diff --git a/src/filters/rclpdf.py b/src/filters/rclpdf.py
index cb95c11e..5471a40b 100755
--- a/src/filters/rclpdf.py
+++ b/src/filters/rclpdf.py
@@ -469,7 +469,40 @@ class PDFExtractor:
if annotsfield:
self.em.setfield("pdfannot", annotsfield)
return html
-
+
+ def _patch_meta(self, html):
+ '''This fixes https://gitlab.freedesktop.org/poppler/poppler/-/issues/136'''
+
+ if not _mswindows:
+ pdfinfo = rclexecm.which("pdfinfo")
+ if not pdfinfo:
+ pdfinfo = rclexecm.which("poppler/pdfinfo")
+
+ if not pdfinfo:
+ return html
+
+ info = subprocess.check_output([pdfinfo, '-isodates', self.filename])
+
+ meta = {}
+ for line in info.split(b'\n'):
+ try:
+ key, value = line.strip().split(b':', 1)
+ meta[key.strip()] = value.strip().replace(b'"', b'\"')
+ except ValueError:
+ pass
+
+ if b'CreationDate' in meta:
+ meta[b'date'] = meta[b'CreationDate']
+
+ title = meta.get('Title')
+ head = [b'%s' % title] if title else []
+ head += [ b'' % x for x in meta.items() ]
+
+ start = html.index(b'') + 6
+ end = html.index(b'')
+ return html[:start] + b"".join(head) + html[end:]
+
+
def _selfdoc(self):
'''Extract the text from the pdf doc (as opposed to attachment)'''
self.em.setmimetype('text/html')
@@ -483,6 +516,8 @@ class PDFExtractor:
"UTF-8", "-eol", "unix", "-q",
self.filename, "-"])
+ html = self._patch_meta(html)
+
html, isempty = self._fixhtml(html)
#self.em.rclog("ISEMPTY: %d : data: \n%s" % (isempty, html))