| 1 |
#cython: language_level=3 |
| 2 |
|
| 3 |
''' |
| 4 |
Created on June 12th 2018 |
| 5 |
|
| 6 |
@author: coissac/mercier |
| 7 |
''' |
| 8 |
|
| 9 |
|
| 10 |
import types |
| 11 |
import re |
| 12 |
import sys |
| 13 |
import os |
| 14 |
import glob |
| 15 |
|
| 16 |
from obitools3.files.universalopener cimport uopen |
| 17 |
from obitools3.utils cimport tostr |
| 18 |
from obitools3.dms.obiseq cimport Nuc_Seq |
| 19 |
from .embl_genbank_features import extractTaxon |
| 20 |
|
| 21 |
from libc.stdlib cimport free, malloc, realloc |
| 22 |
from libc.string cimport strcpy, strlen |
| 23 |
|
| 24 |
|
| 25 |
_featureMatcher = re.compile(b'^FEATURES.+\n(?=ORIGIN(\s*))',re.DOTALL + re.M) |
| 26 |
|
| 27 |
_headerMatcher = re.compile(b'^LOCUS.+(?=\nFEATURES)', re.DOTALL + re.M) |
| 28 |
_seqMatcher = re.compile(b'^ORIGIN.+(?=//\n)', re.DOTALL + re.M) |
| 29 |
_cleanSeq1 = re.compile(b'ORIGIN(\s*)\n') |
| 30 |
_cleanSeq2 = re.compile(b'[ \n0-9]+') |
| 31 |
_acMatcher = re.compile(b'(?<=^ACCESSION ).+',re.M) |
| 32 |
_deMatcher = re.compile(b'(?<=^DEFINITION ).+\n( .+\n)*',re.M) |
| 33 |
_cleanDe = re.compile(b'\n *') |
| 34 |
|
| 35 |
|
| 36 |
def genbankParser(bytes text): |
| 37 |
|
| 38 |
cdef Nuc_Seq seq |
| 39 |
|
| 40 |
try: |
| 41 |
header = _headerMatcher.search(text).group() |
| 42 |
|
| 43 |
ft = _featureMatcher.search(text).group() |
| 44 |
|
| 45 |
s = _seqMatcher.search(text).group() |
| 46 |
s = _cleanSeq1.sub(b'', s) |
| 47 |
s = _cleanSeq2.sub(b'', s) |
| 48 |
|
| 49 |
acs = _acMatcher.search(text).group() |
| 50 |
acs = acs.split() |
| 51 |
ac = acs[0] |
| 52 |
acs = acs[1:] |
| 53 |
|
| 54 |
de = _deMatcher.search(header).group() |
| 55 |
de = _cleanDe.sub(b' ',de).strip().strip(b'.') |
| 56 |
|
| 57 |
tags = {} |
| 58 |
extractTaxon(ft, tags) |
| 59 |
|
| 60 |
seq = Nuc_Seq(ac, |
| 61 |
s, |
| 62 |
definition=de, |
| 63 |
quality=None, |
| 64 |
offset=-1, |
| 65 |
tags=tags) |
| 66 |
|
| 67 |
except Exception as e: |
| 68 |
print("\nCould not import sequence id:", text.split()[1], "(error raised:", e, ")") |
| 69 |
# Do not raise any Exception if you need the possibility to resume the generator |
| 70 |
# (Python generators can't resume after any exception is raised) |
| 71 |
return None |
| 72 |
|
| 73 |
return seq |
| 74 |
|
| 75 |
|
| 76 |
def genbankIterator_file(lineiterator, |
| 77 |
int skip=0, |
| 78 |
only=None, |
| 79 |
firstline=None, |
| 80 |
int buffersize=100000000 |
| 81 |
): |
| 82 |
cdef int lines_to_skip, ionly, read |
| 83 |
cdef Nuc_Seq seq |
| 84 |
cdef char* entry = NULL |
| 85 |
cdef size_t entry_buffer_size |
| 86 |
cdef int entry_len |
| 87 |
cdef int line_len |
| 88 |
|
| 89 |
entry_buffer_size = 2048 |
| 90 |
|
| 91 |
entry = <char*> malloc(entry_buffer_size*sizeof(char)) |
| 92 |
|
| 93 |
if only is None: |
| 94 |
ionly = -1 |
| 95 |
else: |
| 96 |
ionly = int(only) |
| 97 |
|
| 98 |
if isinstance(lineiterator, (str, bytes)): |
| 99 |
lineiterator=uopen(lineiterator) |
| 100 |
if isinstance(lineiterator, LineBuffer): |
| 101 |
iterator = iter(lineiterator) |
| 102 |
else: |
| 103 |
if hasattr(lineiterator, "readlines"): |
| 104 |
