querysource 4.5.0__tar.gz → 4.5.1__tar.gz

{querysource-4.5.0 → querysource-4.5.1}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: querysource
-Version: 4.5.0
+Version: 4.5.1
 Summary: Aiohttp web service for querying several databases easily
 Author-email: Jesus Lara <jesuslarag@gmail.com>
 License: BSD-3-Clause

{querysource-4.5.0 → querysource-4.5.1}/generated/SharepointSource.json

@@ -4,6 +4,13 @@
   "description": "Download a single file from a SharePoint document library.",
   "usage": "Authenticates via Microsoft Graph client credentials (client_id, client_secret, tenant_id) and downloads the specified file from the given SharePoint site and directory path.",
   "attributes": [
+    {
+      "name": "url",
+      "type": "Any",
+      "default": "",
+      "required": false,
+      "description": ""
+    },
     {
       "name": "credentials.client_id",
       "type": "Any",
@@ -73,6 +80,13 @@
       "default": {},
       "required": false,
       "description": ""
+    },
+    {
+      "name": "source.url",
+      "type": "Any",
+      "default": "",
+      "required": false,
+      "description": ""
     }
   ],
   "json_schema": {
@@ -80,6 +94,7 @@
     "type": "object",
     "title": "SharepointSource",
     "properties": {
+      "url": {},
       "credentials": {
         "type": "object",
         "properties": {
@@ -97,7 +112,8 @@
           "filename": {},
           "directory": {},
           "sheet_name": {},
-          "pd_args": {}
+          "pd_args": {},
+          "url": {}
         }
       }
     },

{querysource-4.5.0 → querysource-4.5.1}/querysource/_version.py

@@ -18,7 +18,7 @@ version_tuple: tuple[int | str, ...]
 commit_id: str | None
 __commit_id__: str | None
 
-__version__ = version = '4.5.1.dev0+gfd354d0c9.d20260619'
-__version_tuple__ = version_tuple = (4, 5, 1, 'dev0', 'gfd354d0c9.d20260619')
+__version__ = version = '0.1.dev1+gf6d3cc545.d20260625'
+__version_tuple__ = version_tuple = (0, 1, 'dev1', 'gf6d3cc545.d20260625')
 
-__commit_id__ = commit_id = 'gfd354d0c9'
+__commit_id__ = commit_id = 'gf6d3cc545'

{querysource-4.5.0 → querysource-4.5.1}/querysource/datasources/handlers/datasource.py

@@ -14,6 +14,10 @@ from asyncdb.exceptions import (
     StatementError
 )
 from navigator.views import BaseView
+from navigator_auth.decorators import (
+    is_authenticated,
+    user_session
+)
 from ...conf import default_dsn
 from ...utils.functions import anonymize
 from ...utils.parseqs import ParseDict
@@ -173,6 +177,7 @@ def _item_get(item, key, default=None):
     return getattr(item, key, default)
 
 
+@user_session()
 class DatasourceView(BaseView):
     """API View for managing datasources.
     """

{querysource-4.5.0 → querysource-4.5.1}/querysource/queries/multi/sources/sharepoint.py

@@ -43,6 +43,25 @@ class SharepointSource(ThreadSource):
                 "directory": "Shared Documents/General/Schedule"
             }
         }
+
+    Alternatively, pass the full file ``url`` (copied straight from the
+    browser/Share dialog — sharing links and subsites included) and the file is
+    resolved directly via the Microsoft Graph ``/shares`` endpoint. Only the
+    auth credentials are required; ``tenant_name``/``site``/``directory`` are
+    not needed::
+
+        {
+            "credentials": {
+                "client_id": "SHAREPOINT_APP_ID",
+                "client_secret": "SHAREPOINT_APP_SECRET",
+                "tenant_id": "SHAREPOINT_TENANT_ID"
+            },
+            "source": {
+                "url": "https://<tenant>.sharepoint.com/:x:/r/sites/<site>/<subsite>/<path>/<file>.xlsx?..."
+            }
+        }
+
+    ``sheet_name`` and ``pd_args`` still apply when reading the downloaded file.
     """
 
