Think you need to tell the response header that it a pdf file ( you may have a decorator that does it ) but this is the relevant part of the code that I use to return a pdf file
from cherrypy import response data = pdfFile.getValue() response.headers["Content-disposition"] = "inline; filename=pdfile.pdf" response.headers["Content-Length"] = len(data) response.headers["Content-type"] = "application/pdf" return data Chris On Feb 4, 8:11 am, Remi Jolin - SysGroup <[email protected]> wrote: > le 03.02.2009 03:26 Jim Steil a écrit: > > > Remi Jolin - SysGroup wrote: > >> Hello, > > >> le 02.02.2009 17:27 Jim Steil a écrit: > > >>> Hi > > >>> I use the following code to display PDF files. This has worked for me > >>> for a couple years. All of a sudden, this stopped working for some of > >>> our PDFs. Some display correctly, but some do not. This happens only > >>> on my production box (Win 2003), but works fine on my test/dvlp > >>> environment (WinXP). I don't even know where to begin trying to debug > >>> this. No errors are thrown. In the browser, it's as if the document > >>> displayed, but it really didn't. The title has changed to the one that > >>> should be displayed, but the page never changes. It is not still > >>> processing, but just seems to have forgotten to display the PDF. This > >>> is happening on all clients trying to connect, whether using IE or > >>> Firefox. > > >>> def documentDisplay(self, *args, **kw): > >>> documentId = args[0] > >>> doc = Document.get(documentId) > > >>> pdfFile = StringIO.StringIO() > >>> resourceDir = cherrypy.config.get('motion.resourcedir') > >>> f = open('%sdocuments/%s' % (resourceDir, doc.documentName), 'rb') > > >>> while 1: > >>> data = f.read(1024*8) > >>> if not data:break > >>> pdfFile.write(data) > >>> f.close() > > >>> pdf = pdfFile.getvalue() > >>> pdfFile.close() > > >>> return(pdf) > > >> You could first check the length of the 'pdf' variable... Is it coherent > >> with the actual length of the file ? > > >> Are you PDFs static or generated dynamically on the fly just before > >> displaying them. In that case it could be a "flush" issue of the > >> generated file not closed (so not flushed) to the disk while you try to > >> read it. > > >> And I was just wondering why you use a StringIO and not directly > >> pdf = f.read() > > > Remi, thanks for your reply. > > > The short answer is because that is what I copied from someone else's > > code. I implemented the change you suggested, but it made no > > difference. > > I did not meant it would change anything... > > > I should have mentioned that this is happening when I display PDFs > > from a number of different creation methods. I use this method to > > display PDFs from a library of documents that we have but I also > > generate PDFs on the fly using ReportLab. Again, it is an > > inconsistent problem. After more digging, it appears as though it is > > related to the size of the document that I'm trying to display. I > > have successfully displayed all documents 976 KB or smaller but cannot > > display any files that are 1050 KB or larger. Anyone have any idea > > where I can begin debugging this issue? > > Ok, but again, can you trace the length of 'ftp' (it should give the > size of the file) to check if it comes from the file or the transmission. > > And can you tell us a bit more about the expose decorator you use and > anything you use to modify the http header... > > > > > After my first post I noticed I didn't have the same TG environment on > > both systems. I've upgrade both systems to TG 1.0.8 and still have > > the same behavior, works in my dvlp environment, but not production. > > Again, this has worked in production for a over a year and all of a > > sudden quit working. No changes were made to the box except for > > Windows patches. > > > -Jim > > --~--~---------~--~----~------------~-------~--~----~ You received this message because you are subscribed to the Google Groups "TurboGears" group. To post to this group, send email to [email protected] To unsubscribe from this group, send email to [email protected] For more options, visit this group at http://groups.google.com/group/turbogears?hl=en -~----------~----~----~----~------~----~------~--~---
