Cleaning auto Cave creation from survex file detection

1 files changed, 54 insertions, 43 deletions

diff --git a/parsers/survex.py b/parsers/survex.py
index ef69759..fe7d211 100644
--- a/parsers/survex.py
+++ b/parsers/survex.py
@@ -317,16 +317,16 @@ class LoadingSurvex:
     insp = ""
     callcount = 0
     caverncount = 0
-    ignoreprefix = ["surface", "kataster", "fixedpts", "gpx", "deprecated"]
-    ignorenoncave = [
-        "caves-1623",
-        "caves-1623/2007-NEU",
-        "caves-1626",
-        "caves-1624",
-        "caves-1627",
-        "fixedpts/gps/gps00raw",
-        "",
-    ]
+    ignoreprefix = ["surface", "kataster", "fixedpts", "gpx", "deprecated", "dummy_"]
+    # ignorenoncave = [
+        # "caves-1623",
+        # "caves-1623/2007-NEU",
+        # "caves-1626",
+        # "caves-1624",
+        # "caves-1627",
+        # "fixedpts/gps/gps00raw",
+        # "",
+    # ]
     TREE = "tree"
     ODDS = "oddments"
     svxpass = TREE
@@ -1165,42 +1165,60 @@ class LoadingSurvex:
 
     def IdentifyCave(self, cavepath, svxid, depth):
         """Given a file path for a survex file, e.g. /1626/107/107.svx, or a survex-block path,
-        return the cave object       
+        return the cave object 
+            kataster
+            fixedpts/gps
+            and everything at top level, directly in caves-1623/ not in a subdir
+        NOTE self.cavelist is a superset of GCaveLookup, which already contians both uppercase and lowercase aliases
         """
-        path = cavepath.lower()
-        if path in self.caveslist: # primed with GCaveLookup
-            return self.caveslist[path]
+        if cavepath == "caves-1623/99ob02":
+            for key in self.caveslist:
+                cave = self.caveslist[key]
+                if type(cave) != Cave:
+                    print(f"BAD CAVE TYPE '{key}'  --  {type(cave)}'{cave}'")
+            for key in self.caveslist:
+                cave = self.caveslist[key]
+                print(f"{key}  --  Cave<{cave}>")
+                
+        for i in self.ignoreprefix:
+            if cavepath.lower().startswith(i):
+                message = (f" - {cavepath} starts with <ignoreprefix> (while creating '{svxid}.svx' )")
+                return False
+                
+        if cavepath in self.caveslist: # primed with GCaveLookup
+            return self.caveslist[cavepath]
+
+        rx_svxcollection = re.compile(r"(?i)caves-(\d\d\d\d)/(.*)$") 
         #     rx_cave = re.compile(r"(?i)caves-(\d\d\d\d)/([-\d\w]+|\d\d\d\d-?\w+-\d+)")
         path_match = self.rx_cave.search(cavepath) 
         if path_match:
             sluggy = f"{path_match.group(1)}-{path_match.group(2)}"
-            guesses = [sluggy.lower()] # full 1626-107 search, don;t use short-forms
-            for g in guesses:
-                if g in self.caveslist:
-                    self.caveslist[cavepath] = self.caveslist[g] # set "caves-1626/107/107" as index to cave 1626-107
-                    return self.caveslist[g]
-            cave = create_new_cave(cavepath) # uses the pending stuff to create pending cave descriptions
+            seek = [sluggy, sluggy.replace("1623-","")] # to catch '2023-kt-02' etc
+            for s in seek:
+                if s in self.caveslist:
+                    self.caveslist[cavepath] = self.caveslist[s] # set "caves-1626/107/107" as index to cave 1626-107
+                    return self.caveslist[s]
+            
+            cave = create_new_cave(cavepath, f"Make cave found in survex file {svxid}") # uses the pending code to create pending cave descriptions
             self.caveslist[cavepath] = cave
-            message = f"    ! MAKING cave for {cavepath=} {svxid=}"
-            stash_data_issue(parser="survex", message=message, url=None, sb=(svxid))
-            if not cavepath.startswith("caves-1624") or cavepath.startswith("caves-1626"):
+            message = f"\n    ! MAKING cave {sluggy} for {cavepath=} {svxid=} (not reporting this for 1624 or 1626)"
+            # stash_data_issue(parser="survex", message=message, url="/survexfile/{svxid}.svx", sb=(svxid))
+            if not (cavepath.startswith("caves-1624") or cavepath.startswith("caves-1626")):
                 print(message, file=sys.stderr)
+            return cave
         else:
-            # isn't all this pointless...??
-            if self.is_it_already_pending(cavepath, svxid, depth): # but pending will already have been created as Cave objects
-                pass
+            path_match = rx_svxcollection.search(svxid) 
+            if path_match:
+                message = f"    ! Recognised survex file which is not a cave at {svxid=}"
+                # stash_data_issue(parser="survex", message=message, url=None, sb=(svxid))
+                # print(message, file=sys.stderr)
+                return False
             else:
-                # It is too late to add it to the pending caves list here, they were already 
-                # processed in parsers/caves.py So we have to do a bespoke creation.
-                cave = create_new_cave(svxid)
-                message = f" ! Warning: cave identifier '{caveid}'or {id} (guessed from file path)  is not a known cave.  Need to add to expoweb/cave_data/pendingcaves.txt ?  In '{includelabel}.svx' at depth:[{len(depth)}]."
+                message = f" ! ERROR: no cave at '{svxid}.svx  - is not a known cavename format. "
                 print("\n" + message)
                 print("\n" + message, file=sys.stderr)
-                print(f"{self.pending}", end="", file=sys.stderr)
-                stash_data_issue(parser="survex", message=message, url=None, sb=(includelabel))
-                    
-                print(f'    ! No regex (standard identifier) cave match for {cavepath.lower()}', file=sys.stderr)
-            return None
+                stash_data_issue(parser="survex", message=message, url="{svxid}.svx", sb=(svxid))
+                return None
             
     def is_it_already_pending(self, headpath, includelabel, depth):
         """Ignore surface, kataser and gpx *include survex files"""
@@ -1222,14 +1240,7 @@ class LoadingSurvex:
             # print("\n"+message)
             # print("\n"+message,file=sys.stderr)
             return True
-        for i in self.ignoreprefix:
-            if headpath.startswith(i):
-                message = (
-                    f" - {headpath} starts with <ignoreprefix> (while creating '{includelabel}' sfile & sdirectory)"
-                )
-                # print("\n"+message)
-                # print("\n"+message,file=sys.stderr)
-                return True
+
         caveid = f"{headpath[6:10]}-{headpath[11:]}".upper()
         if caveid in self.pending:
             # Yes we didn't find this cave, but we know it is a pending one. So not an error.