make more robust against data error

databaseReset.py

@@ -268,7 +268,7 @@ class JobQueue:
         self.results["TOTAL"] = []        
         for i in range(len(self.results["date"])):
             self.format_date(i)
-            print(i, self.compute_total(i))
+            # print(i, self.compute_total(i))
             self.results["TOTAL"].append(self.compute_total(i))         
        
         for j in self.results_order:

parsers/logbooks.py

@@ -58,7 +58,7 @@ LOGBOOK_PARSER_SETTINGS = {
 LOGBOOKS_DIR = "years" # subfolder of settings.EXPOWEB
 
 ENTRIES = {
-    "2024": 125,
+    "2024": 126,
     "2023": 131,
     "2022": 94,
     "2019": 55,
@@ -66,7 +66,7 @@ ENTRIES = {
     "2017": 74,
     "2016": 87,
     "2015": 80,
-    "2014": 67,
+    "2014": 68,
     "2013": 52,
     "2012": 76,
     "2011": 71,

parsers/people.py

@@ -137,10 +137,15 @@ def load_people_expos():
         nick = ""
 
         rawlastname = personline[header["Lastname"]].strip()
-        matchlastname = re.match(r"^([\w&;\s]+)(?:\(([^)]*)\))?", rawlastname)
-        lastname = matchlastname.group(1).strip()
+        if rawlastname == "":
+            print(f"MISSING SURNAME FIELD for {name} - check against similar names in the list to see what you have done.")
+        if matchlastname := re.match(r"^([\w&;\s]+)(?:\(([^)]*)\))?", rawlastname):
+            lastname = matchlastname.group(1).strip()
+        else:
+            print(f"MATCH FAIL {personline=}\n     {slug=}\n     {name=}\n     {rawlastname=}")
+            exit(1)
 
-        splitnick = re.match(r"^([\w&;\s]+)(?:\(([^)]*)\))?", plainname)
+        splitnick = re.match(r"^([\w&;\s\-]+)(?:\(([^)]*)\))?", plainname)
         fullname = splitnick.group(1) # removes Nickname in brackets, but also cuts hyphenated names
         nick = splitnick.group(2) or "" 
 

parsers/survex.py

@@ -22,6 +22,12 @@ It also scans the Loser repo for all the svx files, which it loads individually
 """
 
 todo = """
+- Rewrite regexes to use .groupdict instead of .group, 
+  easier to understand and maintain
+  https://mathspp.com/blog/til/re-match-groupdict
+  
+- replace hard-coded instuments list with reding an editable textfile in expoweb.
+
 - Obscure bug in the *team inheritance and rootblock initialization needs tracking down,
   probably in the team cache which should NOT be global, but should be an instance variable of 
   LoadingSurvex
@@ -75,10 +81,6 @@ class SurvexLeg:
     clino = 0.0
     
         
-
-
-
-
 def datewallet(w, earliest):
     """Gets the date of the youngest survexblock associated with the wallet
     REFACTOR this to do the whole date-getting task
@@ -163,7 +165,7 @@ def get_people_on_trip(survexblock):
     
     return list(set(people))
 
-# THIS SHOULD NOT BE GLOBAL ! SHould be per instance of file loader
+# THIS SHOULD NOT BE GLOBAL ! Should be per instance of file loader
 trip_person_record = {}  # indexed by (survexblock, personexpedition) - so never needs cleaning out
 trip_team_cache = {}  #  indexed by  survexblock, so never needs cleaning out
 def put_person_on_trip(survexblock, personexpedition, tm):