     def __init__(
@@ -82,6 +101,27 @@ class SharepointSource(ThreadSource):
         self._sheet_name = source.get('sheet_name', 0)
         # pd_args: extra kwargs forwarded to pd.read_excel / pd.read_csv (e.g. skiprows, header, usecols).
         self._pd_args: dict = source.get('pd_args', {})
+        # url: full SharePoint file URL, including sharing links such as
+        # ".../:x:/r/sites/<site>/<subsite>/<path>/<file>.xlsx?d=...&web=1".
+        # When set, the file is resolved directly via the Microsoft Graph
+        # /shares endpoint (server-side), so tenant_name/site/directory do NOT
+        # need to be configured and subsites work transparently.
+        self._url: str = source.get('url', '') or options.get('url', '') or ''
+
+    @staticmethod
+    def _encode_share_url(url: str) -> str:
+        """Encode a SharePoint URL into a Graph ``shares`` id.
+
+        Per https://learn.microsoft.com/graph/api/shares-get the share id is
+        ``"u!"`` followed by the unpadded base64url encoding of the UTF-8 URL.
+        Works for canonical and sharing URLs alike — Graph resolves the share
+        itself, so no client-side path parsing (and no subsite ambiguity) is
+        involved.
+        """
+        import base64  # noqa: PLC0415
+
+        b64 = base64.urlsafe_b64encode(url.strip().encode('utf-8')).decode('ascii')
+        return 'u!' + b64.rstrip('=')
 
     def _parse_file_content(self, content: bytes) -> pd.DataFrame:
         """Parse raw bytes as Excel or CSV depending on the filename extension."""
@@ -115,6 +155,11 @@ class SharepointSource(ThreadSource):
                 **self._pd_args,
             )
         df = df.infer_objects()
+        # A spreadsheet can yield non-string column names (date/number header
+        # cells, or blank/merged headers -> NaN). Those break JSON output
+        # ("Dict key must be str") and DB writes, so coerce any non-str header.
+        if isinstance(df, pd.DataFrame):
+            df.columns = [c if isinstance(c, str) else str(c) for c in df.columns]
         return df
 
     async def fetch(self) -> pd.DataFrame:
@@ -141,12 +186,6 @@ class SharepointSource(ThreadSource):
                 "pip install httpx"
             ) from exc
 
-        if not self._tenant_name or self._tenant_name == 'SHAREPOINT_TENANT_NAME':
-            raise ValueError(
-                "SharePoint tenant_name must be configured (via credentials or "
-                "navconfig SHAREPOINT_TENANT_NAME)."
-            )
-
         # The msgraph/kiota SDK uses platform.version() to build the User-Agent
         # header. On Linux the string has a trailing space which httpx rejects
         # as an illegal header value. Patch it before the client is constructed.
@@ -163,71 +202,93 @@ class SharepointSource(ThreadSource):
             scopes = ["https://graph.microsoft.com/.default"]
             client = GraphServiceClient(credentials=credential, scopes=scopes)
 
-            # Resolve the site ID
-            site_host = f"{self._tenant_name}.sharepoint.com" if self._tenant_name else None
-            if site_host and self._site:
-                site = await client.sites.by_site_id(
-                    f"{site_host}:/sites/{self._site}:"
-                ).get()
-            else:
-                raise RuntimeError(
-                    "SharePoint tenant_name and site must be specified to locate the site."
+            if self._url:
+                # Resolve the file directly from its (sharing) URL via /shares.
+                # Graph handles subsites, document libraries, folders and
+                # sharing-link tokens server-side — nothing is parsed here.
+                shares_id = self._encode_share_url(self._url)
+                item = await (
+                    client.shares.by_shared_drive_item_id(shares_id)
+                    .drive_item.get()
                 )
+                if item is None or not getattr(item, 'id', None):
+                    raise RuntimeError(
+                        f"Could not resolve SharePoint file from url: {self._url}"
+                    )
+                # Use the resolved item name so extension detection works.
+                if not self._filename:
+                    self._filename = getattr(item, 'name', '') or 'download'
+            else:
+                if not self._tenant_name or self._tenant_name == 'SHAREPOINT_TENANT_NAME':
+                    raise ValueError(
+                        "SharePoint tenant_name must be configured (via credentials "
+                        "or navconfig SHAREPOINT_TENANT_NAME), or pass a full 'url'."
+                    )
+                # Resolve the site ID
+                site_host = f"{self._tenant_name}.sharepoint.com" if self._tenant_name else None
+                if site_host and self._site:
+                    site = await client.sites.by_site_id(
+                        f"{site_host}:/sites/{self._site}:"
+                    ).get()
+                else:
+                    raise RuntimeError(
+                        "SharePoint tenant_name and site must be specified to locate the site."
+                    )
 
-            site_id = site.id
+                site_id = site.id
 
-            # Get drives (document libraries) for the site
-            drives_response = await client.sites.by_site_id(site_id).drives.get()
-            drives = drives_response.value if drives_response else []
+                # Get drives (document libraries) for the site
+                drives_response = await client.sites.by_site_id(site_id).drives.get()
+                drives = drives_response.value if drives_response else []
 
