Add support for extra-chroms (string chromosomes)

.gitignore

@@ -11,3 +11,33 @@ html/
 *.egg-info
 __pycache__
 dist
+
+# editor settings
+.vscode/
+.spyproject/
+
+# cmake output files from CMake.gitignore
+CMakeLists.txt.user
+CMakeCache.txt
+CMakeFiles
+CMakeScripts
+Testing
+Makefile
+cmake_install.cmake
+install_manifest.txt
+compile_commands.json
+CTestTestfile.cmake
+_deps
+
+# Additional CMake files
+CPackConfig.cmake
+CPackSourceConfig.cmake
+
+# file objects
+*.a
+
+# compiled binaries
+tests/bed_test
+tests/bim_test
+tests/fam_test
+tests/plinkio_test

README.md

@@ -20,25 +20,43 @@ Project rationales:
 
 Installing this library is easy, just **configure** and **make**. This will also install Python bindings for the active interpeter.
 
-*NEWS* The python extension can now be installed by
+*NEWS* The python extension can now be installed 
+[from pypi](https://pypi.org/project/plinkio/) by:
 
     pip install plinkio
 
+To compile and install from source code:
+
+    python setup.py build
+    python setup.py install
+
+You require [tox](https://pypi.org/project/tox/) to test plinkio properly. 
+By calling:
+
+    tox
+
+in your project directory, you will download dependencies, compile the library 
+and run tests. If you get stuck with compiled libraries, adding `-r` option
+will reset the test environment (this means wiping out the testing environment 
+and re-initialize it from scratch by downloading dependencies and compiling again)
+
 ### Installing to a standard location
 
-    mkdir build
-    cd build
-    ../configure
-    make && make check && sudo make install
+[CMake](https://cmake.org/) is required in order to compile the C libraries (mind
+to the final `.` after *CMake*, which stands for your current *project* local 
+directory):
 
-You can also pass the --disable-tests flag to **configure** to avoid building the unit tests and the dependency to libcmockery. Note howerver, in this case **make check** will not do anything.
+    cmake .
+    make
+    make test
+    make install
 
 ### Installing to a custom location
 
-    mkdir build
-    cd build
-    ../configure --prefix=/path/to/plinkio
-    make && make check && make install
+    cmake -DCMAKE_INSTALL_PREFIX:PATH=/path/to/plinkio .
+    make
+    make test
+    make install
 
 ### Linking to your program
 
@@ -56,7 +74,7 @@ The genotypes are coded 0, 1, 2, and 3. The numbers 0-2 represent the number of
 
 ## Using in C
 
-For specific information look at http://mfranberg.github.com/libplinkio/index.html
+For specific information look at https://mfranberg.github.io/libplinkio/index.html
 
 The following C program prints the genotypes of all individuals. Note, that it is not recommended to run this program on a big plink file since it will fill your screen with data.
 
@@ -175,9 +193,9 @@ struct pio_locus_t
     size_t pio_id;
 
     /**
-     * Chromosome number starting from 1.
+     * Chromosome as strings.
      */
-    unsigned char chromosome;
+    char *chromosome;
 
     /**
      * Name of the SNP.
@@ -302,7 +320,7 @@ class Sample:
 class Locus:
     def __init__(self, chromosome, name, position, bp_position, allele1, allele2):
         ##
-        # Chromosome number starting from 1
+        # Chromosome string
         #
         self.chromosome = chromosome
 

py-plinkio/cplinkio.c

@@ -393,11 +393,11 @@ int parse_locus(PyObject *py_locus, struct pio_locus_t *locus)
     PyObject *allele1_object;
     PyObject *allele2_object;
 
+    PyObject *chromosome_string;
     PyObject *name_string;
     PyObject *allele1_string;
     PyObject *allele2_string;
 
-    int chromosome;
     float position;
     int bp_position;
 
@@ -410,16 +410,16 @@ int parse_locus(PyObject *py_locus, struct pio_locus_t *locus)
     allele1_object = PyObject_GetAttrString( py_locus, "allele1" );
     allele2_object = PyObject_GetAttrString( py_locus, "allele2" );
 
-    chromosome = PyInt_AsLong( chromosome_object );
+    chromosome_string = PyObject_Str( chromosome_object );
     name_string = PyObject_Str( name_object );
     position = PyFloat_AsDouble( position_object );
     bp_position = PyInt_AsLong( bp_position_object );
     allele1_string = PyObject_Str( allele1_object );
     allele2_string = PyObject_Str( allele2_object );
 
