From 4b827f07932c9c18df08e75a2615b01f0ab0bc74 Mon Sep 17 00:00:00 2001
From: Ahmet Melek <39141206+ahmetmeleq@users.noreply.github.com>
Date: Wed, 5 Jul 2023 22:37:40 +0100
Subject: [PATCH] fix: local connector output filename when a single file is
 being processed (#879)

* fix string processing error for _output_filename

* Add docstring and type hint, update CHANGELOG, update version

* update test fixture

* simple code change commit to retrigger ci checks

* update test fixture - after brew install tesseract-lang

* Update ingest test fixtures (#882)

Co-authored-by: ahmetmeleq <ahmetmeleq@users.noreply.github.com>

* correct CHANGELOG

* correct CHANGELOG

---------

Co-authored-by: Unstructured-DevOps <111007769+Unstructured-DevOps@users.noreply.github.com>
Co-authored-by: ahmetmeleq <ahmetmeleq@users.noreply.github.com>
---
 .pre-commit-config.yaml                           |  1 +
 CHANGELOG.md                                      |  3 ++-
 .../english-and-korean.png.json                   |  0
 unstructured/__version__.py                       |  2 +-
 unstructured/ingest/connector/local.py            | 15 +++++++++++----
 5 files changed, 15 insertions(+), 6 deletions(-)
 rename test_unstructured_ingest/expected-structured-output/local-single-file/{example-docs => }/english-and-korean.png.json (100%)

diff --git a/.pre-commit-config.yaml b/.pre-commit-config.yaml
index bf1e2fe3d..88da6a35c 100644
--- a/.pre-commit-config.yaml
+++ b/.pre-commit-config.yaml
@@ -8,6 +8,7 @@ repos:
       - id: check-json
       - id: check-xml
       - id: end-of-file-fixer
+        exclude: \.json$
         include: \.py$
       - id: trailing-whitespace
       - id: mixed-line-ending
diff --git a/CHANGELOG.md b/CHANGELOG.md
index 1d569477b..2331647ac 100644
--- a/CHANGELOG.md
+++ b/CHANGELOG.md
@@ -1,4 +1,4 @@
-## 0.8.0-dev0
+## 0.8.0-dev1
 
 ### Enhancements
 
@@ -8,6 +8,7 @@
 
 ### Fixes
 * Fix KeyError when `isd_to_elements` doesn't find a type
+* Fix _output_filename for local connector, allowing single files to be written correctly to the disk
 
 ### BREAKING CHANGES
 
diff --git a/test_unstructured_ingest/expected-structured-output/local-single-file/example-docs/english-and-korean.png.json b/test_unstructured_ingest/expected-structured-output/local-single-file/english-and-korean.png.json
similarity index 100%
rename from test_unstructured_ingest/expected-structured-output/local-single-file/example-docs/english-and-korean.png.json
rename to test_unstructured_ingest/expected-structured-output/local-single-file/english-and-korean.png.json
diff --git a/unstructured/__version__.py b/unstructured/__version__.py
index b41f41048..5cee9d168 100644
--- a/unstructured/__version__.py
+++ b/unstructured/__version__.py
@@ -1 +1 @@
-__version__ = "0.8.0-dev0"  # pragma: no cover
+__version__ = "0.8.0-dev1"  # pragma: no cover
diff --git a/unstructured/ingest/connector/local.py b/unstructured/ingest/connector/local.py
index b2d5c8000..6d4901e1e 100644
--- a/unstructured/ingest/connector/local.py
+++ b/unstructured/ingest/connector/local.py
@@ -51,11 +51,18 @@ class LocalIngestDoc(BaseIngestDoc):
         pass
 
     @property
-    def _output_filename(self):
-        return (
-            Path(self.standard_config.output_dir)
-            / f"{self.path.replace(f'{self.config.input_path}/', '')}.json"
+    def _output_filename(self) -> Path:
+        """Returns output filename for the doc
+        If input path argument is a file itself, it returns the filename of the doc.
+        If input path argument is a folder, it returns the relative path of the doc.
+        """
+        input_path = Path(self.config.input_path)
+        basename = (
+            f"{Path(self.path).name}.json"
+            if input_path.is_file()
+            else f"{Path(self.path).relative_to(input_path)}.json"
         )
+        return Path(self.standard_config.output_dir) / basename
 
 
 class LocalConnector(BaseConnector):