Re: [Podofo-users] PdfWriter as a base class
Hi, You're right. I missed that completely. The reason is that i didn't compile the code. I just assumed that a method named "Write" should override a similar method in the base class. Of course it's possible to rename the method like you suggested. However, I feel that this method should indeed override the base class' Write. After looking at the code for the method, I realized that it duplicates some of the code from the base class. I believe that this should be avoided. In addition, it seems like PdfSigIncWriter is writing all the objects of the document (I'm not 100% sure of that). This is certainly not needed. Bottom line: I can certainly do a rename but I think that it would be better to have a bit of a different architecture. Ilan On 5/15/2014 00:08, zyx wrote: On Mon, 2014-05-12 at 15:15 +0300, Ilan Zisser wrote: I didn't see this. My revision didn't include PdfSigIncWriter.h. The reason for the warning is obvious, missing "virtual" before the method Write. Hi, it's not that simple, unfortunately. The problem isn't "missing" 'virtual' qualifier (you are not required to repeat 'virtual qualifier on methods which the base class has defined as virtual), the problem is that the function prototype is different from the parent class' Write. Using both patches shows only almost-the-same warning: In file included from podofo-trunk/src/base/PdfSigIncWriter.h:37:0, from podofo-trunk/src/base/PdfSigIncWriter.cpp:34: podofo-trunk/src/base/PdfWriter.h:103:18: warning: ‘virtual void PoDoFo::PdfWriter::Write(PoDoFo::PdfOutputDevice*)’ was hidden [-Woverloaded-virtual] virtual void Write( PdfOutputDevice* pDevice ); ^ In file included from podofo-trunk/src/base/PdfSigIncWriter.cpp:34:0: podofo-trunk/src/base/PdfSigIncWriter.h:50:18: warning: by ‘virtual void PoDoFo::PdfSigIncWriter::Write(PoDoFo::PdfOutputDevice*, PoDoFo::pdf_int64)’ [-Woverloaded-virtual] virtual void Write( PdfOutputDevice* pDevice, pdf_int64 prevOffset); ^ I guess the best option will be to rename the method in the PdfSigIncWriter.h, thus the names, and the prototypes, will not clash. This involves much more changes around the code. I cannot think of any nice name for the rename, thus either IncWrite(...) or WriteWithOffset(...), whichever sounds better to people. By the way, please provide patches with full paths (call `svn diff` from the root directory of the checkout) and merge all the changes into one patch, not in series of patches - both is easier for maintaining the changes. Thanks for your understanding. Bye, zyx -- Accelerate Dev Cycles with Automated Cross-Browser Testing - For FREE Instantly run your Selenium tests across 300+ browser/OS combos. Get unparalleled scalability from the best Selenium testing platform available Simple to use. Nothing to install. Get started now for free. http://p.sf.net/sfu/SauceLabs___ Podofo-users mailing list Podofo-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/podofo-users
Re: [Podofo-users] PdfWriter as a base class
On Thu, 2014-05-15 at 11:12 +0300, Ilan Zisser wrote: Of course it's possible to rename the method like you suggested. However, I feel that this method should indeed override the base class' Write. After looking at the code for the method, I realized that it duplicates some of the code from the base class. I believe that this should be avoided. Hi, I agree, it is always better to avoid code duplication. In addition, it seems like PdfSigIncWriter is writing all the objects of the document (I'm not 100% sure of that). This is certainly not needed. To be honest, I didn't check the code and what it does, I know it's for signing PDF documents in an incremental way, but nothing more of it. Bottom line: I can certainly do a rename but I think that it would be better to have a bit of a different architecture. Sure, if you can come with a better solution, then I'm all for it. Bye, zyx -- http://www.litePDF.cz i...@litepdf.cz -- Accelerate Dev Cycles with Automated Cross-Browser Testing - For FREE Instantly run your Selenium tests across 300+ browser/OS combos. Get unparalleled scalability from the best Selenium testing platform available Simple to use. Nothing to install. Get started now for free. http://p.sf.net/sfu/SauceLabs ___ Podofo-users mailing list Podofo-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/podofo-users
Re: [Podofo-users] Podofo, Freetype and MacOS Homebrew
Arno Uli, Which patch will it be? /Kunda On Wed, May 14, 2014 at 10:47 AM, zyx z...@litepdf.cz wrote: On Wed, 2014-05-14 at 08:03 -0500, Kunda Loves Scribus wrote: Thanks so much for having the conversation. I'm checking in to see if there is any progress in the assessment, tweeking and integration of patches which can fix these issues that have cropped up for OSX Homebrew Package Manager builds? Hi, to be honest, I lost track what is left and what should be checked into current sources for the Mac build. I would prefer if you three could get the changes together, thus they will work for all of you and I'll commit them to sources afterwards. The best if you could work with current trunk. It's one of two patches I'm aware of, which are pending of inclusion on this message list. Bye, zyx -- http://www.litePDF.cz i...@litepdf.cz -- Accelerate Dev Cycles with Automated Cross-Browser Testing - For FREE Instantly run your Selenium tests across 300+ browser/OS combos. Get unparalleled scalability from the best Selenium testing platform available Simple to use. Nothing to install. Get started now for free. http://p.sf.net/sfu/SauceLabs___ Podofo-users mailing list Podofo-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/podofo-users
Re: [Podofo-users] Podofo, Freetype and MacOS Homebrew
Am 15.05.2014 um 23:25 schrieb Kunda Loves Scribus scribus.u...@gmail.com: Arno Uli, Which patch will it be? Well, my patch seems to comprise all the others, *but* it needs Lua 5.2, which Homebrew doesn't seem to offer. I feel that Homebrew should be updated to Lua 5.2, rather than Podofo implementing a workaround for code that is now outdated for more than 2 years (!) - but I never use package managers (among other things, exactly because of problems like this ...), so that's just my personal take. So someone who feels the urgent need to use an outdated Lua version would have to add to my patch. Bye Uli smime.p7s Description: S/MIME cryptographic signature -- Accelerate Dev Cycles with Automated Cross-Browser Testing - For FREE Instantly run your Selenium tests across 300+ browser/OS combos. Get unparalleled scalability from the best Selenium testing platform available Simple to use. Nothing to install. Get started now for free. http://p.sf.net/sfu/SauceLabs___ Podofo-users mailing list Podofo-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/podofo-users
