support c and c++

README.md

@@ -167,7 +167,7 @@ You can experiment with your own sentence transformer models with the `--model`
 ## Bugs and limitations
 
 - Currently, the `.embeddings` index is not updated when repository files change. As a temporary workaround, `sem embed` can be re-ran occasionally.
-- Supported languages: `{ 'python', 'javascript', 'typescript', 'ruby', 'go', 'rust', 'java' }`
+- Supported languages: `{ 'python', 'javascript', 'typescript', 'ruby', 'go', 'rust', 'java', 'c', 'c++' }`
 - Supported text editors for opening results in: `{ 'vscode', 'vim' }`
 
 ## License

setup.py

@@ -5,7 +5,7 @@
 
 setuptools.setup(
     name='semantic-code-search',
-    version='0.1.3',
+    version='0.2.0',
     author='Kiril Videlov',
     author_email='kiril@codeball.ai',
     description='Search your codebase with natural language.',

src/semantic_code_search/embed.py

@@ -7,6 +7,7 @@
 import numpy as np
 from tree_sitter import Tree
 from tree_sitter_languages import get_parser
+from tqdm import tqdm
 
 
 def _supported_file_extensions():
@@ -17,7 +18,11 @@ def _supported_file_extensions():
         '.java': 'java',
         '.js': 'javascript',
         '.ts': 'typescript',
-        '.py': 'python'
+        '.py': 'python',
+        '.c': 'c',
+        '.h': 'c',
+        '.cpp': 'cpp',
+        '.hpp': 'cpp',
     }
 
 
@@ -52,7 +57,8 @@ def _extract_functions(nodes, fp, file_content, relevant_node_types):
 
 def _get_repo_functions(root, supported_file_extensions, relevant_node_types):
     functions = []
-    for fp in [root + '/' + f for f in os.popen('git -C {} ls-files'.format(root)).read().split('\n')]:
+    print('Extracting functions from {}'.format(root))
+    for fp in tqdm([root + '/' + f for f in os.popen('git -C {} ls-files'.format(root)).read().split('\n')]):
         if not os.path.isfile(fp):
             continue
         with open(fp, 'r') as f:

src/semantic_code_search/prompt.py

@@ -12,6 +12,7 @@
 from prompt_toolkit.lexers import PygmentsLexer
 from prompt_toolkit.widgets.base import Frame
 from pygments.lexers.c_cpp import CLexer
+from pygments.lexers.c_cpp import CppLexer
 from pygments.lexers.go import GoLexer
 from pygments.lexers.javascript import JavascriptLexer, TypeScriptLexer
 from pygments.lexers.jvm import JavaLexer, KotlinLexer
@@ -53,6 +54,10 @@ def _syntax_highlighting(text, file):
         lexer = PhpLexer
     elif file.endswith('rs'):
         lexer = RustLexer
+    elif file.endswith('c') or file.endswith('h'):
+        lexer = CLexer
+    elif file.endswith('cpp') or file.endswith('hpp'):
+        lexer = CppLexer
 
     pigment = PygmentsLexer(lexer, sync_from_start=True)
     lex_func = pigment.lex_document(Document(text.replace('\t', '    ')))