Fix text replacement until end of paragraph, improve tests (#9)

Author: reindervdw-cmi

pyproject.toml

@@ -1,6 +1,6 @@
 [tool.poetry]
 name = "cmi_docx"
-version = "0.1.3"
+version = "0.1.4"
 description = ".docx utilities"
 authors = ["Reinder Vos de Wael <reinder.vosdewael@childmind.org>"]
 license = "LGPL-2.1"

src/cmi_docx/document.py

@@ -36,14 +36,13 @@ def find_in_runs(self, needle: str) -> list[run.FindRun]:
             needle: The text to find.
 
         Returns:
-            The indices of the text in the document.
+            The locations of the text in the runs.
         """
-        run_finds: list[run.FindRun] = []
-        for document_paragraph in self.all_paragraphs:
-            run_finds.extend(
-                paragraph.ExtendParagraph(document_paragraph).find_in_runs(needle)
-            )
-        return run_finds
+        return [
+            finder
+            for para in self.all_paragraphs
+            for finder in paragraph.ExtendParagraph(para).find_in_runs(needle)
+        ]
 
     def replace(self, needle: str, replace: str) -> None:
         """Finds and replaces text in a Word document.

src/cmi_docx/paragraph.py

@@ -64,14 +64,18 @@ def find_in_runs(self, needle: str) -> list[run.FindRun]:
         Returns:
             The indices of the text in the paragraph.
         """
+        if len(needle) == 0:
+            return []
+
         run_finds: list[run.FindRun] = []
         run_lengths = [len(run.text) for run in self.paragraph.runs]
         cumulative_run_lengths = list(itertools.accumulate(run_lengths))
         for occurence in self.find_in_paragraph(needle).character_indices:
             start_run = bisect.bisect_right(cumulative_run_lengths, occurence[0])
             end_run = bisect.bisect_right(
-                cumulative_run_lengths, occurence[1], lo=start_run
+                cumulative_run_lengths[:-1], occurence[1], lo=start_run
             )
+
             start_index = (
                 occurence[0] - cumulative_run_lengths[start_run - 1]
                 if start_run > 0

tests/test_document.py

@@ -1,6 +1,7 @@
 """Tests for the document module."""
 
 import docx
+import pytest
 
 from cmi_docx import document
 
@@ -35,28 +36,58 @@ def test_find_in_runs() -> None:
     assert actual[2].character_indices == (14, 19)
 
 
-def test_replace() -> None:
+@pytest.mark.parametrize(
+    ("runs", "needle", "replace", "expected"),
+    [
+        (
+            ["Hello, world!"],
+            "Hello",
+            "Goodbye",
+            "Goodbye, world!",
+        ),
+        (
+            ["Hello, world!", " Hello!"],
+            "Hello",
+            "Goodbye",
+            "Goodbye, world! Goodbye!",
+        ),
+        (
+            ["Hello {{", "FULL_NAME}}"],
+            "{{FULL_NAME}}",
+            'Shizuka "Lea" Sakai',
+            'Hello Shizuka "Lea" Sakai',
+        ),
+        (
+            ["This is James", " Bond ", "007!"],
+            "James Bond 007",
+            "Patrick",
+            "This is Patrick!",
+        ),
+        (
+            ["This is Alec", " Travelyan ", "006!"],
+            "This is Alec Travelyan 006!",
+            "",
+            "",
+        ),
+        (
+            ["This", " is ", "Patrick!"],
+            "",
+            "Nonsense",
+            "This is Patrick!",
+        ),
+    ],
+)
+def test_replace(runs: list[str], needle: str, replace: str, expected: str) -> None:
     """Test replacing text in a document."""
     doc = docx.Document()
-    doc.add_paragraph("Hello, world!")
-    extend_document = document.ExtendDocument(doc)
-
-    extend_document.replace("Hello", "Goodbye")
-
-    assert doc.paragraphs[0].text == "Goodbye, world!"
-
-
-def test_replace_across_runs() -> None:
-    """Test replacing text across runs in a document."""
-    doc = docx.Document()
-    paragraph = doc.add_paragraph("Hello, world!")
-    paragraph.add_run(" Maintain, World!")
-    paragraph.add_run(" Goodbye, World!")
+    paragraph = doc.add_paragraph(runs[0])
+    for run in runs[1:]:
+        paragraph.add_run(run)
     extend_document = document.ExtendDocument(doc)
 
-    extend_document.replace("world! Maintain, World! Goodbye", "Goodbye")
+    extend_document.replace(needle, replace)
 
-    assert doc.paragraphs[0].text == "Hello, Goodbye, World!"
+    assert doc.paragraphs[0].text == expected
 
 
 def test_insert_paragraph_by_object() -> None: