SeqAn3 3.1.0
The Modern C++ library for sequence analysis.
format_embl.hpp
Go to the documentation of this file.
1// -----------------------------------------------------------------------------------------------------
2// Copyright (c) 2006-2021, Knut Reinert & Freie Universität Berlin
3// Copyright (c) 2016-2021, Knut Reinert & MPI für molekulare Genetik
4// This file may be used, modified and/or redistributed under the terms of the 3-clause BSD-License
5// shipped with this file and also available at: https://github.com/seqan/seqan3/blob/master/LICENSE.md
6// -----------------------------------------------------------------------------------------------------
7
13#pragma once
14
15#include <seqan3/std/algorithm>
16#include <iterator>
17#include <seqan3/std/ranges>
18#include <string>
19#include <string_view>
20#include <vector>
21
40
41namespace seqan3
42{
74{
75public:
79 format_embl() noexcept = default;
80 format_embl(format_embl const &) noexcept = default;
81 format_embl & operator=(format_embl const &) noexcept = default;
82 format_embl(format_embl &&) noexcept = default;
83 format_embl & operator=(format_embl &&) noexcept = default;
84 ~format_embl() noexcept = default;
85
87
89 static inline std::vector<std::string> file_extensions
90 {
91 { "embl" },
92 };
93
94protected:
96 template <typename stream_type, // constraints checked by file
97 typename seq_legal_alph_type,
98 typename seq_type, // other constraints checked inside function
99 typename id_type,
100 typename qual_type>
101 void read_sequence_record(stream_type & stream,
103 seq_type & sequence,
104 id_type & id,
105 qual_type & SEQAN3_DOXYGEN_ONLY(qualities))
106 {
107 auto stream_view = detail::istreambuf(stream);
108 auto stream_it = std::ranges::begin(stream_view);
109
110 std::string idbuffer;
111 std::ranges::copy(stream_view | detail::take_until_or_throw(is_cntrl || is_blank),
112 std::cpp20::back_inserter(idbuffer));
113 if (idbuffer != "ID")
114 throw parse_error{"An entry has to start with the code word ID."};
115
116 if constexpr (!detail::decays_to_ignore_v<id_type>)
117 {
119 {
120 std::ranges::copy(idbuffer | views::char_to<std::ranges::range_value_t<id_type>>, std::cpp20::back_inserter(id));
121 do
122 {
123 std::ranges::copy(stream_view | detail::take_until_or_throw(is_char<'S'>)
124 | views::char_to<std::ranges::range_value_t<id_type>>,
125 std::cpp20::back_inserter(id));
126 id.push_back(*stream_it);
127 ++stream_it;
128 } while (*stream_it != 'Q');
129 id.pop_back(); // remove 'S' from id
130 idbuffer = "SQ";
131 }
132 else
133 {
134 // ID
135 detail::consume(stream_view | detail::take_until(!is_blank));
136
137 // read id
138 if (options.truncate_ids)
139 {
140 std::ranges::copy(stream_view | detail::take_until_or_throw(is_blank || is_char<';'> || is_cntrl)
141 | views::char_to<std::ranges::range_value_t<id_type>>,
142 std::cpp20::back_inserter(id));
143 }
144 else
145 {
146 std::ranges::copy(stream_view | detail::take_until_or_throw(is_char<';'>)
147 | views::char_to<std::ranges::range_value_t<id_type>>,
148 std::cpp20::back_inserter(id));
149 }
150 }
151 }
152
153 // Jump to sequence
154 if (idbuffer !="SQ")
155 {
156 do
157 {
158 detail::consume(stream_view | detail::take_until_or_throw(is_char<'S'>));
159 ++stream_it;
160 } while (*stream_it != 'Q');
161 }
162 detail::consume(stream_view | detail::take_line_or_throw); //Consume line with infos to sequence
163
164 // Sequence
165 auto constexpr is_end = is_char<'/'> ;
166 if constexpr (!detail::decays_to_ignore_v<seq_type>)
167 {
168 auto seq_view = stream_view | std::views::filter(!(is_space || is_digit)) // ignore whitespace and numbers
169 | detail::take_until_or_throw(is_end); // until //
170
171 auto constexpr is_legal_alph = char_is_valid_for<seq_legal_alph_type>;
172 std::ranges::copy(seq_view | std::views::transform([is_legal_alph] (char const c) // enforce legal alphabet
