4 files changed, 101 insertions, 0 deletions
diff --git a/gitlab/description_parser.py b/gitlab/description_parser.py
new file mode 100755
index 00000000..2615b636
--- /dev/null
+++ b/gitlab/description_parser.py
@@ -0,0 +1,39 @@
+import re
+from tomlkit import string
+
+def remove_comments(description):
+    return re.sub(r'<!--(.|\n)*?-->', '', description)
+
+def get_headline_content(description, headline):
+    pattern = rf'## {headline}\s+(.*?)(?=##\s|\Z)'
+
+    match = re.search(pattern, description, re.DOTALL)
+    if match:
+        return string(match.group(1).strip(), multiline=True)
+    else:
+        return "n/a"
+
+def get_bullet_point(description, headline, category):
+    pattern = rf'{headline}(?:(?:.|\n)+?){category}:\s+(?:`)?(.+?)(?:`)?(?=\s)(?:\n|$)'
+
+    match = re.search(pattern, description)
+    if match:
+        return match.group(1).strip()
+    else:
+        return "n/a"
+
+def parse_description(desc):
+    desc = remove_comments(desc)
+
+    result = {
+        "host-os": get_bullet_point(desc, "Host", "Operating system"),
+        "host-arch": get_bullet_point(desc, "Host", "Architecture"),
+        "qemu-version": get_bullet_point(desc, "Host", "QEMU version"),
+        "guest-os": get_bullet_point(desc, "Emulated", "Operating system"),
+        "guest-arch": get_bullet_point(desc, "Emulated", "Architecture"),
+        "description": get_headline_content(desc, "Description of problem"),
+        "reproduce": get_headline_content(desc, "Steps to reproduce"),
+        "additional": get_headline_content(desc, "Additional information")
+    }
+
+    return result
diff --git a/gitlab/downloader.py b/gitlab/downloader.py
new file mode 100755
index 00000000..2b73f1cf
--- /dev/null
+++ b/gitlab/downloader.py
@@ -0,0 +1,38 @@
+from requests import get
+from description_parser import parse_description
+from output import output_issue
+
+project_id = 11167699
+per_page = 100
+url = f"https://gitlab.com/api/v4/projects/{project_id}/issues?per_page={per_page}"
+
+def pages_iterator(first):
+    current = first
+    while current.links.get('next'):
+        current.raise_for_status()
+        yield current
+        current = get(url = current.links.get('next').get('url'))
+    current.raise_for_status()
+    yield current
+
+def main():
+    for response in pages_iterator(get(url = url)):
+        print(f"Current page: {response.headers['x-page']}")
+
+        data = response.json()
+        for i in data:
+            issue = {
+                "id": i['iid'],
+                "title": i['title'],
+                "state": i['state'],
+                "created_at": i['created_at'],
+                "closed_at": i['closed_at'] if i['closed_at'] else "n/a",
+                "labels": i['labels'],
+                "url": i['web_url']
+            }
+
+            issue = issue | parse_description(i['description'])
+            output_issue(issue)
+
+if __name__ == "__main__":
+    main()
diff --git a/gitlab/output.py b/gitlab/output.py
new file mode 100755
index 00000000..745081ae
--- /dev/null
+++ b/gitlab/output.py
@@ -0,0 +1,23 @@
+from tomlkit import dumps
+from os import path, makedirs
+
+def find_label(labels, keyword):
+    match = next((s for s in labels if f"{keyword}:" in s), None)
+    if not match:
+        return f"{keyword}_missing"
+    return match.replace(": ", "_")
+
+def write_file(file_path, string):
+    makedirs(path.dirname(file_path), exist_ok=True)
+    with open(file_path, "w") as file:
+        file.write(string)
+
+def output_issue(issue):
+    labels = issue['labels']
+    issue_id = issue['id']
+    toml_string = dumps(issue)
+
+    target_label = find_label(labels, "target")
+    host_label = find_label(labels, "host")
+    accel_label = find_label(labels, "accel")
+    write_file(f"issues/{target_label}/{host_label}/{accel_label}/{issue_id}.toml", toml_string)
diff --git a/gitlab/requirements.txt b/gitlab/requirements.txt
new file mode 100644
index 00000000..8141b831
--- /dev/null
+++ b/gitlab/requirements.txt
@@ -0,0 +1 @@
+tomlkit