From 7e1d59377aa46a1158da830119a4489fb2e97e5d Mon Sep 17 00:00:00 2001
From: Erik Arvstedt <erik.arvstedt@gmail.com>
Date: Fri, 11 May 2018 14:01:21 +0200
Subject: [PATCH] Add inotify support

---
 paperless.conf.example                        |  2 +
 requirements.txt                              |  1 +
 .../management/commands/document_consumer.py  | 47 ++++++++++++++++++-
 src/paperless/settings.py                     |  2 +
 4 files changed, 50 insertions(+), 2 deletions(-)

diff --git a/paperless.conf.example b/paperless.conf.example
index 45c532fe1..0727ac29d 100644
--- a/paperless.conf.example
+++ b/paperless.conf.example
@@ -165,6 +165,8 @@ PAPERLESS_PASSPHRASE="secret"
 #PAPERLESS_CONVERT_DENSITY=300
 
 
+# (This setting is ignored on Linux where inotify is used instead of a
+# polling loop.)
 # The number of seconds that Paperless will wait between checking
 # PAPERLESS_CONSUMPTION_DIR.  If you tend to write documents to this directory
 # rarely, you may want to use a higher value than the default (10).
diff --git a/requirements.txt b/requirements.txt
index c90a8eaa9..d51a39039 100644
--- a/requirements.txt
+++ b/requirements.txt
@@ -20,6 +20,7 @@ flake8==3.5.0
 fuzzywuzzy==0.15.0
 gunicorn==19.7.1
 idna==2.6
+inotify_simple==1.1.7; sys_platform == 'linux'
 langdetect==1.0.7
 mccabe==0.6.1
 more-itertools==4.1.0
diff --git a/src/documents/management/commands/document_consumer.py b/src/documents/management/commands/document_consumer.py
index 9234617e1..a0e2f00fe 100644
--- a/src/documents/management/commands/document_consumer.py
+++ b/src/documents/management/commands/document_consumer.py
@@ -1,6 +1,7 @@
 import datetime
 import logging
 import os
+import sys
 import time
 
 from django.conf import settings
@@ -9,6 +10,11 @@ from django.core.management.base import BaseCommand, CommandError
 from ...consumer import Consumer, ConsumerError, make_dirs
 from ...mail import MailFetcher, MailFetcherError
 
+try:
+    from inotify_simple import INotify, flags
+except ImportError:
+    pass
+
 
 class Command(BaseCommand):
     """
@@ -53,6 +59,11 @@ class Command(BaseCommand):
             action="store_true",
             help="Run only once."
         )
+        parser.add_argument(
+            "--no-inotify",
+            action="store_true",
+            help="Don't use inotify, even if it's available."
+        )
 
     def handle(self, *args, **options):
 
@@ -60,6 +71,8 @@ class Command(BaseCommand):
         directory = options["directory"]
         loop_time = options["loop_time"]
         mail_delta = options["mail_delta"] * 60
+        use_inotify = (not options["no_inotify"]
+                       and "inotify_simple" in sys.modules)
 
         try:
             self.file_consumer = Consumer(consume=directory)
@@ -70,14 +83,20 @@ class Command(BaseCommand):
         make_dirs(self.ORIGINAL_DOCS, self.THUMB_DOCS)
 
         logging.getLogger(__name__).info(
-            "Starting document consumer at {}".format(directory)
+            "Starting document consumer at {}{}".format(
+                directory,
+                " with inotify" if use_inotify else ""
+            )
         )
 
         if options["oneshot"]:
             self.loop_step(mail_delta)
         else:
             try:
-                self.loop(loop_time, mail_delta)
+                if use_inotify:
+                    self.loop_inotify(mail_delta)
+                else:
+                    self.loop(loop_time, mail_delta)
             except KeyboardInterrupt:
                 print("Exiting")
 
@@ -101,3 +120,27 @@ class Command(BaseCommand):
             self.mail_fetcher.pull()
 
         self.file_consumer.consume_new_files()
+
+    def loop_inotify(self, mail_delta):
+        directory = self.file_consumer.consume
+        inotify = INotify()
+        inotify.add_watch(directory, flags.CLOSE_WRITE | flags.MOVED_TO)
+
+        # Run initial mail fetch and consume all currently existing documents
+        self.loop_step(mail_delta)
+        next_mail_time = self.mail_fetcher.last_checked + mail_delta
+
+        while True:
+            # Consume documents until next_mail_time
+            while True:
+                delta = next_mail_time - time.time()
+                if delta > 0:
+                    for event in inotify.read(timeout=delta):
+                        file = os.path.join(directory, event.name)
+                        if os.path.isfile(file):
+                            self.file_consumer.try_consume_file(file)
+                else:
+                    break
+
+            self.mail_fetcher.pull()
+            next_mail_time = self.mail_fetcher.last_checked + mail_delta
diff --git a/src/paperless/settings.py b/src/paperless/settings.py
index 314edeb03..15e0674bc 100644
--- a/src/paperless/settings.py
+++ b/src/paperless/settings.py
@@ -246,6 +246,8 @@ SCRATCH_DIR = os.getenv("PAPERLESS_SCRATCH_DIR", "/tmp/paperless")
 # This is where Paperless will look for PDFs to index
 CONSUMPTION_DIR = os.getenv("PAPERLESS_CONSUMPTION_DIR")
 
+# (This setting is ignored on Linux where inotify is used instead of a
+# polling loop.)
 # The number of seconds that Paperless will wait between checking
 # CONSUMPTION_DIR.  If you tend to write documents to this directory very
 # slowly, you may want to use a higher value than the default.