Merge pull request #200 from h2020charisma/lightnovo_tsv

Support for TSV file format

src/ramanchada2/io/experimental/lightnovo_tsv.py

@@ -0,0 +1,21 @@
+from typing import Dict, List, Tuple
+
+import pandas
+
+
+def lightnovo_tsv(lines: List[str]) -> Tuple[pandas.DataFrame, Dict]:
+    for i, ll in enumerate(lines):
+        if not ll.startswith(('DeviceSN', 'LaserWavelength_nm', 'Exposure_ms',
+                              'GainMultiplier', 'LaserCurrent_mA', 'LaserPower_mW',
+                              'RepetitionCount', 'Tags')):
+            start_spe = i
+            break
+    else:
+        raise ValueError('The input is not lightnovo tsv format')
+    meta = dict([i.split('\t', 1) for i in lines[:start_spe]])
+    spe_lines = [ll.strip().split('\t') for ll in lines[start_spe:]]
+
+    data = pandas.DataFrame.from_records(data=spe_lines, columns=['Position', 'Amplitude']
+                                         ).apply(pandas.to_numeric, errors='coerce'
+                                                 ).dropna(axis=0)
+    return data, meta

src/ramanchada2/io/experimental/read_txt.py

@@ -5,6 +5,7 @@
 
 from .bw_format import bw_format
 from .neegala_format import neegala_format
+from .lightnovo_tsv import lightnovo_tsv
 from .rruf_format import rruf_format
 
 """ There are 4 types of TXT data files that can be distinguished by their first line:
@@ -35,6 +36,12 @@ def read_txt(data_in: TextIO) -> Tuple[NDArray, NDArray, Dict]:
         intensities = data['Processed_Data'].to_numpy()
         meta['@axes'] = ['Raman_Shift']
         meta['@signal'] = 'Processed_Data'
+    elif lines[0].startswith('DeviceSN\t'):
+        data, meta = lightnovo_tsv(lines)
+        positions = data['Position'].to_numpy()
+        intensities = data['Amplitude'].to_numpy()
+        meta['@axes'] = ['']
+        meta['@signal'] = ''
     else:  # assume two column spectrum
         meta = dict()
         if lines[0].startswith('#'):

src/ramanchada2/spectrum/creators/from_local_file.py

@@ -21,7 +21,8 @@ def from_local_file(
         in_file_name: str,
         filetype: Union[None, Literal['spc', 'sp', 'spa', '0', '1', '2',
                                       'wdf', 'ngs', 'jdx', 'dx',
-                                      'txt', 'txtr', 'csv', 'prn', 'rruf', 'spe', 'cha']] = None,
+                                      'txt', 'txtr', 'tsv', 'csv', 'prn',
+                                      'rruf', 'spe', 'cha']] = None,
         backend: Union[None, Literal['native', 'rc1_parser']] = None,
         custom_meta: Dict = {}):
     """
@@ -32,7 +33,7 @@ def from_local_file(
             Path to a local file containing a spectrum.
         filetype:
             Specify the filetype. Filetype can be any of: `spc`, `sp`, `spa`, `0`, `1`, `2`, `wdf`, `ngs`, `jdx`, `dx`,
-            `txt`, `txtr`, `csv`, `prn`, `rruf`, `spe` (Princeton Instruments) or `None`.
+            `txt`, `txtr`, `tsv`, `csv`, `prn`, `rruf`, `spe` (Princeton Instruments) or `None`.
             `None` used to determine by extension of the file.
         backend:
             `native`, `rc1_parser` or `None`. `None` means both.
@@ -50,7 +51,7 @@ def load_native():
             ft = filetype
         if ft in {'cha'}:
             return from_chada(filename=in_file_name)
-        elif ft in {'txt', 'txtr', 'prn', 'rruf'}:
+        elif ft in {'txt', 'txtr', 'prn', 'rruf', 'tsv'}:
             with open(in_file_name) as fp:
                 x, y, meta = read_txt(fp)
         elif ft in {'csv'}:

src/ramanchada2/spectrum/creators/from_stream.py

@@ -20,12 +20,13 @@
 def from_stream(in_stream: Union[io.TextIOBase, io.BytesIO, io.BufferedReader],
                 filetype: Union[None, Literal['spc', 'sp', 'spa', '0', '1', '2',
                                               'wdf', 'ngs', 'jdx', 'dx',
-                                              'txt', 'txtr', 'csv', 'prn', 'rruf', 'spe']],
+                                              'txt', 'txtr', 'tsv', 'csv', 'prn',
+                                              'rruf', 'spe']],
                 filename: Optional[str] = None,
                 backend: Union[None, Literal['native', 'rc1_parser']] = None,
                 ):
     def load_native():
-        if filetype in {'txt', 'txtr', 'prn', 'rruf'}:
+        if filetype in {'txt', 'txtr', 'tsv', 'prn', 'rruf'}:
             if isinstance(in_stream, io.TextIOBase):
                 fp = in_stream
             else: