[Bioc-devel] topGO and cat() and print() statements in program code

[Witold E Wolski]

I did replace the call to cat in topGO with with a call to a package
private .cat2message function (which makes some adjustments to the
passed arguments so that the messages look as before). The patched
package is available from here.
https://github.com/wolski/topGO
Upstream commits made on 3 February by h.bengtsson are merged.
AFAIK all is working OK (checks pass and my code using topGO is still working)


Witold






On 4 February 2016 at 11:56, Witold E Wolski <wewolski at gmail.com> wrote:
> Hi Dan,
>
> Excellent suggestion. This is very similar to how I am plan to do the
> refactoring. Write a function cat2message which has a signature like
> cat but uses message in the implementation and than replace all calls
> to cat in the topGO package.
>
> best
>
> On 4 February 2016 at 00:56, Dan Tenenbaum <dtenenba at fredhutch.org> wrote:
>> Another dirty hack is to redefine cat() in your own code to do nothing:
>>
>> cat <- function(....) {}
>>
>> Dan
>>
>>
>> ----- Original Message -----
>>> From: "James W. MacDonald" <jmacdon at uw.edu>
>>> To: "Witold E Wolski" <wewolski at gmail.com>
>>> Cc: "bioc-devel" <bioc-devel at r-project.org>
>>> Sent: Tuesday, February 2, 2016 6:50:21 AM
>>> Subject: Re: [Bioc-devel] topGO and cat() and print() statements in program   code
>>
>>> I can't speak to the issue of changing somebody else's code without forking
>>> (which you are free to do), or getting their OK. But do note that there are
>>> usually ways around this. First, you can use include = FALSE in your chunk
>>> options statement, which will run all the code, but silence everything.
>>> This isn't a good use case if you need to print, but that can usually be
>>> split out. Something like
>>>
>>> ```r{noisypart, include = FALSE}
>>>
>>> noisy code goes here
>>>
>>> ```
>>>
>>> ```r{quietpart, echo = FALSE, fig.cap = ""}
>>>
>>> plots go here
>>>
>>> ```
>>>
>>> An alternative is to use GOstats, which may be less noisy, but which still
>>> has 18 calls to cat() (vs 121 for topGO) and 2 calls to print() (vs 29 for
>>> topGO).
>>>
>>> Best,
>>>
>>> Jim
>>>
>>>
>>>
>>> On Tue, Feb 2, 2016 at 5:19 AM, Witold E Wolski <wewolski at gmail.com> wrote:
>>>
>>>> Hi,
>>>>
>>>> I am using the very usefull package topGO to generate a report (R
>>>> markdown). There is not much to complain about topGO (on the contrary)
>>>> except that the package uses cat instead of message to display
>>>> progress information. which ruins the report.
>>>>
>>>> Also the bioconductor package guidelines state:
>>>> cat() or print() are used only when displaying an object to the user,
>>>> e.g., in a show method.
>>>>
>>>> This makes it difficult to integrate topGO.
>>>>
>>>> I did contact the maintainer asking to update topGO. However I did not
>>>> get an reply. It seems that the maintainer is occupied with other
>>>> problems. I did offer to replace the cat with message for these
>>>> functions I am using myself. No reply.
>>>>
>>>> So what I am wondering is... If I do the corrections, and would like
>>>> to commit the code... Sure I could create a branch but since I do not
>>>> have write access to svn no chance to push (commit) it for review.
>>>> And who is going to review it if the maintainer does not have time?
>>>>
>>>> best
>>>> Witold
>>>>
>>>>
>>>>
>>>>
>>>>
>>>> --
>>>> Witold Eryk Wolski
>>>>
>>>> _______________________________________________
>>>> Bioc-devel at r-project.org mailing list
>>>> https://stat.ethz.ch/mailman/listinfo/bioc-devel
>>>>
>>>
>>>
>>>
>>> --
>>> James W. MacDonald, M.S.
>>> Biostatistician
>>> University of Washington
>>> Environmental and Occupational Health Sciences
>>> 4225 Roosevelt Way NE, # 100
>>> Seattle WA 98105-6099
>>>
>>>       [[alternative HTML version deleted]]
>>>
>>> _______________________________________________
>>> Bioc-devel at r-project.org mailing list
>>> https://stat.ethz.ch/mailman/listinfo/bioc-devel
>
>
>
> --
> Witold Eryk Wolski



-- 
Witold Eryk Wolski