TODO: weird docs handling

Signed-off-by: Alfred Wingate <parona@protonmail.com>

.pre-commit-config.yaml

@@ -1,15 +1,15 @@
 repos:
   - repo: https://github.com/psf/black
-    rev: 23.11.0
+    rev: 24.2.0
     hooks:
       - id: black
 
   - repo: https://github.com/PyCQA/isort
-    rev: 5.12.0
+    rev: 5.13.2
     hooks:
       - id: isort
 
   - repo: https://github.com/astral-sh/ruff-pre-commit
-    rev: v0.1.4
+    rev: v0.3.0
     hooks:
       - id: ruff

TODO

@@ -42,11 +42,7 @@ euscan
 - remote-id type deb repository:
 -- find out how to get download url (not sure it's possible)
 
-### remote-id
+### bugs or unwanted behavior
-- Propose new remote-id: deb
-  e.g.: <remote-id type="deb">
-            http://mysite.com/deb/dists/stable/main/binary-i386/Packages
-        </remote-id>
-- Propose new remote-id: freecode
-  e.g.: <remote-id type="freecode">projectname</remote-id>
 
+- Parsing docs and accepting 404's
+-- net-analyzer/sensu

src/euscan/__init__.py

@@ -51,7 +51,7 @@ BLACKLIST_PACKAGES = [
 ]
 
 SCANDIR_BLACKLIST_URLS = [
-    "mirror://rubygems/(.*)",  # Not browsable
+    "https://rubygems.org/(.*)",  # Not browsable
     "mirror://gentoo/(.*)",  # Directory too big
     "https://dev.gentoo.org/(.*)",  # There shouldn't be releases here
     # Waste of time to go through

src/euscan/handlers/berlios.py

@@ -1,59 +0,0 @@
-# Copyright 2011 Corentin Chary <corentin.chary@gmail.com>
-# Copyright 2020-2023 src_prepare group
-# Distributed under the terms of the GNU General Public License v2
-
-import re
-import urllib.error
-import urllib.parse
-import urllib.request
-
-import portage
-
-from euscan import output
-from euscan.handlers.url import process_scan as url_scan
-from euscan.helpers import regex_from_template
-
-HANDLER_NAME = "berlios"
-CONFIDENCE = 90
-PRIORITY = 90
-
-
-berlios_regex = r"mirror://berlios/([^/]+)/([^/]+)"
-
-
-def can_handle(pkg, url=None):
-    if not url:
-        return False
-
-    cp, ver, rev = portage.pkgsplit(pkg.cpv)
-    if ver not in url:
-        return False
-
-    return re.search(berlios_regex, url)
-
-
-def scan_url(pkg, url, options):
-    output.einfo("Using BerliOS handler")
-
-    cp, ver, rev = portage.pkgsplit(pkg.cpv)
-
-    project, filename = re.search(berlios_regex, url).groups()
-
-    project_page = "http://developer.berlios.de/projects/%s" % project
-    content = urllib.request.urlopen(project_page).read()
-
-    project_id = re.search(r"/project/filelist.php\?group_id=(\d+)", content).group(1)
-
-    base_url = (
-        "http://developer.berlios.de/project/filelist.php?group_id=%s" % project_id
-    )
-
-    file_pattern = regex_from_template(filename.replace(ver, "${PV}"))
-
-    result = url_scan(pkg, base_url, file_pattern)
-
-    ret = []
-    for found_url, pv, _, _ in result:
-        found_url = found_url.replace("prdownload", "download")
-        ret.append((found_url, pv, HANDLER_NAME, CONFIDENCE))
-    return ret

src/euscan/handlers/freecode.py

@@ -1,53 +0,0 @@
-# Copyright 2011 Corentin Chary <corentin.chary@gmail.com>
-# Copyright 2020-2023 src_prepare group
-# Distributed under the terms of the GNU General Public License v2
-
-import re
-import urllib.error
-import urllib.parse
-import urllib.request
-
-import portage
-
-from euscan import helpers, mangling, output
-
-HANDLER_NAME = "freecode"
-CONFIDENCE = 100
-PRIORITY = 90
-
-
-def can_handle(pkg, url=None):
-    return False
-
-
-def scan_pkg(pkg, options):
-    cp, ver, rev = portage.pkgsplit(pkg.cpv)
-
-    package = options["data"].strip()
-
-    output.einfo("Using FreeCode handler: " + package)
-
-    fp = urllib.request.urlopen("http://freecode.com/projects/%s/releases" % package)
-    content = str(fp.read())
-
-    result = re.findall(
-        r'<a href="/projects/%s/releases/(\d+)">([^<]+)</a>' % package, content
-    )
-
-    ret = []
-    for release_id, up_pv in result:
-        pv = mangling.mangle_version(up_pv, options)
-        if helpers.version_filtered(cp, ver, pv):
-            continue
-        fp = urllib.request.urlopen(
-            f"http://freecode.com/projects/{package}/releases/{release_id}"
-        )
-        content = str(fp.read())
-        download_page = re.findall(r'<a href="(/urls/[^"]+)"', content)[0]
-        fp = urllib.request.urlopen("http://freecode.com%s" % download_page)
-        content = str(fp.read())
-        url = re.findall(
-            r'In case it doesn\'t, click here: <a href="([^"]+)"', content
-        )[0]
-        ret.append((url, pv, HANDLER_NAME, CONFIDENCE))
-    return ret

