add attachment doc path lookup

CHANGELOG.md

@@ -19,6 +19,9 @@ and this project adheres to [Semantic Versioning](http://semver.org/).
 - BetterBibtex citation key output with the `lookup` command line action.
 - Integration tests with mock Zotero database. Now all tests are self
   contained.
+- An API and command line to give attachment (paper document) paths based on an
+  item key.  This can be combined with the item ID to citation ID mapping if
+  starting with a BetterBibtex citation ID.
 
 
 ## [0.9.1] - 2024-01-02

resources/app.conf

@@ -1,7 +1,7 @@
 #@meta {desc: 'command line config', date: '2024-09-18'}
 
 [cli]
-apps = list: prog_cli, ${cli_config_default:apps}, app, capp, papp
+apps = list: prog_cli, ${cli_config_default:apps}, app, qapp, capp, papp
 app_removes = list: override_cli
 cleanups = list: prog_cli, ${cli_config_default:cleanups}, app, app_decorator, capp, papp
 doc = This project exports your local Zotero library to a usable HTML website.
@@ -60,14 +60,25 @@ option_overrides = dict: {
     'metavar': 'DIR'}}
 mnemonic_overrides = dict: {'print_structure': 'print'}
 
+[qapp]
+class_name = zensols.zotsite.QueryApplication
+resource = instance: zs_resource
+
+[qapp_decorator]
+option_excludes = set: resource
+option_overrides = dict: {
+  'key': {'short_name': 'k'}}
+mnemonic_overrides = dict: {
+  'find_path': 'docpath'}
+
 [capp]
 class_name = zensols.zotsite.CiteApplication
 resource = instance: zs_resource
 
 [capp_decorator]
 option_excludes = set: resource
+option_overrides = dict: {
+  'key': {'short_name': 'k'}}
 
 [papp]
 class_name = zensols.zotsite.PrototypeApplication
-export_app = instance: app
-cite_app = instance: capp

resources/zotero.sql

@@ -9,35 +9,45 @@ select c.collectionId c_id, ci.itemId c_iid,
     where c.libraryId = ? and
           c.collectionName like ?
 
-
 -- name=select_items_attachments
-select c.collectionId c_id, c.parentCollectionId c_pid,
-           c.collectionName c_name,
+select c.collectionId c_id, c.parentCollectionId c_pid, c.collectionName c_name,
        it.itemId i_id, ia.parentItemId i_pid, it.key, iy.typeName type,
        ia.contentType content_type, ia.path,
        itn.title n_title, itn.note n_note, itn.parentItemId n_pid
   from items it, itemTypes iy
-      left join itemAttachments ia on it.itemId = ia.itemId
-      left join collectionItems ci on ci.itemId = it.itemId
-      left join collections c on c.collectionId = ci.collectionId
-      left join itemNotes itn on it.itemId = itn.itemId
+       left join itemAttachments ia on it.itemId = ia.itemId
+       left join collectionItems ci on ci.itemId = it.itemId
+       left join collections c on c.collectionId = ci.collectionId
+       left join itemNotes itn on it.itemId = itn.itemId
   where it.itemTypeId = iy.itemTypeId and
-      it.itemId not in (select itemId from deletedItems)
+        it.itemId not in (select itemId from deletedItems)
   order by ci.orderIndex;
 
 -- name=select_item_metadata
 select f.fieldName name, iv.value
   from items i, itemTypes it, itemData id, itemDataValues iv, fields f
   where i.itemTypeId = it.itemTypeId and
-      i.itemId = id.itemId and
-      id.valueId = iv.valueId and
-      id.fieldId = f.fieldId and
-      i.itemId = ? and
-      i.itemId not in (select itemId from deletedItems);
+        i.itemId = id.itemId and
+        id.valueId = iv.valueId and
+        id.fieldId = f.fieldId and
+        i.itemId = ? and
+        i.itemId not in (select itemId from deletedItems);
 
