fix rust project

binh-vu · Sep 10, 2023 · 6c92848 · 6c92848
1 parent 8e44dd5
commit 6c92848
Show file tree

Hide file tree

Showing 5 changed files with 65 additions and 28 deletions.
diff --git a/Cargo.lock b/Cargo.lock
diff --git a/Cargo.toml b/Cargo.toml
@@ -1,6 +1,6 @@
 [package]
 name = "kgdata"
-version = "3.2.1"
+version = "3.2.2"
 edition = "2021"
 readme = "README.md"
 homepage = "https://github.com/binh-vu/kgdata"
@@ -24,11 +24,11 @@ flate2 = { version = "1.0.24", features = [
     "zlib-ng",
 ], default-features = false }
 glob = "0.3.1"
-hashbrown = { version = "0.13.2", features = ["serde", "rayon"] }
+hashbrown = { version = "0.14.0", features = ["serde", "rayon"] }
 log = "0.4.17"
 ord_subset = "3.1.1"
 petgraph = "0.6.3"
-pyo3 = { version = "0.19.1", features = [
+pyo3 = { version = "0.19.2", features = [
     "anyhow",
     "multiple-pymethods",
     "hashbrown",

diff --git a/pyproject.toml b/pyproject.toml
@@ -1,6 +1,6 @@
 [project]
 name = "kgdata"
-version = "5.0.0a8"
+version = "5.0.0a9"
 description = "Library to process dumps of knowledge graphs (Wikipedia, DBpedia, Wikidata)"
 readme = "README.md"
 authors = [{ name = "Binh Vu", email = "binh@toan2.com" }]

diff --git a/scripts/build.sh b/scripts/build.sh
@@ -70,8 +70,9 @@ wikidata_dataset entity_metadata
 wikidata_dataset entity_all_types
 wikidata_dataset entity_degrees
 wikidata_dataset entity_labels
-wikidata_dataset entity_pagerank
 wikidata_dataset entity_types_and_degrees
+wikidata_dataset entity_outlinks
+wikidata_dataset entity_pagerank
 
 # ======================================================================
 # WIKIPEDIA Datasets
@@ -94,6 +95,8 @@ wikidata_db classes
 wikidata_db properties
 wikidata_db entities
 wikidata_db entity_labels
+wikidata_db entity_metadata
+wikidata_db entity_outlinks
 wikidata_db entity_redirections
 wikidata_db wp2wd
 wikidata_db entity_pagerank
diff --git a/scripts/download-data.ipynb b/scripts/download-data.ipynb
@@ -48,7 +48,7 @@
     {
      "data": {
       "application/vnd.jupyter.widget-view+json": {
-       "model_id": "3964f81555994ad5ae2c05e8cdfb6695",
+       "model_id": "3f953b8eee6746389eab7561355e3422",
        "version_major": 2,
        "version_minor": 0
       },
@@ -82,12 +82,12 @@
     {
      "data": {
       "application/vnd.jupyter.widget-view+json": {
-       "model_id": "0ff6421164ee4b888e0c48a7838ab6fc",
+       "model_id": "77e8475eac214e21a9f850ac58249d9b",
        "version_major": 2,
        "version_minor": 0
       },
       "text/plain": [
-       "HBox(children=(Button(description='2023-06-19', style=ButtonStyle()), Button(description='2023-07-03', style=B…"
+       "HBox(children=(Button(description='2023-08-21', style=ButtonStyle()), Button(description='2023-07-31', style=B…"
       ]
      },
      "metadata": {},
@@ -96,7 +96,7 @@
     {
      "data": {
       "application/vnd.jupyter.widget-view+json": {
-       "model_id": "cd99b0b3d8bd4928b06043f856b1c49b",
+       "model_id": "e0d7e19bc0b849be8146cec96885ef1f",
        "version_major": 2,
        "version_minor": 0
       },
@@ -208,6 +208,18 @@
     "wd_files"
    ]
   },
+  {
+   "cell_type": "code",
+   "execution_count": 9,
+   "id": "cb09fe48-4c4f-4b32-ae59-04a18b9ae0c4",
+   "metadata": {
+    "tags": []
+   },
+   "outputs": [],
+   "source": [
+    "wdjobs = []"
+   ]
+  },
   {
    "cell_type": "code",
    "execution_count": 8,
@@ -228,7 +240,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 9,
+   "execution_count": 10,
    "id": "7cd579aa",
    "metadata": {},
    "outputs": [
@@ -238,7 +250,7 @@
        "[DumpFile(date=datetime.date(2023, 6, 20), url='https://dumps.wikimedia.org/other/enterprise_html/runs/20230620/enwiki-NS0-20230620-ENTERPRISE-HTML.json.tar.gz')]"
       ]
      },
-     "execution_count": 9,
+     "execution_count": 10,
      "metadata": {},
      "output_type": "execute_result"
     }
@@ -250,7 +262,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 10,
+   "execution_count": 11,
    "id": "63c65e42",
    "metadata": {},
    "outputs": [],
@@ -271,7 +283,22 @@
    "execution_count": null,
    "id": "39cdeb3b",
    "metadata": {},
-   "outputs": [],
+   "outputs": [
+    {
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "17d1fee843d6488ab7f3ed75b7f6e070",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "Download enwiki-NS0-20230620-ENTERPRISE-HTML.json.tar.gz: 0.00B [00:00, ?B/s]"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    }
+   ],
    "source": [
     "jobs = wdjobs + wpjobs\n",
     "with WGet.start() as wget:\n",