-    if( chromosome == -1 && PyErr_Occurred( ) )
+    if( chromosome_string == NULL )
     {
-        PyErr_SetString( PyExc_TypeError, "Error chromosome field must be an integer." );
+        PyErr_SetString( PyExc_TypeError, "Error chromosome field must be a string" );
         ret = 0;
     }
     else if( name_string == NULL )
@@ -449,14 +449,15 @@ int parse_locus(PyObject *py_locus, struct pio_locus_t *locus)
     }
 
     /* The strings wont get freed by plinkio so remove const qualifier */
-    locus->chromosome = PyInt_AsLong( chromosome_object );
+    locus->chromosome = (char *) PyString_AsString( chromosome_string );
     locus->name = (char *) PyString_AsString( name_string );
     locus->position = PyFloat_AsDouble( position_object );
     locus->bp_position = PyInt_AsLong( bp_position_object );
     locus->allele1 = (char *) PyString_AsString( allele1_string );
     locus->allele2 = (char *) PyString_AsString( allele2_string );
 
 locus_error:
+    Py_DECREF( chromosome_string );
     Py_DECREF( name_string );
     Py_DECREF( allele1_string );
     Py_DECREF( allele2_string );
@@ -639,7 +640,7 @@ plinkio_get_loci(PyObject *self, PyObject *args)
     {
         struct pio_locus_t *locus = pio_get_locus( &c_plink_file->file, i );
 
-        PyObject *args = Py_BuildValue( "BsfLss",
+        PyObject *args = Py_BuildValue( "ssfLss",
                                         locus->chromosome,
                                         locus->name,
                                         locus->position,

py-plinkio/plinkio/plinkfile.py

@@ -206,7 +206,7 @@ class Locus:
     def __init__(self, chromosome, name, position, bp_position, allele1, allele2):
         # pylint: disable = too-many-arguments
         ##
-        # Chromosome number starting from 1
+        # Chromosome string
         #
         self.chromosome = chromosome
 

py-plinkio/tests/write_test.py

@@ -9,9 +9,19 @@ def test_read_write():
     with tempfile.TemporaryDirectory() as temp_dir:
         plink_prefix = os.path.join(temp_dir, "test")
 
-        samples = [Sample("fid1", "iid1", "0", "0", 0, 0), Sample("fid2", "iid2", "0", "0", 0, 1)]
-        loci = [Locus(1, "chr1:1", 1.0, 1, "A", "C"), Locus(2, "chr1:2", 2.0, 2, "G", "T")]
-        rows = [[0, 1], [1, 2]]
+        samples = [
+            Sample("fid1", "iid1", "0", "0", 0, 0),
+            Sample("fid2", "iid2", "0", "0", 0, 1),
+        ]
+
+        loci = [
+            Locus("1", "chr1:1", 1.0, 1, "A", "C"),
+            Locus("2", "chr1:2", 2.0, 2, "G", "T"),
+            Locus("X", "chrX:3", 1.0, 3, "A", "G"),
+            Locus("Contig123456", "Contig123456:4", 1.0, 4, "T", "C"),
+        ]
+
+        rows = [[0, 1], [1, 2], [1, 1], [0, 0]]
 
         writer = plinkfile.create(plink_prefix, samples)
 

setup.py

@@ -39,7 +39,7 @@
     # Versions should comply with PEP440.  For a discussion on single-sourcing
     # the version across setup.py and the project code, see
     # https://packaging.python.org/en/latest/single_source_version.html
-    version="0.9.8",
+    version="0.9.9.dev0",
     description="A library for parsing plink genotype files",
     long_description=long_description,
     # The project's main homepage.

src/bim.c

@@ -38,6 +38,10 @@ utarray_locus_dtor(void *element)
 {
     struct pio_locus_t *locus = (struct pio_locus_t *) element;
 
+    if( locus->chromosome != NULL )
+    {
+        free( locus->chromosome );
+    }
     if( locus->name != NULL )
     {
         free( locus->name );
@@ -103,7 +107,7 @@ bim_write(struct pio_bim_file_t *bim_file, struct pio_locus_t *locus)
     if( write_locus( bim_file->fp, locus ) == PIO_OK )
     {
         locus_copy.pio_id = bim_num_loci( bim_file );
-        locus_copy.chromosome = locus->chromosome;
+        locus_copy.chromosome = strdup( locus->chromosome );
         locus_copy.name = strdup( locus->name );
         locus_copy.position = locus->position;
         locus_copy.bp_position = locus->bp_position;

src/bim_parse.c

@@ -99,31 +99,6 @@ parse_str(const char *field, size_t length, pio_status_t *status)
     }
 }
 
