removing metadata modification

tutorcruncher · May 24, 2017 · 9beb7f4 · 9beb7f4
1 parent 93ee5c1
commit 9beb7f4
Show file tree

Hide file tree

Showing 6 changed files with 15 additions and 71 deletions.
diff --git a/HISTORY.rst b/HISTORY.rst
@@ -3,6 +3,10 @@
 History
 -------
 
+v0.33.0 (2017-05-24)
+....................
+* remove pdf metadata modification as it can break some pdf viewers
+
 v0.32.0 (2017-05-24)
 ....................
 * set ``cache_dir`` for ``generate_pdf`` by default

diff --git a/benchmark/invoice.html b/benchmark/invoice.html
@@ -2,7 +2,7 @@
 <html lang="en">
 <head>
   <meta charset="UTF-8">
-  <title>Invoice INV-1</title>
+  <title>Invoice INV-123</title>
   <link rel="stylesheet" href="https://secure.tutorcruncher.com/static/css/libraries.css">
   <link rel="stylesheet" href="https://secure.tutorcruncher.com/static/css/pdf_styles.css">
 </head>

diff --git a/benchmark/run.py b/benchmark/run.py
@@ -17,9 +17,6 @@ def go_sync():
     for i in range(count):
         pdf = generate_pdf(
             html,
-            title='Benchmark',
-            author='Samuel Colvin',
-            subject='Mock Invoice',
             page_size='A4',
             zoom='1.25',
             margin_left='8mm',
@@ -41,9 +38,6 @@ async def go_async():
     async def gen(i_):
         pdf = await apydf.generate_pdf(
             html,
-            title='Benchmark',
-            author='Samuel Colvin',
-            subject='Mock Invoice',
             page_size='A4',
             zoom='1.25',
             margin_left='8mm',

diff --git a/pydf/version.py b/pydf/version.py
@@ -1,3 +1,3 @@
 from distutils.version import StrictVersion
 
-VERSION = StrictVersion('0.32.0')
+VERSION = StrictVersion('0.33.0')
diff --git a/pydf/wkhtmltopdf.py b/pydf/wkhtmltopdf.py
@@ -1,5 +1,4 @@
 import asyncio
-import re
 import subprocess
 import tempfile
 
@@ -47,20 +46,6 @@ def _convert_args(**py_args):
     return cmd_args
 
 
-def _set_meta_data(pdf_content, **kwargs):
-    fields = [
-        ('Title', kwargs.get('title')),
-        ('Author', kwargs.get('author')),
-        ('Subject', kwargs.get('subject')),
-        ('Creator', kwargs.get('creator')),
-        ('Producer', kwargs.get('producer')),
-    ]
-    metadata = '\n'.join(f'/{name} ({value})' for name, value in fields if value)
-    if metadata:
-        pdf_content = re.sub(b'/Title.*\n.*\n/Producer.*', metadata.encode(), pdf_content, count=1)
-    return pdf_content
-
-
 class AsyncPydf:
     def __init__(self, *, max_processes=20, loop=None, cache_dir=DFT_CACHE_DIR):
         self.semaphore = asyncio.Semaphore(value=max_processes, loop=loop)
@@ -69,14 +54,7 @@ def __init__(self, *, max_processes=20, loop=None, cache_dir=DFT_CACHE_DIR):
             Path.mkdir(cache_dir)
         self.cache_dir = cache_dir
 
-    async def generate_pdf(self,
-                           html,
-                           title=None,
-                           author=None,
-                           subject=None,
-                           creator=None,
-                           producer=None,
-                           **cmd_args):
+    async def generate_pdf(self, html, **cmd_args):
         cmd_args = [WK_PATH] + _convert_args(cache_dir=self.cache_dir, **cmd_args)
         async with self.semaphore:
             p = await asyncio.create_subprocess_exec(
@@ -94,24 +72,10 @@ async def generate_pdf(self,
                 stderr = await p.stderr.read()
                 raise RuntimeError('error running wkhtmltopdf, command: {!r}\n'
                                    'response: "{}"'.format(cmd_args, stderr.strip()))
-
-            return _set_meta_data(
-                pdf_content,
-                title=title,
-                author=author,
-                subject=subject,
-                creator=creator,
-                producer=producer,
-            )
+            return pdf_content
 
 
 def generate_pdf(html, *,
-                 title: str=None,
-                 author: str=None,
-                 subject: str=None,
-                 creator: str=None,
-                 producer: str=None,
-                 # from here on arguments are passed via the commandline to wkhtmltopdf
                  cache_dir: Path=DFT_CACHE_DIR,
                  grayscale: bool=False,
                  lowquality: bool=False,
@@ -186,15 +150,7 @@ def generate_pdf(html, *,
     if p.returncode != 0 and pdf_content[:4] != b'%PDF':
         raise RuntimeError('error running wkhtmltopdf, command: {!r}\n'
                            'response: "{}"'.format(cmd_args, p.stderr.strip()))
-
-    return _set_meta_data(
-        pdf_content,
-        title=title,
-        author=author,
-        subject=subject,
-        creator=creator,
-        producer=producer,
-    )
+    return pdf_content
 
 
 def _string_execute(*args):

diff --git a/tests/test_sync.py b/tests/test_sync.py
@@ -11,23 +11,13 @@ def test_generate_pdf_with_html():
     assert 'Is this thing on?\n\n\x0c' == text
 
 
-def test_generate_pdf_with_html_meta_data():
-    pdf_content = generate_pdf(
-        '<html><body>Is this thing on?</body></html>',
-        title='title foobar',
-        subject='the subject',
-        author='Samuel Colvin',
-        creator='this is the creator'
-    )
+def test_pdf_title():
+    pdf_content = generate_pdf('<html><head><title>the title</title></head><body>hello</body></html>')
     assert pdf_content[:4] == b'%PDF'
-    beginning = pdf_content.decode('utf8', 'ignore')[:300]
-    print(beginning)
-    assert """
-<<
-/Title (title foobar)
-/Author (Samuel Colvin)
-/Subject (the subject)
-/Creator (this is the creator)""" in beginning
+    text = pdf_text(pdf_content)
+    title = 'the title'.encode('utf-16be')
+    assert b'\n/Title (\xfe\xff%s)\n' % title in pdf_content
+    assert 'hello\n\n\x0c' == text
 
 
 def test_unicode():