xle/troggle-unchained
2
0
Fork 0
forked from expo/troggle
Code Issues 4 Pull Requests Projects 1 Wiki Activity

attempt to refactor *team/*dte interation

Browse Source
This commit is contained in:
Philip Sargent
2025-12-19 01:38:50 +00:00
parent c58857d1e1
commit 3a93d10d4f
1 changed files with 25 additions and 22 deletions
Download Patch File Download Diff File Expand all files Collapse all files

47
parsers/survex.py
View File

@@ -168,9 +168,9 @@ def get_people_on_trip(survexblock):
return list(set(people)) 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, even though they are globally unique
trip_person_record = {} # indexed by (survexblock, personexpedition) - so never needs cleaning out trip_person_record = {} # a dict indexed by tuples (survexblock, personexpedition) = 1
trip_team_cache = {} # indexed by survexblock, so never needs cleaning out trip_team_cache = {} # a dict of lists indexed by survexblock
def put_person_on_trip(survexblock, personexpedition, tm): def put_person_on_trip(survexblock, personexpedition, tm):
"""Uses a cache to avoid a database query if it doesn't need to. """Uses a cache to avoid a database query if it doesn't need to.
Only used for a single person""" Only used for a single person"""
@@ -197,6 +197,7 @@ def put_person_on_trip(survexblock, personexpedition, tm):
if survexblock not in trip_team_cache: if survexblock not in trip_team_cache:
trip_team_cache[survexblock] = [] trip_team_cache[survexblock] = []
trip_team_cache[survexblock].append(personrole) trip_team_cache[survexblock].append(personrole)
print(f"-- trip_team_cache {survexblock}, {trip_team_cache[survexblock]}, {personrole}")
trip_person_record[(survexblock, personexpedition)] = 1 trip_person_record[(survexblock, personexpedition)] = 1
return False return False
@@ -228,17 +229,17 @@ def add_to_pending(survexblock, tm):
if survexblock not in person_pending_cache: if survexblock not in person_pending_cache:
person_pending_cache[survexblock] = set() person_pending_cache[survexblock] = set()
person_pending_cache[survexblock].add(tm) person_pending_cache[survexblock].add(tm)
print(f"-- person_pending_cache {survexblock}, {person_pending_cache[survexblock]}, {tm}")
def get_team_pending(survexblock): def get_team_pending(survexblock):
"""A set of *team names before we get to the *date line in a survexblock """A set of *team names added at the end of the survex block
""" """
global person_pending_cache global person_pending_cache
if survexblock in person_pending_cache: if survexblock in person_pending_cache:
teamnames = person_pending_cache[survexblock] # a set of names teamnames = person_pending_cache[survexblock] # a set of names
person_pending_cache[survexblock] = () person_pending_cache[survexblock] = set() #re zero the cache
return teamnames return teamnames
return return
@@ -634,9 +635,9 @@ class LoadingSurvex:
# so we can't validate whether the person was on expo or not. # so we can't validate whether the person was on expo or not.
# we will have to attach them to the survexblock anyway, and then do a # we will have to attach them to the survexblock anyway, and then do a
# later check on whether they are valid when we get the date. # later check on whether they are valid when we get the date.
# big changes 24/2/2025 when Olly changed survex behaviour, and unilaterally edited hundreds of # refactor this to collect names before and after a *date, and commit them as
# survex files to match. So troggle has to change to match the revised, stricter syntax. # a bulk update only at the END of the survexblock
if not tm: # i.e. null person inthe *team if not tm: # i.e. null person inthe *team
return # ignore: troggle does not need to know. Survex want to though. return # ignore: troggle does not need to know. Survex want to though.
@@ -863,13 +864,24 @@ class LoadingSurvex:
print(self.insp + message) print(self.insp + message)
stash_data_issue(parser='survex', message=message, url=None, sb=(survexblock.survexfile.path)) stash_data_issue(parser='survex', message=message, url=None, sb=(survexblock.survexfile.path))
return expo return expo
def LoadSurvexDate(self, survexblock, line): def LoadSurvexDate(self, survexblock, line):
"""We now have a valid date for this survexblock, so we now know the expo """We now have a valid date for this survexblock, so we now know the expo
it relates to and can use GetPersonExpeditionNameLookup(expo) to check whether it relates to and can use GetPersonExpeditionNameLookup(expo) to check whether
the people are correct. the people are correct.
Note that a *team line can come before AND after a *date line""" Note that a *team line can come before AND after a *date line
REFACTOR this so that the team setting is independent of the *date line.
Keep the team stored in a pending list, and commit to database only at the end of
the survexblock in a bulk_update.
This has turned out to be more fiddly than expected and one
attempt has already been aborted.
caves-1623/161/triassic/henricat.svx has the *date after all the team members
in each survexblock
"""
def setdate_on_survexblock(year): def setdate_on_survexblock(year):
"""Either *date comes before any *team, in which case there are no prior """Either *date comes before any *team, in which case there are no prior
@@ -887,7 +899,7 @@ class LoadingSurvex:
print(self.insp + message) print(self.insp + message)
stash_data_issue(parser='survex', message=message, url=None, sb=(survexblock.survexfile.path)) stash_data_issue(parser='survex', message=message, url=None, sb=(survexblock.survexfile.path))
if teamnames := get_team_pending(survexblock): # WALRUS https://docs.python.org/3/whatsnew/3.8.html#assignment-expressions if teamnames := get_team_pending(survexblock):
for tm in teamnames: for tm in teamnames:
if known_foreigner(tm): if known_foreigner(tm):
message = f"- *team {expo.year} '{tm}' known foreigner *date (misordered) {survexblock.survexfile.path} ({survexblock}) in '{line}'" message = f"- *team {expo.year} '{tm}' known foreigner *date (misordered) {survexblock.survexfile.path} ({survexblock}) in '{line}'"
@@ -908,7 +920,7 @@ class LoadingSurvex:
) )
oline = line oline = line
perps = get_people_on_trip(survexblock) # What, you don't know Judge Dredd slang ? perps = get_people_on_trip(survexblock) # perps used for diagnostic error messages only: they are to blame
if len(line) > 10: if len(line) > 10:
if line[10] == "-": # ie a range, just look at first date if line[10] == "-": # ie a range, just look at first date
@@ -1771,15 +1783,6 @@ class LoadingSurvex:
if edulcni: if edulcni:
self.ProcessEdulcniLine(edulcni) self.ProcessEdulcniLine(edulcni)
# def get_cave(self, path):
# """Read the file path to a survex file and guesses the cave
# """
# path_match = re.search(r"caves-(\d\d\d\d)/(\d+|\d\d\d\d-?\w+-\d+)/", path)
# if path_match:
# pos_cave = f"{path_match.group(1)}-{path_match.group(2)}"
# cave = getCaveByReference(pos_cave)
# return cave
# return None
def LinearLoad(self, survexblock, path, collatefilename): def LinearLoad(self, survexblock, path, collatefilename):
"""Loads a single survex file. Usually used to import all the survex files which have been collated """Loads a single survex file. Usually used to import all the survex files which have been collated