173 {
174 if (!is_legal_alph(c))
175 {
176 throw parse_error{std::string{"Encountered an unexpected letter: "} +
177 "char_is_valid_for<" +
178 detail::type_name_as_string<seq_legal_alph_type> +
179 "> evaluated to false on " +
180 detail::make_printable(c)};
181 }
182 return c;
183 })
184 | views::char_to<std::ranges::range_value_t<seq_type>>, // convert to actual target alphabet
185 std::cpp20::back_inserter(sequence));
186 }
187 else
188 {
189 detail::consume(stream_view | detail::take_until(is_end));
190 }
191 //Jump over // and cntrl
192 ++stream_it;
193 ++stream_it;
194 ++stream_it;
195 }
196
198 template <typename stream_type, // constraints checked by file
199 typename seq_type, // other constraints checked inside function
200 typename id_type,
201 typename qual_type>
202 void write_sequence_record(stream_type & stream,
203 sequence_file_output_options const & options,
204 seq_type && sequence,
205 id_type && id,
206 qual_type && SEQAN3_DOXYGEN_ONLY(qualities))
207 {
208 seqan3::detail::fast_ostreambuf_iterator stream_it{*stream.rdbuf()};
209
210 [[maybe_unused]] size_t sequence_size = 0;
211
212 if constexpr (!detail::decays_to_ignore_v<seq_type>)
213 sequence_size = std::ranges::distance(sequence);
214
215 // ID
216 if constexpr (detail::decays_to_ignore_v<id_type>)
217 {
218 throw std::logic_error{"The ID field may not be set to ignore when writing embl files."};
219 }
220 else
221 {
222 if (ranges::empty(id)) //[[unlikely]]
223 throw std::runtime_error{"The ID field may not be empty when writing embl files."};
224
226 {
227 stream_it.write_range(id);
228 }
229 else
230 {
231 stream_it.write_range(std::string_view{"ID "});
232 stream_it.write_range(id);
233 stream_it.write_range(std::string_view{"; "});
234 stream_it.write_number(sequence_size);
235 stream_it.write_range(std::string_view{" BP.\n"});
236 }
237 }
238
239 // Sequence
240 if constexpr (detail::decays_to_ignore_v<seq_type>) // sequence
241 {
242 throw std::logic_error{"The SEQ field may not be set to ignore when writing embl files."};
243 }
244 else
245 {
246 if (std::ranges::empty(sequence)) //[[unlikely]]
247 throw std::runtime_error{"The SEQ field may not be empty when writing embl files."};
248
249 // write beginning of sequence record
250 stream_it.write_range(std::string_view{"SQ Sequence "});
251 stream_it.write_number(sequence_size);
252 stream_it.write_range(std::string_view{" BP;\n"});
253
254 // write sequence in chunks of 60 bp's with a space after 10 bp's
255 auto char_sequence = sequence | views::to_char;
256 auto it = std::ranges::begin(char_sequence);
257 size_t written_chars{0};
258 uint8_t chunk_size{10u};
259
260 while (it != std::ranges::end(char_sequence))
261 {
262 auto current_end = it;
263 size_t steps = std::ranges::advance(current_end, chunk_size, std::ranges::end(char_sequence));
264
265 using subrange_t = std::ranges::subrange<decltype(it), decltype(it), std::ranges::subrange_kind::sized>;
266 it = stream_it.write_range(subrange_t{it, current_end, chunk_size - steps});
267 stream_it = ' ';
268 written_chars += chunk_size;
269
270 if (written_chars % 60 == 0)
271 {
272 stream_it.write_number(written_chars);
273 stream_it.write_end_of_line(options.add_carriage_return);
274 }
275 }
276
277 // fill last line
278 auto characters_in_last_line = sequence_size % 60;
279 auto number_of_padding_needed = 65 - characters_in_last_line - characters_in_last_line / chunk_size;
280 stream_it.write_range(views::repeat_n(' ', number_of_padding_needed));
281 stream_it.write_number(sequence_size);
282 stream_it.write_end_of_line(options.add_carriage_return);
283
284 // write end-of-record-symbol
285 stream_it.write_range(std::string_view{"//"});
286 stream_it.write_end_of_line(options.add_carriage_return);
287 }
288 }
289};
290
291} // namespace seqan
The <algorithm> header from C++20's standard library.
