From c361a04ca56eeb0b44d6d01ce7d4decf48aa1aea Mon Sep 17 00:00:00 2001
From: Abram Claycomb <aclaycomb@usgs.gov>
Date: Mon, 13 May 2019 17:12:23 -0600
Subject: [PATCH] add one second parsing, tests for headers, simplify for
readability
---
geomagio/pcdcp/PCDCPFactory.py | 1 +
geomagio/pcdcp/PCDCPParser.py | 26 ++++++++++-----------
test/pcdcp_test/PCDCPParser_test.py | 35 ++++++++++++++++++++++++++++-
3 files changed, 48 insertions(+), 14 deletions(-)
diff --git a/geomagio/pcdcp/PCDCPFactory.py b/geomagio/pcdcp/PCDCPFactory.py
index fb3d9ea8c..05d3bdbd9 100644
--- a/geomagio/pcdcp/PCDCPFactory.py
+++ b/geomagio/pcdcp/PCDCPFactory.py
@@ -10,6 +10,7 @@ from .PCDCPWriter import PCDCPWriter
# pattern for pcdcp file names
PCDCP_FILE_PATTERN = '%(obs)s%(y)s%(j)s.%(i)s'
+# note: seconds files end in .raw, not .sec
class PCDCPFactory(TimeseriesFactory):
diff --git a/geomagio/pcdcp/PCDCPParser.py b/geomagio/pcdcp/PCDCPParser.py
index f82a51a1d..4d3f21f21 100644
--- a/geomagio/pcdcp/PCDCPParser.py
+++ b/geomagio/pcdcp/PCDCPParser.py
@@ -29,7 +29,11 @@ class PCDCPParser(object):
def __init__(self):
"""Create a new PCDCP parser."""
# header fields
+ self.header_fields = ['station', 'year', 'yearday', 'date',
+ 'orientation', 'resolution', 'Version']
self.header = {}
+ # resolution (float)
+ self.resolution = 0.0
# array of channel names
self.channels = []
# timestamps of data (datetime.datetime)
@@ -64,11 +68,10 @@ class PCDCPParser(object):
Adds value to ``self.header``.
"""
- self.header['header'] = line
- self.header['station'] = line[0:3]
- self.header['year'] = line[5:9]
- self.header['yearday'] = line[11:14]
- self.header['date'] = line[16:25]
+ self.header = dict(zip(self.header_fields,
+ line.split(None, len(self.header_fields))))
+
+ self.resolution = float(self.header['resolution'].split('nT')[0])
return
@@ -78,13 +81,9 @@ class PCDCPParser(object):
Adds time to ``self.times``.
Adds channel values to ``self.data``.
"""
- t, d1, d2, d3, d4 = self._parsedata
-
- t.append(line[0:4])
- d1.append(int(line[5:13]))
- d2.append(int(line[14:22]))
- d3.append(int(line[23:31]))
- d4.append(int(line[32:40]))
+ values = line.split()
+ for (value, column) in zip(values, self._parsedata):
+ column.append(value)
def _post_process(self):
"""Post processing after data is parsed.
@@ -98,7 +97,8 @@ class PCDCPParser(object):
data = numpy.array(data, dtype=numpy.float64)
# filter empty values
data[data == NINES] = numpy.nan
- data = numpy.divide(data, 100)
+ data[data == NINES_RAW] = numpy.nan
+ data = numpy.multiply(data, self.resolution)
self.data[channel] = data
self._parsedata = None
diff --git a/test/pcdcp_test/PCDCPParser_test.py b/test/pcdcp_test/PCDCPParser_test.py
index e31a4f8ba..192cd911d 100644
--- a/test/pcdcp_test/PCDCPParser_test.py
+++ b/test/pcdcp_test/PCDCPParser_test.py
@@ -18,8 +18,24 @@ BOU 2015 001 01-Jan-15 HEZF 0.01nT File Version 2.00
0008 2086278 -5571 4745734 5237808
"""
+PCDCP_EXAMPLE_SECOND = \
+"""
+BOU 2015 001 01-Jan-15 HEZF 0.001nT File Version 2.00
+00000 20861520 -57095 47457409 52377630
+00001 20861533 -57096 47457397 52377650
+00002 20861554 -57077 47457391 52377650
+00003 20861578 -57068 47457389 52377680
+00004 20861600 -57068 47457384 52377660
+00005 20861640 -57047 47457388 52377690
+00006 20861654 -57039 47457378 52377650
+00007 20861699 -57026 47457377 52377690
+00008 20861721 -56995 47457365 52377680
+00009 20861743 -56977 47457350 52377680
+00010 20861750 -56968 47457349 52377690
+"""
+
-def test__parse_header():
+def test_parse_header():
"""pcdcp_test.PCDCPParser_test.test_parse_header()
Call the _parse_header method with a header.
@@ -33,3 +49,20 @@ def test__parse_header():
assert_equals(parser.header['station'], 'BOU')
assert_equals(parser.header['year'], '2015')
assert_equals(parser.header['yearday'], '001')
+ assert_equals(parser.header['resolution'], '0.01nT')
+
+def test_parse_header_sec():
+ """pcdcp_test.PCDCPParsser_test.test_parse_header_sec()
+
+ Call the _parse_header method with a pcdcp seconds file '.raw'
+ header. Verify the header name and value are split correctly.
+ """
+ parser = PCDCPParser()
+ parser._parse_header('BOU 2015 001 01-Jan-15 HEZF 0.001nT' +
+ ' File Version 2.00')
+
+ assert_equals(parser.header['date'], '01-Jan-15')
+ assert_equals(parser.header['station'], 'BOU')
+ assert_equals(parser.header['year'], '2015')
+ assert_equals(parser.header['yearday'], '001')
+ assert_equals(parser.header['resolution'], '0.001nT')
\ No newline at end of file
--
GitLab