-            # Parse directory into (library_name, subfolder).
-            # Single segment (e.g. "Tests") → Documents library + subfolder "Tests".
-            # Multi-segment (e.g. "Documents/Tests") → first segment is library,
-            # rest is subfolder.  "Shared Documents" normalised to "Documents".
-            _dir = (self._directory or '').replace('\\', '/').strip().strip('/')
-            if not _dir:
-                library_name, subfolder = 'Documents', ''
-            else:
-                _parts = _dir.split('/')
-                if len(_parts) == 1:
-                    library_name, subfolder = 'Documents', _parts[0]
+                # Parse directory into (library_name, subfolder).
+                # Single segment (e.g. "Tests") → Documents library + subfolder "Tests".
+                # Multi-segment (e.g. "Documents/Tests") → first segment is library,
+                # rest is subfolder.  "Shared Documents" normalised to "Documents".
+                _dir = (self._directory or '').replace('\\', '/').strip().strip('/')
+                if not _dir:
+                    library_name, subfolder = 'Documents', ''
                 else:
-                    library_name = _parts[0]
-                    subfolder = '/'.join(_parts[1:])
-                    if library_name.lower() == 'shared documents':
-                        library_name = 'Documents'
-
-            # Find the matching drive (case-insensitive)
-            drive = None
-            for d in drives:
-                if d.name and d.name.lower() == library_name.lower():
-                    drive = d
-                    break
-            if drive is None and drives:
-                drive = drives[0]  # fallback to first drive
-
-            if drive is None:
-                raise RuntimeError(
-                    f"No document library found for site '{self._site}' "
-                    f"matching '{library_name}'."
-                )
+                    _parts = _dir.split('/')
+                    if len(_parts) == 1:
+                        library_name, subfolder = 'Documents', _parts[0]
+                    else:
+                        library_name = _parts[0]
+                        subfolder = '/'.join(_parts[1:])
+                        if library_name.lower() == 'shared documents':
+                            library_name = 'Documents'
 
-            drive_id = drive.id
+                # Find the matching drive (case-insensitive)
+                drive = None
+                for d in drives:
+                    if d.name and d.name.lower() == library_name.lower():
+                        drive = d
+                        break
+                if drive is None and drives:
+                    drive = drives[0]  # fallback to first drive
 
-            # Navigate to the subfolder and find the file
-            file_path = f"{subfolder.rstrip('/')}/{self._filename}" if subfolder else self._filename
-            item = await (
-                client.drives.by_drive_id(drive_id)
-                .items.by_drive_item_id(f"root:/{file_path}:")
-                .get()
-            )
+                if drive is None:
+                    raise RuntimeError(
+                        f"No document library found for site '{self._site}' "
+                        f"matching '{library_name}'."
+                    )
 
-            if item is None or not hasattr(item, 'id'):
-                raise RuntimeError(
-                    f"File '{self._filename}' not found in SharePoint directory "
-                    f"'{self._directory}'."
+                drive_id = drive.id
+
+                # Navigate to the subfolder and find the file
+                file_path = f"{subfolder.rstrip('/')}/{self._filename}" if subfolder else self._filename
+                item = await (
+                    client.drives.by_drive_id(drive_id)
+                    .items.by_drive_item_id(f"root:/{file_path}:")
+                    .get()
                 )
 
+                if item is None or not hasattr(item, 'id'):
+                    raise RuntimeError(
+                        f"File '{self._filename}' not found in SharePoint directory "
+                        f"'{self._directory}'."
+                    )
+
             # Get the download URL from the item's additional data
             download_url = (
                 item.additional_data.get('@microsoft.graph.downloadUrl')

{querysource-4.5.0 → querysource-4.5.1}/querysource/version.py

@@ -6,7 +6,7 @@ __description__ = (
     'QuerySource Query parser and generator.'
     'Supports SQL, NoSQL, MongoDB, Cassandra, Elasticsearch, etc.'
 )
-__version__ = '4.5.0'
+__version__ = '4.5.1'
 __url__ = 'https://github.com/phenobarbital/querysource'
 __copyright__ = 'Copyright (c) 2018-2025 Jesus Lara'
 __author__ = 'Jesus Lara'

{querysource-4.5.0 → querysource-4.5.1}/querysource.egg-info/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: querysource
-Version: 4.5.0
+Version: 4.5.1
 Summary: Aiohttp web service for querying several databases easily
 Author-email: Jesus Lara <jesuslarag@gmail.com>
 License: BSD-3-Clause

{querysource-4.5.0 → querysource-4.5.1}/querysource.egg-info/SOURCES.txt

@@ -125,6 +125,8 @@ querysource.egg-info/dependency_links.txt
 querysource.egg-info/entry_points.txt
 querysource.egg-info/not-zip-safe
 querysource.egg-info/requires.txt
+querysource.egg-info/scm_file_list.json
+querysource.egg-info/scm_version.json
 querysource.egg-info/top_level.txt
 querysource/auth/__init__.py
 querysource/auth/_resource_types.py