iterator = iter(LineBuffer(lineiterator, buffersize)) |
| 105 |
elif hasattr(lineiterator, '__next__'): |
| 106 |
iterator = lineiterator |
| 107 |
else: |
| 108 |
raise Exception("Invalid line iterator") |
| 109 |
|
| 110 |
skipped = 0 |
| 111 |
read = 0 |
| 112 |
|
| 113 |
if firstline is None: |
| 114 |
line = next(iterator) |
| 115 |
else: |
| 116 |
line = firstline |
| 117 |
|
| 118 |
while True: |
| 119 |
|
| 120 |
if ionly >= 0 and read >= ionly-1: |
| 121 |
break |
| 122 |
|
| 123 |
while skipped < skip: |
| 124 |
line = next(iterator) |
| 125 |
try: |
| 126 |
while line[:2] != b'//': |
| 127 |
line = next(iterator) |
| 128 |
line = next(iterator) |
| 129 |
except StopIteration: |
| 130 |
break |
| 131 |
skipped += 1 |
| 132 |
|
| 133 |
try: |
| 134 |
entry_len = 0 |
| 135 |
while line[:2] != b'//': |
| 136 |
line_len = strlen(line) |
| 137 |
while (entry_len + line_len) >= entry_buffer_size: |
| 138 |
entry_buffer_size*=2 |
| 139 |
entry = <char*>realloc(entry, entry_buffer_size) |
| 140 |
strcpy(entry+entry_len, line) |
| 141 |
entry_len+=line_len |
| 142 |
line = next(iterator) |
| 143 |
# Add last line too because need the // flag to parse |
| 144 |
line_len = strlen(line) |
| 145 |
while (entry_len + line_len) >= entry_buffer_size: |
| 146 |
entry_buffer_size*=2 |
| 147 |
entry = <char*>realloc(entry, entry_buffer_size) |
| 148 |
strcpy(entry+entry_len, line) |
| 149 |
line = next(iterator) |
| 150 |
except StopIteration: |
| 151 |
break |
| 152 |
|
| 153 |
seq = genbankParser(entry) |
| 154 |
|
| 155 |
yield seq |
| 156 |
read+=1 |
| 157 |
|
| 158 |
# Last sequence if not empty lines |
| 159 |
if entry.strip(): |
| 160 |
seq = genbankParser(entry) |
| 161 |
yield seq |
| 162 |
|
| 163 |
free(entry) |
| 164 |
|
| 165 |
|
| 166 |
def genbankIterator_dir(dir_path, |
| 167 |
int skip=0, |
| 168 |
only=None, |
| 169 |
firstline=None, |
| 170 |
int buffersize=100000000 |
| 171 |
): |
| 172 |
path = dir_path |
| 173 |
read = 0 |
| 174 |
read_files = 0 |
| 175 |
files = [filename for filename in glob.glob(os.path.join(path, b'*.gbff*'))] |
| 176 |
files.extend([filename for filename in glob.glob(os.path.join(path, b'*.seq*'))]) # new genbank extension |
| 177 |
files = list(set(files)) |
| 178 |
for filename in files: |
| 179 |
if read==only: |
| 180 |
return |
| 181 |
print("Parsing file %s (%d/%d)" % (tostr(filename), read_files+1, len(files))) |
| 182 |
f = uopen(filename) |
| 183 |
if only is not None: |
| 184 |
only_f = only-read |
| 185 |
else: |
| 186 |
only_f = None |
| 187 |
for seq in genbankIterator_file(f, skip=skip, only=only_f, buffersize=buffersize): |
| 188 |
yield seq |
| 189 |
read+=1 |
| 190 |
read_files+=1 |
| 191 |
|
| 192 |
|
| 193 |
def genbankIterator(obj, |
| 194 |
int skip=0, |
| 195 |
only=None, |
| 196 |
firstline=None, |
| 197 |
int buffersize=100000000 |
| 198 |
): |
| 199 |
if type(obj) == bytes or type(obj) == str : |
| 200 |
return genbankIterator_dir(obj, skip=skip, only=only, firstline=firstline, buffersize=buffersize) |
| 201 |
else: |
| 202 |
return genbankIterator_file(obj, skip=skip, only=only, firstline=firstline, buffersize=buffersize) |
| 203 |
|
| 204 |
|