Fix issue with custom_composition reader #843
Conversation
|
I changed the line 405 in model_reader.py, it still gives the error message. |
|
@pfreddy, @wkerzendorf - I think, the PR is ready. It passes our testing suite offline and it create data frames with the correct shape from the test_csv.dat file from @pfreddy. Whenever Tardis is happy and someone reviewed it, we should consider merging. |
|
Travis is happy! |
| df = pd.read_csv(fname, comment='#', | ||
| delimiter=delimiter, skiprows=skip_rows) | ||
| sep=delimiter, skiprows=skip_rows, index_col=0) |
There was a problem hiding this comment.
let's rename it into skiprows to be in line with pandas
|
@pfreddy is checking your PR. |
|
Ok, I've revisited the issue and found some indexing problems in the parser which are hopefully fixed now. On my laptop, I can run your setup @pfreddy without problems. Also the testing suite runs through. If Travis is happy, I'll address your stylistic suggestion, @wkerzendorf . |
Currently, the
custom_compositionreader is not properly working since it expected an additional column containing the cell index, similar to the file struture for thesimple_asciireader.This PR addresses this issue by re-inserting the cell index as the first column into the
custom_compositionfiles.