Документ взят из кэша поисковой машины. Адрес оригинального документа : http://www.stsci.edu/~bsimon/weblog/y2014/m12/index.html
Дата изменения: Thu Jan 1 00:16:50 2015
Дата индексирования: Mon Apr 11 11:18:07 2016
Кодировка:

Поисковые слова: arp 220
Archive for December

Archive for December

More

More

More

More

More

More

More

More

More

Closer

I fixed the bugs that caused the map_point changes to fail tests. Then I removed the no longer needed dimensional variables from the p structure and got those changes to pass all the tests. I then checked the changes into git. More

Controlled

I put the code under subversion control in a new branch called refactor. I started coding the overlap checking, not finished yet. More

New overlap code

I finished the code that computes the line and image overlap, wrote the unit tests for the new code and got them to last. I am now integrating the new code into drizzlepac. More

Partial Updates

i finished integrating the new overlap calculation code and got the tests to work. However, there is an odd problem with the blot tests that is intermittent. I will need to check that more. I added xmax and ymax to the tdriz calling sequence and deleted ystart. I modified the code of dobox to only ... More

Even Closer

I got the code which does partial updates to work. And I modified the code of the different drizzle kernels to loop over the whole image rather than a single line. More

Cleanup Time

The code change which processes each drizzle kernel over an entire image rather than a line at a time now passes the unit tests and was checked in. I also moved the computed quantities in the p structure into local variables in the functions where they are used. Those changes were also tested and ... More

Pruning

I spent the day removing more unneeded fields from the p structure and arguments fro tdriz and tblot. Most of the code works, but there are still problems with the tblot. More

Keywords

I fixed the problem with tblot (argument list mismatch) and started converting tdriz and tblot to use keyword arguments. I have the C code converted, will do the Python code tomorrow morning. More

Nearly Done

I got the keywords argument changes done and got the tests to pass. I improved the error checking in the top level code and go the tests to pass again. Both changes were checked into subversion. More

Done for Now

I added some error checking to the code that computes overlaps and edited the comments in the files that I modified most. All changes were checked into subversion. This ends the modifications I plan to make to drizzlepac. All other changes will be requested by the customer. More

Holding Steady

I studied the code for adrizzle.py and sgarea while waiting to discuss the transition to a new project. More

Astropy

I downloaded the astropy library template, added the drizzle code, and pushed the revised version to github. More

Link Link

I got the astropy version of the drizzle library to comple, link, and pass the unit tests. Next task is to write the Python libraries that call the C code by trimming down the existing Python code. More

Moar Code

I'm coding the python methods which call the cdrizzle code. More

Yikes

I spent the day coding the new class wrapping the C drizzle code, appropriately named Drizzle. Not finished yet. More

Back to Test

I finished the python wrapper class, modified the tests which run drizzle to call the new wrapper, and started debugging them. The sticking point now is the differences between the old stwcs and the newer astropy wcs. I also fixed a bug in the astropy spherical geometry package caused by a version ... More

Testing Testing

I worked on resolving the remaining problems with the astro-sphere library. I ran into issues with out of date test files. I returned to working on getting the drizzle code to pass test and made some progress. More

Sticking Point

I'm working through the problems with drizzle. The sticking point is writing out the file, as copying the existing code doesn't work. Will have to try something new. More

Plugging Away

I've spent the day trying to get the fits i/o in the python code to work. I succeeded, but the new code is giving answers significantly different than the old (adrizzle.) So I will have to investigate why on Monday. More

One Down

I got the drizlle class code to pass the unit tests. I am now working on adding support for blotting. More

Another Down

I added support for blotting to the drizzle class and got the tests to pass. There's one exception: I haven't scaled the resultant CD matrix. I also tested Nadia's fix for the spherical geometry test files and merged the fix into the astropy package. That finishes the work for the astropy version ... More

My Insatiable Nature

I finished figuring out the new overlap area compute code and modified the C code. This code is not yet integrated into cdrizzle. I modified the fits file handling code in the python wrapper, using the existing code in ablot as a guide. This code is not yet debugged. More