Age | Commit message (Collapse) | Author |
|
R/qtl2 bundles can contain more than one file, of the same type. When
errors are encountered in any of the files, we need to be able to
inform the user which file it is, in addition to the line and column
number.
|
|
|
|
|
|
|
|
This reverts commit c213b0010c3ddc8d3215adab65bd489a9b884e30.
After a larger discussion with Arthur, Rob and Pjotr present, it was
verified that the strict checks are important. This reverts the commit
that allowed for looser checks.
|
|
|
|
|
|
|
|
|
|
The number columns in each contents line should be equal to the nember
of columns in the header line.
|
|
|
|
Remove the old code that relied on exceptions to parse errors in the
uploaded files.
|
|
Parse the files with the new functions that return error objects
instead of raising exceptions
|
|
This commit adds a number of functions that return the error object
when an error is found, or `None` otherwise. It avoids the use of
exceptions as control flow constructs.
|
|
Advance the seek position, once we have yielded up an error to causing
an infinite loop in certain conditions, where the `parse_errors`
function ends up resuming the gile in the same position once it
experiences an error.
|
|
To avoid processing all the items in an iterable, the `take` function
is added in this commit. It realised a limited number (specified at
call time) of items from the iterable given.
|
|
Build a function to collect all the parsing errors into a "sequence"
of dict objects containing the issues found.
|
|
Derive a "correct" sample file from an existing sample file with
errors for testing with large files.
Fix issue caught by test.
|
|
|
|
* Implement remaining file parsing tests and some helpers functions
needed for ensuring the tests work.
|
|
|
|
* Improve tests that ensure parsing fails in case the file has errors
* Add strains.csv file
* Implement minimum viable functionality that passes the implemented tests
|
|
Add dummy failing tests and a stub for the parsing of the files
|
|
Change the exception name to be more descriptive.
|
|
* Add tests to check for validity of the headers
* Add stubs for the tests
|
|
|
|
|