Merge pull request carlobaldassi#15 from carlobaldassi/check_desc

Add check_description flag

Author: carlobaldassi

.github/workflows/ci.yml

@@ -43,6 +43,6 @@ jobs:
       - uses: julia-actions/julia-buildpkg@v1
       - uses: julia-actions/julia-runtest@v1
       - uses: julia-actions/julia-processcoverage@v1
-      - uses: codecov/codecov-action@v3
+      - uses: codecov/codecov-action@v4
         with:
           files: lcov.info

codecov.yml

@@ -0,0 +1,2 @@
+codecov: 
+ token: b60713dd-5429-4e88-8a1e-d3a1f45a34f6

docs/src/index.md

@@ -72,8 +72,8 @@ The FASTA format which is assumed by this module is as follows:
    at the beginning or end of the description
 5. Empty lines are ignored (note however that lines containing whitespace will still trigger an error)
 
-When writing, description lines longer than 80 characters will trigger a warning message; sequence data is
-formatted in lines of 80 characters each; extra whitespace is silently discarded.
+When writing, description lines longer than 80 characters will trigger a warning message (this can be optionally
+disabled); sequence data is formatted in lines of 80 characters each; extra whitespace is silently discarded.
 No other restriction is put on the content of the sequence data, except that the `>` character is
 forbidden.
 

src/FastaIO.jl

@@ -307,14 +307,15 @@ mutable struct FastaWriter
     entry::Int
     own_f::Bool
     at_start::Bool
+    check_description::Bool
     function FastaWriter(io::IO)
-        fw = new(io, false, 0, 0, false, 0, 1, false, true)
+        fw = new(io, false, 0, 0, false, 0, 1, false, true, true)
         finalizer(close, fw)
         return fw
     end
     function FastaWriter(filename::AbstractString, mode::AbstractString = "w")
         fopen = endswith(filename, ".gz") ? gzopen : open
-        fw = new(fopen(filename, mode), false, 0, 0, false, 0, 1, true, true)
+        fw = new(fopen(filename, mode), false, 0, 0, false, 0, 1, true, true, true)
         finalizer(close, fw)
         return fw
     end
@@ -351,6 +352,9 @@ file.
 
 The `FastaWriter` object has an `entry::Int` field which stores the number of the entry which is
 currently being written.
+
+After creating the object, you can set the `check_description` field to `false` to disable the warning
+given when description lines are too long.
 """
 function FastaWriter(f::Function, args...)
     fw = FastaWriter(args...)
@@ -442,7 +446,7 @@ function write(fw::FastaWriter, c)
         error("character '>' not allowed in sequence data (entry $(fw.entry) of FASTA input)")
     end
     if fw.pos == 80
-        if !fw.in_seq
+        if !fw.in_seq && fw.check_description
             @warn("description line longer than 80 characters (entry $(fw.entry) of FASTA input)")
         else
             write(fw.f, '\n')
@@ -553,11 +557,13 @@ function writefastaseq(io::IO, seq, entry::Int, nl::Bool = true)
 end
 
 """
-    writefasta([io::IO = stdout], data)
+    writefasta([io::IO = stdout], data; check_description=true)
 
 This version of the function writes to an already opened `IO` stream, defaulting to `stdout`.
+
+Set the keyword `check_description=false` to disable the warning message given when description lines are too long.
 """
-function writefasta(io::IO, data)
+function writefasta(io::IO, data; check_description::Bool=true)
     entry = 0
     for (desc, seq) in data
         entry += 1
@@ -567,18 +573,18 @@ function writefasta(io::IO, data)
         if findfirst(==('\n'), desc) ≢ nothing
             error("newlines are not allowed within description (entry $entry of FASTA input)")
         end
-        if length(desc) > 79
+        if length(desc) > 79 && check_description
             @warn("description line longer than 80 characters (entry $entry of FASTA input)")
         end
         println(io, ">", desc)
         entry_chars = writefastaseq(io, seq, entry)
         entry_chars > 0 || error("empty sequence data (entry $entry of FASTA input)")
     end
 end
-writefasta(data) = writefasta(stdout, data)
+writefasta(data; kw...) = writefasta(stdout, data; kw...)
 
 """
-    writefasta(filename::String, data, [mode::String = "w"])
+    writefasta(filename::String, data, [mode::String = "w"]; check_description=true)
 
 This function dumps data to a FASTA file, auto-formatting it so to follow the specifications detailed in
 the section titled [The FASTA format](@ref). The `data` can be anything which is iterable and which produces
@@ -597,11 +603,13 @@ If the `filename` ends with `.gz`, the result will be a gzip-compressed file.
 
 The `mode` flag determines how the `filename` is open; use `"a"` to append the data to an existing
 file.
+
+Set the keyword `check_description=false` to disable the warning message given when description lines are too long.
 """
-function writefasta(filename::AbstractString, data, mode::AbstractString = "w")
+function writefasta(filename::AbstractString, data, mode::AbstractString = "w"; check_description=true)
     fopen = endswith(filename, ".gz") ? gzopen : open
     fopen(filename, mode) do f
-        writefasta(f, data)
+        writefasta(f, data; check_description)
     end
 end
 

test/Project.toml

@@ -1,3 +1,4 @@
 [deps]
 GZip = "92fee26a-97fe-5a0c-ad85-20a5f3185b63"
-Test = "8dfed614-e22c-5e08-85e1-65c5234f0b40"
+Logging = "56ddb016-857b-54e1-b83d-db4d58db5568"
+Test = "8dfed614-e22c-5e08-85e1-65c5234f0b40"

test/runtests.jl

@@ -3,6 +3,7 @@ module FastaTests
 using FastaIO
 using GZip
 using Test
+using Logging
 
 const fastadata_ascii = Any[
     ("A0ADS9_STRAM/3-104",
@@ -202,4 +203,63 @@ end
     end
 end
 
+outfile = joinpath(@__DIR__, "long_desc_test_out.fasta.gz")
+
+@testset "desc length checks" begin
+    longdesc = ">" * "X"^100
+    data = "DATA"
+
+    FastaWriter(devnull) do fw
+        b = IOBuffer()
+        with_logger(SimpleLogger(b, Logging.Debug)) do
+            write(fw, longdesc)
+            write(fw, data)
+        end
+        s = String(take!(b))
+        @test !isempty(s)
+
+        fw.check_description = false
+        b = IOBuffer()
+        with_logger(SimpleLogger(b, Logging.Debug)) do
+            write(fw, longdesc)
+            write(fw, data)
+        end
+        s = String(take!(b))
+        @test isempty(s)
+    end
+
+    b = IOBuffer()
+    with_logger(SimpleLogger(b, Logging.Debug)) do
+        writefasta(devnull, [(longdesc, data)])
+    end
+    s = String(take!(b))
+    @test !isempty(s)
+
+    b = IOBuffer()
+    with_logger(SimpleLogger(b, Logging.Debug)) do
+        writefasta(devnull, [(longdesc, data)], check_description=false)
+    end
+    s = String(take!(b))
+    @test isempty(s)
+
+    try
+        b = IOBuffer()
+        with_logger(SimpleLogger(b, Logging.Debug)) do
+            writefasta(outfile, [(longdesc, data)])
+        end
+        s = String(take!(b))
+        @test !isempty(s)
+
+        b = IOBuffer()
+        with_logger(SimpleLogger(b, Logging.Debug)) do
+            writefasta(outfile, [(longdesc, data)], check_description=false)
+        end
+        s = String(take!(b))
+        @test isempty(s)
+    finally
+        isfile(outfile) && rm(outfile)
+    end
+
 end
+
+end # module FastaTests