 -- name=select_item_creators
 select c.firstName, c.lastName
   from itemCreators ic, creators c
   where ic.creatorID = c.creatorID and
-    ic.itemID = ?
-    order by ic.orderIndex;
+        ic.itemID = ?
+  order by ic.orderIndex;
+
+-- name=select_item_paths
+select it.key, ia.path
+  from items it, itemTypes iy
+      left join itemAttachments ia on it.itemId = ia.itemId
+      left join collectionItems ci on ci.itemId = it.itemId
+      left join collections c on c.collectionId = ci.collectionId
+      left join itemNotes itn on it.itemId = itn.itemId
+  where it.itemTypeId = iy.itemTypeId and
+        it.itemId not in (select itemId from deletedItems) and
+        path is not null
+  order by ci.orderIndex;

src/python/zensols/zotsite/app.py

@@ -3,7 +3,7 @@
 """
 __author__ = 'Paul Landes'
 
-from typing import Dict, Any
+from typing import Iterable, Dict, Any
 from dataclasses import dataclass, field
 import re
 import logging
@@ -74,6 +74,43 @@ def print_structure(self):
         self.site_creator.print_structure()
 
 
+@dataclass
+class QueryApplication(object):
+    """Query the Zotero database.
+
+    """
+    resource: Resource = field()
+    """Zotsite first class objects."""
+
+    def find_path(self, key: str = None):
+        """Output paths in the form ``<item key>=<path>``.
+
+        :param key: key in format ``<library ID>_<item key>``, standard input
+                    if not given, or ``all`` for every entry
+
+        """
+        def strip_lib_id(s: str) -> str:
+            """See TODO in :obj:`.ZoteroDatabase.paths`."""
+            return re.sub(r'^1_', '', s)
+
+        paths: Dict[str, str] = self.resource.zotero_db.paths
+        keys: Iterable[str]
+        if key is None:
+            keys = map(lambda s: s.strip(), sys.stdin.readlines())
+            keys = map(strip_lib_id)
+        elif key == 'all':
+            keys = paths.keys()
+        else:
+            keys = [strip_lib_id(key)]
+        if len(keys) == 1:
+            print(paths[keys[0]])
+        else:
+            key: str
+            for key in keys:
+                path: Path = paths[key]
+                print(f'{key}={path}')
+
+
 @dataclass
 class CiteApplication(object):
     """Map Zotero keys to BetterBibtex citekeys.
@@ -85,7 +122,7 @@ class CiteApplication(object):
     def lookup(self, format: str = '{itemKey}={citationKey}', key: str = None):
         """Look up a citation key and print out BetterBibtex field(s).
 
-        :param key: key in format ``<libraryID>_<citationKey>``, standard input
+        :param key: key in format ``<library ID>_<item key>``, standard input
                     if not given, or ``all`` for every entry
 
         :param format: the format of the output or ``json`` for all fields
@@ -114,9 +151,6 @@ class PrototypeApplication(object):
     CLI_META = {'is_usage_visible': False}
 
     config_factory: ConfigFactory = field()
-    export_app: ExportApplication = field()
-    cite_app: CiteApplication = field()
 
     def proto(self):
-        lib = self.export_app.resource.zotero_db.get_library()
-        print(lib)
+        self.config_factory('qapp').find_path(item_key='1_VBAMCMJX')

src/python/zensols/zotsite/db.py

@@ -6,9 +6,11 @@
 from typing import Tuple, List, Dict, Any, Union, Optional
 from dataclasses import dataclass, field
 import logging
+import re
 from pathlib import Path
 from zensols.db import DbPersister
-from . import Collection, Library, Item, Note, Name
+from . import Collection, Library, Item, Note, Name, ZoteroApplicationError
+
 
 logger = logging.getLogger(__name__)
 
@@ -188,3 +190,23 @@ def get_library(self) -> Library:
         finally:
             # deallocate pooled connection (configured in ``obj.conf``)
             self._persister.conn_manager.dispose_all()
+
+    @property
+    def paths(self) -> Dict[str, Path]:
+        """The paths of items by key.
+
+        """
+        def map_row(r: Tuple[Any, ...]) -> Tuple[str, Path]:
+            key: str = r['key']
+            path: Path = storage_dir / key / re.sub(pat, '', r['path'])
+            return key, path
+
+        pat: re.Pattern = re.compile(r'^storage:')
+        storage_dir: Path = self._data_dir / 'storage'
+        try:
+            # TODO: add library and collection_like to params and SQL
+            rows: Tuple[Dict[str, Any], ...] = self._select('item_paths')
+            return dict(map(map_row, rows))
+        except Exception as e:
+            raise ZoteroApplicationError(
+                f'Could not access Zotero database: {e}') from e