From 4b41ec15445978842553c60fd5545526d9647dda Mon Sep 17 00:00:00 2001
From: Claude <noreply@anthropic.com>
Date: Tue, 25 Aug 2026 23:24:34 +0000
Subject: [PATCH] Fix resume section detection on text with leading whitespace

_detect_sections() and _strip_markdown() anchored their regexes to the
exact start of a line, so section headers and markdown headers indented
with leading spaces/tabs (common in PDF-extracted text and in-code
fixtures) were never matched. Allow optional leading whitespace before
the anchored patterns.

Fixes #54
---
 ingestion/parsers/resume_parser.py | 10 +++++-----
 tests/unit/test_resume_parser.py   | 15 ---------------
 2 files changed, 5 insertions(+), 20 deletions(-)

diff --git a/ingestion/parsers/resume_parser.py b/ingestion/parsers/resume_parser.py
index bd64123..4e96695 100644
--- a/ingestion/parsers/resume_parser.py
+++ b/ingestion/parsers/resume_parser.py
@@ -98,7 +98,7 @@ def _parse_markdown(self, content: str) -> ParseResult:
     def _strip_markdown(self, content: str) -> str:
         """Remove markdown syntax from content."""
         # Remove markdown headers
-        text = re.sub(r"^#+\s+", "", content, flags=re.MULTILINE)
+        text = re.sub(r"^[ \t]*#+\s+", "", content, flags=re.MULTILINE)
 
         # Remove markdown links [text](url)
         text = re.sub(r"\[([^\]]+)\]\(([^\)]+)\)", r"\1", text)
@@ -131,10 +131,10 @@ def _detect_sections(self, text: str) -> list[str]:
         for section in SECTION_HEADERS:
             # Look for section header patterns
             patterns = [
-                rf"^{re.escape(section)}\s*$",
-                rf"^{re.escape(section)}\s*[:|-]",
-                rf"\n{re.escape(section)}\s*$",
-                rf"\n{re.escape(section)}\s*[:|-]",
+                rf"^[ \t]*{re.escape(section)}\s*$",
+                rf"^[ \t]*{re.escape(section)}\s*[:|-]",
+                rf"\n[ \t]*{re.escape(section)}\s*$",
+                rf"\n[ \t]*{re.escape(section)}\s*[:|-]",
             ]
 
             for pattern in patterns:
diff --git a/tests/unit/test_resume_parser.py b/tests/unit/test_resume_parser.py
index aae644f..372a2d7 100644
--- a/tests/unit/test_resume_parser.py
+++ b/tests/unit/test_resume_parser.py
@@ -17,9 +17,6 @@ def parser(self):
         """Create a ResumeParser instance."""
         return ResumeParser()
 
-    @pytest.mark.xfail(
-        strict=True, reason="issue #54: resume section detection fails on leading whitespace"
-    )
     def test_parse_single_column_resume_text(self, parser, sample_resume_text):
         """Test parsing a standard single-column resume text."""
         result = parser.parse(sample_resume_text)
@@ -36,9 +33,6 @@ def test_parse_single_column_resume_text(self, parser, sample_resume_text):
             "skills" in s for s in detected_lower
         )
 
-    @pytest.mark.xfail(
-        strict=True, reason="issue #54: resume section detection fails on leading whitespace"
-    )
     def test_parse_resume_no_work_experience(self, parser):
         """Test parsing a resume with no work experience section - handles gracefully."""
         resume_no_work = """
@@ -87,9 +81,6 @@ def test_parse_multipage_pdf(self, parser):
             assert "Page 2 Content" in result.text
             assert "Page 3 Content" in result.text
 
-    @pytest.mark.xfail(
-        strict=True, reason="issue #54: resume section detection fails on leading whitespace"
-    )
     def test_parse_markdown_resume(self, parser):
         """Test parsing a Markdown resume."""
         markdown_resume = """
@@ -132,9 +123,6 @@ def test_parse_invalid_list_content(self, parser):
             exc_info.value
         )
 
-    @pytest.mark.xfail(
-        strict=True, reason="issue #54: resume section detection fails on leading whitespace"
-    )
     def test_detect_sections(self, parser):
         """Test section detection in resume text."""
         text = """
@@ -155,9 +143,6 @@ def test_detect_sections(self, parser):
         assert any("education" in s for s in sections_lower)
         assert any("skills" in s for s in sections_lower)
 
-    @pytest.mark.xfail(
-        strict=True, reason="issue #54: resume section detection fails on leading whitespace"
-    )
     def test_strip_markdown_syntax(self, parser):
         """Test markdown syntax stripping."""
         markdown_text = """