There are certainly many ways to skin this cat. I would not trust your original approach from bad experiences in the past. THings may have changed. Hey, things have changed! CFMail in CF8 has ability to delete attachments after being sent???
Here is a fairly recent thread on this topic http://www.houseoffusion.com/groups/cf-talk/thread.cfm/threadid:54028 There are some links in there discussing inline attachments either using CFMail or dropping to Java mail. inline attachments, no file saving needed at all http://blog.pengoworks.com/index.cfm/2007/10/26/Using-CFMAIL-to-send-attachments-stored-in-memory HTH Douglas Knudsen http://www.cubicleman.com this is my signature, like it? On Mon, Nov 10, 2008 at 9:29 PM, Peyton Todd <[EMAIL PROTECTED]>wrote: > I was at my client's site today, and adopted the easiest solution, which > was: (a) writing the PDFs to a directory outside the web root and reading > them from there; (b) naming the PDFs with the job number instead of the > reporter's initials; and (c) deleting any existing PDF by that name before > the process begins (except in one rare, non-mission-critical situation, > where, as a test, I decided to wait and see if it bites me). > > Ajas, assuming that I delete any PDF by the same name before I begin, your > scenario of writing any change PDF content to a database and building the > PDF from there only after the updates to the database are performed does > solve the problem I posed to you, but I still have a question regarding your > step 4, which I repeat below for ease of reference: > > step 4. get the latest content from the table and > a. create pdf using <cfdocument> #pdfcontent#</cfdocument> or > b. send it as an attachment. > > My question: If I create the PDF from the saved data column, and if the > root cause is caching (which remains theoretical), couldn't cfemail still > simply go with the cached copy in cases where the same job number is being > updated? Yes, one would suppose that deleting the file would have led a > smart cache-management program to clear its namesake from the cache, but > then a smart cache-management program ought to have known better than to go > with the cached copy of a PDF which gets re-written directly, without the > intermediary of a database, should it not? (Keeping in mind - I say again - > that the caching explanation may be false anyway.) > > Dawn, my page already re-generates the PDF every time on the fly. But maybe > I don't understand your comment: since you say it patches the security hole, > is it your idea that the PDF never gets written to disk? > > Howard, if I were to sent the office manager a link to the PDF instead of > attaching the PDF to his email, how would I do that? Now that I am writing > the PDFs to a directory outside the web root, how could a link reach that > directory? And if I continue to place the PDFs in a subdirectory of the web > root, how could I secure it? > > Doug and Dean, have either of you convinced the other that either the > caching theory or the asynchronous theory could be correct, or that neither > could be? If the problem goes away, then of course it's academic. But since > I'm an academic myself... > > Peyton > > P.S. Believe me, I'm not making all this up. The problem is a real one. The > solution I put in today means this problem would now occur - if it does > still occur - only in the rare case when a reporter sends a job report and > soon afterwards corrects it and sends it again, as opposed to the frequent > case where the reporter reports two different jobs one after the other. > > > > -----Original Message----- > From: Ajas Mohammed > Sent: Nov 10, 2008 9:20 AM > To: discussion@acfug.org > Subject: Re: [ACFUG Discuss] Another CFMAIL Problem > > Hi, > > As per your quote, > > then wouldn't it be possible that cfmail could grab the old value from the > SQL Server column (presumably via its 'query=' parameter?) before my SQL > Server update statement stuffs the new value into it? > > Well, once you store something, lets say beforeupdate_identColumn.pdf , now > you have a file mapping to identity column. So lets say user changes or > modifies something, then you do update *first* based of identity column. > Then you either create that pdf or send it as an attachment. Makes sense? My > feeling is that once you have anything in DB, you are the master of things. > You can do pretty much anything like update, track changes, come back at > later time and still create docs which might be from way back in time and > what not. Bottom line, you are in control. > > Here is simple logic : > > step 1. create pdf content variable . I guess you could do this, > <cfsavecontent variable="pdfcontent"><table>blah blah > </table></cfsavecontent> > > step 2. insert into db. insert into pdf_table(pdf_content) values > ('#pdfcontent#') ( what i mean here is to save pdf text with table > formattting etc in the column. > > step 3. Now whenever your ready you could > a. lets say user updates pdfcontent, then you do udpate query first, so > you have latest content. > > step 4. get the latest content from the table and > a. create pdf using <cfdocument> #pdfcontent#</cfdocument> or > b. send it as an attachment. > > this way you will know for sure, if it was right content or not. Right now > seems to me, its more of guess work and there is no way to make sure you are > sending the so called cache vs updated content. > > hope that helps. > > Ajas Mohammed. > > > > On Mon, Nov 10, 2008 at 12:23 AM, Dawn Hoagland <[EMAIL PROTECTED]>wrote: > >> IF you can send a link to the server AND he always needs the latest >> version (and assuming it doesn't take long to generate the PDF) - I would >> send a link to a page that generates it on the fly every time and allow him >> to download that version when needed - as previously mentioned by Howard. >> This eliminates any issues to do with email latency (or the possibility of >> caching) and gives the opportunity to close some security holes created by >> leaving content in an unsecured location. >> >> Dawn >> >> >> On Sun, Nov 9, 2008 at 8:26 PM, Peyton Todd <[EMAIL PROTECTED]>wrote: >> >>> Doug, your