Enable type checking of shared and individual fields

- Fixes #257
broadinstitute · Sep 19, 2014 · 1143ab2 · 1143ab2
1 parent 526008e
commit 1143ab2
Show file tree

Hide file tree

Showing 3 changed files with 38 additions and 0 deletions.
diff --git a/gamgee/variant.cpp b/gamgee/variant.cpp
@@ -108,6 +108,14 @@ bool Variant::has_filter(const std::string& filter) const {
 /******************************************************************************
  * Individual field API                                                       *
  ******************************************************************************/
+uint8_t Variant::individual_field_type(const std::string& tag) const {
+  return individual_field_type(get_field_index(tag));
+}
+
+uint8_t Variant::individual_field_type(const int32_t index) const {
+  return bcf_hdr_id2type(m_header.get(), BCF_HL_FMT, index);
+}
+
 IndividualField<IndividualFieldValue<int32_t>> Variant::integer_individual_field(const std::string& tag) const {
   return integer_individual_field(get_field_index(tag)); 
 }
@@ -166,6 +174,14 @@ IndividualField<IndividualFieldValue<std::string>> Variant::individual_field_as_
 /******************************************************************************
  * Shared field API                                                           *
  ******************************************************************************/
+uint8_t Variant::shared_field_type(const std::string& tag) const {
+  return shared_field_type(get_field_index(tag));
+}
+
+uint8_t Variant::shared_field_type(const int32_t index) const {
+  return bcf_hdr_id2type(m_header.get(), BCF_HL_INFO, index);
+}
+
 bool Variant::boolean_shared_field(const std::string& tag) const {
   return find_shared_field(tag) != nullptr;
 }

diff --git a/gamgee/variant.h b/gamgee/variant.h
@@ -50,6 +50,11 @@ class Variant {
   VariantFilters filters() const;                                                                             ///< returns a vector-like object with all the filters for this record
   bool has_filter(const std::string& filter) const;                                                           ///< checks for the existence of a filter in this record
 
+  // type checking functions: returns BCF_HT_FLAG, BCF_HT_INT, BCF_HT_REAL, BCF_HT_STR from htslib/vcf.h
+  uint8_t shared_field_type(const std::string& tag) const;                                                    ///< returns the type of this shared (info) field
+  uint8_t shared_field_type(const int32_t index) const;                                                       ///< returns the type of this shared (info) field
+  uint8_t individual_field_type(const std::string& tag) const;                                                ///< returns the type of this individual (format) field
+  uint8_t individual_field_type(const int32_t index) const;                                                   ///< returns the type of this individual (format) field
 
   // individual field getters (a.k.a "format fields")
   IndividualField<Genotype> genotypes() const;                                                                 ///< special getter for the Genotype (GT) field. Returns a random access object with all the values in a given GT tag for all samples contiguous in memory. @warning Only int8_t GT fields have been tested. @warning Missing GT fields are untested. @warning creates a new object but makes no copies of the underlying values.

diff --git a/test/variant_reader_test.cpp b/test/variant_reader_test.cpp
@@ -179,6 +179,14 @@ void check_individual_field_api(const Variant& record, const uint32_t truth_inde
   BOOST_CHECK(pl_float != record.individual_field_as_float("GQ"));
   BOOST_CHECK(as_string != record.individual_field_as_string("PL"));
 
+  // check types
+  BOOST_CHECK_EQUAL(record.individual_field_type("GQ"), BCF_HT_INT);
+  BOOST_CHECK_EQUAL(record.individual_field_type("AF"), BCF_HT_REAL);
+  BOOST_CHECK_EQUAL(record.individual_field_type("AS"), BCF_HT_STR);
+  BOOST_CHECK_EQUAL(record.individual_field_type(header.field_index("GQ")), BCF_HT_INT);
+  BOOST_CHECK_EQUAL(record.individual_field_type(header.field_index("AF")), BCF_HT_REAL);
+  BOOST_CHECK_EQUAL(record.individual_field_type(header.field_index("AS")), BCF_HT_STR);
+
   for(auto i=0u; i != record.n_samples(); ++i) {
     BOOST_CHECK_EQUAL(gq_int[i][0], truth_gq[truth_index][i]);
     BOOST_CHECK_EQUAL(gq_int_idx[i][0], truth_gq[truth_index][i]);
@@ -258,6 +266,15 @@ void check_shared_field_api(const Variant& record, const uint32_t truth_index) {
   BOOST_CHECK(missing(record.integer_shared_field(-1)));
   BOOST_CHECK(missing(record.float_shared_field(-1)));  
   BOOST_CHECK(missing(record.string_shared_field(-1))); 
+  // type checking
+  BOOST_CHECK_EQUAL(record.shared_field_type("VALIDATED"), BCF_HT_FLAG);
+  BOOST_CHECK_EQUAL(record.shared_field_type("AN"), BCF_HT_INT);
+  BOOST_CHECK_EQUAL(record.shared_field_type("AF"), BCF_HT_REAL);
+  BOOST_CHECK_EQUAL(record.shared_field_type("DESC"), BCF_HT_STR);
+  BOOST_CHECK_EQUAL(record.shared_field_type(header.field_index("VALIDATED")), BCF_HT_FLAG);
+  BOOST_CHECK_EQUAL(record.shared_field_type(header.field_index("AN")), BCF_HT_INT);
+  BOOST_CHECK_EQUAL(record.shared_field_type(header.field_index("AF")), BCF_HT_REAL);
+  BOOST_CHECK_EQUAL(record.shared_field_type(header.field_index("DESC")), BCF_HT_STR);
   // check type conversions in the unforgiving API
   BOOST_CHECK_THROW(record.float_shared_field("AN"), runtime_error);
   BOOST_CHECK_THROW(record.string_shared_field("AN"), runtime_error);