From 55608d5c578a1803ff95a9493addf3e4e1554619 Mon Sep 17 00:00:00 2001
From: peterrrock2 <27579114+peterrrock2@users.noreply.github.com>
Date: Thu, 19 Sep 2024 22:04:37 -0600
Subject: [PATCH] Make some path characters illegal and update TODO

---
 .gitignore              |  4 ++++
 TODO.md                 | 10 ++++++++++
 gerrydb/exceptions.py   |  3 +++
 gerrydb/repos/base.py   | 22 +++++++++++++++++++++-
 gerrydb/repos/column.py | 11 ++++++++---
 gerrydb/schemas.py      |  5 ++++-
 6 files changed, 50 insertions(+), 5 deletions(-)
 create mode 100644 TODO.md

diff --git a/.gitignore b/.gitignore
index dbce267..32964df 100644
--- a/.gitignore
+++ b/.gitignore
@@ -160,3 +160,7 @@ cython_debug/
 #.idea/
 
 .DS_Store
+
+
+# Ignore all of the backup files that might be floating around
+*.tar
\ No newline at end of file
diff --git a/TODO.md b/TODO.md
new file mode 100644
index 0000000..89f57b3
--- /dev/null
+++ b/TODO.md
@@ -0,0 +1,10 @@
+# For Docs
+
+[ ] Show how to make a column with the write context
+    [ ] Show how this all works with the the load_dataframe method
+
+
+# Ideas
+[ ] Maybe cache all of the columns in the database on the local machine so
+    validation can be done locally? Periodically checks for columns can be done
+    to ensure that the local cache is up to date.
\ No newline at end of file
diff --git a/gerrydb/exceptions.py b/gerrydb/exceptions.py
index 2f7cbbd..b973331 100644
--- a/gerrydb/exceptions.py
+++ b/gerrydb/exceptions.py
@@ -39,3 +39,6 @@ class CacheInitError(CacheError):
 
 class ViewLoadError(GerryDBError):
     """Raised when a view cannot be loaded (e.g. from a GeoPackage)."""
+    
+class GerryPathError(GerryDBError):
+    """Raised when an invalid path is provided. Generally, this means invalid characters are present"""
\ No newline at end of file
diff --git a/gerrydb/repos/base.py b/gerrydb/repos/base.py
index 7e0828a..0e4b285 100644
--- a/gerrydb/repos/base.py
+++ b/gerrydb/repos/base.py
@@ -7,7 +7,13 @@
 import httpx
 import pydantic
 
-from gerrydb.exceptions import OnlineError, RequestError, ResultError, WriteContextError
+from gerrydb.exceptions import (
+    OnlineError,
+    RequestError,
+    ResultError,
+    WriteContextError,
+    GerryPathError,
+)
 from gerrydb.schemas import BaseModel
 
 if TYPE_CHECKING:
@@ -104,6 +110,14 @@ def write_context_wrapper(*args, **kwargs):
     return write_context_wrapper
 
 
+INVALID_PATH_SUBSTRINGS = set(
+    {
+        "..",
+        " ",
+    }
+)
+
+
 def normalize_path(path: str, case_sensitive_uid: bool = False) -> str:
     """Normalizes a path (removes leading, trailing, and duplicate slashes, and
     lowercases the path if `case_sensitive` is `False`).
@@ -111,6 +125,12 @@ def normalize_path(path: str, case_sensitive_uid: bool = False) -> str:
     Some paths, such as paths containing GEOIDs, are case-sensitive in the last
     segment. In these cases, `case_sensitive` should be set to `True`.
     """
+    for item in INVALID_PATH_SUBSTRINGS:
+        if item in path:
+            raise GerryPathError(
+                f"Invalid path: '{path}'. Please remove the following substring: '{item}'"
+            )
+
     if case_sensitive_uid:
         path_list = path.strip().split("/")
         return "/".join(
diff --git a/gerrydb/repos/column.py b/gerrydb/repos/column.py
index 363a90c..48e40c7 100644
--- a/gerrydb/repos/column.py
+++ b/gerrydb/repos/column.py
@@ -11,6 +11,7 @@
     namespaced,
     online,
     write_context,
+    normalize_path,
 )
 from gerrydb.schemas import (
     Column,
@@ -68,6 +69,7 @@ def create(
         Returns:
             Metadata for the new column.
         """
+        path = normalize_path(path)
         response = self.ctx.client.post(
             f"{self.base_url}/{namespace}",
             json=ColumnCreate(
@@ -106,8 +108,9 @@ def update(
         Returns:
             The updated column.
         """
+        clean_path = normalize_path(f"{self.base_url}/{namespace}/{path}")
         response = self.ctx.client.patch(
-            f"{self.base_url}/{namespace}/{path}",
+            clean_path,
             json=ColumnPatch(aliases=aliases).dict(),
         )
         response.raise_for_status()
@@ -164,9 +167,10 @@ def set_values(
             raise ValueError("Either `path` or `col` must be provided.")
 
         path = col.path if col is not None else path
+        clean_path = normalize_path(f"{self.base_url}/{namespace}/{path}")
 
         response = self.ctx.client.put(
-            f"{self.base_url}/{namespace}/{path}",
+            clean_path,
             json=[
                 ColumnValue(
                     path=(
@@ -219,6 +223,7 @@ async def async_set_values(
             raise ValueError("Either `path` or `col` must be provided.")
 
         path = col.path if col is not None else path
+        clean_path = normalize_path(f"{self.base_url}/{namespace}/{path}")
 
         ephemeral_client = client is None
         if ephemeral_client:
@@ -238,7 +243,7 @@ async def async_set_values(
             for geo, value in values.items()
         ]
         response = await client.put(
-            f"{self.base_url}/{namespace}/{path}",
+            clean_path,
             json=json,
         )
 
diff --git a/gerrydb/schemas.py b/gerrydb/schemas.py
index e463264..8467f5e 100644
--- a/gerrydb/schemas.py
+++ b/gerrydb/schemas.py
@@ -2,6 +2,7 @@
 
 This file should be kept in sync with the server-side version.
 """
+
 from datetime import datetime
 from enum import Enum
 from typing import Any, Optional, Union
@@ -16,7 +17,9 @@
 UserEmail = constr(max_length=254)
 
 # constr is a constrained string, so this is some path that needs to satisfy this regex
-GerryPath = constr(regex=r"[a-z0-9][a-z0-9-_/]*") # must start with lowercase or digit, then followed by any lowercase, digit, hyphen, underscore, slash
+GerryPath = constr(
+    regex=r"[a-z0-9][a-z0-9-_/]*"
+)  # must start with lowercase or digit, then followed by any lowercase, digit, hyphen, underscore, slash
 NamespacedGerryPath = constr(regex=r"[a-z0-9/][a-z0-9-_/]*")
 
 NATIVE_PROJ = pyproj.CRS("EPSG:4269")