From b2b04691d6ba00a472d674fbf6ee02cd37337ba5 Mon Sep 17 00:00:00 2001
From: pictuga <contact@pictuga.com>
Date: Tue, 25 Jan 2022 22:36:34 +0100
Subject: [PATCH] Ability to pass custom data_files location

---
 README.md      |  1 +
 morss/feeds.py |  3 ++-
 morss/util.py  | 23 ++++++++++++++---------
 morss/wsgi.py  |  2 +-
 4 files changed, 18 insertions(+), 11 deletions(-)

diff --git a/README.md b/README.md
index 8a663ce..712e094 100644
--- a/README.md
+++ b/README.md
@@ -474,6 +474,7 @@ debugging.
 - `IGNORE_SSL=1`: to ignore SSL certs when fetch feeds and articles
 - `DELAY` (seconds) sets the browser cache delay, only for HTTP clients
 - `TIMEOUT` (seconds) sets the HTTP timeout when fetching rss feeds and articles
+- `DATA_PATH`: to set custom file location for the `www` folder
 
 When parsing long feeds, with a lot of items (100+), morss might take a lot of
 time to parse it, or might even run into a memory overflow on some shared
diff --git a/morss/feeds.py b/morss/feeds.py
index 3563216..a229521 100644
--- a/morss/feeds.py
+++ b/morss/feeds.py
@@ -65,7 +65,8 @@ def parse_rules(filename=None):
             # for each rule
 
             if rules[section][arg].startswith('file:'):
-                file_raw = open(data_path(rules[section][arg][5:])).read()
+                path = data_path('www', rules[section][arg][5:])
+                file_raw = open(path).read()
                 file_clean = re.sub('<[/?]?(xsl|xml)[^>]+?>', '', file_raw)
                 rules[section][arg] = file_clean
 
diff --git a/morss/util.py b/morss/util.py
index 5de28e5..2f8a705 100644
--- a/morss/util.py
+++ b/morss/util.py
@@ -15,32 +15,37 @@
 # You should have received a copy of the GNU Affero General Public License along
 # with this program. If not, see <https://www.gnu.org/licenses/>.
 
+import os
 import os.path
 import sys
 
 
-def pkg_path(path=''):
-    return os.path.join(os.path.dirname(__file__), path)
+def pkg_path(*path_elements):
+    return os.path.join(os.path.dirname(__file__), *path_elements)
 
 
 data_path_base = None
 
 
-def data_path(path=''):
+def data_path(*path_elements):
     global data_path_base
 
+    path = os.path.join(*path_elements)
+
     if data_path_base is not None:
         return os.path.join(data_path_base, path)
 
     bases = [
-        os.path.join(sys.prefix, 'share/morss/www'), # when installed as root
-        pkg_path('../../../share/morss/www'), 
-        pkg_path('../../../../share/morss/www'),
-        pkg_path('../share/morss/www'), # for `pip install --target=dir morss`
-        pkg_path('../www'), # when running from source tree
-        pkg_path('../..'), # when running on `.cgi` subdir on Apache
+        os.path.join(sys.prefix, 'share/morss'), # when installed as root
+        pkg_path('../../../share/morss'), 
+        pkg_path('../../../../share/morss'),
+        pkg_path('../share/morss'), # for `pip install --target=dir morss`
+        pkg_path('..'), # when running from source tree
     ]
 
+    if 'DATA_PATH' in os.environ:
+        bases.append(os.environ['DATA_PATH'])
+
     for base in bases:
         full_path = os.path.join(base, path)
 
diff --git a/morss/wsgi.py b/morss/wsgi.py
index ef829e8..ee88803 100644
--- a/morss/wsgi.py
+++ b/morss/wsgi.py
@@ -169,7 +169,7 @@ def cgi_file_handler(environ, start_response, app):
     if re.match(r'^/?([a-zA-Z0-9_-][a-zA-Z0-9\._-]+/?)*$', url):
         # if it is a legitimate url (no funny relative paths)
         try:
-            path = data_path(url)
+            path = data_path('www', url)
             f = open(path, 'rb')
 
         except IOError: