Refactor _parse_xy_xy to standardize output as 3D arrays

nmrglue/fileio/bruker.py

@@ -1904,7 +1904,12 @@ def read_jcamp(filename):
     with open(filename, 'r') as f:
         while True:     # loop until end of file is found
 
-            line = f.readline().rstrip()    # read a line
+            try:
+                line = f.readline().rstrip()    # read a line
+            except Exception as e:
+                warn("Unable read line, leave it as a comment")
+                line = "$$"
+
             if line == '':      # end of file found
                 break
 

nmrglue/fileio/jcampdx.py

@@ -37,14 +37,14 @@ def _getkey(keystr):
             .replace("-", "").replace("_", "").replace("/", ""))
 
 
-def _readrawdic(filename):
+def _readrawdic(filename, read_err=None):
     '''
     Reads JCAMP-DX file to key-value dictionary, from which
     actual data is separated later.
     '''
 
     dic = {"_comments": []}  # create empty dictionary
-    filein = open(filename, 'r')
+    filein = open(filename, 'r', errors=read_err)
 
     currentkey = None
     currentvaluestrings = []
@@ -151,6 +151,8 @@ def _detect_format(dataline):
     firstvalue_re = re.compile(
         "(\s)*([+-]?\d+\.?\d*|[+-]?\.\d+)([eE][+-]?\d+)?(\s)*")
 
+    xy_re = re.compile('^[0-9\.]+,[ ]?[0-9\.]+')
+
     index = firstvalue_re.match(dataline).end()
     if index is None:
         return -1
@@ -165,6 +167,10 @@ def _detect_format(dataline):
         return 1
     if firstchar in _SQZ_DIGITS:
         return 1
+
+    if re.search(xy_re, dataline):
+        return 2
+
     return 0
 
 
@@ -347,10 +353,35 @@ def _parse_pseudo(datalines):
     return data
 
 
+def _parse_xy_xy(datalines):
+    pts = []
+    len_group_data = 0
+    for dataline in datalines:
+        if not dataline:
+            continue
+        xy_re = re.compile('[^ ][0-9\.]+, [0-9\.]+')
+        group_data = re.findall(xy_re, dataline)
+        len_group_data = len(group_data)
+        if len_group_data == 0:
+            xy_re = re.compile('[^ ][0-9\.]+,[0-9\.]+;')
+            group_data = re.findall(xy_re, dataline)
+
+        for data in group_data:
+            clean_data = data.replace(', ', ',')
+            clean_data = clean_data.replace(';', '')
+            x, y = clean_data.split(',')
+            pts.append([float(x), float(y)])
+    return [pts]
+
+
 def _parse_data(datastring):
     '''
     Creates numpy array from datalines
     '''
+    probe_data = datastring[80:320]
+    if ',' in probe_data and not('.' in probe_data): # fix comma as decimal points
+        datastring = datastring.replace(',', '.')
+
     datalines = datastring.split("\n")
     headerline = datalines[0]
 
@@ -364,6 +395,11 @@ def _parse_data(datastring):
         data = _parse_pseudo(datalines)
     elif mode == 0:
         data = _parse_affn_pac(datalines)
+    elif mode == 2:
+        if headerline == '(X++(Y..Y))':
+            data = _parse_affn_pac(datalines)
+        else:
+            data = _parse_xy_xy(datalines)
     else:
         return None
     if data is None:
@@ -417,7 +453,7 @@ def find_yfactors(dic):
     return (factor_r, factor_i)
 
 
-def _getdataarray(dic):
+def _getdataarray(dic, show_all_data=False):
     '''
     Main function for data array parsing, input is the
     raw dictionary from _readrawdic
@@ -447,19 +483,23 @@ def _getdataarray(dic):
                     idatalist.append(data)
                 else:
                     rdatalist.append(data)
-            if len(rdatalist) > 1:
-                warn("NTUPLES: multiple real arrays, returning first one only")
-            if len(idatalist) > 1:
-                warn("NTUPLES: multiple imaginary arrays, \
-                     returning first one only")
-            if rdatalist:
-                if idatalist:
-                    data = [rdatalist[0], idatalist[0]]
-                else:
-                    data = rdatalist[0]
+
+            if show_all_data:
+                data = { 'real': rdatalist, 'imaginary': idatalist }
             else:
-                if idatalist:
-                    data = [None, idatalist[0]]
+                if len(rdatalist) > 1:
+                    warn("NTUPLES: multiple real arrays, returning first one only")
+                if len(idatalist) > 1:
+                    warn("NTUPLES: multiple imaginary arrays, \
+                         returning first one only")
+                if rdatalist:
+                    if idatalist:
+                        data = [rdatalist[0], idatalist[0]]
+                    else:
+                        data = rdatalist[0]
+                else:
+                    if idatalist:
+                        data = [None, idatalist[0]]
 
     if data is None:  # XYDATA
         try:
@@ -472,11 +512,27 @@ def _getdataarray(dic):
         except KeyError:
             warn("XYDATA not found ")
 
+    if data is None:  # PEAK TABLE
+        try:
+            valuelist = dic["PEAKTABLE"]
+            if len(valuelist) == 1:
+                data, datatype = _parse_data(valuelist[0])
+            else:
+                warn("Multiple PEAKTABLE arrays in JCAMP-DX file, \
+                     returning first one only")
+        except KeyError:
+            warn("PEAKTABLE not found ")
+
     # apply YFACTOR to data if available
     if is_ntuples:
         yfactor_r, yfactor_i = find_yfactors(dic)
         if yfactor_r is None or yfactor_r is None:
             warn("NTUPLES: YFACTORs not applied, parsing failed")
+        elif show_all_data:
+            for i, _ in enumerate(data['real']):
+                data['real'][i] = data['real'][i] * yfactor_r
+            for i, _ in enumerate(data['imaginary']):
+                data['imaginary'][i] = data['imaginary'][i] * yfactor_i
         else:
             data[0] = data[0] * yfactor_r
             data[1] = data[1] * yfactor_i
@@ -492,7 +548,7 @@ def _getdataarray(dic):
     return data
 
 
-def read(filename):
+def read(filename, show_all_data=False, read_err=None):
     """
     Read JCAMP-DX file
 
@@ -515,13 +571,14 @@ def read(filename):
     # first read everything (including data array) to "raw" dictionary,
     # in which data values are read as raw strings including whitespace
     # and newlines
-    dic = _readrawdic(filename)
+    dic = _readrawdic(filename, read_err)
 
     # find and parse NMR data array from raw dic
-    data = _getdataarray(dic)
+    data = _getdataarray(dic, show_all_data)
 
     # remove data tables from dic
     try:
+        dic['XYDATA_OLD'] = dic["XYDATA"]
         del dic["XYDATA"]
     except KeyError:
         pass