stop raster output from being resampled #55
Conversation
|
Thanks. Since the tests fail, I think we should make this an optional parameter for the |
|
Could you please also rebase to current |
|
Hadn't thought of putting this under the "rasterResolution" flag; seems like the sensible option. |
…RasterResolution)
… pixels am working on OS X, so didn't notice this before!
… the test builds run from Linux, so convert to an X11 PNG
|
sorry about all those commits; I forgot a couple of issues and OS X and Linux appear to produce different PNGs which cause the tests to fail. everything seems to be working now! |
|
Thanks, I'm fine with the commits. Could you please add documentation, and an item to |
|
|
||
| res = getOption('tikzRasterResolution') | ||
| if (is.null(res)) res = NA; | ||
| if (is.na(res)) interpolate = FALSE; |
There was a problem hiding this comment.
No, now I see that else would be wrong. Sorry.
There was a problem hiding this comment.
yes, maybe a comment explaining what I'm doing would help?
There was a problem hiding this comment.
It would make things clearer. On the other hand, does the setting of interpolate make a difference at all if res is NA?
There was a problem hiding this comment.
Yes, it seems to make a difference. I get something very different from what I put in without this
|
@krlmlr, I've had a tidy up of the code based on your comments. |
stop raster output from being resampled
|
Thanks! |
| tools::file_path_sans_ext(fileName), | ||
| '_ras', rasterCount, '.png') | ||
|
|
||
| res = getOption('tikzRasterResolution') |
There was a problem hiding this comment.
you can use res = getOption('tikzRasterResolution', NA) and delete the three lines below
There was a problem hiding this comment.
hadn't noticed that parameter before—thanks! Have pushed an appropriate change to my noresample repo
There was a problem hiding this comment.
Could you please open a new pull request?
There was a problem hiding this comment.
sorry, I'm sure when I get people hacking my repos I'll appreciate all these ways of moving code around!
There was a problem hiding this comment.
I'm sure you will ;-) Merging with a few clicks is very comfortable indeed.
|
I'm now revisiting this code due to reentrance problems (#132), my approach is to write the raster "as is" using @smason: Do you think this makes sense? |
|
Yes, assuming I'm interpreting you correctly, it sounds reasonable. Whenever I've sent a raster to output in R I've wanted it preserved exactly, and any resampling would be very undesirable. Further, I've not noticed any other R device I've used resampling any raster given to it. |
|
Thanks for your fast reply. I've noticed that R seems to be doing all the resampling it needs already: In particular, the base_raster test looks almost unchanged when resampling is turned off; this was a pleasant surprise for me. |
|
Wrong: It's done by PGF, via |
this is in reference to issue #54