feat: Migrate over fsspec connectors <- Ingest test fixtures update (#…

…3148) This pull request includes updated ingest test fixtures. Please review and merge if appropriate. Co-authored-by: rbiseck3 <rbiseck3@users.noreply.github.com>
Unstructured-IO · Jun 4, 2024 · b506d68 · b506d68
1 parent 61a9a49
commit b506d68
Show file tree

Hide file tree

Showing 3 changed files with 32 additions and 32 deletions.
diff --git a/test_unstructured_ingest/expected-structured-output/gcs/ideas-page.html.json b/test_unstructured_ingest/expected-structured-output/gcs/ideas-page.html.json
@@ -10,14 +10,14 @@
       ],
       "filetype": "text/html",
       "data_source": {
-        "url": "gs://utic-test-ingest-fixtures/nested-1/nested/ideas-page.html",
-        "version": "310890354306462681752199911957569001015",
+        "url": "gs://utic-test-ingest-fixtures/ideas-page.html",
+        "version": "199523943725186047835150971481714294476",
         "record_locator": {
           "protocol": "gs",
           "remote_file_path": "gs://utic-test-ingest-fixtures/"
         },
-        "date_created": "1687304893.75",
-        "date_modified": "1687304893.75"
+        "date_created": "1687304971.038",
+        "date_modified": "1687304971.038"
       }
     }
   }

diff --git a/test_unstructured_ingest/expected-structured-output/gcs/nested-1/fake-text.txt.json b/test_unstructured_ingest/expected-structured-output/gcs/nested-1/fake-text.txt.json
@@ -9,14 +9,14 @@
       ],
       "filetype": "text/plain",
       "data_source": {
-        "url": "gs://utic-test-ingest-fixtures/nested-2/fake-text.txt",
-        "version": "198731266903969902154134165613731741332",
+        "url": "gs://utic-test-ingest-fixtures/nested-1/fake-text.txt",
+        "version": "180263070579038859328651626981788275889",
         "record_locator": {
           "protocol": "gs",
           "remote_file_path": "gs://utic-test-ingest-fixtures/"
         },
-        "date_created": "1687304904.192",
-        "date_modified": "1687304904.192"
+        "date_created": "1687304893.303",
+        "date_modified": "1687304893.303"
       }
     }
   },
@@ -30,14 +30,14 @@
       ],
       "filetype": "text/plain",
       "data_source": {
-        "url": "gs://utic-test-ingest-fixtures/nested-2/fake-text.txt",
-        "version": "198731266903969902154134165613731741332",
+        "url": "gs://utic-test-ingest-fixtures/nested-1/fake-text.txt",
+        "version": "180263070579038859328651626981788275889",
         "record_locator": {
           "protocol": "gs",
           "remote_file_path": "gs://utic-test-ingest-fixtures/"
         },
-        "date_created": "1687304904.192",
-        "date_modified": "1687304904.192"
+        "date_created": "1687304893.303",
+        "date_modified": "1687304893.303"
       }
     }
   },
@@ -51,14 +51,14 @@
       ],
       "filetype": "text/plain",
       "data_source": {
-        "url": "gs://utic-test-ingest-fixtures/nested-2/fake-text.txt",
-        "version": "198731266903969902154134165613731741332",
+        "url": "gs://utic-test-ingest-fixtures/nested-1/fake-text.txt",
+        "version": "180263070579038859328651626981788275889",
         "record_locator": {
           "protocol": "gs",
           "remote_file_path": "gs://utic-test-ingest-fixtures/"
         },
-        "date_created": "1687304904.192",
-        "date_modified": "1687304904.192"
+        "date_created": "1687304893.303",
+        "date_modified": "1687304893.303"
       }
     }
   },
@@ -72,14 +72,14 @@
       ],
       "filetype": "text/plain",
       "data_source": {
-        "url": "gs://utic-test-ingest-fixtures/nested-2/fake-text.txt",
-        "version": "198731266903969902154134165613731741332",
+        "url": "gs://utic-test-ingest-fixtures/nested-1/fake-text.txt",
+        "version": "180263070579038859328651626981788275889",
         "record_locator": {
           "protocol": "gs",
           "remote_file_path": "gs://utic-test-ingest-fixtures/"
         },
-        "date_created": "1687304904.192",
-        "date_modified": "1687304904.192"
+        "date_created": "1687304893.303",
+        "date_modified": "1687304893.303"
       }
     }
   },
@@ -93,14 +93,14 @@
       ],
       "filetype": "text/plain",
       "data_source": {
-        "url": "gs://utic-test-ingest-fixtures/nested-2/fake-text.txt",
-        "version": "198731266903969902154134165613731741332",
+        "url": "gs://utic-test-ingest-fixtures/nested-1/fake-text.txt",
+        "version": "180263070579038859328651626981788275889",
         "record_locator": {
           "protocol": "gs",
           "remote_file_path": "gs://utic-test-ingest-fixtures/"
         },
-        "date_created": "1687304904.192",
-        "date_modified": "1687304904.192"
+        "date_created": "1687304893.303",
+        "date_modified": "1687304893.303"
       }
     }
   },
@@ -114,14 +114,14 @@
       ],
       "filetype": "text/plain",
       "data_source": {
-        "url": "gs://utic-test-ingest-fixtures/nested-2/fake-text.txt",
-        "version": "198731266903969902154134165613731741332",
+        "url": "gs://utic-test-ingest-fixtures/nested-1/fake-text.txt",
+        "version": "180263070579038859328651626981788275889",
         "record_locator": {
           "protocol": "gs",
           "remote_file_path": "gs://utic-test-ingest-fixtures/"
         },
-        "date_created": "1687304904.192",
-        "date_modified": "1687304904.192"
+        "date_created": "1687304893.303",
+        "date_modified": "1687304893.303"
       }
     }
   }

diff --git a/test_unstructured_ingest/expected-structured-output/gcs/nested-2/nested/ideas-page.html.json b/test_unstructured_ingest/expected-structured-output/gcs/nested-2/nested/ideas-page.html.json
@@ -10,14 +10,14 @@
       ],
       "filetype": "text/html",
       "data_source": {
-        "url": "gs://utic-test-ingest-fixtures/nested-1/nested/ideas-page.html",
-        "version": "310890354306462681752199911957569001015",
+        "url": "gs://utic-test-ingest-fixtures/nested-2/nested/ideas-page.html",
+        "version": "113813498010717860141768546590661839404",
         "record_locator": {
           "protocol": "gs",
           "remote_file_path": "gs://utic-test-ingest-fixtures/"
         },
-        "date_created": "1687304893.75",
-        "date_modified": "1687304893.75"
+        "date_created": "1687304904.586",
+        "date_modified": "1687304904.586"
       }
     }
   }