src/euscan/handlers/github.py

@@ -1,66 +0,0 @@
-# Copyright 2011 Corentin Chary <corentin.chary@gmail.com>
-# Copyright 2020-2023 src_prepare group
-# Distributed under the terms of the GNU General Public License v2
-
-import json
-import re
-import urllib.error
-import urllib.parse
-import urllib.request
-
-import portage
-
-from euscan import helpers, mangling, output
-
-HANDLER_NAME = "github"
-CONFIDENCE = 100
-PRIORITY = 90
-
-
-def can_handle(pkg, url=None):
-    return url and url.startswith("mirror://github/")
-
-
-def guess_package(cp, url):
-    match = re.search("^mirror://github/(.*?)/(.*?)/(.*)$", url)
-
-    assert match
-    return (match.group(1), match.group(2), match.group(3))
-
-
-def scan_url(pkg, url, options):
-    "http://developer.github.com/v3/repos/downloads/"
-
-    user, project, filename = guess_package(pkg.cpv, url)
-
-    # find out where version is expected to be found
-    cp, ver, rev = portage.pkgsplit(pkg.cpv)
-    if ver not in filename:
-        return
-
-    # now create a filename-matching regexp
-    # XXX: supposedly replace first with (?P<foo>...)
-    # and remaining ones with (?P=foo)
-    fnre = re.compile("^%s$" % re.escape(filename).replace(re.escape(ver), "(.*?)"))
-
-    output.einfo(
-        f"Using github API for: project={project} user={user} filename={filename}"
-    )
-
-    dlreq = urllib.request.urlopen(
-        f"https://api.github.com/repos/{user}/{project}/downloads"
-    )
-    dls = json.load(dlreq)
-
-    ret = []
-    for dl in dls:
-        m = fnre.match(dl["name"])
-
-        if m:
-            pv = mangling.mangle_version(m.group(1), options)
-            if helpers.version_filtered(cp, ver, pv):
-                continue
-
-            url = mangling.mangle_url(dl["html_url"], options)
-            ret.append((url, pv, HANDLER_NAME, CONFIDENCE))
-    return ret

src/euscan/handlers/google_code.py

@@ -1,47 +0,0 @@
-# Copyright 2011 Corentin Chary <corentin.chary@gmail.com>
-# Copyright 2020-2023 src_prepare group
-# Distributed under the terms of the GNU General Public License v2
-
-import re
-
-import portage
-
-from euscan import output
-from euscan.handlers.url import process_scan as url_scan
-from euscan.helpers import regex_from_template
-
-HANDLER_NAME = "google-code"
-CONFIDENCE = 90
-PRIORITY = 90
-
-
-package_name_regex = r"http://(.+).googlecode.com/files/.+"
-
-
-def can_handle(pkg, url=None):
-    if not url:
-        return False
-
-    cp, ver, rev = portage.pkgsplit(pkg.cpv)
-    if ver not in url:
-        return False
-
-    return re.match(package_name_regex, url)
-
-
-def scan_url(pkg, url, options):
-    output.einfo("Using Google Code handler")
-
-    cp, ver, rev = portage.pkgsplit(pkg.cpv)
-
-    package_name = re.match(package_name_regex, url).group(1)
-    base_url = "http://code.google.com/p/%s/downloads/list" % package_name
-
-    file_pattern = regex_from_template(url.split("/")[-1].replace(ver, "${PV}"))
-
-    result = url_scan(pkg, base_url, file_pattern)
-
-    ret = []
-    for url, pv, _, _ in result:
-        ret.append((url, pv, HANDLER_NAME, CONFIDENCE))
-    return ret

src/euscan/handlers/pypi.py

@@ -17,11 +17,11 @@ PRIORITY = 90
 
 
 def can_handle(pkg, url=None):
-    return url and url.startswith("mirror://pypi/")
+    return url and url.startswith("https://files.pythonhosted.org/packages/source/p/")
 
 
 def guess_package(cp, url):
-    match = re.search(r"mirror://pypi/\w+/(.*)/.*", url)
+    match = re.search(r"https://files.pythonhosted.org/packages/source/p/(.*)/.*", url)
     if match:
         return match.group(1)
 

src/euscan/handlers/rubygems.py

@@ -1,5 +1,5 @@
 # Copyright 2011 Corentin Chary <corentin.chary@gmail.com>
-# Copyright 2020-2023 src_prepare group
+# Copyright 2020-2024 src_prepare group
 # Distributed under the terms of the GNU General Public License v2
 
 import json
@@ -18,11 +18,11 @@ PRIORITY = 90
 
 
 def can_handle(pkg, url=None):
-    return url and url.startswith("mirror://rubygems/")
+    return url and url.startswith("https://rubygems.org/")
 
 
 def guess_gem(cpv, url):
-    match = re.search("mirror://rubygems/(.*).gem", url)
+    match = re.search("https://rubygems.org/gems/(.*).gem", url)
     if match:
         cpv = "fake/%s" % match.group(1)