def test_find_keyword_sets():
"""Check the indexing of list of strings into Eclipse keywords"""
assert find_keyword_sets(["PORO", "0 1 2 3", "4 5 6", "/"]) == [(0, 3)]
# Missing slash, then nothing found:
assert find_keyword_sets(["PORO", "0 1 2 3", "4 5 6"]) == []
# MORE!!
# Keyword with no data, will be found, but untouched by compression
kw_nodata = ["PORO", "/"]
kw_sets = find_keyword_sets(kw_nodata)
assert kw_sets == [(0, 1)]
assert compress_multiple_keywordsets(kw_sets, kw_nodata) == kw_nodata
def test_formatting():
"""Test that compressed output is only 79 characters wide"""
numbers = " ".join([str(number) for number in np.random.rand(1, 100)[0]])
filelines = ["PORO", numbers, "/"]
formatted = compress_multiple_keywordsets(find_keyword_sets(filelines), filelines)
assert max([len(line) for line in formatted]) <= 79
# But, some keywords will not tolerate random
# newlines in their data-section, at least the multi-record keywords.
# So we should never wrap a line with a slash in it:
filelines = ["VFPPROD", " FOO" * 30 + " /"]
# If this is fed through eclcompress, it will be wrapped due to its
# length:
formatted = compress_multiple_keywordsets(find_keyword_sets(filelines), filelines)
assert len(formatted) > 2
def test_eclcompress():
"""Test a given set of lines, and ensure that the output
can be parsed by opm.io"""
kwsets = find_keyword_sets(FILELINES)
compressed = compress_multiple_keywordsets(kwsets, FILELINES)
compressedstr = "\n".join(compressed)
assert opm.io.Parser().parse_string(compressedstr, OPMIO_PARSECONTEXT)
def test_grid_grdecl():
"""A typical grid.grdecl file must be able to do compression on the
COORDS/ZCORN keywords, while conserving the other two"""
filelines = """
SPECGRID
214 669 49 1 F /
GDORIENT
INC INC INC DOWN RIGHT /
ZCORN
1 1 1 1 1 1 /
""".split(
"\n"
)
kwsets = find_keyword_sets(filelines)
assert (
compress_multiple_keywordsets(kwsets, filelines)
== """
SPECGRID
214 669 49 1 F /
GDORIENT
INC INC INC DOWN RIGHT /
ZCORN
6*1 /
""".split(
"\n"
)
)
def test_multiplerecords():
"""Test compression on keywords with multiple records,
for which eclcompress only supports compressing the first records
Conservation of the remainder of the keyword is critical to test.
"""
filelines = [
"EQUALS",
" MULTZ 0.017101 1 40 1 64 5 5 / nasty comment without comment characters",
"/",
]
kwsets = find_keyword_sets(filelines)
assert compress_multiple_keywordsets(kwsets, filelines) == [
"EQUALS",
" MULTZ 0.017101 1 40 1 64 2*5 / nasty comment without comment characters",
"/",
]
filelines = [
"EQUALS",
"1 1 / nasty comment/",
"2 2 / foo",
"3 3 /",
"/",
"PERMX",
"1 1 /",
]
kwsets = find_keyword_sets(filelines)
assert compress_multiple_keywordsets(kwsets, filelines) == [
"EQUALS",
" 2*1 / nasty comment/", # (only compressing first record)
"2 2 / foo",
"3 3 /",
"/",
"PERMX",
" 2*1 /",
]
filelines = ["EQUALS", "1 1//", "2 2 / foo", "/"]
kwsets = find_keyword_sets(filelines)
assert compress_multiple_keywordsets(kwsets, filelines) == [
"EQUALS",
" 2*1 //",
"2 2 / foo",
"/",
]
def test_eclcompress():
"""Test a given set of lines, and ensure that the output
can be parsed by opm.io"""
kwsets = find_keyword_sets(FILELINES)
compressed = compress_multiple_keywordsets(kwsets, FILELINES)
compressedstr = "\n".join(compressed)
# Feed the compressed string into opm.io. OPM hopefully chokes on whatever
# Eclipse would choke on (and hopefully not on more..)
parsecontext = opm.io.ParseContext(
[("PARSE_MISSING_DIMS_KEYWORD", opm.io.action.ignore)]
)
assert opm.io.Parser().parse_string(compressedstr, parsecontext)
def test_eclkw_regexp(tmpdir):
"""Test that custom regular expressions can be supplied to compress
otherwise unknown (which implies no compression) keywords"""
tmpdir.chdir()
uncompressed_str = "G1\n0 0 0 0 0 0 0 0 0 0 0 0 0\n/"
# Nothing is found by default here.
assert not find_keyword_sets(uncompressed_str.split())
# Only if we specify a regexp catching this odd keyword name:
kw_sets = find_keyword_sets(uncompressed_str.split(), eclkw_regexp="G1")
kwend_idx = len(uncompressed_str.split()) - 1
assert kw_sets == [(0, kwend_idx)]
assert compress_multiple_keywordsets(kw_sets, uncompressed_str.split()) == [
"G1",
" 13*0",
"/",
]
with open("g1.grdecl", "w") as f_handle:
f_handle.write(uncompressed_str)
# Alternative regexpes that should also work with this G1:
kw_sets = find_keyword_sets(
uncompressed_str.split(), eclkw_regexp="[A-Z]{1-8}$"
) == [(0, kwend_idx)]
kw_sets = find_keyword_sets(
uncompressed_str.split(), eclkw_regexp="[A-Z0-9]{2-8}$"
) == [(0, kwend_idx)]
sys.argv = ["eclcompress", "g1.grdecl", "--eclkw_regexp", "G1"]
main()
compressed = open("g1.grdecl").read()
assert "File compressed with eclcompress" in compressed
assert "13*0" in compressed
def test_eclkw_regexp(tmp_path, mocker):
"""Test that custom regular expressions can be supplied to compress
otherwise unknown (which implies no compression) keywords"""
os.chdir(tmp_path)
uncompressed_str = "G1\n0 0 0 0 0 0 0 0 0 0 0 0 0\n/"