-/**
- * Parses a chromosome number and returns it.
- *
- * @param field Csv field.
- * @param length Length of the field.
- * @param status Status of the conversion.
- *
- * @return The parsed csv field, or 0 if it could
- *         not be parsed.
- */
-static unsigned char
-parse_chr(const char *field, size_t length, pio_status_t *status)
-{
-    char *endptr;
-    unsigned char chr = (unsigned char) strtol( field, &endptr, 10 );
-    if( length > 0 && ( endptr == NULL || *endptr == '\0' ) )
-    {
-        *status = PIO_OK;
-        return chr;
-    }
-
-    *status = PIO_ERROR;
-    return 0;
-}
-
 /**
  * Parses a genetic distance (float).
  *
@@ -204,7 +179,7 @@ new_field(void *field, size_t field_length, void *data)
     switch( state->field )
     {
         case 0:
-            state->cur_locus.chromosome = parse_chr( buffer, field_length, &status );
+            state->cur_locus.chromosome = parse_str( buffer, field_length, &status );
             break;
         case 1:
             state->cur_locus.name = parse_str( buffer, field_length, &status );
@@ -286,7 +261,7 @@ pio_status_t
 write_locus(FILE *bim_fp, struct pio_locus_t *locus)
 {
     int bytes_written = fprintf( bim_fp,
-                    "%d\t%s\t%f\t%lld\t%s\t%s\n",
+                    "%s\t%s\t%f\t%lld\t%s\t%s\n",
                     locus->chromosome,
                     locus->name,
                     locus->position,

src/plinkio/bim.h

@@ -27,9 +27,9 @@ struct pio_locus_t
     size_t pio_id;
 
     /**
-     * Chromosome number starting from 1.
+     * Chromosome as strings.
      */
-    unsigned char chromosome;
+    char *chromosome;
 
     /**
      * Name of the SNP.

tests/bim_test.c

@@ -36,11 +36,30 @@ test_parse_position(void **state)
 void
 test_parse_chr(void **state)
 {
-    const char *TEST_STRING = "16";
+    const char *TEST_STRING1 = "16";
     pio_status_t status;
 
-    assert_int_equal( parse_chr( TEST_STRING, strlen( TEST_STRING ), &status ), 16 );
+    char *chrom1 = parse_str( TEST_STRING1, strlen( TEST_STRING1 ), &status );
+
+    assert_string_equal( chrom1, TEST_STRING1 );
+    assert_int_equal( status, PIO_OK );
+    free(chrom1);
+
+    const char *TEST_STRING2 = "X";
+
+    char *chrom2 = parse_str( TEST_STRING2, strlen( TEST_STRING2 ), &status );
+
+    assert_string_equal( chrom2, TEST_STRING2 );
+    assert_int_equal( status, PIO_OK );
+    free(chrom2);
+
+    const char *TEST_STRING3 = "Contig123456";
+
+    char *chrom3 = parse_str( TEST_STRING3, strlen( TEST_STRING3 ), &status );
+
+    assert_string_equal( chrom3, "Contig123456" );
     assert_int_equal( status, PIO_OK );
+    free(chrom3);
 }
 
 /**
@@ -58,15 +77,15 @@ test_parse_multiple_loci(void **state)
     assert_int_equal( bim_num_loci( &bim_file ), 2 );
 
     locus = *bim_get_locus( &bim_file, 0 );
-    assert_int_equal( locus.chromosome, 1 ); 
+    assert_string_equal( locus.chromosome, "1" ); 
     assert_string_equal( locus.name, "rs1" );
     assert_true( fabs( locus.position - 0.0 ) <= 1e-6 );
     assert_int_equal( locus.bp_position, 1234567 );
     assert_string_equal( locus.allele1, "A" );
     assert_string_equal( locus.allele2, "C" );
 
     locus = *bim_get_locus( &bim_file, 1 );
-    assert_int_equal( locus.chromosome, 1 ); 
+    assert_string_equal( locus.chromosome, "1" ); 
     assert_string_equal( locus.name, "rs2" );
     assert_true( fabs( locus.position - 0.23 ) <= 1e-6 );
     assert_int_equal( locus.bp_position, 7654321 );