guess that the email gets sent before the new PDF gets built >>> rings true to me. It would explain the problem on my client's server, as >>> well as the fact that it's not happening to me right now testing the system >>> working on my own PC as localhost. It jogs my memory that my coworker >>> experienced a similar problem back about 2001, when I was programming >>> ColdFusion in New York City. (?New York City!?? Get a rope!) >>> >>> Dean, doesn't it make sense to you that the e-mail could get sent before >>> the second version of the file gets built, so it merely picks up what it >>> finds on the disk , namely, the earlier version of the file? Actually, come >>> to think of it, that would not be an instance of caching. >>> >>> Ajas, I don't know, but it seems to me that saving the PDF into a SQL >>> Server column might have the same problem as that described by Doug: if >>> these operations can be performed asynchronously, then wouldn't it be >>> possible that cfmail could grab the old value from the SQL Server column >>> (presumably via its 'query=' parameter?) before my SQL Server update >>> statement stuffs the new value into it? >>> >>> Howard, No, I didn't mean the office manager might get mixed up on which >>> file is which - he's reasonably computer literate, and knows perfectly well >>> that the one on top was sent later (Whoops! It just occurred to me that that >>> might not always be true if there's a glitch in the e-mail process...). >>> Anyway, though, what I did mean is that it's simply the older version of the >>> file that gets attached to the e-mail. Your idea of sending a link seems >>> like a good one, and I may try that but ... >>> >>> >>> Naming the PDF by the Job Number instead of by the reporter's ID is >>> necessary of regardless of whether I attach the file or send a link. In >>> either case, I'd be writing the file to disk. If I were to delete any old >>> copy beforehand, what are the chances that (if this asynchrous story is >>> true) cfmail would fail to find the file when it looks for it and throw an >>> error? >>> >>> Peyton >>> >>> -----Original Message----- >>> From: Howard Fore >>> Sent: Nov 8, 2008 8:12 PM >>> To: discussion@acfug.org >>> Subject: Re: [ACFUG Discuss] Another CFMAIL Problem >>> >>> I think what he means is that if a recorder creates a file and sends it, >>> then corrects it and sends again, the recipient could open the first one and >>> think it is the most current version when it isn't. It wasn't identified as >>> an issue by Peyton but I could see how it could cause someone an issue. My >>> response to that would be to not mail the PDF as an attachment but rather >>> email a link to the user. The link would either lead to a real PDF, stored >>> on disk, or to a CFM file that generates the PDF when it is requested. That >>> would get rid of any caching issue as the file requested is always the >>> latest version. >>> On Sat, Nov 8, 2008 at 6:19 PM, Dean H. Saxe < >>> [EMAIL PROTECTED]> wrote: >>> >>>> What "things" get asynchronous? If what you say is true, email would >>>> regularly go out without complete attachments. I think you're way off on >>>> this one. >>>> >>> -- >>> Howard Fore, [EMAIL PROTECTED] >>> "The universe tends toward maximum irony. Don't push it." - Jeff Atwood >>> >>> ------------------------------------------------------------- >>> To unsubscribe from this list, manage your profile @ >>> http://www.acfug.org?fa=login.edituserform >>> >>> For more info, see http://www.acfug.org/mailinglists >>> Archive @ http://www.mail-archive.com/discussion%40acfug.org/ >>> List hosted by FusionLink <http://www.fusionlink.com> >>> ------------------------------------------------------------- >>> >>> ------------------------------------------------------------- >>> To unsubscribe from this list, manage your profile @ >>> http://www.acfug.org?falogin.edituserform >>> >>> For more info, see http://www.acfug.org/mailinglists >>> Archive @ http://www.mail-archive.com/discussion%40acfug.org/ >>> List hosted by http://www.fusionlink.com >>> ------------------------------------------------------------- >>> >> >> >> ------------------------------------------------------------- >> To unsubscribe from this list, manage your profile @ >> http://www.acfug.org?fa=login.edituserform >> >> For more info, see http://www.acfug.org/mailinglists >> Archive @ http://www.mail-archive.com/discussion%40acfug.org/ >> List hosted by FusionLink <http://www.fusionlink.com> >> ------------------------------------------------------------- >> > > > > -- > <Ajas Mohammed /> > http://ajashadi.blogspot.com > We cannot become what we need to be, remaining what we are. > No matter what, find a way. Because thats what winners do. > You can't improve what you don't measure. > Quality is never an accident; it is always the result of high intention, > sincere effort, intelligent direction and skillful execution; it represents > the wise choice of many alternatives. > > ------------------------------------------------------------- > To unsubscribe from this list, manage your profile @ > http://www.acfug.org?fa=login.edituserform > > For more info, see http://www.acfug.org/mailinglists > Archive @ http://www.mail-archive.com/discussion%40acfug.org/ > List hosted by FusionLink <http://www.fusionlink.com> > ------------------------------------------------------------- > > ------------------------------------------------------------- > To unsubscribe from this list, manage your profile @ > http://www.acfug.org?falogin.edituserform > > For more info, see http://www.acfug.org/mailinglists > Archive @ http://www.mail-archive.com/discussion%40acfug.org/ > List hosted by http://www.fusionlink.com > ------------------------------------------------------------- > ------------------------------------------------------------- To unsubscribe from this list, manage your profile @ http://www.acfug.org?fa=login.edituserform For more info, see http://www.acfug.org/mailinglists Archive @ http://www.mail-archive.com/discussion%40acfug.org/ List hosted by http://www.fusionlink.com -------------------------------------------------------------