# Nothing is found by default here.
assert not find_keyword_sets(uncompressed_str.split())
# Only if we specify a regexp catching this odd keyword name:
kw_sets = find_keyword_sets(uncompressed_str.split(), eclkw_regexp="G1")
kwend_idx = len(uncompressed_str.split()) - 1
assert kw_sets == [(0, kwend_idx)]
assert compress_multiple_keywordsets(kw_sets, uncompressed_str.split()) == [
"G1",
" 13*0",
"/",
]
Path("g1.grdecl").write_text(uncompressed_str, encoding="utf8")
# Alternative regexpes that should also work with this G1:
kw_sets = find_keyword_sets(
uncompressed_str.split(), eclkw_regexp="[A-Z]{1-8}$"
) == [(0, kwend_idx)]
kw_sets = find_keyword_sets(
uncompressed_str.split(), eclkw_regexp="[A-Z0-9]{2-8}$"
) == [(0, kwend_idx)]
mocker.patch("sys.argv", ["eclcompress", "g1.grdecl", "--eclkw_regexp", "G1"])
main()
compressed = Path("g1.grdecl").read_text(encoding="utf8")
assert "File compressed with eclcompress" in compressed
assert "13*0" in compressed
def test_whitespace(tmpdir):
"""Ensure excessive whitespace is not added"""
kw_string = """
MULTIPLY
'PORO' 2 /
/"""
filelines = kw_string.splitlines()
assert (compress_multiple_keywordsets(find_keyword_sets(filelines),
filelines) == filelines)
# Test the same when the string is read from a file:
tmpdir.chdir()
Path("test.inc").write_text(kw_string)
eclcompress("test.inc")
compressed_lines = Path("test.inc").read_text().splitlines()
# The compressed output should have only two header lines added and one
# empty lines after the header added:
assert len(compressed_lines) == len(filelines) + 3
def test_include_statement():
"""A file with an INCLUDE statement has been tricky
not to destroy while compressing"""
filelines = ["INCLUDE", " '../include/grid/grid.grdecl' /"]
kwsets = find_keyword_sets(filelines)
assert compress_multiple_keywordsets(kwsets, filelines) == filelines
def test_compress_multiple_keywordsets():
"""Test compression of sample lines"""
filelines = ["PORO", "0 0 0 3", "4 5 6", "/ postslashcomment"]
kwsets = find_keyword_sets(filelines)
assert compress_multiple_keywordsets(kwsets, filelines) == [
"PORO",
" 3*0 3 4 5 6",
"/ postslashcomment",
]
filelines = ["PORO", "0 0 0 3", "4 5 6", "/"]
kwsets = find_keyword_sets(filelines)
assert compress_multiple_keywordsets(kwsets, filelines) == [
"PORO",
" 3*0 3 4 5 6",
"/",
]
filelines = ["PORO", "0 0 0 3", "4 5 6 /"]
kwsets = find_keyword_sets(filelines)
assert compress_multiple_keywordsets(kwsets, filelines) == [
"PORO",
" 3*0 3 4 5 6 /",
]
filelines = ["PORO", "0 0 0 3", "4 5 6 / postslashcomment"]
kwsets = find_keyword_sets(filelines)
assert compress_multiple_keywordsets(kwsets, filelines) == [
"PORO",
" 3*0 3 4 5 6 / postslashcomment",
]
filelines = ["PORO", "0 0 0 3 4 5 6 / postslashcomment"]
kwsets = find_keyword_sets(filelines)
assert compress_multiple_keywordsets(kwsets, filelines) == [
"PORO",
" 3*0 3 4 5 6 / postslashcomment",
]
filelines = ["PORO", "0 0 /", "PERMX", "1 1 /"]
kwsets = find_keyword_sets(filelines)
assert compress_multiple_keywordsets(kwsets, filelines) == [
"PORO",
" 2*0 /",
"PERMX",
" 2*1 /",
]
filelines = ["PORO", "0 0 /", "", "PERMX", "1 1 /"]
kwsets = find_keyword_sets(filelines)
assert compress_multiple_keywordsets(kwsets, filelines) == [
"PORO",
" 2*0 /",
"",
"PERMX",
" 2*1 /",
]
filelines = ["-- comment", "PORO", "0 0", "/"]
kwsets = find_keyword_sets(filelines)
assert compress_multiple_keywordsets(kwsets, filelines) == [
"-- comment",
"PORO",
" 2*0",
"/",
]
filelines = ["-- nastycomment with / slashes", "PORO", "0 0", "/"]
kwsets = find_keyword_sets(filelines)
assert compress_multiple_keywordsets(kwsets, filelines) == [
"-- nastycomment with / slashes",
"PORO",
" 2*0",
"/",
]