Core alphabet concept and free function/type trait wrappers.
Provides seqan3::views::char_to.
The EMBL format.
Definition: format_embl.hpp:74
static std::vector< std::string > file_extensions
The valid file extensions for this format; note that you can modify this value.
Definition: format_embl.hpp:90
void read_sequence_record(stream_type &stream, sequence_file_input_options< seq_legal_alph_type > const &options, seq_type &sequence, id_type &id, qual_type &qualities)
Read from the specified stream and back-insert into the given field buffers.
Definition: format_embl.hpp:101
void write_sequence_record(stream_type &stream, sequence_file_output_options const &options, seq_type &&sequence, id_type &&id, qual_type &&qualities)
Write the given fields to the specified stream.
Definition: format_embl.hpp:202
format_embl() noexcept=default
Defaulted.
Provides various utility functions.
Provides various transformation traits used by the range module.
Provides seqan3::dna5, container aliases and string literals.
Provides seqan3::detail::fast_ostreambuf_iterator.
auto const to_char
A view that calls seqan3::to_char() on each element in the input range.
Definition: to_char.hpp:63
auto const char_to
A view over an alphabet, given a range of characters.
Definition: char_to.hpp:67
constexpr auto is_blank
Checks whether c is a blank character.
Definition: predicate.hpp:145
constexpr auto is_digit
Checks whether c is a digital character.
Definition: predicate.hpp:269
constexpr auto is_char
Checks whether a given letter is the same as the template non-type argument.
Definition: predicate.hpp:65
constexpr auto is_space
Checks whether c is a space character.
Definition: predicate.hpp:128
constexpr auto is_cntrl
Checks whether c is a control character.
Definition: predicate.hpp:92
decltype(detail::transform< trait_t >(list_t{})) transform
Apply a transformation trait to every type in the list and return a seqan3::type_list of the results.
Definition: traits.hpp:471
constexpr auto repeat_n
A view factory that repeats a given value n times.
Definition: repeat_n.hpp:91
The generic concept for a (biological) sequence.
Provides various utility functions.
Provides seqan3::detail::istreambuf.
The main SeqAn3 namespace.
Definition: cigar_operation_table.hpp:2
SeqAn specific customisations in the standard namespace.
Provides character predicates for tokenisation.
The <ranges> header from C++20's standard library.
Provides seqan3::views::repeat_n.
Provides seqan3::sequence_file_input_format and auxiliary classes.
Provides seqan3::sequence_file_input_options.
Provides seqan3::sequence_file_output_format and auxiliary classes.
Provides seqan3::sequence_file_output_options.
Thrown if there is a parse error, such as reading an unexpected character from an input stream.
Definition: exception.hpp:48
The options type defines various option members that influence the behaviour of all or some formats.
Definition: input_options.hpp:27
bool embl_genbank_complete_header
Read the complete_header into the seqan3::field::id for embl or genbank format.
Definition: input_options.hpp:31
bool truncate_ids
Read the ID string only up until the first whitespace character.
Definition: input_options.hpp:29
The options type defines various option members that influence the behaviour of all or some formats.
Definition: output_options.hpp:26
bool add_carriage_return
The default plain text line-ending is "\n", but on Windows an additional carriage return is recommend...
Definition: output_options.hpp:43
bool embl_genbank_complete_header
Complete header given for embl or genbank.
Definition: output_options.hpp:46
Provides seqan3::detail::take_line and seqan3::detail::take_line_or_throw.
Provides seqan3::views::take_until and seqan3::views::take_until_or_throw.
Provides seqan3::views::to_char.
Provides traits to inspect some information of a type, for example its name.