc is the best way to go. "Wholly within R" basically means wrappers for Pandoc in R.

Originally posted on 2014-01-10 03:43:56

You have to specify a field "o: foo.tex" in the config.

Originally posted on 2014-01-30 02:20:42