Use new BioGenerics macros #111
Conversation
jakobnissen
force-pushed
the
highlevel
branch
from
b18a857
to
ce53c60
Compare
| @@ -14,22 +11,26 @@ PrecompileTools = "aea7be01-6a6a-4083-8856-8a6e6704d82a" | |||
| StringViews = "354b36f9-a18e-4713-926e-db85100087ba" | |||
| TranscodingStreams = "3bb67fe8-82b1-5028-8e26-92a6c54297fa" | |||
|
|
|||
| [weakdeps] | |||
| BioSequences = "7e6ae17a-c86d-528c-b3b9-7f778a29fe59" | |||
There was a problem hiding this comment.
Oh man, great use of this functionality
docs/Project.toml
Outdated
| BioSequences = "7e6ae17a-c86d-528c-b3b9-7f778a29fe59" | ||
| Documenter = "e30172f5-a6a5-5a46-863b-614d45cd2de4" | ||
| FASTX = "c2308a5c-f048-11e8-3e8a-31650f418d12" | ||
|
|
There was a problem hiding this comment.
Do we need CodecZlib here as well? Might be nice if we want doctests for compression
There was a problem hiding this comment.
That's a good idea.
Actually, maby of the examples are not doctested because it requires a compressed file to be present. I should just add a few small dummy files for test purposes.
There was a problem hiding this comment.
Test dep on BioFormatSpecimens? That's what it's there for
docs/src/files.md
Outdated
| A `Reader` and a `Writer` are structs that wrap an IO, and allows efficient reading/writing of FASTX `Record`s. | ||
| For FASTA, use `FASTA.Reader` and `FASTA.Writer`, and for FASTQ - well I'm sure you've guessed it. | ||
| For FASTA, use `FASTAReader` and `FASTAWriter`, and for FASTQ - well I'm sure you've guessed it. |
There was a problem hiding this comment.
Looks like these changes are duplicates from #113 . If you branched from there, just be on the lookout for merge conflicts.
docs/src/files.md
Outdated
| FASTAReader(GzipDecompressorStream(open("seqs.fna.gz"; lock=false))) | ||
| ``` | ||
|
|
||
| To use rdr `rdr` and `wtr` macros with `do`-syntax, use the `defer` function. |
There was a problem hiding this comment.
Are we married to defer? Might be something useful to bikeshed. I agree we shouldn't use with given the new use in Base, but defer doesn't seem intuitive from a user perspective (though I understand where it's coming from).
There was a problem hiding this comment.
I'm completely open to suggestions. I like defer because it's short, but it's a little opaque.
There was a problem hiding this comment.
In my mind, I really want to be able to use existing stuff rather than introducing a new function name. If we do need a new function name, the only thing better I can think of is apply, but it's not much better.
The primary reason not to just make these iterables so they one can use eg map and foreach is the need to close the handle, right?
How does Base.eachline handle this? It allows you to pass a file name, iterates through lines without needing an explicit open, and doesn't need to be closed. But I guess that's only reading, not writing.
To be clear, I don't think any of this should be blocking. defer is probably fine, and as-is this is a really big increase in usability. I don't want the perfect to be the enemy of the really good here.
There was a problem hiding this comment.
I really want to be able to use existing stuff rather than introducing a new function name
I agree. Not sure what it should be, though.
Base.eachline closes the handle when the iterator is exhausted. This is quite elegant, but only works for readers, not for writers. We could auto-close readers, but I'm a bit wary of guiding users towards not closing their readers, when they really do need to close their writers.
Mention the most important things first in the documentation: How to read and write files.
The new BioGenerics `@rdr_str` and `@wtr_str` macros, as well as the `defer` keyword has been a longstanding FASTX todo. This change adds this new functionality.
jakobnissen
force-pushed
the
highlevel
branch
from
ce53c60
to
c403439
Compare
jakobnissen
changed the title
Codecov ReportAttention:
Additional details and impacted files@@ Coverage Diff @@
## master #111 +/- ##
==========================================
- Coverage 89.78% 89.52% -0.26%
==========================================
Files 15 15
Lines 695 697 +2
==========================================
Hits 624 624
- Misses 71 73 +2
Flags with carried forward coverage won't be shown. Click here to find out more.
☔ View full report in Codecov by Sentry. |
The new BioGenerics
@rdr_strand@wtr_strmacros, as well as thedeferkeyword has been a longstanding FASTX todo.This change adds this new functionality.
TODO: