aboutsummaryrefslogtreecommitdiff
blob: 6ed8e0184943b776122aa43c878e891494fc1da4 (plain)
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
import xml.etree.ElementTree as ET
from flask import json
import requests
from .. import app, db
from .models import Category, Maintainer, Package, PackageVersion

proj_url = "https://api.gentoo.org/metastructure/projects.xml"
pkg_url_base = "https://packages.gentoo.org/"
http_session = requests.session()

def get_project_data():
    data = http_session.get(proj_url)
    if not data:
        print("Failed retrieving projects.xml")
        return
    root = ET.fromstring(data.text)
    projects = {}
    # Parsing is based on http://www.gentoo.org/dtd/projects.dtd as of 2016-11-10
    if root.tag.lower() != 'projects':
        print("Downloaded projects.xml root tag isn't 'projects'")
        return
    for proj_elem in root:
        if proj_elem.tag.lower() != 'project':
            print("Skipping unknown <projects> subtag <%s>" % proj_elem.tag)
            continue
        proj = {}
        for elem in proj_elem:
            tag = elem.tag.lower()
            if tag in ['email', 'name', 'url', 'description']:
                proj[tag] = elem.text
            elif tag == 'member':
                member = {}
                if 'is-lead' in elem.attrib and elem.attrib['is-lead'] == '1':
                    member['is_lead'] = True
                for member_elem in elem:
                    member_tag = member_elem.tag.lower()
                    if member_tag in ['email', 'name', 'role']:
                        member[member_tag] = member_elem.text
                if 'email' in member:
                    # TODO: Sync the members (it's valid as email is given) - maybe at the end, after we have synced the project data, so we can add him to the project directly
                    if 'members' not in proj:
                        proj['members'] = []
                    proj['members'].append(member)
                    pass
            elif tag == 'subproject':
                if 'ref' in elem.attrib:
                    if 'subprojects' not in proj:
                        proj['subprojects'] = []
                    # subprojects will be a list of (subproject_email, inherit-members) tuples where inherit-members is True or False. TODO: Might change if sync code will want it differently
                    proj['subprojects'].append((elem.attrib['ref'], True if ('inherit-members' in elem.attrib and elem.attrib['inherit-members'] == '1') else False))
                else:
                    print("Invalid <subproject> tag inside project %s - required 'ref' attribute missing" % proj['email'] if 'email' in proj else "<unknown>")
            else:
                print("Skipping unknown <project> subtag <%s>" % tag)
        if 'email' in proj:
            projects[proj['email']] = proj
        else:
            print("Skipping incomplete project data due to lack of required email identifier: %s" % (proj,))
    return projects

def sync_projects():
    projects = get_project_data()
    existing_maintainers = {}
    # TODO: Use UPSERT instead (on_conflict_do_update) if we can rely on postgresql:9.5
    for maintainer in Maintainer.query.all():
        existing_maintainers[maintainer.email] = maintainer
    for email, data in projects.items():
        if email in existing_maintainers:
            print ("Updating project %s" % email)
            existing_maintainers[email].is_project = True
            if 'description' in data:
                existing_maintainers[email].description = data['description']
            if 'name' in data:
                existing_maintainers[email].name = data['name']
            if 'url' in data:
                existing_maintainers[email].url = data['url']
        else:
            print ("Adding project %s" % email)
            new_maintainer = Maintainer(email=data['email'], is_project=True, description=data['description'], name=data['name'], url=data['url'])
            db.session.add(new_maintainer)
    db.session.commit()


def sync_categories():
    url = pkg_url_base + "categories.json"
    data = http_session.get(url)
    # TODO: Handle response error (if not data)
    categories = json.loads(data.text)
    existing_categories = {}
    # TODO: Use UPSERT instead (on_conflict_do_update) if we can rely on postgresql:9.5
    for cat in Category.query.all():
        existing_categories[cat.name] = cat
    for category in categories:
        if category['name'] in existing_categories:
            existing_categories[category['name']].description = category['description']
        else:
            new_cat = Category(name=category['name'], description=category['description'])
            db.session.add(new_cat)
    db.session.commit()

def sync_packages():
    for category in Category.query.all():
        existing_packages = category.packages.all()
        data = http_session.get(pkg_url_base + "categories/" + category.name + ".json")
        if not data:
            print("No JSON data for category %s" % category.name) # FIXME: Better handling; mark category as inactive/gone?
            continue
        packages = json.loads(data.text)['packages']
        # TODO: Use UPSERT instead (on_conflict_do_update)
        existing_packages = {}
        for pkg in Package.query.all():
            existing_packages[pkg.name] = pkg
        for package in packages:
            if package['name'] in existing_packages:
                continue # TODO: Update description once we keep that in DB
            else:
                new_pkg = Package(category_id=category.id, name=package['name'])
                db.session.add(new_pkg)
    db.session.commit()

def sync_versions():
    for package in Package.query.all():
        data = http_session.get(pkg_url_base + "packages/" + package.full_name + ".json")
        if not data:
            print("No JSON data for package %s" % package.full_name) # FIXME: Handle better; e.g mark the package as removed if no pkgmove update
            continue
        from pprint import pprint
        pprint(json.loads(data.text))
        break