From 575aeb3d18c9bca3541f34d35c7345a6521224e2 Mon Sep 17 00:00:00 2001
From: Gertjan van den Burg <gertjanvandenburg@gmail.com>
Date: Thu, 23 Jan 2020 20:22:18 +0000
Subject: Reorder the input validation (fixes #21)

We used to assume that if a source is not a local file,
then it must be a URL. Now, we check if a source is a
URL and if not, also check if it's a local file. If neither,
then we can raise an error.
---
 paper2remarkable/ui.py    | 20 +++++++++++++-------
 paper2remarkable/utils.py | 12 ++++++++++--
 setup.py                  |  1 +
 3 files changed, 24 insertions(+), 9 deletions(-)

diff --git a/paper2remarkable/ui.py b/paper2remarkable/ui.py
index 6343077..032bf99 100644
--- a/paper2remarkable/ui.py
+++ b/paper2remarkable/ui.py
@@ -14,7 +14,7 @@ import sys
 from . import __version__, GITHUB_URL
 
 from .providers import providers, LocalFile
-from .utils import follow_redirects
+from .utils import follow_redirects, is_url
 
 
 def parse_args():
@@ -97,16 +97,21 @@ def exception(msg):
 
 def main():
     args = parse_args()
+    cookiejar = None
 
-    if LocalFile.validate(args.input):
+    if is_url(args.input):
+        # input is a url
+        url, cookiejar = follow_redirects(args.input)
+        provider = next((p for p in providers if p.validate(url)), None)
+    elif LocalFile.validate(args.input):
         # input is a local file
         provider = LocalFile
     else:
-        # input is a url
-        url = args.input
-        # follow all redirects of the url
-        url = follow_redirects(url)
-        provider = next((p for p in providers if p.validate(url)), None)
+        # not a proper URL or non-existent file
+        exception(
+            "Couldn't figure out what source you mean. If it's a "
+            "local file, make sure it exists."
+        )
 
     if provider is None:
         exception("Input not valid, no provider can handle this source.")
@@ -122,6 +127,7 @@ def main():
         pdfcrop_path=args.pdfcrop,
         pdftk_path=args.pdftk,
         gs_path=args.gs,
+        cookiejar=cookiejar,
     )
 
     prov.run(args.input, filename=args.filename)
diff --git a/paper2remarkable/utils.py b/paper2remarkable/utils.py
index 1bf261e..79421df 100644
--- a/paper2remarkable/utils.py
+++ b/paper2remarkable/utils.py
@@ -9,6 +9,7 @@ Copyright: 2019, G.J.J. van den Burg
 """
 
 import PyPDF2
+import regex
 import requests
 import string
 import subprocess
@@ -110,8 +111,7 @@ def upload_to_remarkable(filepath, remarkable_dir="/", rmapi_path="rmapi"):
     remarkable_dir = remarkable_dir.rstrip("/")
     if remarkable_dir:
         status = subprocess.call(
-            [rmapi_path, "mkdir", remarkable_dir],
-            stdout=subprocess.DEVNULL,
+            [rmapi_path, "mkdir", remarkable_dir], stdout=subprocess.DEVNULL,
         )
         if not status == 0:
             raise RemarkableError(
@@ -128,3 +128,11 @@ def upload_to_remarkable(filepath, remarkable_dir="/", rmapi_path="rmapi"):
             "Uploading file %s to reMarkable failed" % filepath
         )
     logger.info("Upload successful.")
+
+
+def is_url(string):
+    # pattern adapted from CleverCSV
+    pattern = "((https?|ftp):\/\/(?!\-))?(((([\p{L}\p{N}]*\-?[\p{L}\p{N}]+)+\.)+([a-z]{2,}|local)(\.[a-z]{2,3})?)|localhost|(\d{1,3}\.\d{1,3}\.\d{1,3}\.\d{1,3}(\:\d{1,5})?))(\/[\p{L}\p{N}_\/()~?=&%\-\#\.:]*)?(\.[a-z]+)?"
+    string = string.strip(" ")
+    match = regex.fullmatch(pattern, string)
+    return match is not None
diff --git a/setup.py b/setup.py
index f54170a..bddbd24 100644
--- a/setup.py
+++ b/setup.py
@@ -25,6 +25,7 @@ REQUIRED = [
     "unidecode>=1.1",
     "titlecase>=0.12",
     "PyPDF2>=1.26",
+    "regex>=2018.11"
 ]
 
 docs_require = []
-- 
cgit v1.2.3