Merge pull request #4 from h2020charisma/bugfix

Fix logbook encoding problems and invalid date in header

VERSION

@@ -1 +1 @@
-0.0.1
+0.0.2

code/spc_io/high_level/spc.py

@@ -2,9 +2,11 @@
 
 import io
 from typing import Union, List, Literal, ForwardRef
+import logging
 from datetime import datetime
 import numpy.typing as npt
 import numpy as np
+import pandas as pd
 from pydantic import validate_arguments
 from spc_io.low_level.headers.fxytype import Fxtype, Fytype
 from spc_io.low_level.spc_raw import SpcRaw
@@ -17,7 +19,7 @@
 from .log_book import LogBook
 from .even_axis import EvenAxis
 
-import pandas as pd
+logger = logging.getLogger(__name__)
 
 SPC = ForwardRef('SPC')
 
@@ -132,12 +134,16 @@ def from_spc_raw(cls, spc_raw: SpcRaw):
                            text=spc_raw.log_book.txt_as_dict())
         main_header = spc_raw.main_header
         fdate = main_header.fdate
-        date = datetime(year=fdate.year,
-                        month=fdate.month,
-                        day=fdate.day,
-                        hour=fdate.hour,
-                        minute=fdate.min,
-                        )
+        try:
+            date = datetime(year=fdate.year,
+                            month=fdate.month,
+                            day=fdate.day,
+                            hour=fdate.hour,
+                            minute=fdate.min,
+                            )
+        except ValueError as e:
+            logger.warning(repr(e))
+            date = datetime.fromtimestamp(0)
 
         if not main_header.ftflgs.TXYXYS:
             if main_header.ftflgs.TXVALS:

code/spc_io/low_level/headers/logstc.py

@@ -6,7 +6,11 @@
 
 class LogBookBase:
     def txt_as_dict(self):
-        return dict([i.split(b'=') for ii in self.txt.split(b'\r\n') for i in ii.split(b'\n\r') if b'=' in i])
+        return dict([i.decode(errors='surrogateescape').split('=')
+                     for ii in self.txt.split(b'\r\n')
+                     for i in ii.split(b'\n\r')
+                     if b'=' in i
+                     ])
 
     def disk_as_bytes(self):
         return bytearray(string_at(addressof(self.disk), sizeof(self.disk)))
@@ -33,7 +37,7 @@ def new_header_and_logbook_from_data(cls, *,
                                          binary: bytes = b'',
                                          txt: Union[bytes, Dict] = b''):
         if isinstance(txt, dict):
-            txt = b'\r\n'.join([f'{k}={v}'.encode() for k, v in txt.items()])
+            txt = b'\r\n'.join([f'{k}={v}'.encode(errors='surrogateescape') for k, v in txt.items()])
         txt += b'\x00'  # NUL terminate
         logsizd = len(disk)+len(binary)+len(txt)+sizeof(cls)
         logsizm = ((logsizd // 4096) + 1) * 4096

code/spc_io/low_level/spc_raw.py

@@ -147,7 +147,8 @@ def calcuate_offsets(self):
                 size += sizeof(sub._xarray)
             size += sizeof(sub._yarray)
             ofs += size
-            self.dirs.append(Ssfstc(ssfposn=ofs, ssfsize=size, ssftime=sub.z))
+            if sub.z is not None:
+                self.dirs.append(Ssfstc(ssfposn=ofs, ssfsize=size, ssftime=sub.z))
 
         # Directory offset
         ofs += sizeof(self.log_book)