Move API header warning to init of pipeline
Browse files- pipeline.py +20 -17
pipeline.py
CHANGED
@@ -9,17 +9,6 @@ import requests
|
|
9 |
import torch
|
10 |
from transformers import Pipeline
|
11 |
|
12 |
-
API_HEADERS = {"Accept": "application/vnd.github+json"}
|
13 |
-
if os.environ.get("GITHUB_TOKEN") is None:
|
14 |
-
print(
|
15 |
-
"[!] Consider setting GITHUB_TOKEN environment variable to avoid hitting rate limits\n"
|
16 |
-
"For more info, see:"
|
17 |
-
"https://docs.github.com/authentication/keeping-your-account-and-data-secure/creating-a-personal-access-token"
|
18 |
-
)
|
19 |
-
else:
|
20 |
-
API_HEADERS["Authorization"] = f"Bearer {os.environ['GITHUB_TOKEN']}"
|
21 |
-
print("[+] Using GITHUB_TOKEN for authentication")
|
22 |
-
|
23 |
|
24 |
def extract_code_and_docs(text: str):
|
25 |
"""Extract code and documentation from a Python file.
|
@@ -53,11 +42,11 @@ def extract_code_and_docs(text: str):
|
|
53 |
return code_set, docs_set
|
54 |
|
55 |
|
56 |
-
def get_topics(repo_name):
|
57 |
api_url = f"https://api.github.com/repos/{repo_name}"
|
58 |
print(f"[+] Getting topics for {repo_name}")
|
59 |
try:
|
60 |
-
response = requests.get(api_url, headers=
|
61 |
response.raise_for_status()
|
62 |
except requests.exceptions.HTTPError as e:
|
63 |
print(f"[-] Failed to get topics for {repo_name}: {e}")
|
@@ -71,19 +60,19 @@ def get_topics(repo_name):
|
|
71 |
return topics
|
72 |
|
73 |
|
74 |
-
def download_and_extract(repos):
|
75 |
extracted_info = {}
|
76 |
for repo_name in repos:
|
77 |
extracted_info[repo_name] = {
|
78 |
"funcs": set(),
|
79 |
"docs": set(),
|
80 |
-
"topics": get_topics(repo_name),
|
81 |
}
|
82 |
|
83 |
download_url = f"https://api.github.com/repos/{repo_name}/tarball"
|
84 |
print(f"[+] Extracting functions and docstrings from {repo_name}")
|
85 |
try:
|
86 |
-
response = requests.get(download_url, headers=
|
87 |
response.raise_for_status()
|
88 |
except requests.exceptions.HTTPError as e:
|
89 |
print(f"[-] Failed to download {repo_name}: {e}")
|
@@ -107,6 +96,20 @@ def download_and_extract(repos):
|
|
107 |
|
108 |
|
109 |
class RepoEmbeddingPipeline(Pipeline):
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
110 |
def _sanitize_parameters(self, **kwargs):
|
111 |
_forward_kwargs = {}
|
112 |
if "max_length" in kwargs:
|
@@ -118,7 +121,7 @@ class RepoEmbeddingPipeline(Pipeline):
|
|
118 |
if isinstance(inputs, str):
|
119 |
inputs = (inputs,)
|
120 |
|
121 |
-
extracted_infos = download_and_extract(inputs)
|
122 |
|
123 |
return extracted_infos
|
124 |
|
|
|
9 |
import torch
|
10 |
from transformers import Pipeline
|
11 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
12 |
|
13 |
def extract_code_and_docs(text: str):
|
14 |
"""Extract code and documentation from a Python file.
|
|
|
42 |
return code_set, docs_set
|
43 |
|
44 |
|
45 |
+
def get_topics(repo_name, headers=None):
|
46 |
api_url = f"https://api.github.com/repos/{repo_name}"
|
47 |
print(f"[+] Getting topics for {repo_name}")
|
48 |
try:
|
49 |
+
response = requests.get(api_url, headers=headers)
|
50 |
response.raise_for_status()
|
51 |
except requests.exceptions.HTTPError as e:
|
52 |
print(f"[-] Failed to get topics for {repo_name}: {e}")
|
|
|
60 |
return topics
|
61 |
|
62 |
|
63 |
+
def download_and_extract(repos, headers=None):
|
64 |
extracted_info = {}
|
65 |
for repo_name in repos:
|
66 |
extracted_info[repo_name] = {
|
67 |
"funcs": set(),
|
68 |
"docs": set(),
|
69 |
+
"topics": get_topics(repo_name, headers=headers),
|
70 |
}
|
71 |
|
72 |
download_url = f"https://api.github.com/repos/{repo_name}/tarball"
|
73 |
print(f"[+] Extracting functions and docstrings from {repo_name}")
|
74 |
try:
|
75 |
+
response = requests.get(download_url, headers=headers, stream=True)
|
76 |
response.raise_for_status()
|
77 |
except requests.exceptions.HTTPError as e:
|
78 |
print(f"[-] Failed to download {repo_name}: {e}")
|
|
|
96 |
|
97 |
|
98 |
class RepoEmbeddingPipeline(Pipeline):
|
99 |
+
def __init__(self, *args, **kwargs):
|
100 |
+
super().__init__(*args, **kwargs)
|
101 |
+
self.API_HEADERS = {"Accept": "application/vnd.github+json"}
|
102 |
+
if os.environ.get("GITHUB_TOKEN") is None:
|
103 |
+
print(
|
104 |
+
"[!] Consider setting GITHUB_TOKEN environment variable to avoid hitting rate limits\n"
|
105 |
+
"For more info, see:"
|
106 |
+
"https://docs.github.com/authentication/keeping-your-account-and-data-secure/creating-a-personal-access-token"
|
107 |
+
)
|
108 |
+
else:
|
109 |
+
self.API_HEADERS["Authorization"] = f"Bearer {os.environ['GITHUB_TOKEN']}"
|
110 |
+
print("[+] Using GITHUB_TOKEN for authentication")
|
111 |
+
|
112 |
+
|
113 |
def _sanitize_parameters(self, **kwargs):
|
114 |
_forward_kwargs = {}
|
115 |
if "max_length" in kwargs:
|
|
|
121 |
if isinstance(inputs, str):
|
122 |
inputs = (inputs,)
|
123 |
|
124 |
+
extracted_infos = download_and_extract(inputs, headers=self.API_HEADERS)
|
125 |
|
126 |
return extracted_infos
|
127 |
|