File size: 6,369 Bytes
6f6019f
0c52573
6f6019f
b33162a
6f6019f
 
 
 
dc07873
b33162a
0c52573
b33162a
0c52573
b33162a
 
0c52573
b33162a
0c52573
b33162a
0c52573
 
0a9a29e
0c52573
 
dc07873
b33162a
0c52573
0a9a29e
dc07873
6f6019f
0c52573
b33162a
6f6019f
 
 
 
 
b33162a
6f6019f
1aeb49e
 
 
 
b33162a
1aeb49e
 
 
 
b33162a
6f6019f
 
 
 
66a05ca
6f6019f
66a05ca
1aeb49e
66a05ca
1aeb49e
 
b33162a
0a9a29e
 
 
b33162a
1aeb49e
 
 
 
66a05ca
 
0a9a29e
1aeb49e
 
 
0a9a29e
1aeb49e
 
 
6f6019f
b33162a
1aeb49e
 
 
b33162a
1aeb49e
 
 
 
66a05ca
 
0a9a29e
1aeb49e
 
 
0a9a29e
1aeb49e
 
 
 
66a05ca
1aeb49e
0a9a29e
1aeb49e
 
 
 
 
66a05ca
 
0a9a29e
1aeb49e
 
 
0a9a29e
1aeb49e
 
 
dc07873
66a05ca
6f6019f
0a9a29e
6f6019f
dc07873
 
0a9a29e
dc07873
0a9a29e
66a05ca
0a9a29e
 
 
66a05ca
0a9a29e
 
dc07873
 
6f6019f
0a9a29e
6f6019f
dc07873
 
 
6f6019f
 
dc07873
 
 
 
 
 
0a9a29e
dc07873
6f6019f
dc07873
 
 
 
 
0a9a29e
dc07873
 
 
 
6f6019f
0a9a29e
66a05ca
6f6019f
0a9a29e
6f6019f
 
 
 
 
 
dc07873
0a9a29e
6f6019f
0a9a29e
dc07873
0a9a29e
dc07873
0a9a29e
dc07873
0a9a29e
 
 
66a05ca
0a9a29e
dc07873
0a9a29e
dc07873
0a9a29e
 
 
 
 
 
b33162a
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
import ast
from langchain.schema import Document 

def chunk_python_code_with_metadata(source_code, source):
    """
    Entry point method to process the Python file.
    It invokes the iterate_ast function.
    """
    documents = []
    print(f"Processing file: {source}")

    iterate_ast(source_code, documents, source)

    # Determine usage based on the source path
    if source.startswith("kadi_apy/lib/"):
        usage = "library"
    elif source.startswith("kadi_apy/cli/"):
        usage = "cli_library"
    elif source.startswith("doc/"):
        usage = "doc"
    else:
        usage = "undefined"

    # Add metadata for usage to all documents
    for doc in documents:
        doc.metadata["source"] = source
        doc.metadata["usage"] = usage  # Add the determined usage metadata
        print(doc)
    return documents


def iterate_ast(source_code, documents, source):
    """
    Parses the AST of the given Python file and delegates
    handling to specific methods based on node types.
    """
    # Parse the source code into an abstract syntax tree (AST)
    tree = ast.parse(source_code, filename=source)

    first_level_nodes = list(ast.iter_child_nodes(tree))

    # Check if there are no first-level nodes
    if not first_level_nodes:
        handle_no_first_level_node_found(documents, source_code, source)
        return

    all_imports = all(isinstance(node, (ast.Import, ast.ImportFrom)) for node in first_level_nodes)
    if all_imports:
        handle_first_level_imports_only(documents, source_code, source)
    
    # Iterate over first-level nodes
    for first_level_node in ast.iter_child_nodes(tree):
        if isinstance(first_level_node, ast.ClassDef):
            handle_first_level_class(first_level_node, documents, source_code)
        elif isinstance(first_level_node, ast.FunctionDef):
            handle_first_level_func(first_level_node, documents, source_code)
        elif isinstance(first_level_node, ast.Assign):
            handle_first_level_assign(first_level_node, documents, source_code)


def handle_first_level_imports_only(documents, source_code, source):
    """
    Handles cases where the first-level nodes are only imports.
    """
    if source.endswith("__init__.py"):
        type = "__init__-file"
    else:
        type = "undefined"

    # Create metadata without imports
    metadata = {"type": type}

    # Create and store a Document with the full source code
    doc = Document(
        page_content=source_code,
        metadata=metadata
    )
    documents.append(doc)


def handle_no_first_level_node_found(documents, source_code, source):
    """
    Handles cases where no top-level nodes are found in the AST.
    """
    if source.endswith("__init__.py"):
        type = "__init__-file"
    else:
        type = "undefined"

    # Create metadata without imports
    metadata = {"type": type}

    # Create and store a Document with the full source code
    doc = Document(
        page_content=source_code,
        metadata=metadata
    )
    documents.append(doc)


def handle_first_level_assign(assign_node, documents, source_code):
    """
    Handles assignment statements at the first level of the AST.
    """
    assign_start_line = assign_node.lineno
    assign_end_line = assign_node.end_lineno
    assign_source = '\n'.join(source_code.splitlines()[assign_start_line-1:assign_end_line])

    # Create metadata without imports
    metadata = {"type": "Assign"}

    # Create and store Document for the assignment
    doc = Document(
        page_content=assign_source,
        metadata=metadata
    )
    documents.append(doc)


def handle_first_level_class(class_node, documents, source_code):
    """
    Handles classes at the first level of the AST.
    """
    class_start_line = class_node.lineno
    class_body_lines = [child.lineno for child in class_node.body if isinstance(child, ast.FunctionDef)]
    class_end_line = min(class_body_lines, default=class_node.end_lineno) - 1
    class_source = '\n'.join(source_code.splitlines()[class_start_line-1:class_end_line])

    # Create metadata without imports
    metadata = {
        "type": "class",
        "class": class_node.name,
        "visibility": "public"
    }

    # Create and store Document for the class
    doc = Document(
        page_content=class_source,
        metadata=metadata
    )
    documents.append(doc)

    # Handle methods within the class
    for second_level_node in ast.iter_child_nodes(class_node):
        if isinstance(second_level_node, ast.FunctionDef):
            method_start_line = (
                second_level_node.decorator_list[0].lineno
                if second_level_node.decorator_list else second_level_node.lineno
            )
            method_end_line = second_level_node.end_lineno
            method_source = '\n'.join(source_code.splitlines()[method_start_line-1:method_end_line])

            visibility = "internal" if second_level_node.name.startswith("_") else "public"

            doc = Document(
                page_content=method_source,
                metadata={
                    "type": "method",
                    "method": second_level_node.name,
                    "visibility": visibility,
                    "class": class_node.name
                }
            )
            documents.append(doc)


def handle_first_level_func(function_node, documents, source_code):
    """
    Handles functions at the first level of the AST.
    """
    function_start_line = (
        function_node.decorator_list[0].lineno
        if function_node.decorator_list else function_node.lineno
    )
    function_end_line = function_node.end_lineno
    function_source = '\n'.join(source_code.splitlines()[function_start_line-1:function_end_line])

    visibility = "internal" if function_node.name.startswith("_") else "public"

    is_command = any(
        decorator.id == "apy_command"
        for decorator in function_node.decorator_list
        if hasattr(decorator, "id")
    )

    metadata = {
        "type": "command" if is_command else "function",
        "visibility": visibility
    }
    if is_command:
        metadata["command"] = function_node.name
    else:
        metadata["method"] = function_node.name

    doc = Document(
        page_content=function_source,
        metadata=metadata
    )
    documents.append(doc)