ETag support and error handling

For files served from storage.zeit.de (epub and mobi), the E-Tag used by
the server is just a md5sum.
Unless `--reload` is specified, the script will now compute the md5
hash and download the file only if it was did not match.

This method does not work for the PDF files, therefore the download is
skipped if the file already exists.

If any request fails, the script will now abort and print the HTTP
status code.

zeitdownload.py

@@ -5,6 +5,7 @@ import cgi
 import sys
 import re
 import os.path
+import hashlib
 from argparse import ArgumentParser
 
 parser = ArgumentParser(description='Download "Die Zeit" in multiple formats from the premium subscription service')
@@ -35,6 +36,19 @@ if formats == None:
     print("No formats specified, all done.")
     sys.exit(0)
 
+
+# Src: https://stackoverflow.com/questions/22058048/hashing-a-file-in-python#22058673
+def md5sum(path):
+    BUF_SIZE = 4 * 1024 * 1024 # 4 MiB
+    md5 = hashlib.md5()
+    with open(path, 'rb') as f:
+        while True:
+            data = f.read(BUF_SIZE)
+            if not data:
+                break
+            md5.update(data)
+    return md5.hexdigest()
+
 RELEASE_XPATH = '//p[@class="epaper-info-release-date"]'
 DOWNLOAD_XPATH = "//a[contains(text(), '{}')]"
 DATE_REGEX = r"^\d{2}\.\d{2}\.\d{4}$"
@@ -88,13 +102,27 @@ for fmt in formats:
     date = "-".join(latest_release.split(".")[::-1])
     filename = 'die_zeit_' + date + "." + fmt
 
+    request_headers = {}
     if os.path.exists(filename) and not forcereload:
-        print("File already exits. If you want to download anyway, use --reload")
+        # Somehow E-Tags do not work for PDF
+        if fmt == 'pdf':
+            print(f"File {filename} already exits. If you want to download anyway, use --reload")
+            continue
+        else:
+            request_headers["If-None-Match"] = '"' + md5sum(filename) + '"'
+
+    url = "https://epaper.zeit.de" + link \
+            if not link.startswith('https') else link
+    print(f"Downloading {fmt} from {url}...")
+    response = s.get(url, headers=request_headers)
+
+    if response.status_code == 304:
+        print("  => Skipped, file did not change")
         continue
 
-    print(f"Downloading {fmt} from {link}...")
-    response = s.get("https://epaper.zeit.de" + link 
-            if not link.startswith('https') else link)
+    if response.status_code != 200:
+        print(f"Request for {url} returned status {response.status_code}", file=sys.stderr)
+        sys.exit(-1)
 
     with open(filename, 'wb') as file:
         file.write(response.content)